Alcohol and the irrelevance of inequality

Almost every day, we’re fed another piece of inequality in the news – another example of the gap between the advantaged and the disadvantaged, another case where better-off parents have secured the successes of their children over the less well-off. Coming home from a fabulous BSA event on inequality last week, I felt flooded with inequalities and injustices of different forms from different countries, which seemed to form an unpickable web of unfairness.

Which is why, sometimes, we need to remind ourselves that not everything in life is linked to generalised advantage. One example is that higher-income women (unlike men) are less likely to have a partner, as Diederik has discussed on the blog.  But closer to my own experience, the examples I know best are about the very British behaviour of binge-drinking, which is the subject of this post…

Binge-drinking across society

If you pay attention to the popular press in the UK – particularly the Daily Mail and the Daily Express – then you’d be forgiven for thinking that binge-drinking was dominated by poorer people, the ‘chavs’ of the public imagination. But these aren’t the people who binge-drink the most. The chart below shows the picture using the latest official data, and defining binge-drinking as having more than double the daily recommended limits of 3 standard drinks for women and 4 standard drinks for men. (A standard drink is roughly one shot of spirits or a half-pint of beer). For both men and women, greater household income goes along with greater chances of binge-drinking – although the relative increase is greater among women.

Binge-drinking by income in the UK 2009, taken from the General Lifestyles Survey

According to Wilkinson & Pickett’s The Spirit Level, it’s only for strongly class-patterned negative outcomes that we’d expect to see greater problems in more unequal countries. And this seems to be the case here – there’s simply no link between most measures of drinking and societal income inequality as measured by the Gini coefficient. This holds true for drunkenness among 11-15 year olds and overall levels of adult consumption (as argued by the interesting if sometimes flawed critique of The Spirit Level by Peter Saunders; see page 105) – and even this unusual comparison of villages in a pre-modern society. More informally, we know that relative rates of drunkenness (if not alcohol consumption per se) are high in the more equal Nordic countries, while the high-inequality UK and the US have completely different drinking cultures to each other.

The reappearance of inequality

Yet this doesn’t mean that alcohol and inequality researchers should go their separate ways – for there are three ways in which social advantage does make a reappearance.  Firstly, the UK is pretty unusual among high-income countries in showing that higher-income, higher-class men binge-drink more than others. While similar results are found in low—and middle-income countries, in most European countries it is low-educated men that are likely to drink the most. The picture for women is even more culturally variable; better-educated women are often heavier drinkers per se but it is unclear how this links to getting drunk. The relationship of gender, class, alcohol and societal factors is complex here, if enormously interesting.

Secondly, even in the UK, moderate drinking is much much more common in higher-income, higher-class people, reflecting the near-universal finding that people with more money are more likely to drink, and (in some countries) to drink more often. In the UK, 74% of people in households earning £1,000/wk or more drank last week, compared to only 46% in households on less than £200/wk. Among men, the chances of binge-drinking on any given drinking occasion is nearly always higher among the lower-educated. So even if binge-drinking is not heavily patterned by social advantage, other patterns of drinking are.

Finally, even though risky drinking is found across British society, the worst alcohol-related problems are much more common among disadvantaged groups – at least among men. The figure below (from Harrison & Gardiner 1999) covers deaths directly due to drinking, and men show a steep gradient across classes I-IV and an even steeper increase in risk for the lowest class. For women, though, the picture is more complex – if anything there is a reverse tendency where higher-class women are more likely to die from alcohol, with the exception of the highest rate being found in class III-M (of manual workers).

Alcohol deaths by social class in England, taken from Harrison & Gardiner 1999

A further sign of the class-linked nature of alcohol deaths comes from Finland, where a recent drop in alcohol prices led to increases in alcohol-related mortality, which was larger – if only slightly larger – among those with low education, income or social class.

