By Tara Hoke
In February 2014 the auto manufacturer General Motors (GM) issued a recall of several models of vehicles produced between 2003 and 2007. The reason for the recall was a faulty ignition switch that could be easily bumped from the "run" position into the "accessory" position, causing the vehicle to stall and disabling air bags and other safety features. In the subsequent months, significant attention has been devoted to the 10-year span during which members of GM's staff and leadership were aware of the faulty switch but failed to take action to address it. It is now known that the switch was a factor in dozens of serious accidents and at least 19 fatalities. An independent report commissioned by GM (Report to Board of Directors of General Motors Company regarding Ignition Switch Recalls, by A.R. Valukas, May 29, 2014) documents the individual and organizational lapses that prevented GM from properly addressing this defect and serves as another example of the need to emphasize engineering ethics within the corporate culture.
Situation
The ignition switch that would become the subject of this recall was plagued with problems almost from its origin. Intended to be less expensive and less prone to failure than the prior model, the prototype performed poorly in initial tests and ultimately required a complete electrical redesign. It was noted after the redesign that the new switch failed to meet GM's mechanical specifications for torque in that less force was required to turn the key than was called for in the design. Incorrectly believing that this would not affect performance and fearing that further changes might cause even more problems with the electrical system, a GM design engineer approved the deviation from the specifications. This decision did not require review or approval by other GM personnel, and the engineer did not advise GM management of the deviation.
Soon after vehicles equipped with the new switch hit the market, engineers at GM were made aware of problems created by the low torque. By 2004 GM had received numerous complaints from consumers reporting "moving stalls," caused when an incidental knee brush or a heavy key fob pulled the ignition switch out of the "run" position while the car was in use. The issue was referred to a committee of GM engineers and managers for review, but that body decided that the problem was not a safety issue because drivers could still maneuver their stalled vehicles off the road. Proposed fixes were circulated among several GM committees, but each was ultimately rejected as being too costly to address a matter of "customer convenience."
None of the committees charged with reviewing the switch problem ever considered other systems that might be affected by the vehicle stall, including air bags designed not to deploy in the "off" or "accessory" positions so as to avoid deployment in cases involving a parked car. In situations in which the ignition switch jolted out of position during or immediately before an accident, the resulting stall prevented this crucial safety feature from engaging just when it was most needed.
Even as reports accumulated at GM of cases in which air bags had failed to deploy in front-impact collisions, GM failed to make the connection between the ignition switch and the air bag failures. The air bag failures were viewed as a rare, if mysterious, issue, and investigation of the problem languished among a series of teams and committees operating without an express directive to conduct a thorough investigation of the failures.
Meanwhile, outside reports making the crucial connection between the two problems did not reach the necessary hands at GM. A 2007 accident investigation linking air bag failure to the faulty ignition switch was transmitted to GM's legal department, where it languished until just before the recall. Similar conclusions were reached in a university-funded study and other publicly available documents, yet there is no evidence that anyone at GM either sought out or was aware of this information.
Further complicating GM's understanding of the issue, the engineer responsible for the ignition switch approved a change in its design in 2007 to address the issue of low torque. However, he failed to let anyone know that the part had been modified. (He would later claim he had no memory of approving the change.) When GM investigators observed that air bag failures were occurring only in older models, they mistakenly concluded that the ignition switch could not be the source of the problem, believing that the same switch was still in use in the trouble-free newer models. These investigators, however, failed to make any direct examination of the ignition switches. Indeed, it was only when a plaintiff's expert witness compared X-rays of the pre- and post-2007 ignition switches that the design change was finally revealed.
Question
What ethical lessons can be drawn from the GM ignition switch recall?
Discussion
While the events leading up to the recall underscore the importance of good communication between departments, proper documentation, and an efficient problem-solving process, the most important lesson from an ethical perspective is that engineers must be personally accountable for the consequences of their decisions.
Canon 1 of ASCE's Code of Ethics states in part that engineers "shall hold paramount the safety, health, and welfare of the public . . . in the performance of their professional duties." Category (a) in the guidelines to practice for this canon is fairly sweeping: "Engineers shall recognize that the lives, safety, health, and welfare of the general public are dependent upon engineering judgments, decisions, and practices incorporated into structures, machines, products, processes, and devices."
The independent review of the GM recall catalogs a series of failures by GM's staff to recognize the safety implications of their decisions. These lapses ranged from that of the engineer who first approved a design that was significantly below specifications to that of the committee of engineers and managers that inexplicably categorized a problem of stalling even at highway speeds as "not a safety issue," all the while failing to consider the effects of this faulty equipment on the vehicles' most vital safety features.
Moreover, the repeated referral of the problem to myriad committees and teams demonstrates a corporate culture in which individuals were not expected or encouraged to assume personal responsibility for safety issues. Staff members interviewed for the report described what they called "the GM salute," that is, standing with arms crossed and fingers pointed out to the sides to indicate that the responsibility for dealing with problems belongs to somebody else. This lack of personal accountability created the inertia that made it possible for a serious risk to the safety of GM consumers to remain unrecognized for more than a decade. (The failure to establish a corporate culture of safety has been a recurring theme in this column, for example, "The Importance of Engineering Ethics," August 2011, pages 42-43, which discussed the explosion of the shuttle Columbia, and "Fostering a Culture of Safety," January 2013, pages 38-39, which discussed the tragedy connected with the drilling rig Deepwater Horizon.)
Unlike physicians, lawyers, and other professionals, engineers do not always have a direct personal interaction with the customers who benefit from their professional services. Yet even when there is a distance between the service provided and its ultimate end use, it is essential for engineers not to lose sight of the fact that their decisions directly affect the well-being of men, women, and children. It is only when an engineer takes personal responsibility that it can truly be said that the ethical obligation to "hold paramount" the health, safety, and welfare of the public has been met, and there is a greater likelihood of this happening when this responsibility figures prominently within a corporation's culture.
Tara Hoke is ASCE’s general counsel and a contributing editor to Civil Engineering.
© ASCE, ASCE News, October, 2014
