A Public Display of BMA Ignorance

By Patrick Basham and John Luik
This article appeared on Spiked (Online) on November 23, 2010.

We can only hope, for the sake of patients, that the medical skills of doctors making policy at the British Medical Association (BMA) are more finely honed than their analytical skills.

The reason for our concern is the BMA’s reaction to the publication of two new studies looking at the effects of bans on tobacco displays in shops. On Friday, the BMA trumpeted these studies as proof that, after Ireland’s tobacco displays were removed in July 2009, ‘fewer young people believed smoking was widespread among their peers’ and tobacco retailers were not financially harmed. Hence, the BMA has called on the UK government fully to implement the existing legislation to ban displays in England.

The BMA intervention raises several questions. For example, proponents usually argue that the sole aim of tobacco display bans is to reduce youth smoking. So why does the BMA’s statement suggest that other goals are being promoted, too? Why is Ireland now held up as a tobacco control nirvana? Most importantly, why does the BMA ignore a plethora of contrary economic evidence on the effects of such bans?

Until very recently, tobacco-control advocates campaigned for a display ban because it would lead to reduced youth smoking, full stop. But the evidence from various jurisdictions that have implemented a display ban suggests that smoking prevalence, especially among adolescents, is at best unaffected by such a ban. Indeed, there is evidence in some places that display bans have coincided with an increase in smoking.

Consequently, ban advocates are quietly and subtly moving the empirical goalposts. They are replacing youth smoking levels as the test of success with a measurement of how many young people perceive that their peers are smoking and then propagating a lower score as ‘evidence’ of the display ban’s effectiveness. It is an intellectually dubious tactic, but left unchallenged it may do the trick, politically. Hence, the new Tobacco Control article by McNeill et al finds, ‘The proportion of youths believing more than a fifth of children their age smoked decreased from 62 per cent to 48 per cent’.

Curiously, the BMA’s press release chose not to highlight the very same article’s finding that: ‘There were no short-term significant changes in prevalence among youths or adults.’ In other words, no one in Ireland has stopped smoking because of the ban: the policy is a failure.

That finding is in line with all of the available evidence. For example, Ireland’s Office of Tobacco Control, which collects the official government data, recently reported that smoking ‘rates in the… youngest age group, 15-17 year olds, did not decline’ between June 2008 and June 2010. Also overlooked by the BMA is the fact that, over the same period, ‘the proportion of occasional smokers has increased by almost two per cent’.

Smoking prevalence data is also available from the EU in its Eurobarometer Survey: 72.3 Tobacco, released in May 2010, and in the report, Smoking Patterns in Ireland: Implications for Policy and Services, 2007, prepared by the Division of Population Health Sciences at the Royal College of Surgeons in Ireland and commissioned by the Irish government’s Department of Health and Children.

Comparing the prevalence of smoking post-ban with prevalence reported in Smoking Patterns in Ireland, it is clear that there has been no post-ban decrease in prevalence. As Professor Luke Clancy, director general of the Research Institute for a Tobacco Free Society, told the Irish Independent about the Eurobarometer figures: ‘There is no evidence of any decline in smoking in this survey.’

So, the claim that simply seeing cigarettes displayed leads to adolescents perceiving cigarettes as easier to obtain — and therefore more appealing to smoke — is very dubious indeed. In fact, very few adolescents obtain their tobacco at a retail venue, so tobacco displays cannot have the effect attributed to them.

As both Emery et al and Harrison et al report, only about five per cent of young people purchase their cigarettes from retail venues. Given that the overwhelming majority of adolescents obtain their tobacco from friends or family, their beliefs about the retail accessibility of tobacco are irrelevant to the decision to smoke. This accounts for the fact that in Ireland, as elsewhere, there is little connection between youth smoking prevalence and the ease of access to tobacco products.

If perceived ease of access to tobacco were a cause of increased smoking, then one would expect to find reduced adolescent smoking where access to tobacco was more tightly controlled. But as Castrucci et al found, strict access does not reduce youth prevalence but simply changes the source of tobacco. As Croghan et al note, despite the widespread restrictions on accessibility in California, the prevalence of smoking was largely unaffected. In short, the argument that tobacco displays cause adolescent smoking through affecting adolescents’ perceptions about the accessibility of tobacco is contradicted by the empirical evidence.

Today, Ireland is the new Canada of tobacco control. For many years, Canada was the public health establishment’s poster child for tobacco control. In tandem, Canadian politicians, public-health bureaucrats and tobacco-control activists led the rest of the world in advocating, and often passing and implementing, the regulations and laws that the tobacco-control movement considers conventional public health wisdom. The basic message was we should all be more like Canada.

Until now. The problem with Canada today is that the evidence on the consequences of display bans — both in terms of smoking prevalence and economic fallout for tobacco retailers — is extremely unhelpful to the pro-ban side (as one of us recently documented for the Institute of Economic Affairs, and as we discuss in a forthcoming Economic Affairs journal article). Hence, the pro-ban side now says the Canadian experience is not applicable to the British one, or any other for that matter.

So, if Ireland did not exist, it would have to be invented; display ban advocates desperately need a new, go-to country case study to justify their ideas. Witness Quinn et al’s dismissal in Tobacco Control of the contrary economic evidence we have put forward, stating that our ‘findings contradict recent reports that sales… are being lost since the removal of… displays and second, that this decline can be attributed to the [display ban] policy’.

In truth, the display ban’s disastrous economic consequences are very real and far more powerful in evidentiary terms than the preliminary sales figures relied upon in the Quinn study. Many small retailers rely on sales of tobacco — and the sales of other items to tobacco purchasers — to keep their heads above water. Tobacco display bans will not only fail to reduce smoking, but such restrictions have already put many retailers out of business.

For example, accounting firm PriceWaterhouseCoopers and HEC Montreal, a Canadian university, report that bankruptcies in the Canadian independent retail sector are at a record level. In 2008, with tobacco display bans either in force or due to come into force in all but one of Canada’s provinces, a record 2,300 corner shops shuttered their doors — that’s one out of every seven shops.

Moreover, the economic hardship caused by display bans goes beyond simply lost tobacco sales revenue. According to the PWC/HEC study, there is a loss in daily traffic of some 1.06million store visits with CAD$2 billion in lost sales of other merchandise, as smokers take their business elsewhere.

Based upon our comparative statistical analysis of the respective Canadian and UK tobacco retail sectors, were the Canadian experience to be replicated in the UK, we calculate this would translate into approximately 10,500 shop closings by the end of the first year.

Revealingly, Quinn et al’s Tobacco Control article does not question the economic realities facing Canadian tobacco retailers. Instead, the authors postulate that such a decline is ‘almost surely a function of’ such phenomena as the global recession. Strangely, their analysis omits the critical facts that the steep decline in the Canadian independent retail sector coincided with the advent of a display ban and predated the onset of Canada’s economic recession, which was far shallower, shorter and, consequently, less destructive than the respective American and British recessions.

The BMA press release constitutes just one element of the public health establishment’s carefully choreographed campaign to derail the commonsense efforts by the UK business secretary Vince Cable to defang the tobacco display ban. Any political traction the BMA’s intervention gains clearly will owe far more to shrewd marketing than to supporting evidence.

Patrick Basham directs the Democracy Institute and is a Cato Institute adjunct scholar. John Luik is a Democracy Institute senior fellow. They are co-authors of Hidden in Plain Sight: Why Tobacco Display Bans Fail.