At first sight there’s an obvious inconsistency between libertarianism and paternalism. The latter usually involves government actions which limit someone’s choices, even if the intention is to promote their own good. How can it be right (asks the libertarian) to restrict an individual’s freedom and autonomy just because the government thinks it knows better?

Indeed, many of the arguments that are being used to justify more and more state intervention are pretty flimsy. Here I particularly recommend Chris Snowdon’s excellent book, ‘Killjoys’, which is a compelling critique of ‘public health paternalism’, and Paul Dolan’s ‘Happy Ever After: Escaping the Myth of the Perfect Life‘.

One of the most common of these ‘nanny state’ arguments is the perceived need to correct some form of externality, such as the costs of smoking or obesity to the NHS. In reality, the proposed measures usually go far further than those that could possibly be justified on the basis of any harm to others. For example, the additional duties on tobacco in the UK are already well above the ‘Pigovian tax’ that would be sufficient to cover the social cost of any negative externalities from smoking.

Why is this? To put it bluntly, smokers and fat people typically die younger and will therefore be less of a drain on the rest of the population in their old age, once other factors such as future savings on unpaid pensions and long-term nursing care are taken into account. If this were just about saving taxpayers’ money (which of course it isn’t), the government should actually be encouraging life-shortening choices, rather than penalising them.

There is also the more fundamental issue of freedom of choice. If someone decides they would be happier following an ‘unhealthy’ diet or pursuing riskier activities, who are we to stop them? Some people might simply prefer to top off a fine meal with a sweet dessert and cigar, then go for a midnight dip, regardless of what others might think is best. The government could have a role to play in helping people make informed choices, but otherwise should leave them be.

Nonetheless, there are areas where exceptions might be excused. In particular, even John Stuart Mill (On Liberty) allowed for the possibility that paternalism could be justified in the case of children. Would we really feel comfortable allowing an 8-year old to buy alcohol, or fireworks? Here at least there is some basis for the presumption that the individual is not always the right person to decide what is in their own best interests. Ideally any intervention might be left to parents or guardians, but it doesn’t seem unreasonable for the state to act as backstop.

Nudge economics provides further examples. As explained by Thaler and Sunstein (Nudge: Improving Decisions about Health, Wealth, and Happiness) “A nudge … is any aspect of the choice architecture that alters people’s behaviour in a predictable way without forbidding any options or significantly changing their economic incentives. To count as a mere nudge, the intervention must be easy and cheap to avoid. Nudges are not mandates. Putting fruit at eye level counts as a nudge. Banning junk food does not.”

This approach can be seen as a form of ‘libertarian paternalism’. Admittedly, many of most commonly-cited examples of nudge economics involve little or no coercion, and the benefits are primarily felt by others. In particular, gentlemen are free to ignore the image of a housefly painted in urinals, which is mainly there to help the person whose job it is clean up afterwards. It is therefore debatable whether this sort of measure is ‘paternalistic’ at all.

Nonetheless, there may still be an increased risk that people find themselves in situations they would not otherwise have chosen. For example, if the default for organ donation is changed so that people are automatically ‘opted in’, some people may end up donating organs after death against their wishes. Even it is too late for them to care, their relatives and friends might still have a view.

What about more coercive policies? A good example here is the compulsory wearing of seatbelts in cars or planes. A purist could say that people should be free to be reckless, and that banning one type of risky activity might be a slippery slope towards more interventions elsewhere. But I’d be more pragmatic. The rules on seatbelts have undoubtedly saved many lives, including those of third parties (passengers wearing seatbelts are less likely to kill the person in front). I would be surprised if many people who have benefited from wearing seatbelts subsequently wish the rules hadn’t been in place.

However, it is also crucial that any intervention is properly evidence-based. It’s not clear, for instance, that requiring cyclists to wear helmets would be proportionate to the risks involved, and it might even be counter-productive. Wearing a helmet might make some cyclists more reckless, or encourage other road users to pass them with less care. Poorly-fitted helmets could obscure vision and actually increase the risk of injury. The compulsory wearing of helmets could also deter some people from taking up a healthy activity in the first place. Who should make these judgements, if not the individual themselves?

Indeed, history is littered with examples of governments and lobby groups getting it badly wrong, such as the support for diesel, bans on e-cigarettes, and the criminalisation of certain recreational drugs. There are also plenty of examples of the authorities changing their minds so often that no-one can have much faith left in what they have to say – consider the endlessly varying, often contradictory advice on ‘safe’ levels of alcohol or meat consumption.

This is on top, of course, of the simple libertarian argument that if a cyclist values the wind in her hair more than any reduction in the risk to her life, what right has the government to force her to wear a helmet? It seems especially dangerous to expect any regulator, however benign, to second guess what someone really wants. Free choices allow for individual experimentation and learning from mistakes (either our own, or those of others). The ‘nanny state’ does not.

In short, my view is that libertarians should sometimes be willing to tolerate a little light paternalism. But this is most definitely not a green light for celebrity campaigners and health puritans to tell the rest of us what to do.

This is an extended version of a blog which was first published (with a counterview from Andy Mayer) by the Institute of Economic Affairs – here

• Behavioural Economics