Social Media Buttons



What's new?

What's new?
My next beer book is fully funded but there's still time to pledge! Click here for details.
My latest @morningad column: what I did on my holidays
Busy autumn with lots of book events lined up. Check to see if there're any near you.

Thursday, 14 January 2010

Answering the neo-prohibitionists, 8 of 10: “Alcohol advertising/promotion must be tightly regulated because it encourages under-age drinking.”

The HSC Report states:

“The current system of controls on alcohol advertising and promotion is failing the young people it is intended to protect. Both the procedures and the scope need to be strengthened.”

Children and alcohol – the truth

The report presents evidence of an increase in alcohol consumption among teenagers. As usual, they present the data selectively, cutting out anything that does not support their case. So here are some Office of National Statistics numbers about kids and booze. In England, among pupils aged 11 to 15:

  • In 2006, 45% of pupils said they had never had a proper alcoholic drink (a whole drink and not just a sip), an increase compared to 39% in 2001
  • In 2007, 20% reported drinking alcohol in the week prior to interview, down from 26% in 2001

So alcohol consumption among children is falling, not rising as we are regularly told. But even if it was rising (which it isn’t) how strong is the ink between underage drinking and exposure to alcohol advertising?

The link to advertising – or rather, the lack of one

When exploring the link between advertising and underage drinking, the main study the HSC Report refers to is discussed as follows:

“The RAND Corporation’s study, Early Adolescent Exposure to Alcohol Advertising and its Relationship to Underage Drinking (2007) found that children exposed to high levels of alcohol advertising were more likely both to drink and to intend to drink than those with low levels of exposure.”

So I Googled this study and found out two interesting things about it.

The first is that research took place in two schools in South Dakota. How relevant this makes it to UK youth is a question for someone better qualified than me. But even if it was two schools in South Yorkshire, I would still question the validity of applying it to British youth generally.

The second interesting thing is this. This study into two South Dakota schools, which the HSC claims proves a link between exposure to advertising and intention to drink, says the following in its conclusion:

“Causal effects are uncertain”

Let’s go through that again.

Here’s what the HSC says about the RAND research:

“[It] found that children exposed to high levels of alcohol advertising were more likely both to drink and to intend to drink than those with low levels of exposure.”

And here’s what RAND itself says about the RAND research:

“Causal effects are uncertain”

The HSC also relied heavily on a detailed study from Sheffield University. The Sheffield Study is where they get their justification for minimum pricing from. So what does the Sheffield Study say about the dangers of drinks advertising?

It says there is “No case for a ban… the evidence was too weak to come to a conclusion.”

Here are a couple more quotes on the subject:

“In terms of interventions, codes and bans are typically designed to protect young people and any effects of bans on adults remain largely unknown. In practice, only France has introduced a comprehensive ban but no convincing evaluation was carried out.”

There is an ongoing methodological debate on how advertising effects can and should be investigated and there is also a need for further research to establish whether advertising definitely influences consumption.”

Just like the RAND study, no causal effect was established between advertising and under-age drinking. It’s clear from reading the testimony given by the author of the Sheffield Study to the HSC that they set out trying to establish such a link. They are specifically looking for evidence of a link. And they can’t find one.

Awareness versus consumption

In the absence of any evidence of a causal link between exposure to advertising and propensity to drink under-age, the HSC falls back on demonstrating that children are aware of alcohol: the implication being that if they are aware it exists, then by definition they want to drink it. Discussing a survey of 13 year olds:

“Nearly half were found to own alcohol branded merchandise and, when shown masked prompts, the vast majority could name the leading brands: Carling (95%), Smirnoff (93%) and WKD (91%). Research conducted as part of the NPRI study illustrates the extent to which teenagers in the UK are aware of alcohol, alcohol brands and related communications.”

Again, there is no actual evidence to suggest that awareness means propensity to drink. If you’re a Liverpool or a Celtic fan and have a replica shirt, it probably has a drinks logo on it. Of course children are aware of alcohol – their parents drink it. If men drink moderately in a family setting, their children are bound to notice alcohol brands. And far from being a dangerous thing to do, experts advocate such a drinking occasion as being responsible.

Believing who you want to believe

The report bases its conclusions on expert testimony. It listened to testimony from experts in the advertising industry, which said alcohol advertising didn’t tempt children to drink, and it dismissed this testimony.

It then listened to testimony from health experts, who said advertising did tempt children to drink, and accepted this testimony without question.

Without wishing to denigrate the great work health experts do, they are not qualified to judge whether children drink because of advertising or because of something else – this is a completely different field from the one they are trained in. But the HSC don’t let this detail bother them.

The HSC analysed documents from the advertising industry. After doing so, Professor Hastings of the Alcohol Health Alliance found that the documents:

“Reveal major shortcomings in the current self regulatory codes covering alcohol advertising. Specifically, the codes do not, as they are supposed to, protect young people from alcohol advertising; prevent the promotion of drunkenness and excess; or the linking of alcohol with social and sexual success”, and the HSC concludes in his evidences that “The result is a regulatory system that is impossible to police and vulnerable to exploitation.”

