Monday, May 29, 2006

Taxing booze again

I have previously argued the case for volumetric rather than value-based taxes on booze. If external costs are identified with alcohol consumption then it makes sense to tax alcohol content not the value of alcoholic products. To some extent this is already done through lower taxes on light beers.

This issue can be directly connected to currently-discussed health and crime issues facing aboriginal Australians. According to an unpublished report discussed in today's Australian, large increases in alcohol sales has fueled a dramatic rise in violence, deaths and hospital admissions among aborigines. A major component of this growth has been in sales of cheap cask and fortified wines. Sales of fortified wines have fallen over the last few years but still remain at very high levels. Moreover, such falling sales are offset, more than four-fold, by increases in bottled and cask wine sales.

Of course one can argue that alcohol consumption is a matter of personal choice and that no externalities are involved. Even though alcohol is a neuro-toxin and has numerous other health effects - and few demonstrable benefits for health - it can be argued that the decision to destroy your brain and liver is rationally taken after a fully-internalised computation of costs and benefits. I think if you believe this you are gullible given the evident scale of the self-control issues involved as evidenced by the existence of groups such as AA.

But even if you dispute the existence of such self-control problems Australia does have a publicly-funded universal-cover national health scheme. Australians also fund police and social services in aboriginal communities where violence and health problems are spiraling out of control. So long as we retain such schemes there are social as well as private costs from excessive intake of alcohol and a standard efficiency-improving argument for taxing heavily to internalise social costs.


Afterthought (1/6): An issue in taxing low-value high-alcohol fortified wines and cask wines might be the incentives to switch towards drinking of metholated spirits, petrol sniffing and perhaps other drugs. I am certain that this types of substitutions will occur but uncertain of the extent. This could limit the case for volumetric taxes - empirical evidence on cross price elasticities is required.

1 comment:

Patrick said...

I second that Harry.

The fact we don't have a content-based tax as it stand is a testament to the lobbying power of alcohol companies in all their forms.

I've worked in the alcohol and other drugs sector and it's something the sector's peak bodies campaign for every year, with never a good result. Meanwhile you can buy a _cask_ of port for six dollars, which has far far far more standard drinks per dollar than anything else out there.