4  A Uniform Tax in the Case of Alcohol

In order to advise on the appropriate rate of tax on alcohol one must derive the relationship between the externalities of alcohol consumption and the amount of revenue that would be raised from an optimal uniform tax on alcohol. Information on the shape of the damage function, the total size of the externality, responsiveness of consumers and the distribution of alcohol consumption is required.

4.1  Shape of the damage function

The externalities of alcohol consumption can be characterised as harm from acute consumption (binge drinking) and harm from long-term drinking (chronic consumption).

Acute alcohol consumption leads to externalities through adverse health and behavioural consequences. A large part of the tangible element of this externality can be proxied by the cost of addition public services. At the extreme, health effects include raised blood pressure, cardiac failure, stroke and foetal damage. Acute alcohol consumption increases the risk of falls and accidents, can lead to violent or anti social behaviour and is associated with adverse workplace outcomes. In Australia it is estimated that half of alcohol related deaths result from the short-term effects of drinking. As the incidence of casualties is greater among young adults, a high number of life years are lost from fatal casualties.[19] Further, the costs of acute alcohol consumption dominate the estimates of external costs.[20]

The risk of adverse consequences increases with the amount consumed in a single drinking session. However, even low to moderate levels of consumption increase the risk of accidents significantly and can generate significant damage.[21] As risks increase with the frequency of consumption and the amount drunk per occasion it is reasonable to assume that these risks increase with average consumption levels.

Long-term external effects are mainly health effects, being the additional cost of use of public services.[22] It is likely that the relationship between long-term alcohol consumption and health damage is J shaped for the population, with low-moderate levels of alcohol consumption conferring beneficial health effects on some consumers and higher levels of long-term alcohol consumption causing health damage.

Long-term heavy consumption of alcohol increases the risk of cancer, liver damage, pancreatitis, hypertension, cardiovascular disorders, brain damage and haemorrhagic stroke.[23] Long-term heavy consumption also results in adverse work place outcomes.[24] It is generally the quantity drunk rather than the drink type that is important.[25]

Low or moderate levels of alcohol consumption can provide some health benefits in terms of reducing the risk of ischaemic heart disease (CHD) and ischaemic stroke.[26] The benefits flow from a regular pattern of drinking and are not gained through binge drinking. Heavy drinking is likely to increase the risk of ischaemia relative to light drinking and abstaining.[27] Ischeamia is a major cause of illness in New Zealand. CHD accounted for about 6369 (23%) deaths in 1997 and stokes accounted for 2566 deaths in 1997, the majority (around 2000) of which were ischaemic.[28]

Although the protective effect of alcohol occurs at all ages, it is of little relevance for younger age groups as these groups have little risk of ischaemia and the protective effect is probably largely gained in the short-term.[29] In 1997 84% of those who died from CHD were aged over 64.[30] The risk of ischaemia can be reduced through other less risky and more cost effective mechanisms such as diet, exercise and taking aspirin. The weight of opinion suggests that particular beverages are not more beneficial than others, as health benefits come from the alcohol component.[31]

Studies differ on the point to which marginal benefits/costs are obtained from alcohol consumption.

For those who are not at risk of ischaemia or who mainly binge drink, there is no external benefit from alcohol consumption and heavy and binge drinking will confer net external costs. Expected damage will be an increasing function of average alcohol consumption.

For those at risk of ischaemia who drink regularly, it is reasonable to assume that significant marginal benefits accrue up to an average of about 2 drinks per day and that net marginal benefits accrue up to an average of one drink per day. Consumption up to about two drinks per day is likely to incur a net external benefit. Consumption past this level is likely to incur net external costs. Past one drink per day, net external damage is likely to increase with average alcohol consumption.[32]