Fewer young US adults reported drinking alcohol in 2021–2023 than two decades earlier. Gallup’s comparison found the share of drinkers aged 18–34 fell from 72% to 62%. These data do not establish that all of Generation Z drinks less than every previous generation.

Age groups were compared across three-year periods
Gallup’s analysis, published on 22 August 2023, combined surveys from 2001–2003, 2011–2013 and 2021–2023. Respondents were asked whether they ever drank alcoholic beverages. The 62% therefore measures the share who drink, rather than the volume consumed.
In 2021–2023, the 18–34 group included both older Generation Z members and millennials. Comparing the same age group at different times does not describe only one birth generation’s habits.
| Age group | 2001–2003 | 2021–2023 |
|---|---|---|
| Aged 18–34 | 72% | 62% |
| Aged 55 and over | 49% | 59% |
Older adults followed a different trend
In the same analysis, the share of drinkers aged at least 55 rose from 49% to 59%. Among 35–54-year-olds, it changed from 67% to 69%. This shows why an overall average or a single generation label is insufficient.
The 2026 survey offers a more recent overall picture
Gallup’s results published on 20 August 2026 found 54% of US adults said they drank alcohol. This matched 2025 and was the lowest figure in Gallup’s series dating back to 1939.
Gallup said drinking participation had fallen across all major age groups since 2023, although changes over the latest year were not statistically significant within those groups. The earlier young-adult comparison and the 2026 all-adult figure therefore answer different questions.
Is cannabis the reason?
These percentages cannot establish how much change was caused by cannabis, health concerns, prices or social habits. A trend and its causes require separate investigation. Specific earlier claims about social media, mental health and cannabis substitution were not sufficiently supported by the cited source.
Also read our comparison of cannabis and alcohol health risks (in Estonian).
Our overview of the Canadian beer-sales study also examines possible substitution. Surveys and sales statistics do not support one universal conclusion about everyone’s habits.
