Should auto recalls be delineated by severity?

The idea has been floated for years, though it has taken more urgency as legislation to ground recalls is bandied about in the House and Senate. The theory goes that creating a system of severity would require vehicles subjected to the most serious recalls be grounded immediately, while others would fall under another classification, ostensibly allowing for a longer time frame to repair.

In theory, it makes sense: Some recalls seem more directly related to the safety of the vehicles’ occupants — such as airbags, steering and powertrain — while others are not, such as sunroofs and detached warning labels.

This is in light of the fact that the entities beholden to the legislation, such as auto dealers, car rental companies and used car consignors, face a considerable burden in grounding and fixing a recalled vehicle. With the unprecedented spike in recalls in the last two years, these operators have been left stranded with unrepairable vehicles (for months, sometimes) when parts aren’t available. It’s a nasty hit to the bottom line of a small business.

Even more pressing from a safety standpoint is the notion of “recall fatigue” — as the sheer number of recalls causes sensory overload, there is a fear that completion rates could be negatively affected. A determination of severity may prompt consumers to act with greater urgency.

But could such a system of recall hierarchy work?

I decided to analyze recall data to get closer to an answer. I downloaded the publically available file of automobile recalls from the National Highway Traffic Safety Administration (NHTSA) from Jan. 1, 2014 to Sept. 9, 2015.

To make the list more manageable, I expunged recalls from tire manufacturers and those covering aftermarket parts. I limited the list to passenger vehicles, light-duty trucks and vans from major auto manufacturers. I grouped recall types where appropriate. This effort produced the enclosed list, which contains 35 types of recalls covering 70,604,442 vehicles.

And then I went about trying to rank the “severity” of the recall.

There are some no-brainer categories when it comes to the safe operation of a vehicle: airbags, electrical systems, brakes, power trains, steering, vehicle speed control, suspension, wheels and engines.

Other categories are a little murkier, such as structure, exterior lighting, equipment-electrical, equipment-adaptive and latches/locks/linkages. The categories that seemed to fall on the side of overly cautious include spare tires, “equipment-other labels” and “visibility-sunroof.”

Delving further into those I deemed “less severe,” I pulled this description from the structure category:

“In the event of an accident, the center storage console may not stay latched. … The storage compartment door may not remain closed in the event of a crash, increasing the risk of occupant injury.”

This description came from the exterior lighting category:

“Due to a wiring incompatibility, the front side marker lamps may not function. … Without the proper illumination of the side marker lamps, the vehicle may be less visible in night-time conditions, increasing the risk of a crash.”

This description is from what I’m calling the “overly cautious” category:

The affected vehicles may have a brake reservoir cap that does not have the required text advising owners which brake fluid the vehicle uses. … Without the correct brake reservoir label on the cap, someone may add the wrong brake fluid, causing damage to the brake system seals, resulting in brake fluid leaks, lengthened stopping distances and increasing the risk of a crash.”

In practical terms, an exploding airbag is riskier than a lack of text on a brake reservoir cap. Still, trying to rank recall severity is akin to playing God. The phrase “increasing the risk of a crash (or injury)” was inherent in almost all recall notices, from the alarming to the seemingly benign. If any component failure increases the risk of a crash, how can it be ranked as more or less safe as others? From a liability standpoint, zero is the acceptable threshold.

Would safety advocacy groups or the general public be comfortable with this ranking? More importantly, would NHTSA — the creator of these notices and the crash-risk language — allow ranking by severity?

Upon taking office last year, Mark Rosekind, NHTSA’s administrator, faced immediate pressure to take a tougher stance on safety, especially in light of the agency’s perceived regulatory lapses regarding the Takata airbag recalls and General Motors’ faulty ignition switches.

In light of this, a ranking of recalls by severity is unlikely to happen. Remember, NHTSA examines scores of safety issues and not all of them become recalls. The ones that make the cut have already passed the agency’s severity threshold.

Too, imagine the consumer who delayed fixing a “non-urgent” recall and was injured or killed as a result of the failure of a recalled component. This is not a chance NHTSA is willing to take.

Whether you’re for or against recall legislation, energies are best spent elsewhere.

From a real-world standpoint, the number of units affected for those destined for the most severe category dwarf those in the “less severe” categories by a 65-1 margin. Even if enacted, such a hierarchy would provide little practical relief indeed.