Price and availability revisited: Do alcohol taxation and sale restrictions also reduce harms for marginalized groups?

Alcohol taxation and sale restrictions are effective policies for reducing alcohol use and related problems overall, yet have the potential to influence subgroups of people differently. In order to better understand who’s most influenced by these taxes and policies, the authors of this study explored how alcohol taxes and alcohol sale restrictions in the United States relate to alcohol consumption and related problems across different racial and gender subgroups.

WHAT PROBLEM DOES THIS STUDY ADDRESS?

For any desirable commodity, price and availability are two major factors that influence its consumption. As price increases and ease of access decreases, consumption goes down. Alcohol taxation and restrictions on where alcohol can be sold are generally effective strategies employed by state and local governments to reduce alcohol consumption and associated harms. However, alcohol taxes may function differently across subgroups reflecting, for example, differences in beverage preferences among different racial/ethnic groups, and income. It has been argued that this approach to reducing alcohol use is unfair to socially disadvantaged individuals who might want to purchase and consume alcohol, because such people are more likely to drink alcohol, and at the same time have less disposable income.

Figure 1.

Though the relationship between alcohol consumption and taxation is well established (e.g., increasing the alcohol price to consumers through increasing alcohol taxes lowers consumption and related harms in a population) the majority of taxation studies have not accounted for other widely used policy approaches, such as restrictions on when and where alcohol can be sold. This is an important gap to address as limiting alcohol availability at off-premise outlets (i.e., to-go sales) may influence distinct subpopulations differently. For instance, for good or bad, because off-premise alcohol outlets are more concentrated in low-income and minority communities, restrictions on sales may disproportionately impact individuals of lower socio-economic status and/or minority racial/ethnic groups, and thus be a racially biased policy. At the same time, greater liquor-outlet density is known to lead to greater alcohol-related mortality, so these restrictions have public health benefit.

In order to better understand the relationships among alcohol consumption, alcohol taxation, alcohol availability, and race/ethnicity, Subbaraman and colleagues examined how state-level beer-, wine-, and hard-liquor-specific taxes and alcohol outlet densities are related to beverage-specific per capita alcohol consumption, as well as beverage-specific volume consumed by gender and race/ethnicity. They also assessed how alcohol taxation and outlet density relate to prevalence of alcohol use disorder and alcohol-related consequences across these subgroups.

HOW WAS THIS STUDY CONDUCTED?

This was a retrospective observational study using multiple state, federal, and alcohol industry data sources to explore population-level and individual-level relationships between alcohol taxation, regulation, and alcohol related problems in the United States. Relationships between alcohol taxes and sale policies, alcohol outlet density (i.e., the number of alcohol stores in a given area), and alcohol use over time were explored using state-level data (i.e., state-wide alcohol consumption statistics), and individual-level data (i.e., survey data from randomly selected individuals in the community). The state-level per capita data, and individual-level survey data complement one another in that the state-level per capita data allowed the authors to examine effects at the population level, while the survey data permitted subgroup analyses.

Outcome measures and data sources

Study outcomes included beverage-specific per capita consumption (i.e., beer, wine, hard-liquor), Diagnostic and Statistical Manual of Mental Disorders IV alcohol dependence diagnosis (now referred to as alcohol use disorder moderate or severe), and alcohol-related consequences across subgroups defined by gender and race/ethnicity. State-level per capita alcohol consumption was assessed using the Beverage Information Group Beer, Wine, and Liquor Handbooks, the National Alcohol Beverage Control Association database, the Alcohol Epidemiologic Data System, and alcohol producer-reported percent alcohol by volume (%ABV) values.

In parallel, individual-level alcohol use outcomes were derived from survey data using the 2000 (n= 7260), 2005 (n= 6631), 2010 (n= 7647) and 2015 (n= 6713) U.S. National Alcohol Surveys (Total N= 28 251). These computer-assisted telephone surveys assessed individuals aged 18+ representing all 50 U.S. states. While these surveys are designed to capture a sample of individuals representative of the total U.S. population, the authors intentionally oversampled Black and Latinx individuals to aid research into these underrepresented groups. Individual-level outcomes included beer-specific, wine-specific, and spirit-specific total volume consumed, alcohol dependence diagnosis, and alcohol-related consequences, all measured for past 12 months.

Alcohol tax and availability data sources

Alcohol tax information was derived from the National Institute on Alcohol Abuse and Alcoholism’s Alcohol Policy Information System, which provides information on state alcohol taxes and tax changes from 2003 to 2015 for beer and hard-liquor. Tax rates for beer and hard-liquor prior to 2003 come from the Prevention Research Center’s state-wide availability data system. Hard-liquor taxation information was derived from the National Alcohol Beverage Control Association database.

