Tuesday, 17 October 2017

Jamie Oliver's sugar tax 'success'

The evidence that taxing sugary drinks has any effect on obesity is - to put it mildly - poor. The great success story is supposed to be Mexico but we now know that per capita consumption of sugary drinks was essentially the same after that tax was introduced (in 2014) as it was before.

The other success story is supposed to be Berkeley, California, where sugary drink sales fell by 9.6 per cent after a soda tax was introduced in March 2015. But as I mentioned last week, you only have to read the study to see what really happened. Sales fell by 0.8 fluid ounces per transaction in Berkeley but they rose by 0.7 fluid ounces per transaction in neighbouring areas. People simply went out of town to do their shopping.



Since the quantities of sugary drinks consumed didn't change after the soda taxes were introduced in Mexico and Berkeley, it is inconceivable that they could have any effect on obesity. The authors of the Berkeley study (who include soda tax fanatic Barry Popkin) admitted that there was no statistically significant change in calorie intake from sugary drinks and that 'caloric intake of untaxed beverages (milk and other diary-based [sic] beverages) increased.'

Things are not looking good for advocates of these taxes, especially after the taxpayers' revolt in Chicago. An element of desperation is creeping in, hence this today...

Jamie Oliver’s 10p tax on sugary drinks sold in his Italian restaurants has resulted in a significant drop in sales, a study has found.

Jamie knows a thing or two about losing sales. He's been closing restaurants left, right and centre this year. However, the study looks at soft drink sales per customer so should not be affected by the general decline of Oliver's businesses. So what's the story?

A study of the effects of the levy, published in the Journal of Epidemiology & Community Health, has found that sales of sugar-sweetened drinks such as colas and lemonades fell by 11% in the first 12 weeks. At the end of six months, sales were 9.3% lower than they had been before the levy was introduced.

The price elasticity of sugary drinks is generally though to be around 0.8-1.2, meaning that a 10 per cent increase in price leads to a fall in demand of roughly 10 per cent. Jamie Oliver's drinks are so expensive - at £2.60 to £3.25 - that a 10p 'tax' only increases the price by 3.5 per cent.

A decline in sales of 9.3 per cent as a result of a 3.5 per cent price rise is implausible. It would imply a price elasticity of about 3.0, ie. three times higher than has been observed elsewhere.

To their credit, the researchers admit that this is not very likely...

Prof Steven Cummins of the department of social and environmental health research at the London School of Hygiene and Tropical Medicine, who carried out the study, acknowledged that the clientele of Oliver’s restaurants tended to be affluent, and that the price hike on a drink costing between £2.60 and £3.25 might not make a lot of difference to them.

“I don’t think the financial element of it is a massive disincentive,” he said. But he likened it to the plastic bag charge, which prompts people to think about having one.

It's worth remembering that Jamie introduced his 'tax' after presenting a ridiculous documentary on Channel 4 that portrayed sugary drinks as something akin to asbestos. This is likely to have had an effect on the kind of morons who admire the man and go to his restaurants. By the same token, it is likely that people who enjoy sugary drinks and don't want to be lectured by a fat-tongued Essex pea-brain would have been less likely to go to his restaurants after Oliver got on his high horse.

In other words, the people who visited his poxy restaurants after he introduced this gimmick were not necessarily the same people who visited before.

The drop in sales at six months of 9.3% was only in the restaurants that previously had higher levels of sales of sweetened drinks. There was a general drop in sales on non-alcoholic beverages, except for fruit juices, which went up. 

This is an important point. The most interesting thing about sugary drinks taxes is seeing what substitution effects take place. The Guardian doesn't give the figures, but the study says that there was a 22 per cent rise in the sale of fruit juice (which has about the same amount of sugar as a fizzy drink), although fruit juice orders from the children's menu fell. Sales of off-menu mixers went up slightly, but sales of diet cola and bottled water went down (by 6-7 per cent).

In fact, the sale of nearly every type of drink went down. It is not clear what, if anything, people were switching to. The authors don't have figures for alcohol sales for some reason, but the rise in the sale of mixers suggests that the sale of spirits may have increased. Alternatively, people could have switched to tap water. Either way, it left Oliver out of pocket.

He [Cummins] said he thought the effect was “entirely transferable” to other less expensive chains. “There is no reason why other restaurants couldn’t do exactly the same,” he said. 

Actually, there is a very good reason. Drink sales are an important revenue stream for restaurants and Oliver seems to have lost them overall. It speaks volumes about 'public health' researchers that Cummins doesn't think this would play a part in a restaurateur's planning.

This study doesn't tell us anything useful about the impact of sugary drink taxes as a government policy. I'll leave it to Kevin McConway, emeritus professor of applied statistics at the Open University, to have the last word:

“The menu was redesigned: it explained that the proceeds of the levy would go to the Children’s Health Fund, new drink products were introduced, and Jamie himself appeared in a television programme about sugar. So we certainly can’t be sure that the fall in consumption of sugary drinks was entirely, or even mainly, caused by the extra 10p.

“The researchers do provide some circumstantial evidence that the 10p played a role in the reduction in consumption, but they (rightly) make it clear that a study like this can’t prove what caused what. Actually, it doesn’t even establish that any of the specific changes at Jamie’s Italian restaurants had anything to do with the lower consumption – for instance, the researchers had no data from any other restaurants, and maybe consumption fell there as well"

“It’s interesting that, in this study, the consumption of fruit juice from the children’s menu fell as well – indeed it fell by rather more than the consumption of the sugar-sweetened drinks, while consumption of fruit juice from the main menu went up. Maybe the numbers of children going to the restaurants changed, relative to the numbers of adults – the researchers couldn’t tell because they had no data on whether customers were adults or children. Maybe things would have been clearer if they had had data over a full year after the change, rather than just from September to February.

Quite so.

1 comment:

Chris Oakley said...

I wrote a comment about this under an article in The Times, which becomes more like the Daily Mail every day. I included a line pointing out that the effect being hyped by whichever tabloid sub editor writes the headlines these days was only seen in some of Mr Oliver's restaurants and that that alone should have caused any self respecting investigator to hold back from sending out a press release containing or at least implying claims that are unjustifiable from the data.