What it all means…

So alcohol is both an example of where inequality doesn’t matter, and (in other ways) also where it does – but to my mind the real significance goes beyond this. Drinking is a behaviour that is both positive and negative; it is at once a display of status, a use of scarce resources, and a pleasurable, social experience – and at the same time a way of coping with distress, a risk factor for more than 60 diseases and one of the most important causes of ill-health globally, and a cause of violence, addiction and spiralling social exclusion.

Alcohol’s relationship with the mechanisms of inequality are therefore complex – with different patterns of drinking and domains of harm showing different relationships to different domains of advantage. Moreover, alcohol is so tightly bound up with cultural factors that these processes work in different ways in different settings. Yet this is what offers the potential to really uncover something at a deep level about how inequality works – how inequality is played out for a behaviour that is finely balanced on the knife-edge between pleasure and problems. My hope is that researchers will return to unpicking the delicate mesh of inequality, not just for figuring out how to reduce alcohol problems, but as a way of getting to the heart of the deeper working of society.

(Further information can be found in the EU Alcohol in Europe report I co-authored with Peter Anderson back in 2006, particularly chapters 4-6.  If you’ve seen something post-2006 that I haven’t linked to, though, then please let me know!).

About Ben Baumberg Geiger

I am a Senior Lecturer in Sociology and Social Policy at the School of Social Policy, Sociology and Social Research (SSPSSR) at the University of Kent. I also helped set up the collaborative research blog Inequalities, where I write articles and short blog posts. I have a wide range of research interests, at the moment focusing on disability, the workplace, inequality, deservingness and the future of the benefits system, and the relationship between evidence and policy. You can find out more about me at
This entry was posted in Articles and tagged , , . Bookmark the permalink.

10 Responses to Alcohol and the irrelevance of inequality

  1. Hi Ben,

    Once again, I’ve learned a ton from your blog post. In the United States the relationship between binge drinking and income is also positive (, although I’m not sure how much of this is explained by the independent effect of race on binge drinking (whites more likely to binge).

    A couple of questions:
    -I just wanted to mention a consistent finding from the US epidemiological literature: there is a positive association between income and DSM-defined alcohol abuse, but a negative association between income and alcohol dependence (

    Does the same relationship hold in the UK?

    -In the United States lower income people are much more likely to report unmet treatment need, but then access to private treatment is spotty among these populations, and the publicly funded treatment system does not consistently deliver comprehensive care. What does alcohol treatment access look like in Britain, and is it patterned at all by class? (I assume NHS is much more comprehensive than US health system).

    Finally, apropos of nothing really, I can’t help mentioning Martin McKee’s interesting work on alcohol attributable mortality in post-Soviet Russia, one of the most sobering (pun intended) illustrations of how large scale economic dislocation can cause major alcohol-related health problems, especially among lower classes. (

    • Ben Baumberg says:

      Thanks Brendan – one day we should really try and do some comparative work on this, given our mutual interests! Anyway – the picture on alcohol abuse/dependence in the UK is similar but slightly different. The main survey on this from 2007 (see Table 9.8) show that ‘mild dependence’ is much higher in those with the highest income, just like alcohol abuse in the US. It’s hard to look in detail at ‘severe dependence’ due to small numbers though, and ‘moderate dependence’ for men shows an erratic pattern; the APMS report says that ‘there was no significant variation in the prevalence of alcohol dependence by income’ (p156).

      BUT – and it’s a big but – interpreting these figures is difficult, as this isn’t DSM-defined, it’s using the ‘The Severity of Alcohol Dependence Questionnaire (SADQ-C)’. You can get ‘mild dependence’ if you drink 2-4 times/mth, have 3-4 drinks on each occasion, and have 6+ drinks less than monthly, together with no other symptoms – which seems a low threshold if you ask me! You could exclude these questions, but I haven’t seen this done; the survey also includes the AUDIT questionnaire, but again I haven’t seen any splits by income. So all-in-all, there’s some simple re-analysis here that would be really useful.

