The Effects of Driving Age, Driver Education, and Curfew Laws on Traffic Fatalities of 15–17 Year Olds1


  • 1

    This paper was written while the author was at Rutgers University and does not in any way reflect the views of the Federal Trade Commission.


This study examines the effect of state driving age, learning permit, driver's education, and curfew laws on 15–17-year-old driver fatality rates. A multivariate regression model is estimated for 47 states and nine years. The minimum legal driving age and curfew laws are found to be important determinants of fatalities. Driver's education and learning permits have smaller effects. The relationship between rates of licensure and driving age, education, and curfew laws is also examined. In each case, a more restrictive policy is found to reduce licensure of 15–17 year olds. The results suggest that the imposition of curfew laws and higher minimum driving ages are particularly effective traffic safety policies.


This study examines the effects of state driving age, learners permit, curfew, and driver's education laws on traffic fatalities of 15–17 year olds. The empirical results indicate that each of the laws has statistically significant effects on both multivehicle and single-vehicle driver fatality rates. An important part of the effect of curfew laws, driving education laws, and, of course, driving age laws appears to occur through their effect on discouraging early licensure.

In interpreting these results, certain potential limitations deserve mention. Correlation does not imply casuality; the effects of the laws might be due to other underlying causes. By including a variable for the 25–29-year-old fatality rate, the estimation equation does include a control for the fatality propensity in a particular state and year. However, the state laws might still reflect a reaction to 15–17 year old fatalities rates or to those factors that affect these rates. To the extent that states react to high 15–17-year-old fatalities by making youth traffic fatalities more stringent, the estimated effects of the laws will tend to be biased downward. Second, the estimates are average effects of laws across states and time. The nature of coverage for each law (e.g., the time of day covered by curfew laws) and the degree of enforcement efforts differs among states. Consequently, effects may vary for a particular law in different states. Finally, this study did not consider experience effects, i.e., the potential increase in fatalities when first year (inexperienced) drivers begin to drive.18 However, early studies by Levy(6) Robertson(5) and Williams et al.(2) indicate that experience effects do not significantly counteract the effects of raising driving ages.

In sum, the results suggest that state laws, especially driving age laws, have important effects on youth fatalities. As earlier work also indicates (e.g., Ref. 2), raising the driving age merits serious consideration as a policy measure. The results on curfew laws are consistent with the earlier study by Preusser et al.(3) and are not surprising considering that young driver's spend a disproportionately larger amount of time than other age groups on the road.19 Unlike many early studies, the results here suggest that mandatory education laws have some salutory effect, where the alternative is the same minimum driving age but without a driver's education requirement. However, as suggested in other studies (e.g., Ref. 4), they are a poor substitute for raising the driving age. A final, albeit limited, measure is raising the age at which a learning permit can be obtained.