Since 1996, states have been implementing and enhancing their graduated driver licensing (GDL) programs. Increased licensing restrictions could steer new drivers to bypass training and licensing altogether. Unlicensed driving is associated with increased fatal crashes and high-risk behaviors that have been shown to adversely affect passenger safety behaviors like restraint use. The objective of this study was to assess the impact of varying state level GDL programs on rates of unlicensed driving and on passenger restraint use. De-identified data from the National Highway Traffic Safety Administration's Fatality Analysis Reporting System from years 1996-2010 was analyzed. Fatal crashes involving drivers (15-24 yrs) and their passengers (15-24 yrs) were included. Using a validated system, each state's GDL laws at a given month were rated as poor, marginal, fair, or good. The association between GDL strength and unlicensed driving was analyzed graphically and by chi-square test. Multivariate logistic regression with generalized estimating equations were undertaken to assess the relationship between GDL strength and passenger restraint use. From January 1996 to December 2010, 26,504 (23.4%) passengers were involved in fatal crashes taking place in states with GDL programs rated poor, 21,366(18.9%) marginal, 33,603 (29.6%) fair, and 31,903 (28.1%) good. Rates of unlicensed driving ranged from 16.4% in state-months rated marginal versus 21.5% in state-months rated good (p<0.001). In the multivariate model, compared to states with poor GDL ratings, each additional rating boost was associated with an increased odds of passenger safety restraint use (OR 1.15, 95% CI 1.13-1.18). Our findings suggest that stronger GDL law can mitigate passenger risk in fatal crashes by encouraging passenger restraint use. Our study provides evidence that stronger legislation in these states may reduce overall risk to young drivers and their passengers.