Rumours abound that the government is secretly quite keen on imposing a new tax on sugary drinks, to attempt to combat obesity. There are strong arguments against such a proposal. These taxes hit the poor hardest and will probably only have a marginal impact on obesity levels. Plus, so-called “sin taxes” of this kind have been abandoned in many places where they have been tried (such as Denmark with its “fat tax”). But one issue with the policy has been much less commented on: how such a tax would further complicate and render incoherent the taxation of food and drink more broadly.

You will probably be aware that many items of food and drink are currently VAT exempt (or “zero rated”, to use the tax jargon). But you would be wrong to assume that this categorisation has any rhyme or reason. If you buy chocolate chip cake decorations, chocolate jaffa cakes, chocolate spread, or drinking chocolate, you will not pay VAT on them. If you buy chocolate ice cream, chocolate button cake decorations, or nuts in a chocolate coating, you will. To take the most egregious example: if you buy a gingerbread man with chocolate eyes and a chocolate mouth, you would pay 20 per cent VAT, but if you go for a gingerbread man with only chocolate eyes, then you will escape the charge.

This, quite simply, is madness, and highlights the poverty of this whole debate. Many of the mooted sugar tax proposals are relatively narrowly targeted at very sugary drinks (i.e. products with added sugar, to exclude fruit juices). But what really matters is calories and the interaction between consumption and burning them. In that regard, in our VAT system, we have many other food and drink products which could be considered high calorie or unhealthy in other ways which are, in effect, subsidised through tax exemptions. A sugar tax, meanwhile, would impose an additional charge on products which are already taxed much more highly than these other foods, but which could in many cases also have damaging effects on our health if not consumed in moderation.

One would think policy-makers would seek to address these inconsistencies in existing taxes first before even considering new “sin” taxes. But evidence from other sectors suggests this is unlikely. After all, we still apply VAT on domestic energy at a reduced rate (effectively an extremely inefficient subsidy), but then impose a raft of green taxes to try to deter its use.

Sugary drinks – such as colas, lemonades and ginger beers – are already charged the full VAT rate. But if we are concerned about obesity, and not just trying to hit the drinks industry (which, ironically, unlike many other parts of the food market, almost always produces low sugar alternatives), imposing an additional charge on a narrow range of drinks alone makes no sense.

That’s because one consequence of taxing products such as sugary drinks will be a degree of substitution to other high calorie drinks, such as fruit juices. These would not be taxed to the same extent under the current proposals. So this makes a mockery of the whole justification for the so-called “sugar tax”. We are told it is needed to reduce the negative social costs associated with obesity. But it makes no sense to tax cola but not fruit juices, and lemonade but not chocolate.

One reason the campaigners are focusing so narrowly is that introducing a broader sugar tax would be incredibly complex – requiring examination of the sugar content of all products. Far simpler to impose an ad valorem tax on certain drinks. But even then, would there be an added sugar threshold? Would there be substitution to greater use of sweeteners? Would all sugary drinks be taxed the same irrespective of the sugar content above a threshold (in turn undermining the whole argument for the tax)?

As well as all the existing strong arguments against such a tax, the desirability of a simple and coherent tax system should be added to the list.

Ryan Bourne is the IEA’s head of Public Policy. This article was first published in City AM.