Auto Insurance and Traffic Fatalities

11/01/2003
Summary of working paper 9602
Featured in print Digest

No-fault limits on driver liability in an accident, by diminishing the possibility of being sued, increase fatalities by about 10 percent.

When states adopt compulsory automobile insurance regulations and no-fault liability laws, some drivers may become a little more lax or relaxed. They may drive slightly faster or a little less cautiously, aware that the financial and legal consequences of an accident are reduced. They may be somewhat less likely to use seat belts and more likely to buy an old car without air bags, or to drink and drive, or simply to drive more often and further. As a result, the number of traffic fatalities increases, according to a study by Alma Cohen and Rajeev Dehejia. In The Effect of Automobile Insurance and Accident Liability Laws on Traffic Fatalities (NBER Working Paper No. 9602), they find that no-fault limits on driver liability in an accident, by diminishing the possibility of being sued, increase fatalities by about 10 percent.

That increase is significant. Traffic accidents claim over 40,000 lives each year in the United States, roughly the same as the total number of Americans killed during the Vietnam War, Cohen and Dehejia note. Americans spend roughly $100 billion each year on automobile insurance premiums, and they bear more than $250 billion of uninsured accident costs each year. So the impact of introducing no-fault insurance has meant an additional 5,160 to 6,450 lives lost in the United States, depending on the year, the authors calculate. Economists call such consequences "moral hazard costs."

For this study, Cohen and Dehejia analyzed data from 50 states and the District of Columbia for 1970-98, a period during which most states adopted compulsory insurance requirements and 16 states adopted no-fault automobile insurance. By 1997, 45 states had compulsory insurance laws, although there were variations in the amount of insurance required and the methods of enforcement.

The introduction of compulsory insurance laws did lead to a reduction in the number of uninsured motorists. Compared to a base level of 12.9 percent, the number of uninsured motorists fell by 2.4 percentage points. The authors note that the introduction of compulsory insurance requirements made no difference to those who were already buying insurance voluntarily, but did have an effect on those individuals who were not previously buying insurance and chose to do so because of compulsory insurance, and on those individuals who chose to remain uninsured. Drivers who remained uninsured despite the new requirement in theory drove more cautiously, for fear of being in an accident, so their number of fatalities likely dropped. But drivers forced to buy insurance and now facing diminished liability in case of an accident might have driven less carefully, increasing accidents and fatalities. Looking at the data, the authors find a 2 percent increase in fatalities for each percentage point decrease in uninsured mot orists. So, compulsory insurance is "not an unmitigated good," the authors write.

These findings may be useful in assessing the social desirability of policy shifts in automobile insurance and no-fault liability rules. Though there are many positive aspects of compulsory insurance (such as greater assurance that the costs of an accident, including vehicle replacement or medical needs, will be compensated to some extent) and no-fault liability rules (such as reducing the court and other legal and administrative costs associated with driving and traffic accidents), the authors suggest that both policies also had unintended negative consequences in terms of increased fatalities. These consequences must also be taken into account.

-- David R. Francis