Believe it or not, people actually like smoking and eating fatty food

Public information campaigns and nutritional labelling are good at informing people about what’s healthy and what isn’t, but don’t seem to have much impact on what they actually eat. That’s what a comprehensive review of 121 'healthy eating' policies found, and I think it should make us rethink more heavy-handed policies to do with unhealthy food, tobacco and alcohol. There are benefits as well as costs to every activity that public health groups want to discourage. We know there are benefits because people do them freely. But we know there are costs as well, like living a shorter and less healthy life.

The liberal view is that each person’s cost-benefit calculation is different, because they enjoy and dislike things differently. In this view there’s no case for stopping people from doing things unless they don’t actually have the information they need to make a judgement. We should want to make people’s lives better as they themselves understand ‘better’, not according to a single measure we’ve decided on, like lifespan.

So telling people that sugar makes them fatter may be a good policy, if they didn't already know that. And policies that do that do seem to make people more informed. But what’s interesting is the impact they have on people’s diets – usually not much, and sometimes an unexpected one.

For example, a 2008 study found that people who used nutrition labels had big increases in fiber and iron intake, but no change to their total fat, saturated fat or cholesterol intake. The UK’s ‘five a day’ campaign about fruit and veg was very successful at getting people to think about eating more fruit and veg, but increased people’s intake by an average of 0.3 portions a day (which was not viewed as being a very good improvement). 44 studies of similar campaigns in the US and EU have shown about the same size effect.

To some people that might make it look like we need to do more. To me it looks as if people view the costs of changing their diet to something less enjoyable or convenient as being quite important, and are willing to forgo some level of health to avoid that.

Maybe this tells us something about cigarette regulation too – there is some evidence that smokers actually overestimate the risk of smoking and some that they underestimate it. If they do overestimate the risks, we’re ‘informing’ people so much that it’s become misleading.

It would be fair to respond to this that people have no real way of doing a proper cost-benefit analysis about eating sugary foods or smoking, but because the state can’t measure the benefits – that is, the pleasure – it is just as limited.

The fact that people do change their habits about iron and fibre, but not fats, suggests that they aren’t ignorant, they just don’t want to eat less fat! If that’s the case and we’re working to improve people’s lives on their terms, there is no case at all for more heavy-handed policies like taxes, ingredients restrictions and advertising bans.