DETROIT -- The General Motors engineer who approved modifying a faulty ignition switch design in 2006 without changing the part number or pushing for a recall told congressional investigators this month that he forgot ever authorizing the change, according to a report published Wednesday.
Ray DeGiorgio, one of two GM employees who have been suspended with pay since April 10 as the company conducted an internal investigation, seemed "very emotional at times" during questioning and "genuinely upset" that the problem has been linked to at least 13 deaths, an unidentified congressional aide told The New York Times.
He told the investigators who interviewed him May 19 that he never knew that the problem could cause airbags to fail in a crash, which GM says has happened 47 times, the newspaper said.
"He came across as if he was just overburdened and just missed it," the staff member was quoted as saying.
A GM spokesman declined to comment Wednesday.
A congressional aide confirmed to Automotive News that staff involved in the GM investigation have met with DeGiorgio and other current and former GM employees.
GM has recalled 2.6 million cars that were built with or may contain the faulty switch design. If the switch slips out of the "run" position, airbags, power steering and power brake assist are disabled.
During an April 2013 deposition, DeGiorgio claimed he did not authorize the 2006 change to the switch. Paperwork that GM obtained from the switch supplier, Delphi Automotive, in October 2013, shows that he signed off on a longer spring and detent plunger, designed to make the switch less flimsy, and a new printed circuit board to resolve an electrical problem.
That revelation prompted several members of Congress to declare during a pair of April hearings that DeGiorgio lied under oath. GM CEO Mary Barra conceded that the evidence suggested DeGiorgio lied but said she was waiting for the company's internal probe to look into the matter further. The results of that probe are expected to be released as soon as next week.
DeGiorgio told the congressional investigators that he had forgotten about the change by the time of the deposition, the Times reported. The story did not address why DeGiorgio agreed to make the change without having GM assign a new part number -- a move that Barra said is a violation of GM protocol and that delayed detection of the problem years later.
No memory problems
For most of the five-hour deposition, which was part of a lawsuit brought by the parents of a Georgia woman who died in a Chevrolet Cobalt crash, DeGiorgio did not speak as if he was unsure of himself or having trouble with his memory. The plaintiffs' lawyer, Lance Cooper, showed proof that the switch had been modified sometime in 2006, but DeGiorgio said at least 10 times that he knew nothing about it, even though he was the primary engineer in charge of the part.
Twice, he said he "certainly" did not authorize or become aware of a redesign, and at one point he declared, "absolutely not." As the questioning wore on, he began hedging some of his answers, saying "I don't recall ever authorizing such a change" at one point.
"I'm pretty sure it was never communicated to us or to me," DeGiorgio said. "As best of my recollection, it was never mentioned, but again, this is going back a few years, so I cannot answer in the absolute."
But he later became more resolute. As the session wrapped up, he seemed to have no doubts whatsoever. "I don't know enough about this change to comment one way or the other," DeGiorgio said, "but I can certainly tell you, I was not aware of this."
The Georgia crash victim's parents, who reached a confidential settlement with GM a month before the company says it obtained the document with DeGiorgio's signature, this month refiled their claim, accusing GM of concealing evidence and DeGiorgio of committing perjury.
DeGiorgio told the congressional investigators he had been more focused on the electrical problem than issues with the switch slipping out of "run," according to the Times' source.
By the time DeGiorgio authorized the modification in April 2006, GM had investigated complaints about weak torque in the ignition switch at least three separate times and issued two service bulletins to dealers about the problem. And a June 2005 e-mail written by a Delphi engineer to two of his colleagues said DeGiorgio was requesting torque testing data for the switch because "Cobalt is blowing up in their face in regards to turning the car off with the driver's knee."
DeGiorgio, who has not spoken publicly about the recall, has been a GM employee since 1991. In the deposition, he said the ignition switch for the Cobalt and Saturn Ion was "one of my first ignition switches."
Gabe Nelson and Mike Colias contributed to this report.