Thousands of lives a year could be saved by a small rise in alcohol prices, doctors\’ leaders said yesterday as they put more pressure on Gordon Brown to tackle Britain\’s binge-drinking "epidemic".
More than a quarter of all drink-related deaths could be prevented by a 10 per cent rise in taxes on beer, wine and spirits, the British Medical Association (BMA) claimed.
You know, that really would be rather surprising.
Sir Charles George, the chairman of the BMA science and education board, said a 10 per cent rise would prevent 29 per cent of alcohol-related deaths in men and 27 per cent in women.
Are we to assume there that a 10% rise in tax (say, a 6 or 7 % rise in final price) will cut consumption by some 28%? That would make alcohol demand amazingly responsive to changes in price. An extremely elastic response.
Does anyone actually know what the elasticity of demand of alcohol is? Have there been any economic studies on this? Or is the BMA talking out of its arse?
Update: Looking here:
For example, a price elasticity of alcohol demand of -0.5 means that a 1-percent increase in price would reduce alcohol consumption by 0.5 percent (or a 10-percent increase in price would reduce consumption by 5 percent). An extensive review of the economic literature on alcohol demand concluded that based on studies using aggregate data (i.e., data that report the amount of alcohol consumed by large groups of people), the price elasticities of demand for beer, wine, and distilled spirits are -0.3, -1.0, and -1.5, respectively
So a 10% rise in price would reduce consumption of beer by 3%, wine by 10% and spirits by 15%. Is that enough to reduce the medical impact by 28%?
Or here, more specifically the elasticity of demand amongst young people:
The finding that drinking by young adults can be considered an addictive behavior has important implications for the effects of price on alcohol consumption. For example, when Grossman and colleagues (1998) used models that ignored the addictive aspects of alcohol consumption to analyze their data, they estimated an average price elasticity of alcohol demand of -0.29.
That is, a 10% rise in prices will reduce consumption in the age group by 2.9%….you don\’t think the BMA has got a little confused do you? Multiplied the effect by 10? Or forgotten to tell us that it\’s a 100% rise in tax that will reduce the effects by 28%?
Or Table 2.1 here. A meta-study of the research.
A summary of the own price elasticity, qp η , information — reported in absolute value
form — is presented in Table 2.1. Estimates are reported for 18 countries, and there are 46 beer
own-price elasticity estimates, bb η , 54 wine own price elasticity estimates, ww η , and 50 spirits
own-price elasticity estimates, ss η . The bb η estimates ranged from highly inelastic (0.09) to
elastic (1.20), with a mean bb η of 0.38. For the ww η the range of estimated values was slightly
greater; (0.05) to (1.80), ww η = 0.77. While the ss η estimates showed the greatest variation,
ranging from; (0.10) to (2.00), ss η = 0.70. The ww η estimate and the ss η estimate appear quite
similar, and statistical tests — details of which are given in Appendix II — indicate the ww η and
the ss η are not statistically different. Using the same approach, it is possible to conclude the
bb η is statistically different to both the ww η and the ss η .
Frequency distributions for the bb η , the ww η , and the ss η are presented in Figures 2.1,
2.2, and 2.3 and the plots clearly show the majority of estimates to be less than one. In
particular, 93 percent of the bb η estimates, 69 percent of the ww η estimates, and 80 percent of
the ss η estimates are less than one. Based on this result, it might seem reasonable to generalise
and conclude: The demand for all alcoholic beverages is inelastic, and beer is the most
inelastic beverage category.
If the price elasticity of alcohol is less than one, ie inelastic, then the BMA results are higly improbable (ie, what I really mean to say is that they\’re speaking out of their arses). A 10% rise in taxation, a 6 or 7 % rise in total price, will lead to a less than 6 or 7% drop in consumption, meaning that, well, at least I think it means that, there\’s no way that there can be a 28% drop in the medical effects of it.