Consumption
The consumption of Alcoholic Beverages, or ABVs causes a variety of diseases in consumers. It is associated with numerous health problems such as chronic diseases (liver cirrhosis, pancreatitis, various cancers, including liver, mouth, throat, larynx, and esophagus; high blood pressure; and psychological disorders). It is also associated with unintentional injuries (motor-vehicle traffic crashes, falls, drowning, burns, and firearm injuries), social phenomenon (violence, such as child maltreatment, homicide, and suicide), harm to a developing fetus if a woman drinks while pregnant, sudden infant death syndrome (SIDS) and finally, a variety of alcohol use disorders.[1]
The public finance dimension
Governments have had historically imposed several taxes on products which are regularly consumed by a large proportion of the population and possess significant addictive properties. From a normative point of view the rationale to impose these kinds of levies is usually related to the objective of preventing the consumption of goods and services that have negative individual and societal externalities by raising their price through taxes. Demand of those goods has inelastic (low sensitivity to price changes) properties, which open the opportunity of imposing high tax rates and public revenues.
Impact on future generations
Mounting negative impacts have the potential to increase national health systems’ demands for more resources. For example, in a recent comparative study for the period between 2020 and 2050, some countries included in this study will require an additional $14 to $76 USD per capita annually in purchasing power parities (PPPs) to cover the fiscal pressure caused by diseases from alcohol consumption, a similar figure to some advanced Asian and European G20 countries (OECD, 2021). The magnitude of the demand for resources is particularly relevant considering LAC countries’ tax raising capacity is lower than the Organization of Economic Co operation and Development (OECD) countries: from 22.9 to 33.4% tax to GDP ratio averages (OECD, 2022).
Huge disparities among countries
Consumption heterogeneity among regions and countries is an important aspect to consider when thinking of excise taxes for ABVs. In the table below, alcohol consumption per capita in adults is displayed. Magnitudes and historical patterns considerably differ among the 12 countries included in this research. There is no regional or subregional aggregation possibility since countries with relatively low consumption belong to different regions (for example, Guatemala and Ecuador). There is though, a potential for identifying three clusters: high (Barbados, Argentina, and Chile), medium (Brazil, Peru, Trinidad and Tobago, Belize, Colombia, and Mexico), and low (Honduras, Ecuador, and Guatemala) consumption.
Chart 1. Alcohol consumption per capita (+ 15 years, liters of pure alcohol 2000-2019).
Source: Own elaboration with data from WHO Global Information System on Alcohol and Health (GISAH).
Consumption patterns
The research on alcoholic beverages brings into the discussion of excise taxes the importance of distinguishing between moderate versus hard users, which underscores the importance of discriminating the leisure and addiction component in tax design (OECD, 2021). The proportion of heavy drinkers’ consumption of total ABVs pure alcohol is and indirect indicator of the magnitude of the ABVs challenge. Heavy drinking should not be taken as a behavior like “binge” or explosive drinking. The first is a regular and cumulative action and the latter an episodic one.[2] For example, in a recent research report for England, a group studied the ABVs industry to find out what was heavy drinkers’ contribution to the profits of companies. These represent 4% of the population yet they account for 30% of total consumption and 23% of the industry revenue. For 2013 and 2014, 77% of ABVs were sold to drinkers consuming above guideline levels: 30% to harmful drinkers and 48% to hazardous drinkers (Bhattacharya, A., et. al., 2018). Prevalence of heavy drinking is displayed in the following graph for the 12 countries considered. Two regional comparative averages are included (OECD, G20). An important aspect to consider is that some LAC countries have similar indicators (Peru) than advanced countries (United States). Intuitively, advanced regions or countries count with more resources from a fiscal and public health standpoint, so it seems plausible that LAC countries, and especially those with a high percentage of adult population with a heavy drinking problem, face an important challenge.
Chart 2. Prevalence of heavy episodic drinking, Percentage of adult population (aged 15+) with at least one occasion of heavy episodic drinking in the past 30 days, 2016 [i]
Source: Own elaboration with data from OECD, 2021.
Gender matters
Differences are also relevant when addressing alcohol beverage consumption, in terms of gender gap. As in the above graph (heavy drinkers) as in the graph below (consumption for drinkers only), the differences among the 12 countries included in this research and other regions and economies are substantial. Extreme cases are Saudi Arabia and South Africa where the proportion of women alcohol consumption are both at the extremes of the sample. Again, as in the case of Peru for heavy drinkers, in this case Brazil comes close to countries which are often associated with high ABVs consumption such as Latvia and other countries situated contiguous to the Baltic Sea (Latvia and Lithuania). Yet, most of the countries in the LAC region are quite similar to the G20 and even higher than the United States and Canada.