Clearly we have to take the ad industry’s testimony with a pinch of salt because they want to make ads. But we also have to take Prof Hastings’ testimony with an equal degree of skepticism.

Prof Hastings is the author of a BMA report which calls for a total ban on alcohol advertising. In response to that report, the Advertising Association commissioned an independent study into the effects of alcohol advertising on children. The full report is here and makes very interesting reading, but here is one extract from it:

Launching the BMA report, its author Gerard Hastings was questioned by journalists on the lack of evidence to support his claims. Mr Hastings insisted that existing ‘cross-sectional’ and ‘longitudinal’ research provided an evidence base for the case against alcohol marketing. Basham & Luik analyse eleven relevant cross-sectional and longitudinal studies of advertising exposure and recall, and conclude that they do not support claims made by the BMA. Not only are they mostly inconclusive, but some have severe methodological problems. Basham & Luik also claim that the results from the case studies tend to be of marginal significance.

Strickland (1982) Alcohol Advertising: orientations and influence, Journal of Advertising 1:307-319 - ‘reducing the amount of advertising for alcoholic beverages is likely to have a negligible impact on the level of consumption among teenagers’.

Chen, M. et al (2005) Alcohol advertising: what makes it attractive to youth? Journal of Health Communication 10: 553-565 – ‘it is possible that alcohol advertising influences young people’s drinking beliefs and behaviours, but the opposite also may be true. That is, young people who are predisposed to drinking may be more attentive to and hold more favourable attitudes toward alcohol advertising’.

The HSC takes Prof Hastings evidence at face value, without question. Yet it makes no mention of an independent, exhaustive study which casts sever doubt on Prof Hasting's methodology and conclusions.

I worked in advertising for 15 years. Contrary to what Prof Hastings implies, the codes covering advertising explicitly forbid appeal to young people (you’re not allowed to shows anyone who looks under 25 in ads), and do not allow the promotion of drinking to excess, or the linking of alcohol with sexual or social success. I have personally tried to argue to get ads made which we as a team felt did not do these things, but were judged by the industry regulator as potentially doing so, and we have lost the argument – every time.

I invite any anti-drink campaigner to create an ad which they think appeals to children, suggests a link between drink and sexual or social success, or promotes excessive drinking, and see if they can get it past the regulators. I would also like them to name examples of ads which have run in the past five years that they believe do any of these things which have not been banned by the regulators.

Short of a total ban on TV advertising, it’s difficult to see how regulation could be any tighter – but of course, a total ad ban is exactly what anti-alcohol campaigners want.

But the HSC ignores evidence of the effectiveness of current regulation, ignores the lack of evidence in support of their own view, and concludes:

“Several of the witnesses to our inquiry disagreed with the drinks industry and regulators. Their argument was that the codes are ineffective because the rules, albeit enforced by the ASA and others are not adequate to protect the young: the quantity rather than the content of advertising has greatest effect. This is particularly true for children. Moreover, advertising has a cumulative effect in the long term—even if sales don’t show immediate response.”

“Several of the witnesses disagreed with the drinks industry?” Did they? Oh well, case closed then.

Then we get the introduction of this claim about it being quantity of advertising rather than quality. There is no factual basis offered to back up this opinion – it emerges out of the blue towards the end of the section on advertising, but nevertheless makes it into the summary as a key finding – a central plank behind the committee’s recommendations for swingeing curbs on alcohol advertising and promotion.

In summary

The HSC says drinking among children is increasing. But recent official figures suggest it is falling.

The HSC simply asserts that advertising encourages young people to drink. But there is no evidence of a causal link, despite people looking very hard to try to find one.

So they imply that there is a link between awareness of alcohol brands and propensity to drink underage, because they can prove awareness. But there’s no evidence of this either.

So after having spent a long time discussing the content of alcohol ads, they then say it’s not the content, but the quantity of it that has an effect. There’s no evidence of this either.

So in the end, they disregard testimony from advertising professionals, and simply choose to believe the testimony of people who want alcohol advertising to be banned, say it is damaging to children, but can produce no evidence to back up their assertion.


Brad said...

Maddening as it is, I love reading your point-by-point takedowns of this nonsense, Pete.

I see correlational data reported all the time by the media (it's the easiest thing to study; finding causation is very hard) but nobody, least of all the media reporting these things, seems to know the difference between correlation and causation.

Correlation could simply be a wild coincidence, it could indicate causation in either direction, or there could be intervening factors. But most of the time, it's never mind all that.

I will say that it seems like, when reporting correlations, the common interpretation uses the "more likely" language ... as in, "children exposed to high levels of alcohol advertising were more likely both to drink and to intend to drink..."