      Not sure about the treatment question – my guess is that alcohol treatment is dominated by lower-income groups, but who still have greater levels of (self-defined) unmet need as its not as easily available as it should be. There was a 2004 official ‘alcohol needs assessment’ for treatment which is interesting, but I don’t think covers income.

      And I agree: Martin McKee’s work is amazing but deeply, deeply depressing…

  2. Wonderful post; I love when things don’t map out neatly onto a linear just-so kind of narrative (there’s the historian in me).

    To this post I would just add that one subgroup for whom structural injustices and devastating oppression seem to be deeply, profoundly linked with high levels of alcohol use is in Native Americans/Alaska Natives. So this supports the second part of your past indicating that among some groups, there does seem to be some significant positive relationship between social inequalities and alcohol use. If one knows anything about what we (Americans) have done and are still doing to Native Americans/Alaska Natives, it is hard to be surprised by the connection to alcohol use, although your post certainly suggests reasons to be cautious in generally inferring any kind of just-so implication.

    I am less familiar with the data for First Peoples in Canada, Latin America, and Australia, but I’d be interested in seeing how that all shakes out as well.

    • Ben Baumberg says:

      A really good point, and a welcome correction to a dimension I completely missed out of the original post.

      Having previously focused more on alcohol in Europe I don’t know the situation for First Peoples in the US/Australia/elsewhere – but from what I’ve picked up, they tend to have massive alcohol problems. It would be really interesting to know why this is the case (my guess is that it’s a mix of the historical (very sudden access to cheap alcohol, possibly supplied by colonial powers) and structural reasons (poverty, exclusion etc), but I’d welcome a more knowledgeable response…)

  3. Pingback: Health Nexus Today / Nexus Santé aujourd'hui » Blog Archive » Health Promotion Headlines from Robyn and Meghan – May 16, 2011

  4. I would think since it is something one is born with and not a choice then it would be the same as alcoholism where it is not a choice but treatment is
    available if one wishes to get treatment.
    Why is the term offensive, just curious. Also if a gay or lesbian person really wants to become heterosexual are
    they able to through counseling? I know the forced counseling often doesn’t work and can be seen as very offensive but I am wondering for those who are gay/lesbian but don’t
    want to be anymore. . I am not gay myself but know a couple that is and
    am just curious. As a kid I would make offensive comments that I know regret
    about gay/lesbian people. It was a prejudice and immature attitude.
    While I don’t believe in gay marriage for religious reasons I do believe in all other rights for gays and lesbians. . So just curious the views on why mental illness is seen as an offensive term for homosexuality. I often think of it as like alcoholism except that alcoholism causes hurt to families even though it is never in

    • Eliot Rosewater says:

      Non-heterosexual orientation is not considered a mental illness since it is not deleterious to the person in question, nor to society at large. It is not a choice, but it is also not something that needs to be corrected.

      Quite apart from the lingering homophobia that I’m afraid still colours your opinion, your post is also completely irrelevant to the topic at hand.

      • Ben Baumberg says:

        As Eliot says, this is getting off-topic, this isn’t the kind of debate we’re aiming for on the site. I haven’t deleted the comments as I don’t like doing this on the site, but please try to keep everything relevant to the site, if we have any more debate on this then I’ll start removing the comments…

        Thanks both for being engaged with the blog though!

  5. Ben Baumberg says:

    Just to flag that the link to the data underlying the graph above is broken – the GHS/GLS 2009 data can now be found at

  6. Nicki Keating says:

    Interesting post – only reading it now. I think your point “Among men, the chances of binge-drinking on any given drinking occasion is nearly always higher among the lower-educated” is the key – there is very much a social gradient among drinkers but when one tries to homogenise a total population (whereby those of low SES are less likely to drink but have more risky drinking patterns when they drink) results in the dilution of the SES relationship. So should the analyses be presented as among drinkers and among the total population to reflect this heterogeneity? I think presenting the total population figures is somewhat misleading of the inequalities relationship. Thoughts?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.