Alcohol consumption in individuals aged 15 and over, calculated from recorded sales and unrecorded data, was estimated at 10 litres of pure alcohol per person in 2018 – this is equivalent to two bottles of wine, or nearly 4 litres of beer, per week per inhabitant in OECD countries. This level of consumption has shown little variation over the last decade. However, this apparent stability masks significant variations both across countries and, within the same country, across different population groups:

• Harmful drinking patterns are highly concentrated. Prior to the pandemic, nearly one in three adults in OECD countries had engaged in binge drinking at least once in the previous month; this corresponds to drinking more than 80% of a bottle of wine, or 1.5 litres of beer on a single occasion. In addition, alcohol is disproportionately consumed by a minority: people who drink heavily make up 4% to 14% of the population, depending on the country, but they consume between a third and a half of all alcohol consumed, according to an analysis of six OECD countries.

• Harmful alcohol drinking among young adults is widespread. More than 60% of teenagers aged 15 drink alcohol and one in five has already experienced drunkenness at least twice. Younger generations are less likely than a decade ago to have experienced drunkenness. Those who have never experienced drunkenness are 30% more likely to perform well at school.

• Significant inequalities exist in patterns of alcohol consumption. Certain population groups, such as women with high levels of education and individuals at the two extremes of the income distribution, are more likely to binge drink.

• COVID‑19 lockdowns have affected both drinking patterns and alcohol sales. Survey data suggest that a higher share of population increased their alcohol consumption and frequency of drinking. Government data from a few countries suggest a 3% to 5% increase in alcohol sales. At the same time, however, a slightly higher share of the population reported a decrease in binge drinking. People also changed the place of consumption: alcohol sales in on-licence premises (e.g. bars, pubs and restaurants) plummeted during lockdowns, while off-premise sales – and in particular online sales – grew significantly.

According to OECD simulations, life expectancy is estimated to be almost a year (0.9) lower in the period 2020‑50 due to medical conditions caused by drinking more than 1 drink per day for women and 1.5 drinks per day for men, which corresponds to a lower-risk threshold because at these levels alcohol may have some protective effects on specific diseases such as ischaemic cardiovascular diseases and diabetes for some age groups. In the same period, diseases and injuries caused by drinking above 1/1.5 drinks per day will cause 1.1 million premature deaths in OECD, European Union (EU27) and Group of 20 (G20) countries. Harmful alcohol consumption is also the cause of injuries such as road traffic crashes and interpersonal violence, foetal alcohol spectrum disorders and chronic diseases such as alcohol dependence, cancers and liver cirrhosis.

Drinking more than 1/1.5 drinks per day leads to additional costs for the health system. These account, on average, for 87% of all treatment costs for dependence, 35% of treatment costs for cirrhosis and a significant share of treatment costs for injuries, cancers and other diseases. According to simulations, on average over 2020‑50 about 2.4% of annual health expenditure will be devoted to treating the diseases caused by drinking above this level. A total of USD 138 billion, adjusted for differences in purchasing power, will be spent annually on treating these diseases across the 52 countries included in the analysis (i.e. OECD, EU27 and G20 countries). This is equivalent to, for instance, the current health spending in Australia or more than twice the current health spending in Belgium.

Adolescents are particularly affected by harmful drinking. Adolescents who have experienced drunkenness show lower life satisfaction; they are up to twice as likely to bully their classmates; and they have lower performance at school, although the causal link cannot be asserted. Students with harmful drinking patterns are less likely to complete higher education, particularly in the case of girls. Lower education outcomes affect the formation of human capital, economic growth and social welfare, and worsen inequalities.

According to OECD simulations, diseases caused by drinking more than 1/1.5 drinks per day will lower participation in the labour market and damage labour productivity over the period 2020‑50, reducing the workforce by an equivalent of 33 million full-time workers per year across the 52 countries analysed, or the equivalent of 0.62% of the total workforce on average across countries.

Combining the effects on life expectancy, health expenditure, employment and productivity, GDP could be 1.6% lower over the period 2020‑50 in OECD countries due to diseases caused by drinking more than 1/1.5 drinks per day, according to simulations. To cover the increased fiscal pressure caused by alcohol-related diseases, each person pays an additional USD 232 in taxes (adjusted for differences across countries in purchasing power) per year in OECD countries.

Countries’ policy responses to tackling harmful alcohol use can be improved by using the most effective intervention in each policy domain and by extending coverage. Too often, the implementation of policies “on the ground” and their effectiveness at the population level is hindered by poor implementation, limited resources or practical problems.

Tackling harmful alcohol consumption requires a combination of policy measures. Those can operate both at the population level (such as communication campaigns and pricing policies) and at the individual level, directly targeting individuals consuming high quantity of alcohol (such as drink-driving policies, counselling in primary care and personalised pharmacological treatment for dependence). All these interventions are effective and cost-effective, but combining them into coherent policy packages offers higher results than implementing single interventions in isolation.

For example, a package of policies built around the “PPPP approach”, including interventions such as limiting the promotion of alcohol to children, better police enforcement to prevent alcohol-related road traffic injuries, upscaling coverage of primary care counselling for patients with harmful alcohol consumption and pricing policies to limit the affordability of alcohol – particularly cheap alcohol, would:

• save 4.6 million life years annually across 48 countries; which broadly corresponds to, for instance, the total life years lost due to lung cancer in the United States each year, or the total life years lost due to cardiovascular diseases in Germany;

• save about USD 28 billion annually (adjusted for purchasing power) in health expenditure, broadly equivalent to 0.5% of health spending, which is equivalent to, for instance, the current health spending in Israel or half the current health spending in Sweden;

• generate savings that are greater than the implementation costs – for every USD 1 invested in a comprehensive policy package, up to USD 16 are returned in economic benefits.

Alcohol production and trade represent an important part of the economy in some countries. While alcohol industry revenues may be affected by policy measures, in either profitable or unprofitable ways, countermeasures exist to minimise additional costs. Comprehensive, well-designed policy packages associated with approaches to reduce consequences for the alcohol industry can get the expected health gains and limit any impact on the industry and related businesses.

While cost-effective, and with an excellent return on investment, measures to tackle harmful alcohol consumption will always require complex trade‑offs – for instance, regarding their impact on the economy and the labour market, as well as which type of consumer the policy aims to target. For example, interventions targeting all consumers are highly effective and efficient, but they affect those to drink at low to moderate levels, as well as those who consume alcohol heavily. On the other hand, interventions targeting only people who engage in risky drinking have a significant short-term to medium-term impact on those people, but they have a lower impact at the population level and higher implementation costs.

Ultimately, it is up to each country to consider the most appropriate mix of policies to implement to address such trade‑offs. This report provides detailed estimates of policy impacts both at the population level and in specific subgroups, thereby supporting an evidence‑based approach to such decisions.