Academic journal article Contemporary Economic Policy

The Impact of Large Container Beer Purchases on Alcohol-Related Fatal Vehicle Accidents

Academic journal article Contemporary Economic Policy

The Impact of Large Container Beer Purchases on Alcohol-Related Fatal Vehicle Accidents

Article excerpt

I. INTRODUCTION

Excessive alcohol consumption has well-known adverse consequences, such as alcohol-related fatal accidents, diseases, injuries, crime, and violence. Given that many of the behaviors related to excessive alcohol consumption have negative externalities, curbing such behavior has meaningful social benefits. Consequently, there are various regulations imposed by the authorities to lower excessive alcohol consumption, such as excise taxes, Minimum Legal Drinking Age (MLDA) Laws, maximum Blood Alcohol Content (BAC) thresholds, hours of sale, etc.

Of particular note, Levitt and Porter (2001) have shown that the negative externalities associated with drunk driving are quite high ($.30 per mile driven while intoxicated). Statistics show that there were 9,878 fatalities in alcohol-related accidents, 31% of total traffic fatalities in 2011 in the United States (National Highway Traffic Safety Administration 2012). Although authorities have been trying to mitigate this social problem by imposing various policies, there is still a need for further examination of the underlying determinates of drunk driving. Understanding what behaviors impact these negative outcomes and how individuals are best incentivized to prevent drunk driving is important to designing optimal public policy.

In this study we explore one unnoticed and potentially important factor affecting consumer behavior of alcohol: beer container sizes. In particular, we aim to examine the relationship between beer container size and alcohol-related fatal accidents. There are many studies that discuss the impacts of policies aiming at reducing drunk driving. However, to our knowledge, this is the first study to investigate the relationship between beer container size purchases and alcohol-related fatal accidents.

Unlike wine and liquor, which are typically consumed by the glass and have a greater shelf life after opened, beer is typically packaged and consumed in serving size containers (e.g., can or bottle). This difference is largely a function of the much shorter time period for which consumption of beer is optimal after opening. In recent years beer manufactures have begun to market beer in larger than standard (12 oz.) containers at a cheaper price per unit (ounce). The combination of the lower per unit price, larger container size, and a short consumption window after opening, may cause people to consume alcohol more quickly, or in greater quantities, or both. Hence, consumption of beer in larger container sizes may lead to greater average levels of intoxication and, subsequently, an increase in alcohol-related fatal vehicle accidents.

To investigate this issue we utilize data from two principal data sources, the Fatality Analysis Reporting System (FARS), which comprises detailed information about all fatal accidents in the United States, and the Nielsen scanner panel data, which provides data on alcohol purchases from retail outlets. Empirically we employ a weighted least squares (WLS) fixed-effects model specification. Our results are consistent with the concern outlined above, as we observe an increase in alcohol-related fatal accidents in locations with increased sales of larger beer containers. In particular, we find that a 10% increase in beer purchases in containers that are greater than 12 oz. in size increases the number of alcohol-related fatal accidents by 1.95%. Additionally, a 10% increase in beer purchases in containers greater than 18 oz. increases the number of alcohol-related fatal accidents by 2.24%.

As we will demonstrate below, these estimates are robust to the inclusion of controls for area and time fixed effects, changes in population, and changes in factors that may influence overall driving risk separate from drinking behavior (e.g., construction, weather, etc.). Furthermore, these estimates are also robust to the estimation method selected (e.g., instrumental variable [IV], weighted IV, ordinary least squares, negative binomial, etc. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.