Some of the vexing questions looming over General Motors' decadelong failure to recall cars with a deadly ignition-switch defect could get answers this week -- or at least a more glaring spotlight.
GM CEO Mary Barra and acting National Highway Traffic Safety Administration chief David Friedman will appear Tuesday, April 1, before a U.S. House committee, followed by testimony Wednesday before a Senate panel. On Thursday, GM faces a deadline to answer a battery of 107 questions from NHTSA about the company's recall of 1.37 million U.S. small cars to replace faulty ignition switches linked to 34 crashes and 12 deaths. That recall was expanded Friday to include 824,000 more U.S. cars.
Amid a wave of lawsuits and media coverage, four central issues have emerged that bear the closest watching this week:
How GM's multiple internal investigations failed to lead to a recall sooner
After GM in 2001 recognized the potential for the switch to slip out of the run position and stall the engine, the problem was reported at least three times over the years in GM's Problem Resolution Tracking System. Known internally as the PRTS, it was described by two former high-level GM engineers as a depository for any issues that could delay a vehicle program.
The switch's potential to shut off the engine should have earned it a "priority 1" designation in the system, reserved for safety concerns or issues that could delay production, said the former engineers. Neither was involved in the handling of the ignition switch, and both asked not to be identified discussing the sensitive issue.
"Any safety-related part that could throw the whole program into the ditch if there's a long lead time to fix it would be priority 1," said one of the former engineers. "Those are high on everybody's radar."
That would have triggered reviews by the engineer responsible for the part, his or her supervisor, possibly the engineer in charge of electrical components and ultimately the vehicle's chief, both sources said.
Yet each time engineers opened a PRTS inquiry into the faulty switch, they either did nothing or settled on remedial action that failed to grasp the safety-critical nature of the problem, GM's report to regulators shows.
For example, GM told regulators that in 2004, its engineers were able to replicate an incident in which a Cobalt lost power because the switch moved out of the run position. Engineers proposed several possible solutions, but the inquiry later was closed without action.
Asked by a plaintiff's attorney during a June 2013 deposition whether GM "made a business decision not to fix the problem," Gary Altman, program engineering manager for the Cobalt in 2004, answered: "That is what happened, yes," according to a transcript.
GM confirmed in a timeline submitted to NHTSA in February that the PRTS was closed "after consideration of the lead time required, cost and effectiveness of these solutions."
Several of NHTSA's 107 questions to GM relate to that apparent dropped ball and likely foreshadow the sort of questions that await Barra. The answers could help explain how much GM's processes were to blame for the delayed recall or whether one or more bad actors foiled earlier action.
Why NHTSA failed to launch an investigation, despite signs that a faulty switch might be causing airbags not to deploy
NHTSA also will be in lawmakers' cross hairs for failing to open a defect investigation into the ignition switch, notwithstanding Friedman's insistence that the agency could have spotted the problem if only GM had shared more information.
"There are as many important questions in this for NHTSA as there are for General Motors," said Sean Kane, president of Safety Research & Strategies, a Rehoboth, Mass., company that has worked with plaintiffs' attorneys on automotive litigation. Kane has criticized NHTSA for lacking a standardized method for determining when an investigation is warranted.
NHTSA officials met with GM employees in March 2007 to discuss passive restraint systems, such as airbags, according to GM's timeline. At that meeting, the officials mentioned a NHTSA-commissioned investigation into a July 2005 crash in Maryland in which the airbags in a 2005 Cobalt did not deploy. They noted the car's ignition was in the "accessory" position at the time of the crash.
The connection between the airbag and the ignition may not have been obvious to NHTSA. Experts say the interaction between the ignition position and airbag deployment depends on the way airbag systems are programmed and varies from company to company.
GM, for its part, seems to have recognized a link. The company started a legal file and assigned an employee to track down crashes in which airbags didn't deploy. But apparently, NHTSA didn't make the same logical leap.
GM's chronology mentions no further contact from the agency. Why not?
Members of Congress will push NHTSA to answer this question during this week's hearings. It will be key in determining how much of the blame for GM's long-delayed recall should rest on the government's shoulders.
Whether and how GM's vehicle-safety protocols have changed
Much of the focus in hearings this week and in the unfolding legal cases will be on decade-old decisions made in the bowels of GM's engineering center.
But overarching those issues are more fundamental questions: Are GM's processes better today? What are the risks of another deadly safety flaw slipping through GM's sprawling product-development enterprise?
Lawmakers, regulators and consumers will want to know whether they can believe Barra's promise, made in a video message to consumers last week, that GM "will not let it ever happen again."
Barra hasn't offered specifics about how GM will improve vehicle safety, beyond naming GM veteran Jeff Boyer in March to the newly created post of global safety chief.
"What we need to do is simplify our process and make the processes more visible, more manageable so that when issues are identified, we can resolve them quickly," Boyer told Bloomberg TV last week.
NHTSA wants specifics, and lawmakers likely will, too.
In NHTSA's demand for more information from GM, due Thursday, it asks GM to "detail the ways in which" the process fell short and to describe "GM's plans to change its process."
The answers will shed light on how far GM's safety system and protocols need to go for Barra to make good on her promise.
Whether GM's internal processes were violated or laws were broken
The two former GM engineers, as well as a third interviewed by Automotive News, said GM's internal processes for evaluating potential safety defects and determining a course of action were rigid and supported by checks and balances.
That makes the failure to recall the cars sooner -- and the number of apparent missed opportunities -- all the more vexing for protocol-minded engineers.
The sources and other observers have zeroed in on a redesign of the faulty ignition switch that was approved in April 2006 for the 2007 Cobalts.
Why wasn't the part number changed for the redesigned part, leaving GM engineers years later fumbling to explain why models after 2007 had different switches from older cars -- and prompting Friday's expanded recall? Friday's action was aimed at ferreting out faulty replacement parts that may have been installed in cars originally built with the correct switch.
GM's account of that decision and others is sparing: It says simply that the supplier proposed the change, and the engineer responsible for the part signed off on it.
NHTSA wants to know more: Who was the engineer who signed off? Why was the change made? Who else knew about it? It has asked for all documents related to the redesign, too.
Those answers ultimately could dictate whether GM's ignition-switch recall leads to new laws or regulations meant to prevent future defects from slipping through.
You can reach Gabe Nelson at email@example.com