Graduation Year

Spring 2012

Document Type

Campus Only Senior Thesis

Degree Name

Bachelor of Arts



Reader 1

Eric Helland

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Rights Information

© 2012 George Posner


If you have to wear a seat belt when you drive, are you safer? Intuitively, it may seem that the answer is yes. After all, if you are wearing a seat belt and get in an accident, you are half as likely to die, and 62% of fatal accident victims were not wearing seat belts at the time of accident. The Centers for Disease Control and Prevention, the National Highway and Transportation Safety Administration, the Governors' Highway Safety Administration, and many other organizations highly recommend wearing seat belts. The NHTSA claims that in 2010 alone, approximately 12,500 deaths were prevented by seat belt use alone. Seat belt laws clearly reduce the chance of death to vehicle occupants in a given auto accident.

In response to these findings, the federal government has made the release of highway funds to states contingent on the passage of state laws mandating seat belt adoption. Laws mandating seat belt use, along with extensive campaigns to raise public awareness, have caused seat belt use to rise from 69% in 1998 to 88% in 2009. As of this writing, laws mandating the use of seat belts when driving have been passed in every state save New Hampshire. Intuitively, this should make roads safer because seat belts make an accident more survivable.

Does wearing a seat belt, however, make that accident more likely to occur in the first place? If a driver wears a seat belt now and I didn't before, does he feel safe enough to take more risk? In this paper, I examine this question using insurance premiums as a proxy for the likelihood of an accident. The rest of the paper is organized as follows. Section 2 offers background information, including a framework with which to interpret a driver’s actions and a review of the relevant literature. Section 3 contains details on the data analyzed. Section 4 covers the results of my preliminary data analysis, model specifications, and robustness checks. Section 5 concludes.

This thesis is restricted to the Claremont Colleges current faculty, students, and staff.