The 131st Annual Meeting (November 15-19, 2003) of APHA

Abstract #68194

Daniel Eisenberg, PhD, NIMH Postdoctoral Scholar, School of Public Health, UC-Berkeley, 140 Warren Hall, MC 7360, Berkeley, CA 94720, 510-642-5652, daniel7@uclink.berkeley.edu

PURPOSE: This study evaluates the effectiveness, in terms of fatal crash reductions, of state-level public policies related to drunk driving in the U.S. Graduated licensing for young drivers and the MADD organization are evaluated for the first time in this type of analysis. METHODS: An OLS regression approach with state and year fixed effects is used for a panel of all 50 states and D.C., 1982-2000. Conventional estimates of drunk driving policy effects might be biased due to the endogeneity of policies. In particular, the adoption of stricter policies is likely to coincide with shifts in unobserved attitudes towards drunk driving, and thus reduced accidents can be improperly attributed to only the policies. This concern is addressed by analyzing the time pattern of policy effects with respect to the date of adoption. A policy “effect” that occurs before policy enactment would be interpreted as possible evidence for an endogeneity problem. RESULTS: For the 0.08 BAC law and graduated licensing, the results suggest that a bias upwards in conventional estimates exists, but the policies are still effective. The estimates for MADD do not imply an effect, but this result could be due to the crudeness of the variable used. Other policies estimated to be effective are dram shop laws, open container laws, and seat belt laws. For young drivers, zero tolerance BAC limits and beer taxes are effective. CONCLUSION: The time pattern analysis suggests that conventional estimates for some policies are overstated, but the policies still appear to be effective.

Learning Objectives:

After attending this presentation, participants will have learned the following:
• how effective various drunk driving polices are in terms of reducing fatal accidents
• why prior estimates of some policies' effectiveness are probably overstated

Presenting author's disclosure statement:
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.