NHTSA's Botched Oversight of GM's Cobalt Crashes

Still, the agency is demanding more money and power


Cobalt Crash

Even when Congress diagnoses a problem correctly, you can pretty much count on it to prescribe exactly the wrong solution. This week, it issued a scathing report about NHTSA's (National Highway Traffic and Safety Administration) mishandling of GM's ignition switch debacle. But instead of holding the federal agency accountable, lawmakers want to reward NHTSA with more money and power. Really.

By GM's own admission, a faulty ignition switch in its 2005 Cobalt and six other models has so far caused more than 30 crashes and 19 deaths. The switch in these cars was loose. So if a car hit a bump, the switch would slip from the run to the accessory position, shutting down the engine (especially if any object weighed down the key chain). This would not only diminish maneuverability, increasing the chances of a crash, but also disable the airbags that protect occupants. Despite many red flags, it took GM nine years to figure this out and finally issue a recall of some two million vehicles.

Why didn't NHTSA, whose primary purpose is to identify such defects before they do too much damage, detect the problem and force GM to act sooner? Basically, because the agency was in over its head, the House Energy and Commerce Committee report revealed.

NHTSA is supposed to be command central for the auto industry. It obtains information from multiple sources — consumer complaints, quarterly reports by manufacturers about accidents, warranty claims and property damage involving their vehicles, police reports of crashes — analyzes it, launches investigations when it detects a trend involving particular models, and orders remedies. Because NHTSA is motivated neither by bottom-line considerations nor hampered by informational gaps, it can monitor automakers better than they can themselves. At least in theory. But in reality, that's not how it works.

The congressional report found that a state trooper who investigated a fatal 2006 crash involving a Cobalt explicitly noted that the faulty ignition switch, which was in the accessory position at the time of the accident, might have prevented the airbag from deploying. But NHTSA dismissed the connection despite three separate investigations into several different accidents involving non-deploying airbags.

Admittedly, it is not an easy forensic exercise to identify the common cause among the millions of accidents every year. And hindsight is always perfect. But what prevented NHTSA from connecting the dots in this case wasn't the difficulty of sorting through a massive amount of data, but the agency's sheer ignorance of advanced airbag technology that it had itself mandated.

Some of the occupants in the fatal accidents involving the Cobalt and its sister cars, including the one investigated by the trooper, were not wearing seatbelts. And because the older generation of air bags were designed not to be deployed for unbelted occupants, NHTSA assumed that's what was causing their failure to deploy in these cars, too. (Old bags inflated with uniform force regardless of severity of impact or size of passenger, something that could cause out-of-position occupants, especially small women and children, unnecessary and grave injuries if deployed in minor accidents.)

What NHTSA stunningly overlooked was that in 2003, it mandated advanced air bag systems that could safely deploy for both belted and unbelted occupants. That's because they were programmed to be "smarter" and adjust the force of inflation based on the size and position of an occupant.

But since these new airbags needed more energy, they had to be powered by the engine, unlike the old system that had its own power supply. This meant that when the engine turned off, the new bags became almost immediately deactivated.

So NHTSA investigators made a double mistake of attributing air bag non-deployment to the lack of seat belt use (which was irrelevant in the new cars) and of failing to understand the true cause (the loss of power resulting from the slip of the ignition switch).

All of this happened because "NHTSA's safety defect investigators' understanding of the systems failed to keep pace with the evolution of the technology," the congressional report found. Hence, the agency was "completely unaware" of the link between engine power and air bag function.

NHTSA Chief David Friedman, defiant and unrepentant during congressional testimony this week, is demanding more funds to add more technical staff in 2015 — as if expansion is the cure for bloopers resulting from ignorance of the agency's own mandates.

Worse, some lawmakers, like Sen. Claire McCaskill (D-Mo.), want him to stop "singing kumbaya with manufacturers" and act instead like a "beat cop" and issue more recalls.

But showering the agency with more money and power will only reward its ineptitude, not prod it to reform. Worse, it will do more to bankrupt the already-struggling automakers and less to improve auto safety. In 2013, 45 percent more cars were recalled than sold. That is galling — especially when one considers that defective cars are responsible for ever fewer accidents, notwithstanding the GM tragedy. Indeed, features such as air bags and seat belts have contributed to cutting driving fatalities from 26 per 100,000 people in 1965 to 10 now. The big remaining culprit in car accidents is driver error. And the most effective way of dealing with that is improving vehicular safety.

But the more resources carmakers are forced to devote to satisfying bureaucratic red tape the less they'll be able to spend on actual safety improvements.

There are no perfect solutions here, but a bigger and more aggressive NHTSA is a cure worse than the disease.

This column originally appeared in The Week. You can find Dalmia's archive here.