Are you still doing Dry January? Well done if so.
I made it to the Friday after New Year – five days in total. Particular congratulations if you made it further than that – or are still going – and you have two small children at home. That’s what broke me.
January – dry or otherwise – feels like an excellent time to explore longer term trends in alcohol consumption. With just a couple of exceptions, these are moving very slowly. As we always argue in this newsletter, the slow trend is as important as the fast one.
Sloe Trends
There are a lot of slow moving trends here. Here’s three of them, in one very dull chart.
Overall, there has been just the smallest decline in the proportion of people who drink over the last decade.1 In 2011, 83% of people drank, at least occasionally. In 2021 that was 79%. There’s been a slightly more notable (and more consistent) decrease in the proportion drinking beyond the recommended 14 units a week – from about one in four in 2011 (26%) to about one in five in 2021 (21%). In turn, that’s causing the average number of units consumed each week to fall slightly, from 13.4 to 11.6 over the decade. All small changes, nothing dramatic.
Faster Trends
But as is often the case with alcohol, the drama isn’t far away. These next charts show the proportion of people in each age band who say they haven’t drunk alcohol at all in the previous 12 months. First up, men:
The trends are mostly consistent and generally glacial. Every age group, with the noble exception of the over 75s, contained more non-drinkers in 2021 than a decade earlier, but usually only slightly more. Amongst those aged 35-44, for example, the proportion who didn’t drink went from 15% to 19%. A four percentage point rise is the norm for most age groups.
Except the youngest. For the first part of the decade the trend for more abstainers was in tune with older groups – minor increases. But over the past few years it’s surged. Before the pandemic, 26% of 16-24 year old men hadn’t drunk in the past 12 months. In 2021 that was 35%. The proportion of teetotal 16-24 year old men doubled in a decade.
There are similar stories to be found in the same data for women.
Some of the trends in female alcohol consumption are a mirror of those found in the male data. The proportion of 16-24 year female teetotallers has also doubled in the last decade (to 42%), with an acceleration in that trend between 2019 and 2021. Increases among most other groups are much smaller.
But older women aren’t like older men. Women aged over 55 are now more likely to drink than those aged under 45. The 55-64 cohort is the least likely to contain any abstainers. Those aged over 75 are seeing relatively rapid decreases in the number than don’t drink: from 35% in 2011 to 26% in 2021.
Millennial parents take note: next time your parents say they can’t babysit, it’s because they’re off down the pub themselves.
What does this mean?
The rise of young teetotallers is what’s causing the biggest stir in the drinks industry. This was underlined by recent polling by YouGov for the Portman Group on the appeal of low and no alcohol alternatives. The survey found that 39% of 18-24s don’t drink alcohol at all (which is right in line with the HSE data) and that 44% of that age group considered themselves regular or occasional consumers of low or no alcohol alternatives.
This raises significant strategic questions for both the on and off trade. On what occasions are low or no alternatives drinks chosen – by drinkers and teetotallers? How does the population of non-drinkers segment? How will those segments evolve in the future? These each warrant detailed investigation.
If the trends demonstrated by the HSE data are correct – and there’s no reason to think they aren’t – then it also suggests that the forces driving teetotalism are changing too. For a decade, the gradual rise in the number of non-drinkers could be attributed to multiple factors, including rising health consciousness, growth in the number of people in the population who might not drink for cultural or religious reasons, stricter licensing laws (making it harder for those under drinking age to drink) and financial pressures.
The other big thing that happened between 2019 and 2021 is the pandemic. It seems pretty likely to me that that event – with hospitality venues closing for a year – has had a notable impact on the trends, particularly for younger people, who are more likely to drink socially and less likely to drink at home.
If this is a factor it might not behave like health or regulatory contributors to teetotalism. They’re pretty consistent – we’re not likely to see a reversal in our understanding of alcohol’s impact on health and I doubt a liberalisation of drinking laws will be in any party’s general election manifesto. The lockdowns were a one-time thing. It’s possible that this younger cohort’s behaviour changes as they get older. As we’ve seen from the data, that can happen at any age.
The different changes by age group are particularly remarkable, and are helping to keep the overall trends in the number of drinkers and average consumption flat. Or to put it another way, the grandparents of 16-24 year olds are courageously doing their drinking for them.
A longer term view
Alcohol is pretty enduring. It’s tempting to see alcohol consumption as similar to smoking rates – the sort of thing that could decline very rapidly through the generations and become almost taboo in society. I think that is very unlikely, partly because there are safe(r) levels of drinking and partly because it is much more closely intertwined with leisure and socialising.
I can’t find very long term data on alcohol consumption in the UK, but the data below, from Gallup in America, show how remarkably consistent alcohol consumption is in the States. In a country that has seen dramatic social change and turbulence there has been virtually no change in the proportion of people who consume alcohol in over 80 years.
In 1939, 58% of Americans drank alcohol. In 2023, 62%. The more things change, eh?
All data here is from the Health Survey for England. It used to report every year with a one-year time delay, but is still getting back up to speed after pandemic-induced disruption. So the data for 2021 (released in 2022 and 2023) is the latest available.
Fascinating! As you say, it would be interesting to see whether the non-drinkers are still going out, or just staying at home.