Sugar taxes still don't work

Adam Briggs did some of the modelling for the sugar levy back in the day and is an advocate of the policy. The 'public health' racket being what it is, he has since been put on the team tasked by the government with evaluating it.

In the meantime, he has written a response to Boris Johnson's criticisms of sin taxes for the Telegraph. Under the headline 'Revealed: The evidence which shows we need the sugar tax to fight the obesity epidemic', Briggs claims that sugar taxes are evidence-based, and yet he is unable to provide any evidence that they have ever had the slightest impact on obesity.

Ten years’ ago, evidence for the impact of soft drink taxes on behaviour was all based on mathematical modelling (hypothetical projections) but now there are over 30 countries with such taxes and the ‘real-world’ data are filtering through.

You'd think, therefore, that sugar tax campaigners could point to at least one country where obesity has declined as a result. Alas, no.

Perhaps the most well-studied country is Mexico, where a soft drink tax of a peso per litre (around a 10 per cent price increase) led to an average 8 per cent reduction in purchases over the first two years. 

But it didn't. A model designed by soda tax campaigners (ie. one of those 'hypothetical projections') claimed that sugary drink consumption would have risen if the tax hadn't been introduced. In reality, per capita consumption of sugary drinks stayed the same.

More recent data from Philadelphia in the US has shown a 40 per cent fall in sales following a 1.5c per ounce tax.

Indeed it does. That's because people started shopping out of town. The tax was bad for business in Philadelphia (as was always the intention), but good for neighbouring areas.

So what was the net effect on calorie consumption?

Due to cross-shopping and compositional changes in demand, we do not detect a significant reduction in calorie and sugar intake.

Oh dear.

Briggs continues:

A systematic analysis of all such real-world evaluations published last month concluded that a 10 per cent price rise in soft drinks leads to – on average – a 10 per cent fall in purchases.

Yes, that's roughly the price elasticity for soft drinks. If the price of any good goes up, consumption will generally go down. That's the law of demand. It does not make a sugar tax a 'genuine evidence-based public health policy', as Briggs claims, nor does it mean that 'as single interventions go, sugary drink taxes are right up there'.

What is remarkable about soda taxes is how rarely they produce a measurable decline in sugary drink consumption, let alone in calorie consumption or obesity.

There is plenty of other real world evidence that Briggs could have cited. None of it supports his case. If sugar taxes are 'right up there' as anti-obesity policies, it is a damning indictment of all the other anti-obesity policies, which are not really anti-obesity policies but anti-smoking policies rebadged by people with no imagination and inappropriately applied to a more complex and totally different issue.

Catch up on a decade's research into sugar taxes here.