GM Passed on Ignition-Switch Fix in 2001, Advocates Say

General Motors Co. chose not to use a more robust ignition-switch part in Chevrolet Cobalts and other small cars while they were being designed, a decision that may have led to deaths, safety advocates said.

GM engineers in 2001 designed an alternative to the spring used in an ignition switch for the 2003 Saturn Ion before it was rejected, according to a letter sent to Chief Executive Officer Mary Barra today by Joan Claybrook, a former head of the U.S. National Highway Traffic Safety Administration, and Clarence Ditlow, executive director of the Center for Auto Safety.

“General Motors picked a smaller and cheaper ignition switch that cost consumers their lives,” Claybrook and Ditlow said in the letter. “Who inside GM made these decisions and at what level?”

GM’s handling of a defect now tied to 13 fatalities in accidents, after car engines lost power and air bags failed to deploy, is under investigation by NHTSA and both chambers of Congress. The largest U.S. automaker announced a first-quarter charge of $1.3 billion for recall-related expenses including repairs to 2.59 million cars.

An internal GM investigation led by Jenner & Block LLC Chairman Anton Valukas is looking into the issues Claybrook and Ditlow raise, James Cain, a GM spokesman, said in an e-mail.

“We are hoping that Mr. Valukas’s findings will be completed within the next 45-60 days,” Cain said.

GM rose 1.8 percent to $33.97 at 4 p.m. in New York.

Spring Swap

The drawing of a longer spring GM eventually adopted for the ignition switch in 2006 was included in documents released by the House Energy and Commerce Committee April 11. A chain of e-mails that explains GM’s part change in April 2006 shows the spring used was designed five years earlier. The new switch probably started showing up in 2007 models, GM told regulators.

The part change, approved by GM engineer Ray DeGiorgio, has become a focal part of investigations by Congress and NHTSA. DeGiorgio authorized the change without setting a new part number, in violation of company protocol and accepted engineering practices, and then testified during a deposition that he hadn’t approved the switch, documents show.

Senator Kelly Ayotte, a New Hampshire Republican, called the move a “criminal deception” during a hearing April 2.

Adversarial Relationship

The automaker’s relationship with its parts makers in the previous decade was adversarial and focused on cutting costs, said Dave Sullivan, an analyst with AutoPacific Inc. in Southfield, Michigan, who worked during that time for a GM switch supplier.

“It boggles my mind that somehow this made it past their production part-approval process,” Sullivan said.

Consumer complaints about the ignition key in the Cobalt, Ion and other models slipping out of the run position when bumped -- a defect later traced to a too-weak spring that didn’t meet GM specifications -- slowed after the part change.

When GM changed the part in 2006, it “did not have to develop a new more robust design because GM engineers had already designed the safer switch that GM previously rejected in 2001,” Claybrook and Ditlow said.

Stouffer, Delphi

The documents show that the shorter spring used in the original defective switch was designed in October 2001, a month after the longer spring, Claybrook said in a comment left on voice mail.

“We know GM designed two different plungers and springs,” Claybrook said. “They made an affirmative decision to use the cheaper one.”

The drawings of the two springs are part of an October 2013 e-mail exchange between a GM engineer, Brian Stouffer, and a member of the legal staff at the automaker’s parts supplier, Delphi Automotive Plc. Stouffer tried to find out why ignition switches in Cobalt sedans sold in 2008 were performing better than those in older models. Delphi manufactured both switches using GM designs.

The Delphi lawyer told Stouffer in the e-mail that DeGiorgio approved a design change for the ignition switch in 2006 that used a part with a longer spring, which made it work better.

GM said in a timeline sent to regulators earlier this year that at least some people in the company didn’t know about the design change that DeGiorgio approved until 2013.

Two Suspensions

Barra should have known about the documents identifying this alternative part before testifying to Congress, Claybrook and Ditlow said in the letter.

GM should make public all documents relevant to the decision to use a shorter, cheaper spring in 2001, Claybrook and Ditlow said.

DeGiorgio and another GM engineer, Gary Altman, were suspended by the Detroit-based automaker April 10, according to people familiar with the matter. The company announced they had put employees on paid leave over their roles in the recall without naming them.

Altman led the engineering team working on the Cobalt and rejected a part fix as too expensive.

Before it's here, it's on the Bloomberg Terminal.