Chart 3. Women alcohol consumption. Total per capita (15+) alcohol consumption (in liters of pure alcohol) for drinkers only, 2016
Source: Own elaboration with data from OECD analysis of WHO (2020) GISAH data, https://www.who.int/data/gho/data/themes/global-information-system-on-alcohol-and-health.
An additional indicator to understand the magnitude of the harmful use of alcohol is the share of dependent drinkers of total population. This intends to disentangle individuals who drink ABVs based on leisure from those addicted. The gender gap, which means the distance between men and woman is very different in the countries where data is available. The gender gap in Chile is by far the highest compared with Mexico and Brazil. Note that Chile resembles more countries like Canada or the OECD average.
Chart 4. Share of dependent drinkers (+15 years), by sex, 2016
Source: Own elaboration with data from OECD Health at a Glance 2019. The term 'alcohol dependence' has replaced 'alcoholism’ and imply that individuals are unable to internalize the idea of cure and disease, relapse and are recurrent drinkers.
Academic & Policy Research
Research on alcoholic beverages harmful consumption is mostly produced by people and organizations directly or indirectly associated with the health sector. Researchers are often medical doctors or people specialized in the health sector which build research cases from their studies and experience. They use an ample variety of perspectives and disciplines which are familiar to the public health community. Within the health sector, alcohol beverages harmful consumption is often labeled as alcohol use disorder, or AUD. AUD is a health condition resulting from unrestrained and harmful drinking.
There is another branch or community of scholars who are interested in knowledge agendas which are mostly interested in the policy dimension, and more specifically the economic and fiscal substratum of AUD. In the present document, our aim is to bring both the core tradition of public health perspective and the policy and fiscal school of research together, since we consider both as complementary. Although pertinent to the purpose of this research, the literature review included in this section cannot be considered exhaustive.[3] In our view is an updated and proper sample of some of the most important research pieces on AUD and the harmful consumption of alcoholic beverages that bridges the gap between both the health and policy-fiscal domains.
Since 2014, the Mexican Community of Public Management for Results, A.C. (hereinafter Mexican Community), has produced pioneering research on a variety of fiscal and economic dimensions of alcohol beverage consumption, both at national and subnational level. It has also produced a variety of international, global comparisons on a variety of types of beverages.[4] Thus, our research experience and accumulation of knowledge is crucial in the design and development of our latest report, which uses artificial intelligence extensively.
The new frontier: Machine Learning
A sizeable number of individuals consume alcohol for leisure, yet in some countries and regions this phenomenon leads to a harmful pattern which has a direct impact on people’s health, their surroundings, health systems, and, ultimately, obligate governments to tackle its negative and generalized impact. It is crucial to understand the ecosystem in which such consumption takes place by differentiating types of beverages (fermented and distilled). Harmful alcohol consumption is associated with the preferred type of beverages rather than alcoholic beverages or pure alcohol consumption in general. Since we believe that the policy dimension is key and governments, along with other organizations, have the mandate and obligation to tackle harmful alcohol consumption, countries should be the basis of the unit of analysis.
A novel way to understand the environment in which harmful consumption of alcohol around the world is embedded is presented in our latest study. Differentiation refers to beverage strength (in terms of pure alcohol content) and their gender impact (female and male consumption). We use a variety of health data and indicators to incorporate the health dimension and social impact in our model as well (disorders and road injuries). We cluster or group countries by their preferred type of beverage, gender, age, health and social indicators and thus place countries by their probability, or likelihood, to belong to a cluster.
Mexican Community proposes an ecosystem analysis based on differentiation (beverage strength and gender) which uses a variety of machine learning methods as practical and useful tools to build global policy inferences. This analysis aims to find affinities, generate identities, assemble associations, and open new spaces to rethink global policies. We aim to provide harmful alcohol consumption phenomenon’s stakeholders with global policy guidelines.
Following we include the results of our model by identifying the clusters for the 3 countries included in other study which will be released in 2025, “Behavioral strategies to reduce harmful alcohol in Mexico: a comparative approach with Germany and the United Kingdom”.
Chart 5. Probabilities
Source: Farfán Mares, G., Esteban Bruera y Pedro Torres, Building global policy inferences for alcoholic beverages: A cluster analysis & machine learning approach, 2024.
[1] For a comprehensive review of diseases see https://www.cdc.gov/alcohol/faqs.htm
[2] Whether both types of consumption are represented by different age groups (or gender) is subject to question.
[3] This section only mentions or cite the research pieces which are most related to our research. An extensive list of references is included at the corresponding section at the end of this paper.
[4] At the end of this report our published research in alcoholic beverages is listed chronologically. Full documents can be accessed on the following link: https://www.comunidadmexicana.org.mx/impuestos-bebidas-alcoholicas
[i] Heavy episodic drinking is defined as consuming at least 60 grams or more of pure alcohol. Percentages in the top (Men) and percentages at the bottom (Women).
Comments