Data on alcohol availability came from two primary sources. The National Alcohol Beverage Control Association Survey Books provided data on alcohol availability policies, including control status and types of off-premise outlets allowed to sell beer and hard-liquor (e.g., grocery stores, gas stations). Data for state-level on-premise outlets came from the Beverage Information Group Beer Handbook and Liquor Handbook.

How were the data analyzed?

The authors explored associations between state-level per capita alcohol consumption, and year and state-level beer tax, hard-liquor tax, and government controlled hard-liquor sales and state-level outlet density. They statistically controlled for year and state-level sales tax, unemployment, poverty, median household income, % population non-Latinx Black, % population Latinx, and % population age 15+.

For individual-level outcomes, the authors explored associations between beverage-specific volume of consumption, and alcohol-related problems on state-level alcohol taxation, government-controlled hard-liquor sales (such as in states/counties where all liquor sales are through state-run stores) and respondent ZIP-code-level density of off-premise beer outlets, off-premise hard-liquor outlets and on-premise bars. They statistically controlled for the effects of age, marital status, education, employment, income, state sales tax rates, region of residence, and year in order to account for these individual factors.

WHAT DID THIS STUDY FIND?

Higher alcohol-specific taxes were associated with reductions in alcohol use.

Greater alcohol-specific taxes were associated with reductions in the use of certain types of alcohol. For instance, a 1% increase in the beer tax was related to a 0.19% decrease in per capita beer consumption, as well as a 0.09% decrease in total per capita consumption. None of the tax or availability variables were related to per capita wine consumption, and state-level general sales tax rates were not related to any of the per capita consumption outcomes.

The effects of state-level alcohol taxes differ depending on race and gender.

For individuals, alcohol use among women of racial minorities appeared to be most influenced by the effects of alcohol taxation. A 1% increase in beer taxes was related to a 0.5% decrease in total volume of alcohol consumed among Black women. Further, a 1% increase in taxes on hard liquor was related to a 1.04% decrease in volume of alcohol consumed among Latina women, and a 0.92% decrease in volume of alcohol consumed among Latino men. State-level general sales tax rates, however, were not related to individual-level alcohol consumption in any subgroup.
Additionally, increased beer taxes were related to slightly lower odds of any drinking among White women, and among Black women, a 1% increase in beer taxation was associated with a 0.40% reduction in volume of beer consumed, and slightly lower odds of having two or more alcohol-related consequences.

State-level alcohol outlet density had a limited influence on state-level per capita alcohol use.

The only state-level alcohol outlet density variable related to any per capita outcome was on-premise beer outlets (presumably cafes or restaurants only serving beer), with a 1% increase in on-premise beer outlets related to a 0.03% decrease in per capita hard-liquor consumption.

State-level alcohol outlet density influenced individual-level alcohol use.

Increased bar density was related to increased odds of drinking among Latina women and increased wine volume among Latino men, with each 1% increase in bar density associated with a 0.39% increase in wine consumption. For White men, a 1% increase in bar density was related to a 0.17% increase in total volume consumed. Notably though, for Black men, outlet density was not associated with alcohol use.
Living in a state where hard liquor sales are done through state-run outlets – a “control state” – was related to a 0.14% decrease in total volume of alcohol consumed. This effect appeared to be especially protective for White men in terms of reductions in any alcohol consumption, beer volume, and odds of consequences. For example, living in a control state decreased odds of any drinking by 17% for White men.
Increased off-premise beer outlet density was related to greater wine and hard-liquor volume consumed. Specifically, each additional alcohol outlet per square mile was associated with 0.22% greater wine consumption, and 0.12% greater hard-liquor consumption among White women.
Finally, a 1% increase in off-premise beer outlet density was related to 0.2% increased total volume alcohol consumed for White women and a 1% increase in off-premise hard-liquor outlet density was related to 0.22% increased total volume consumed for Latina women.

WHAT ARE THE IMPLICATIONS OF THE STUDY FINDINGS?

This is the first study to examine how beverage-specific tax and availability measures are related to beverage-specific consumption in diverse subgroups, and builds on previous work showing strategic alcohol policies can reduce binge drinking, alcohol-related health problems, drink driving and domestic violence. While confirming the effectiveness of policies that increase price and reduce availability on alcohol use and related harms, these results show that the effects of beverage-specific alcohol taxes and alcohol availability policies vary across subgroups defined by gender and race/ethnicity. This highlights the importance of considering in future research and interventions how alcohol taxation and sale policies impact subgroups differently.

