SUV Safety Stat Studies I was quite intrigued by Judith Lewis’ September 23–29 column “Better Mileage on Ice,” especially by the section where she writes: “Fuel efficiency equals death. It’s an equation beloved by the conservative Heritage Foundation and the libertarian Cato Institute, but it stupidly omits one factor: In the majority of those fatal crashes, a smaller car was hit by an SUV, often one that rolled over.” Can Ms. Lewis provide me and her readers with statistics to support her assertion? According to the National Highway Traffic Safety Administration’s most recent “Traffic Data Facts” report, which reports traffic data from 2003, SUVs were involved in only 11.3 percent of all fatal accidents that year, a far cry from a “majority of . . . fatal crashes.”

—Thomas A. Firey Managing editor, Regulation magazine The Cato Institute

Lewis responds: I think the confusion lies in the word involved. When Mr. Firey refers to SUV drivers involved in fatal accidents, he means only the drivers of the SUVs were killed. When I say SUV drivers were involved in fatal accidents, I refer to vehicle-to-vehicle crashes in which the SUV driver, the passenger in the smaller vehicle or the driver of the smaller vehicle was killed. According to the U.S. Department of Transportation and the National Highway Traffic Safety Administration, 81 percent of the time, when an SUV, light truck or van is involved in a fatal accident with a smaller passenger vehicle, it’s someone in that smaller car who dies.This well-known and much-studied phenomenon is often referred to as the “aggressivity” of light trucks. That said, even Mr. Firey’s statistics about SUV fatalities aren’t quite right. According to the National Highway Traffic Safety Administration’s August 2005 report, “Traffic Safety Facts,” almost 14 percent (4,483 out of 32,271) of motorists killed in traffic accidents in 2004 were driving SUVs. If you add light trucks to that, it goes up to 32 percent, and with vans, it’s roughly 39 percent. A more detailed analysis of this phenomenon began almost 10 years ago with a study done for the NHTSA called “The Aggressivity of Light Trucks and Vans.” It found that “collisions between cars and LTVs account for over one-half of all fatalities in light vehicle-to-vehicle crashes. Nearly 60 percent of all fatalities in light-vehicle side impacts occur when the striking vehicle is an LTV.” The study included Ford Broncos, Chevy Blazers and Toyota 4-Runners in the LTV category. Gabler and Hollowell repeated their findings in 1998 and 2000 (as “The Crash Compatibility of Cars and Light Trucks”). Firey responds: With respect to Ms. Lewis, I understand the word involved to mean something like “to include or contain as a part.” Given this understanding, an SUV would be involved in a fatal accident not just if the SUV driver were killed, but if any occupant of any involved vehicle were killed, or if a pedestrian were killed. In that sense, SUVs were involved in only 11.5 percent of all fatal accidents in 2003, the last year for which NHTSA has released complete data. If the emerging class of utility station wagons is included as SUVs, that percentage rises to 12.3 percent. To be clear, this involves the universe of all fatal accidents — both single-vehicle and multiple-vehicle. Whether “SUV-involved” fatal accidents are 11.5 percent or 12.3 percent or even “almost 14 percent,” it clearly is not the case that “In the majority of those fatal crashes, a smaller car was hit by an SUV.” In her response to my letter, Ms. Lewis asserts that she was in fact referring to the much smaller universe of vehicle-to-vehicle fatal accidents. According to NHTSA, there were 16,584 fatal multivehicle accidents in 2003. In that year, 6,757 fatal accidents involved SUVs; 7,206 if sport utility wagons are included as SUVs. Assuming, improbably, that all SUV fatal accidents were multivehicle, it still would not be the case that “In the majority of those fatal crashes, a smaller car was hit by an SUV.” Ms. Lewis does, of course, have considerable empirical evidence on her side in arguing that in a multi-vehicle accident involving an SUV and a lighter car, the lighter car typically sustains more damage and its occupants are more likely to die or be injured than the SUV and its occupants. But this fact does not rebut the “stupidly” asserted stylized fact that heavier passenger vehicles perform better in accidents than lighter vehicles. As indicated by the NHTSA data cited above, the majority of fatal accidents are single-vehicle. It is in these accidents that SUVs and other light trucks significantly outperform passenger cars in protecting their occupants. Of course, this does not close the argument. Legitimate social-justice questions can be raised about a situation in which certain types of vehicles are safer for their occupants but are more dangerous for other motorists. Interestingly, that question suggests that it would be better from a safety perspective for more people to drive SUVs (and enjoy their occupant-protective effects). On the other hand, one of my favorite researchers, UCSD’s Michelle White, has conducted an excellent statistical analysis that indicates that society would be worse off, safetywise, if everyone were to drive SUVs — in part because many non–SUV drivers have driving patterns that do not match well with SUV performance.


All-access pass to the top stories, events and offers around town.

  • Top Stories


All-access pass to top stories, events and offers around town.

Sign Up >

No Thanks!

Remind Me Later >