In March 2021, a study was published in BMJ Open claiming that the UK sugar tax had led to a 10 per cent reduction in the amount of sugar consumed in soft drinks. Although one of its authors admitted that a decline of this magnitude “might sound modest”, it was presented as a win for public health. The preposterous pressure group Action on Sugar called for the tax to be “extended to other categories” and the 10 per cent figure soon found its way into the National Food Strategy and several World Health Organisation reports.

Last week the study was retracted, along with an editorial titled “UK sugar tax hits the sweet spot” that had been published in the British Medical Journal claiming that the tax was “working exactly as intended”.

It turns out that tax has not been not working exactly as intended. In a new version of the study, the authors estimate that the decline in sugar consumption from soft drinks was just 2.7 per cent, barely a quarter of the original figure, and that in contrast to the original study, which claimed that there had been no change in soft drink sales, the volume of soft drinks rose by 2.6 per cent.

The decline in sugar consumption was originally said to be 30 grams per household per week. In the new study, it is estimated to be 8 grams per household per week. That works out at less than two calories per person per day. To get an idea of how little that is, get a slice of bread and take the tiniest nibble off one of the corners. That is the amount of calories reduced by a tax that costs consumers £300 million a year.

The mistake was so big that the journal decided that only a full retraction would suffice

The authors of the study say that they made a mistake with a weighting variable which was “incorrectly calculated as the inverse of what it should have been” and should not have been used in any case because it “it replicatsed [sic] a component of a second weighting variable”. The mistake was so big that the journal decided that only a full retraction would suffice, particularly since it affected “the policy implications of the paper”. I suppose those implications are that a policy that was thought to have a very small effect on calorie intake actually had a negligible effect.

This is not just any old bunch of researchers. They are conducting the official evaluation of the sugar tax and have been given £1.6 million of taxpayers’ money to do so. The same team has since produced a study claiming that the sugar tax reduced obesity among one group of girls but not among another group of girls and not among any boys (and obesity rates actually rose among all age groups), and a study claiming that it reduced tooth decay extractions among a subset of children who never drank many sugary drinks to begin with (I wrote about the latter last month). Almost unbelievably, they have a study in the pipeline that will claim that the sugar tax reduced childhood hospital admissions for asthma by more than 20 per cent!

The retraction casts further doubt on all these claims, partly because it raises serious questions about the team’s competence and partly because the reduction in sugar consumption attributed to the tax is now so tiny that any measurable difference in health outcomes is less plausible than ever.

This strikes me as being all rather interesting and yet there has been very little interest shown in it. The retraction happened over a week ago and yet it has only been reported by the niche website Retraction Watch. The news outlets which reported that “Sugar tax cuts Britain’s intake by 10%” — as The Times misleadingly put it in 2021 — have been strangely quiet. None of the study’s 13 authors have tweeted about it despite telling the BMJ that they are eager to set the record straight. 

The sugar tax was big news when Jamie Oliver was campaigning for it and even bigger news when George Osborne introduced it. Now that it has fallen flat on its face, no one wants to know.