Do Seat Belt Laws Work?

by John Semmens

Many states that have passed mandatory seat belt-use laws have required that evidence of the law's effectiveness be produced for the law to escape automatic expiration. A recently published report-"Arizona Hospital Costs for Seat Belt Use vs. Non-Use 1989, 1990,1991"-from the Governor's Office of Highway Safety purports to be the needed evidence for the extension of Arizona's seat belt law. Unfortunately, these kinds of reports have neither asked nor answered the right questions.

Proving that people suffer more severe and expensive injuries when they're not wearing seat belts belabors the obvious. No credible opponent of seat belt laws has disputed that seat belts can save wearers from death and injury. To present statistics that never were in doubt as the longawaited evidence fails to deal with the unresolved issue of whether requiring seat belt use is good public policy.

Critics of seat belt laws have contended that they alter driver behavior in ways that increase the hazards for other users of the streets and highways. In particular, some drivers wearing seat belts may feel more assured of surviving an accident, and hence tend to drive more aggressively, thus raising the risk of collisions with other vehicles and pedestrians.

In the early 1970s, a few challenges to the presumed safety benefits of increased auto safety regulations appeared in lightly read academic journals. In a 1970 issue of Applied Economics, L. B. Lave and W. W. Weber suggested that mandated safety devices (seat belts, better bumpers, collapsible steering wheels) might lead to faster driving that could offset the safety gains. In 1975, Sam Peltzman's "The Effects of Automobile Safety Regulations" in the Journal of Political Economy hypothesized that safer autos would lead to more aggressive driving that would endanger other users of the roads.

This earlier research has been mostly ignored or dismissed in favor of adherence to more simplistic research that, unsurprisingly, proves that crashtest dummies suffer more damage without safety devices. Crash-test dummies, of course, cannot have their driving behavior altered by a perception of greater crash survivability. Consequently, the research with dummies doesn't refute the hypothesis that driver behavior might be changed and thus negate or reduce some of the anticipated safety gains.

The plausibility of the aggressive driver hypothesis cries out for more research. For example, Hawaii, the state with the most rigorously enforced seat belt law and the highest compliance rate in the nation, has experienced an increase in traffic fatalities and fatality rates since its law went into effect in December 1985.

This is not to say that the seat belts are killing vehicle occupants. Clearly enough crash-test dummies have smashed into enough auto windshields and dashboards to convince all but the most obstinate that wearing a seat belt is probably a good idea. What, then, is going on in Hawaii? Well, we don't know. But the data do not support a smug assurance that forcing people to wear seat belts is without potential undesirable outcomes.

A recent statistical study of states with and without seat belt laws was undertaken by Professor Christopher Garbacz of the University of Missouri-Rolla. This study seems to support the altered driver behavior hypothesis. Dr. Garbacz found that states with seat belt laws saw decreases in traffic fatalities for those covered by the laws (typically drivers and front-seat passengers), but increases in fatalities for rear-seat passengers, cyclists, and pedestrians. Further, the patterns of changes in total traffic fatalities among the states showed no consistent relationship with the existence of a seat belt law in the state.

This suggests a significantly less optimistic interpretation of the impact of seat belt laws than the prevailing orthodoxy would allow. Forcing unwilling motorists to wear seat belts may save their lives and reduce their injuries. Disconcertingly, though, seat belt laws appear to be increasing the hazards for other users of the roads.

Deciding whether this apparent shift in risk is an acceptable cost of a seat belt law is a far different proposition from pretending that there is no significant cost. Policy-makers may be satisfied that the benefits of a seat belt law outweigh the costs. However, a humane public policy demands that those who may ultimately pay the costs be warned of the potential increased risks they face on the streets and highways. To do less is to endanger some of the least protected users of our roads.