Alcohol Taxes and Binge Drinking in U.S. States: The Effects of Including Multiple Tax and Beverage Types

Background: In the U.S., the relationship between alcohol taxes and binge drinking is incompletely understood. Most tax analyses have assessed only specific excise taxes as a proxy for price but not ad valorem excise or sales taxes, which are based on a percent of price, and most states have not adjusted specific excise taxes to match inflation. Methods: The association and goodness-of-fit between state alcohol taxes and binge drinking was assessed using different combinations of tax and beverage types from 2001-2010. Alcohol tax data were obtained from the Alcohol Policy Information System, the Tax Foundation, and WestlawNext. Weighted aggregate and beverage-specific taxes-per-drink were calculated by state and year based on existing tax rates and settings where applied. Binge drinking prevalence was obtained from the Behavioral Risk Factor Surveillance System survey. Results: The correlation between beer specific excise tax and beer combined tax was r = 0.28 (p-value=0.04) in 2010. For all models, there was a significant negative association between tax-per-drink and binge drinking. Increasing the number of tax types improved the goodness-of-fit, particularly in later years (for beer in 2010, â= -0.30, R2 = 0.18 for all 3 tax types vs. â= -0.13, R2=0.01 for only excise taxes among all states). However, including taxes on liquor and wine improved the model fit slightly (â=-0.34, R2=0.33) compared with beer-only models (â=-0.39, R2=0.30 in 2010) among licensed states. Conclusions: Raising alcohol taxes should result in substantial decreases in binge drinking. Future research assessing alcohol taxes should ideally incorporate all taxes.