Smoke gets in the government’s eyes

Professor James Heckman, the Nobel Prize-winning economist, has devoted a decade to understanding what makes young people engage in risky behaviours, such as smoking and illegal drug use, and what can be done to prevent this.

Given that a UK House of Commons committee is currently debating how to prevent youth smoking through legislation that requires shopkeepers to hide all tobacco products from sight, it would be reasonable to assume that the committee would talk to Professor Heckman.

But it hasn’t. And the reason the committee is not doing so speaks volumes about the debased state of public policy debate.

The basis of the UK government’s legislation is the claim that one of the major reasons why young people smoke is that they see tobacco displays in shops. If you find this claim more than a little odd, you are not alone.

When the legislation was debated in the House of Lords, the government was unable to produce any credible and compelling scientific evidence that seeing tobacco displays in shops led anyone to smoke or that removing tobacco displays had led to a decline in youth smoking. Indeed, the international evidence from a variety of countries that had tried this ‘silver bullet’ to stop youth smoking showed that adolescent tobacco use had either increased following a display ban or stayed the same.

In an attempt to bolster their less than robust case, the government touted a study by professor Gerard Hastings, which argued that the more young people were aware of tobacco brands the more likely they were to ‘intend’ to smoke (notice, not smoke, but intend to smoke).

Regrettably, the Hastings study (which was not published in a peer-reviewed journal) has several defects. Since it wasn’t a study establishing cause and effect, it could never counter the commonsense response that the reason young ‘intending’ smokers knew more tobacco brands is because they were young intending smokers. Their smoking led to their interest in brands, not the other way round.

It also failed to provide a demonstrated connection between increased awareness of tobacco brands and tobacco displays in shops, which is, after all, what the legislation is about. Finally, Hastings’ research is contradicted by a host of other studies, including some cited by the government itself, which showed that tobacco brands are not important in the process of taking up smoking.

Where does this leave Heckman? The reason that Hastings addressed the Commons committee rather than Heckman is because Hastings’ research supports the government’s tobacco control policy, while Heckman’s work exposes the government’s nonsensical arguments.

Heckman does this in two ways. First, his work shows that studies like Hastings’ and claims about shop displays causing young people to start smoking are instances of flawed science: they don’t meet the rigorous standards necessary to establish the government’s claim that ‘tobacco displays cause young people to begin smoking’.

Heckman argues that the government and public health advocates consistently make both strong assertions about the supposed causes of youth smoking and push draconian policies like display bans based on simplistic and flawed statistical methods that would not pass muster in other areas of public policy. Their studies never show that the particular focus of a policy, such as tobacco displays, is the cause of a problem such as youth smoking.

Nor do they ever consider and control for the role of other factors in initiating youth smoking. Nor are the studies replicable by other researchers. As a result, public policies and enormous resources are being focused on alleged ‘causal factors’, such as tobacco displays, ‘that have not been scientifically established but merely assumed to affect smoking initiation’.

As a result, public policies for dealing with youth smoking and drug use are put forward without any connection to the problem for which they are intended. This means that by their very nature they miss the mark and fail. For example, if tobacco displays don’t cause young people to smoke, banning them won’t do anything to prevent youth smoking. Such measures are nothing more than window dressing and they fail to come to terms with the root causes of what drives smoking.

Second, Heckman’s work provides real answers to the question of what leads young people to start smoking and what can be done to prevent it.

In a series of recently published research studies, Heckman showed that two factors, both malleable – family environment and school environment – are crucial in the development of two skill sets: cognitive skills and non-cognitive skills, such as motivation, determination, self-esteem, and self-regulation (1).

These two skill sets are both strongly and consistently related to the probability of becoming a regular smoker by age 18. Moreover, the foundations for both of these skill sets are developed at an early age. In short, the higher one’s cognitive and non-cognitive skills, the less likely it is that one will become a smoker.

Heckman’s research finds considerable support in recent research on youth smoking in the UK. Several studies have found that youth smoking is strongly correlated with living in areas of high social and economic deprivation, failing schools and dysfunctional families – precisely the sorts of environments that fail to provide for the development of the cognitive and non-cognitive skills that Heckman finds so important.

The implications of Heckman’s research for smoking policy are therefore enormous. Instead of focusing on things like tobacco displays that are unrelated to youth smoking, the essence of the government’s tobacco strategy ought to be three-fold:

1. designing early childhood interventions to develop crucial cognitive and non-cognitive skills in those areas with the worst performing schools and the highest youth smoking rates (these tend to overlap);
2. removing the causes of social and economic deprivation in those with the highest youth smoking rates;
3. creating novel interventions to address the sources of family dysfunction in those areas with the highest youth smoking rates.

The great benefit of such a tobacco policy is that, in addition to dramatically reducing smoking, it would reap significant benefits in a host of other areas.

It’s a real tragedy that the Commons isn’t hearing from James Heckman on youth smoking. This isn’t just because his analysis cuts through the government’s cant and shows how unsupported its policies, such as a display ban, really are. Far more importantly, his enormous knowledge of what really does cause smoking may help prevent a generation of British kids from lighting up.

Patrick Basham, a Cato Institute adjunct scholar, and John Luik, a Democracy Institute senior fellow, are coauthors of Hidden in Plain Sight: Why Tobacco Display Bans Fail, available from the Democracy Institute.

(1) See ‘An Assessment of Causal Inference in Smoking Initiation Research and a Framework for Future Research, by James Heckman, in Economic Inquiry, January 2008.