I don't disagree that this is a dangerous, overly simplistic and awfully convenient way of interpreting data that fail to explain any causal mechanisms (by the RAND researchers' own admission, in this case). However, the HSC are hardly alone in taking this "more likely" concept and running with it.

The best that intelligent people can do is to keep pointing out, loudly, when these perversions of statistical interpretation have taken place.

David Strange said...

This HSC report sounds an utterly scandalous work of duplicity. Why is there so little criticism of such mendacious drivel in the mass media?

Cooking Lager said...

When an under age drinker, advertising had an effect on me. Not in my decision to drink, but in the brand I chose to drink.

I'd have still had a pint with my mates, it's a right of passage and part of what our culture considers becomming an adult.

On interest is that the older I get to less advertising appears to influence me. Either that or it has to be a little more sophisticated.

The single best advert for beer was the film Smokey and the Bandit. I always wanted a Coors despite its unavailibility at the time in the UK. I drank Bud, the beer of Smokey and the Bandit 2 instead.

Coxy said...

Everyone who has been around kids these days know they much prefer drugs to booze, and are doing alot more of it than twenty or thirty years ago.
That Hoffmeister bear used to pull a bit , didnt he.

BLTP said...

have you see this , it's a new anti drinking campaign aimed towards parents and kids.

Dave said...

The trouble here is that the HSC want to make a moral point, but can't, and so rely on pretty spurious evidence.

I think the moral point is a good one. Lets take football shirts. I think seeing kids advertising drinks is awful. Not because they're likely to start drinking earlier, or harder but just because it seems wrong. Booze and kids don't mix, it really is as simple as that.

The fact that it might or might not contribute to under-age drinking is besides the point and muddles the issue into now where the argument stands on the quality of evidence in a difficult, under-researched and philosophically problematic area.

In the end, the issue was that the drinks industry knew that it would struggle to say the world needed kids to sport their logos without looking like the biggest pile of poisoners, and so recommended it shouldn't happen knowing they didn't have a moral leg to stand on.

Pivní Filosof said...

This links between ads and drinking/smoking make absolutely no sense really. As Coxy says above, kids do more drugs these days, and drugs dont advertise on the telly. Kids drink or smoke because the stuff is available to them.

I'm not saying that we should make booze even harder to get for them. A friend of mine who grew up in the US told me once that when he was a teenager it was easier for him to buy coke or crack than to buy a beer.

DJ said...

I don't think as a kid Advertising influenced me in my decision to try alcohol, it was much more of a peer thing. I think advertising might have steered me to a particular brand but the biggest influence on what we drank was what we could afford and what we could actually get hold of.

lizvater said...

When I was a kid, I loved the alcohol ads. I can still sing the whole of the Double Diamond Works Wonders jingle, can do a fairly accurate rendition of the Harp: Stays Sharp to the Bottom of the Glass tagline and laughed like a drain when Leonard Rossiter threw Cinzano all over Joan Collins. Yet I have never knowingly drunk, or been tempted to drink, any of them. I loved the Silk Cut print ads but didn't feel coerced into smoking.

I would go on, but I've got some cheap leather furniture to buy so that I can bounce up and down on it playing an air guitar.

Anonymous said...

Pete, do you not think you're becoming a little excessive on this subject?
Your posts are becoming a little repetitive and boring...

Pete Brown said...


Then by all means, please don't read them.

I'm acutely aware that this is not typical beer blogging - I'm staying on one topic longer than I ever normally would, and the posts are longer than I would normally write them.

It's probably tedious for some people to read the whole things through in order. But there are two reasons I'm doing them:

- One - nobody else - with the notable exception of a handful of other bloggers such as Phil Mellows, Pub Curmudgeon and a few others - is taking on an issue which is of vital importance. The industry itself, CAMRA, the BBPA, the trade press, the mainstream press - no one else is fighting back.

- Two, it's a complex issue and it takes a lot of work to extract the facts. The idea behind the way I've labelled the posts, numbered them etc, is so that the series of posts becomes a useful resource for anyone who might want to counter a point such as 'binge drinking is increasing' in their own writing - they can come here and get the facts they need. I'm thinking of them more as a reference thing than a narrative to be read from beginning to end.

At the start, I outlined that there were ten points that needed addressing - having done that, I can't really give up halfway through.

But I feel your pain - don't worry, when no.10 is up, I'll get back to talking about the usual shit.

azulbuho said...

Generally excellent work here, Pete. But I feel this section is a little flimsy.

Absence of evidence (re. effect of advertising on drinking) is not evidence of absence. It was wrong of the HSC to claim evidence of a causal link if none has been established, but I wouldn't have any problem whatsoever with banning alcohol ads anyway just in case. We wouldn't be losing out on anything worthwhile, would we?

I should probably admit that I hate all advertising and would like to see it all banned, but that's another argument for another day.