Question: A number of public policies related to alcohol consumption have been instituted over the past couple of decades in an attempt to limit the number of alcohol-related traffic fatalities. These policies include:
• Increasing taxes on beer.
• Raising the minimum legal drinking age (MLDA)
• Automatically suspending driver's licenses of drunk drivers ("per se law").
• Automatically suspending licenses of drivers who refuse to take a blood alcohol test ("im-plied consent law").
• Permitting those injured by drunk drivers to sue those who served alcohol to the driver ("dram shop law").
The data found represent a dataset constructed by Christopher Ruhm for his paper, "Alcohol Policies and Highway Vehicle Fatalities" (Journal of Health Economics, Vol. 15, No. 4 (August 1996), pp. 435-454). The author has graciously pro-vided these data for use here. It contains data for the vehicle fatality rate in each of the 48 con-tiguous United States (i.e., excluding Alaska and Hawaii) in each year between 1982 and 1988, along with each of the policies listed above and other factors that may influence accident rates.
a. Estimate an OLS regression model of the vehicle fatality rate on each of the policy variables along with: the unemployment rate, the percentage of drivers between the ages of 15 to 24, per capita income, the average miles driven, and a dummy variable for each year of the sam-ple (excluding 1). Interpret the results of this regression.
b. Estimate the same regression model, but now include state fixed effects. Conduct an F test to determine whether you can reject the null hypothesis that these state fixed effects are jointly equal to zero.
c. Compare the coefficient estimates in this model compared to that in the OLS regression. Provide an explanation for any differences that emerged.
d. Which model do you prefer? Why?
e. Based on the evidence you have compiled, what policy recommendations would you make regarding approaches to reduce alcohol-related traffic fatalities?