As was widely reported yesterday (e.g., USA Today, Bloomberg, LA Times), the National Highway Traffic Safety Administration (NHTSA) sent Chrysler a letter earlier this week asking it to recall the 1993-2004 Jeep Grand Cherokee and the 2002-2007 Jeep Liberty because they “performed poorly when compared to all but one of the 1993-2007 peer vehicles, particularly in terms of fatalities, fires without fatalities, and fuel leaks in rear end impacts and crashes.” Specifically, the NHTSA’s Office of Defects Investigation said:


In our tentative view, there is a performance defect and a design defect.

The performance defect is that the fuel tanks installed on these vehicles are subject to failure when the vehicles are struck from the rear. Such failure can result in fuel leakage, which in the presence of external ignition sources, can result in fire.

The design defect is the placement of the fuel tanks in the position behind the axle and how they were positioned, including their height above the roadway.


(Spaces added for clarity.) The NHTSA notes that, because of the defects, passengers “have burned to death in rear impact crashes, there have been fires (without fatalities) in these vehicles from rear impact crashes that have, or could have, led to deaths and injuries.” Compared to similar SUVs, the Grand Cherokee and the Liberty had roughly twice as many fatalities per million registered vehicle years (MRVY), a standard measure for vehicle safety over time. When it came to non-fatal fires, the Grand Cherokee was almost ten times as likely to be involved in a fire than similar vehicles, and the Liberty was nearly sixty times as likely.


In the face of that evidence, Chrysler said “no, we won’t recall it.” They put out their own paper claiming “NHTSA used an incomplete and unrepresentative group of comparison vehicles” and arguing that the fatal crashes weren’t representative because they all involved unusually high speed crashes. They also complained that the NHTSA hadn’t recalled other vehicles with higher MRVY rates of fatal rear-impact crashes with fire.


There’s a lot to learn from this battle.


(1) Regulation Makes Cars Safer.


Everyone likes to complain about government regulation, but few people will admit that it makes people safer. Rear impact collisions in vehicles with rear-mounted gas tanks aren’t exactly a new issue: as the NHTSA points out, the “placement of the fuel tanks in the position behind the axle” is essentially the same problem the 1970s Ford Pinto had. Back in 2002, the NHTSA looked into a similar issue regarding the 1992-2001 Ford Crown Victoria, Grand Marquis, and Town Car.

So, why aren’t Jeep models after 2007 affected? Because Federal Motor Vehicle Safety Standard No. 301, “Fuel system integrity,” is a lot stricter for those vehicles manufactured then. Pre-2007 models are tested using a rear impact at 48 km/hr (roughly 30 mph). 2007 and later models are tested using a rear impact at 80 km/hr (roughly 50 mph), with an overlap, to better simulate real-word conditions. The 1993-2004 Jeep Grand Cherokee and 2002-2007 Jeep Liberty pass the old test but fail the new test, hence the safer design in the later models to meet that test.


(2) Comparative Safety Is In The Eye Of The Beholder


As noted above, the NHTSA admits that the affected vehicles passed the relevant federal standards for “fuel system integrity” applicable to cars made back then. So does that get them off the hook? No: as Brett Emison explains in depth, for decades the NHTSA and the U.S. Department of Transportation have made clear that the “FMVSS” standards are a floor, not a ceiling. If a car doesn’t meet them, it can’t be sold. If you do meet them, that doesn’t necessarily mean a car is safe any more than the issuance of a driver’s license means a driver can’t be negligent. (Bonus historical point: in Emison’s post did you catch the name of the firm on that 1981 letter from the NHTSA? It is “Beasley, Hewson, Casey, Colleran, Erbstein & Thistle,” the predecessor to our firm.)


So how do you prove a car is defective? One way is to compare the car to other cars meant to fulfill a similar market. The NHTSA evaluated the Jeep SUVs in part by comparing them to “peer” vehicles, including Toyota 4Runner, Ford Explorer, Jeep Wrangler, Nissan Pathfinder, Chevrolet Blazer, Mitsubishi Montero, Isuzu Rodeo, Isuzu Trooper, Suzuki Sidekick and Suzuki XL-7 from the same years, 1993-2007. Those all sound to me like comparable Sport Utility Vehicle meant to fill the same consumer needs.


