The Home Office says its plans for a minimum price of alcohol are “designed to address the issue of binge drinking.” This is boilerplate managerialist-speak. In the managerialist’s world, there are not problems but issues, and these are not solved, but addressed.
In this instance, though, the choice of language is correct.
The proposed minimum - 38p for a can of lager and £10.71 for a litre of vodka - is equivalent to 20-26p per unit of alcohol.
And the most influential research here - the Sheffield study (pdf) estimates that such a low minimum will have almost zero effect (p109) upon overall alcohol consumption. In fact, it will slightly increase drinking among heavy drinkers aged 18-24 (p121).
This is not because there are generally perverse price elasticities; the study estimates that a more biting minimum would cut consumption.
I suspect there are two mechanisms through which a low minimum price might raise consumption.
One is the Gneezy-Rustichini effect. They have famously described (pdf) how, when nurseries in Haifa introduced fines for parents who were late in picking up their children, the number of late-comers actually increased. This happened because prices can crowd out social norms, and if the price is low enough, its deterrent effect will be weaker than the social norm it displaced.
A similar thing could happen in this case. The low minimum price might be taken as a official vindication of cheap drinking without pricing committed drinkers out of their habits.
Even if higher prices do reduce someone’s consumption, though, they might reduce the wrong people’s. If they deter the moderate drinker from drinking a lot, then one restraint upon his friends’ heavy drinking is removed, with the result that aggregate boozing might increase; “now the lightweight‘s gone we can do some serious drinking.”
The government claims to be interested in behavioural economics. Not interested enough, it would seem.