THE GM RECALL

Outsiders push the new GM into recalls of old GM products

The plaintiff, Jennifer Brooke Melton, died in her 2005 Chevy Cobalt after a March 10, 2011, single-car accident.
PRIMER
Related Links
Related Stories
Related Topics

The biggest crisis for the new General Motors might still be a secret buried in the debris of old GM, if not for persistent digging by a Georgia lawyer, a Mississippi auto mechanic and a Florida engineer.

The lawyer, Lance Cooper, sued GM in June 2011 on behalf of Brooke Melton, a nurse who lost control of her 2005 Chevrolet Cobalt and died on her 29th birthday. The suit linked the crash to black-box data showing that the car’s ignition switch had slipped out of the “run” position, long before GM revealed that correlation publicly.

The case, settled last fall, played a critical role in exposing the defect and compelling GM to issue a recall, more than a decade after company engineers discovered the problem. But it’s only one example of how even the postbankruptcy GM waited to approve recalls for long-standing problems until regulators or someone else outside the company forced the issue.

In fining GM $35 million last week over its handling of the ignition-switch issue, auto-safety regulators issued a broad condemnation of its ability and willingness to fix dangerous defects, citing a flawed safety culture that persisted well past GM’s 2009 bankruptcy and still needs to be remedied.

‘Slow to act’

As recently as March, at the height of the unfolding switch recall crisis, GM planned to address an unrelated problem with the airbags on 1.3 million mid-sized crossovers through a customer-satisfaction campaign. But the National Highway Traffic Safety Administration disagreed, and GM issued a recall instead.

GM also recalled 1.5 million cars for a power-steering defect after the head of the National Legal and Policy Center wrote an open letter to GM CEO Mary Barra asking why there had been no action nearly four years after NHTSA began investigating the problem. A chronology filed with NHTSA shows that GM formed a team to study the issue on March 19, the same day it received the letter, and announced the recall 12 days later.

Last week, GM announced five more recalls covering 3 million vehicles globally, bringing its total for the year to 24 campaigns. The recalls total 12.8 million vehicles worldwide and 11.2 million in the United States, though that counts some vehicles multiple times because they are subject to more than one recall.

One of last week’s recalls, covering 2.7 million mid-sized cars, relates to a brake-light problem that GM first addressed with a service bulletin to dealers in 2008. NHTSA investigated the matter two times, first concluding that the bulletin was sufficient and then opening an inquiry in 2013 in response to 1,305 complaints and 14,447 “potentially related” warranty claims.

A NHTSA official called GM “slow to act” in a July e-mail to Carmen Benavides, GM’s director of field product investigations, listing four examples in which the agency had to prod the automaker into issuing a recall. GM reassigned Benavides to another role in the company last week.

'Inconclusive results'

Experts retained by Cooper are believed to have been the first people outside GM to discover a 2006 redesign of the troublesome ignition switch that had been missing from GM’s own records. Even after GM owned up to the defect in February, Cooper used the information he had gathered to show that the acknowledgment was long overdue — and still didn’t go far enough.

Cooper said he and Melton’s parents are “humbled” that their case helped spur GM to take action on the ignition-switch issue, which dates to 2001. “The most important thing is that, hopefully, it saved lives by getting these cars, to a certain extent, fixed,” he said in an interview.

Documents that GM provided to NHTSA and congressional investigators do not reveal any action on the issue after the automaker’s 2009 bankruptcy until a month after the Melton case was filed.

At that point, GM’s legal staff met with product-investigation employees to begin a “field performance evaluation” of Cobalt crashes in which airbags failed, according to a chronology filed with the NHTSA. GM then assigned an engineer, Brian Stouffer, to investigate the crashes, many of which involved the ignition switch slipping into “accessory” mode.

GM continued studying the issue throughout 2012, poring through design records and testing switches at a salvage yard to understand why the crashes involved only cars built before 2008, but “no explanation was discovered,” it said in the chronology. Studies using various problem-solving techniques produced “inconclusive results,” GM said.

But the automaker did establish a link between the ignition switch and failed airbags, surmising that drivers were turning the cars off by bumping the keys with their knees, according to a September 2012 e-mail written by Stouffer.

'Did not have the time'

Meanwhile, Cooper’s experts were producing more fruitful results. Charlie Miller, owner of an auto repair shop in Merigold, Miss., identified the ignition switch in Melton’s Cobalt as faulty and discovered wide variation in the performance of switches he obtained from junked cars and bought from GM. Mark Hood, a materials engineer in Pensacola, Fla., determined that the switches GM used starting in the 2007 model year contained a different spring and detent plunger -- the pieces designed to hold the key in place.

Cooper arranged a series of depositions with GM employees in the spring of 2013. On April 29, at a hotel attached to the Detroit airport, he confronted Ray DeGiorgio, the engineer responsible for the Cobalt’s ignition switch, with the differences Miller and Hood found. DeGiorgio denied ever approving or knowing about the change.

“We had to take him at his word,” Cooper said last week, even though “it didn’t seem to make sense.”

Two days later, Cooper asked Stouffer why GM’s internal investigation hadn’t found the information that his experts had discovered.

“I would have had to leave my job, try to make arrangements to find a vehicle somewhere and find a part,” Stouffer replied, according to a transcript of the deposition. “I just did not have the time to do that.”

GM’s chronology says that, after learning in “late April 2013 … that others had observed and documented” the switch modification, the company retained “outside engineering resources” to study the switches further. The consultant -- from Exponent, the same firm that Toyota Motor Corp. used to investigate sudden-acceleration complaints -- confirmed the findings of Cooper’s experts, according to a presentation he prepared titled “Melton vs. GM.”

Document from Mexico

It wasn’t until October 2013 -- at least five months after the depositions began and a month after the Meltons reached a confidential settlement with GM -- that Stouffer contacted the switch manufacturer, Delphi Automotive, documents show. Delphi’s product investigations manager initially found no record of the change either but forwarded Stouffer’s request to employees at the Mexico plant where the switch was made.

There, Delphi employee Antero Cuervo discovered paperwork showing that DeGiorgio had signed off on the ignition-switch modifications seven years earlier without changing the part number. Barra called that a violation of GM protocol during an April 1 congressional hearing, and she conceded that DeGiorgio appeared to have lied under oath in denying knowledge of the change.

DeGiorgio and the former project engineering manager of the Cobalt, Gary Altman, have since been suspended with pay. GM has said its internal investigation of the recall will conclude in the next several weeks and that it will not comment on the chronology or documents released by Congress until the investigation is finished.

Seven weeks after getting the Delphi paperwork, following additional testing, the issue reached for the first time the GM committee that decides on recalls. The committee asked for more analysis and approved recalling the Cobalt on Jan. 31.

The news seemed rather routine at first, until Cooper wrote to NHTSA on Feb. 19, asking the agency to investigate the timeliness of GM’s actions. He revealed that GM knew about the problem even before the Cobalt went on sale and that five other nameplates not included in the recall contained the same ignition switch.

GM ultimately expanded the recall to cover 2.6 million cars, from 778,000 initially. In addition to a NHTSA probe, GM is now under investigation by the Justice Department, the Securities and Exchange Commission, Congress and a group of state attorneys general. It also is facing 75 federal lawsuits, court records show.

“If there is ever a case that warranted civil and potential criminal penalties,” Cooper said Feb. 27, “it’s this case.”

You can reach Nick Bunkley at nbunkley@crain.com. -- Follow Nick on Twitter

Have an opinion about this story? Click here to submit a Letter to the Editor, and we may publish it in print.

Or submit an online comment below. (Terms and Conditions)




Rocket Fuel