Chrysler’s response, including a chart titled “Chrysler Group vehicles are as safe as comparable vehicles,” tries to compare the 1993-2004 Jeep Grand Cherokee and 2002-2007 Jeep Liberty with obviously incomparable cars. Here are a couple of them with worse rates rear-impact fatal collisions where the vehicle experienced a post-collision fire than the Jeep Liberty:



NISSAN 240SX 1989-1994

MAZDA RX7 1984-1991/1993-1995


MAZDA GLC 1984-1985

NISSAN 200SX 1984-1988

TOYOTA MR2 1985-1995


EAGLE/AMC ALLIANCE 1984-1987/ENCORE 1984-1986

ISUZU AMIGO 1989-1994


Why did Chrysler try to compare its 1993-2007 SUVs to, for example, the Isuzu Impulse, a compact car from 1984-1989 of which only 13,000 were even made? Probably because the Isuzu Impulse had one of the very worst fatal rear-impact crashes with fire rates of the past 30 years. Indeed, Chrysler admits in the fine print they were comparing their vehicles to “100 vehicles having the highest rate of rear-impact fatal collisions where the vehicle experienced a post-collision fire,” the bulk of which are compact cars manufactured in the 1980s.


I am not convinced, and I hope the NHTSA isn’t convinced, either. Those cars aren’t comparable and, more to the point, they’re not as prevalent as the 1993-2004 Jeep Grand Cherokee and 2002-2007 Jeep Liberty, of which they are more than 2.7 million on the road. Thus, they don’t present the same threat to public safety.


(3) The Pinto Recall Formula Is Alive And Well


The infamous Ford Pinto Memo (you can read online the original 1977 Mother Jones article revealing it) compared the cost of a recall with the cost of simply letting the problem persist and settling the injury and death lawsuits out of court, noting that it didn’t make economic sense to issue a recall that cost more than the settlements would. (If you’re under 40, you probably first learned of it from the movie Fight Club.)


Jalopnik has the details on what kind of modifications that would need to be made to the Jeep SUVs to make them safer. According to the Center for Auto Safety — which filed the initial defect complaint, and was the key driver of this recall — “a recall involving installation of a 3 mm steel skid, a fuel tank check valve and better fuel filler hose would cost Chrysler no more $300 million.” Others in the industry have estimated $500 million. Either way, comparing those estimates to the NHTSA’s has a tentative assessment of 51 fatal rear impact fire crashes, that works out to between $5.88M-$9.80M per death, which is right in the range for the settlement of a product liability claim involving a fiery death in a car accident.


Chrysler might very well have determined that the damage to their brand was going to be done if they agreed with the recall or not, and that it was more economical for them to simply pay for the lawsuits than to issue the recall. Indeed, if Chrysler is right about the crashes largely being “high-energy” impacts, where the other car hit the Jeep at a relative speed greater than 50mph, then Chrysler might win several of those cases, or settle them for a compromised amount.


(4) Chrysler Has A Superficial Point, But One That Raises More Questions Than It Answers


One point in Chrysler’s response jumped out at me:


All but one fatal crash involving the subject Grand Cherokees, and all but four Jeep Liberty incidents, involved high-energy crashes. One highly publicized crash cited by NHTSA involved a tractor-trailer traveling 65 mph and a stationary Grand Cherokee. Crash energy was estimated at more than 23 times the required performance threshold. Seventy-eight percent of Grand Cherokee incidents involved impacts with crash energy that exceeded today’s rear impact fuel system integrity standard requirement which was doubled in the fall of 2008.


This argument has a superficial appeal: no car can be manufactured to be perfectly safe in all conditions, and so the argument that no car can be designed to prevent a fatality when a tractor-trailer going 65 mph hits a stationary car from behind.


Yet, this argument raises the obvious question: why don’t we see the same issues with all of those similar SUVs made in the same years? What’s the difference? Are the Jeep SUVs of those years somehow involved in far more high-energy crashes than other SUVs of those years? There’s no evidence that’s the case; if it was, I can guarantee you Jeep would be shouting it loudly.


Truth is, there isn’t any explanation for the disparity between the Jeep SUVs and other SUVs other than the Grand Cherokee and the Liberty are simply more prone to fatal rear impact fire crashes. That, then, brings us back around to what we mean when we say a car is “defective”: a car that is substantially more likely than comparable vehicles to cause a death or serious injury in the same types of crashes is, quite simply, defective, because it is not performing at the standards of the state of the art.