The efficiency of a volumetric alcohol tax in Australia

Joshua Byrnes, Dennis J. Petrie, Christopher M. Doran, Anthony Shakeshaft

In Australia and elsewhere, fiscal measures such as alcohol taxation are a commonly used intervention and cost-effective strategy to reduce alcohol consumption and associated harm. However, alcohol taxation policies distort the market for alcohol, specifically increasing the marginal cost of alcohol. It is proposed that a volumetric tax, which taxes alcohol equally across all beverage types, is less distortive of consumer preferences and more efficient at reducing alcohol consumption than the current Australian tax model, where taxes are charged at varying amounts per litre of pure alcohol, depending on the beverage type. This paper quantifies the effect of four different alcohol taxation systems, relative to the current Australian system: two different types of volumetric taxation (deadweight loss neutral and tax revenue neutral); the recent strategy trialled in Australia of increasing the tax only on ready-to-drink alcoholic beverages (i.e. premixed spirits); and a tiered tax system, which may be more politically acceptable. A partial equilibrium approach was used to measure taxation revenue, consumer welfare and consumption in alcohol markets. Estimates of taxation revenue, consumer welfare and consumption were first calculated for 2008 and then compared with the four scenarios considered. Relative to the previous alcohol taxation scheme in Australia, the taxation strategy that increased the tax solely on ready-to-drink alcoholic beverages increased taxation revenue by 479 million Australian dollars ($A), reduced pure alcohol consumption by 754 000 litres and increased the net deadweight loss of taxation by $A62 million. For a tax-neutral approach, for the same level of taxation revenue as is currently generated, a volumetric tax could substantially reduce the cost of taxation (as described by the net loss in consumer welfare) by $A177 million and reduce pure alcohol consumption by 4 68 000 litres. Under a deadweight loss-neutral scenario, for the same amount of deadweight loss generated from the previous taxation scenario, taxation revenue could be increased by $A1153 million, in addition to reducing pure alcohol consumption by 4 316 000 litres. A tiered taxation regime, as modelled here, could decrease pure alcohol consumption by 2 616 000 litres whilst increasing taxation revenue by $A1101 million. However, this scenario would also increase the deadweight loss of taxation by $A113 million. From these scenarios, it can be shown that, for the same tax revenue, consumer welfare can be reduced or, for the same level of loss to consumer welfare, taxation revenue can be increased. Both these scenarios result in a reduction of pure alcohol consumption.