Women, for instance appear to be more likely to reduce their alcohol consumption in response to even slightly higher prices due to greater alcohol taxation. Additionally, the protective effects of reduced alcohol consumption associated with living in a state with government-controlled hard liquor sales or a neighborhood with lower bar density was greater among White men than other groups. Notably too, while lower density of alcohol outlets appeared to be protective for most subgroups, it was not protective for Black men, which may speak to other sociopolitical factors that may be differentially affecting this population.

The authors of this paper theorized that alcohol policy directly affects alcohol consumption. Though causation cannot be proven with the kind of cross-sectional data used in this study, a great deal of converging evidence including sophisticated policy analyses showing positive effects of increased price and reduced availability on alcohol use, increase confidence that policy differences were directly influencing alcohol consumption.

Finally, though the percentage changes in alcohol use in relation to alcohol taxes and availability appear small, these changes nevertheless could have major impact on public health and safety. Slight increases in price to consumers through alcohol taxation and sale restrictions are not in and of themselves going to fix America’s massive alcohol use problem, and the outsized influence of the alcohol industry, but rather should be seen as public health tools to complement other community- and individual-level approaches to help individuals reduce alcohol consumption and mitigate related harms.

LIMITATIONS

As noted by the authors:

Survey data may be unreliable and/or subject to biases, including variability in social acceptability of binge drinking between states and over time.
The authors did not examine the wine tax, as it is highly correlated with the beer tax, and substantially more people drink beer than wine; future studies could examine differential effects of wine taxes.
The authors did not include laws related to drunk driving, such as blood alcohol content limits and license revocation policies, which vary across states and may have differential effects across subgroups examined here.

BOTTOM LINE

For individuals and families seeking recovery: This study explored associations between alcohol consumption, and alcohol taxation and restrictions on sales across gender and racial/ethnic subgroups. The authors found that alcohol taxation and restrictions on sales effect subgroups differently. For instance, women appear to be more likely to reduce their alcohol consumption in response to greater alcohol taxation. Further, living in a state with government-controlled hard-liquor sales or a neighborhood with lower bar density is more likely to lead to reduced alcohol consumption among White men, but appeared to have no influence on alcohol consumption among Black men. Though it is difficult to prove causality between alcohol outlet density and alcohol-related harms, an association is evident. Many individuals don’t have the luxury of choosing where they live, but given the choice, individuals in, or seeking alcohol use disorder recovery would do well to avoid living in areas with high alcohol availability that promotes easier access to it.
For treatment professionals and treatment systems: This study explored associations between alcohol consumption, and alcohol taxation and restrictions on sales across gender and racial/ethnic subgroups. The authors found evidence indicating that alcohol taxation and restrictions on sales effect subgroups differently. For instance, women appear to be more likely to reduce their alcohol consumption in response to greater alcohol taxation. Further, living in a state with government-controlled hard-liquor sales or a neighborhood with lower bar density is more likely to lead to reduced alcohol consumption among White men, but appeared to have no influence on alcohol consumption among Black men. Though it is difficult to prove causality between alcohol outlet density and alcohol-related harms, an association is evident. Helping individuals in, or seeking alcohol use disorder recovery find housing in areas with lower alcohol outlet density will likely improve alcohol use outcomes.
For scientists: This study explored associations between alcohol consumption, and alcohol taxation and restrictions on sales across gender and racial/ethnic subgroups. The authors found evidence indicating that alcohol taxation and restrictions on sales effect subgroups differently. For instance, women appear to be more likely to reduce their alcohol consumption in response to greater alcohol taxation. Further, living in a state with government-controlled hard-liquor sales or a neighborhood with lower bar density is more likely to lead to reduced alcohol consumption among White men, but appeared to have no influence on alcohol consumption among Black men. This important study answers some questions but gives rise to many more. In particular, additional research is needed to more deeply explore the relationship between alcohol outlet density and alcohol consumption and related problems.
For policy makers: This study explored associations between alcohol consumption, and alcohol taxation and restrictions on sales across gender and racial/ethnic subgroups. The authors found evidence indicating that alcohol taxation and restrictions on sales effect subgroups differently. For instance, women appear to be more likely to reduce their alcohol consumption in response to higher prices caused by greater alcohol taxation. Further, living in a state with government-controlled hard liquor sales or a neighborhood with lower bar density is more likely to lead to reduced alcohol consumption among White men, but appeared to have no influence on alcohol consumption among Black men. These findings offer clear evidence of the differential effects of alcohol policies on various gender and racial/ethnic subgroups. Though alcohol taxation and regulation on alcohol sales appear to be effective tools for reducing alcohol consumption among certain sectors of the population, they are not protective for all.