168 Alcohol in our lives and measuring the impact of changes in policy. For example, ACC may wish to ascertain the portion of its benefit payments properly attributable to alcohol- related activities. This would enable it to target better preventative measures. Such reports can also be useful for measuring the costs that may be recovered 10.25 by the imposition of alcohol excise tax. Quantifying the magnitude of alcohol-related harm, and putting a dollar value 10.26 on it, is not an exact science and is sensitive to which harms are being considered. It is inevitably a value-laden exercise. In New Zealand there have been three recent attempts to do so. In 1997, economist Dr Brian Easton estimated the social costs of alcohol misuse to be $16.1 billion in 1990 ($24.5 billion in 2005 dollar terms).¹⁷ Dr Easton’s estimate covered all social costs including the human costs of morbidity and mortality arising from alcohol consumption. In the same year, but using quite a different method, University of Otago academics estimated the social cost of alcohol abuse to be between $1 billion and $4 billion in 1999 (a midpoint of approximately $3.8 billion in 2005 dollar terms).¹⁸ A Treasury working paper in 2002 reanalysed the University of Otago study using different assumptions to estimate the external costs (i.e. those not included in the cost of alcohol) of harmful alcohol use to be most likely more than $735 million annually in 1991.¹⁹ Most recently the Ministry of Health and the Accident Compensation Corporation 10.27 (ACC) commissioned Business and Economic Research Limited (BERL) to estimate the social costs of alcohol-related harm in New Zealand, including costs to the government sector.²⁰ BERL estimated the social costs of the harmful use of alcohol in 2005/06 10.28 (expressed in 2008 dollar terms) to be $5.3 billion, of which 76 per cent ($3.7 billion) were tangible costs.²¹ Tangible costs are those which have a monetary value and are borne by private individuals (for example, increased car insurance costs, lost wages, reduced productivity and medical treatment) and by government (for example, crime costs, road crashes, police and justice resources, healthcare costs and ACC costs). The tangible costs represent approximately 2.5 per cent of New Zealand’s GDP. 10.29 This figure compares with an average tangible cost of alcohol to the European Union of 1.3 per cent of GDP (based on analysis of 21 social cost studies with a range of 0.9 per cent-2.4 per cent of GDP).²² A recent Australian study, using the same methodology as in the BERL study, estimated tangible social harm of alcohol at 1.4 per cent of GDP,²³ while a USA estimate was 2.3 per cent of GDP, and a Canadian study estimated tangible costs at 0.9-1.3 per cent of GDP.²⁴ The BERL report has been strongly criticised by two economists (Eric Crampton 10.30 and Matt Burgess) on the grounds that BERL failed to include any of the economic benefits associated with alcohol consumption and also inflated the costs in a number of ways, including counting as a cost the money spent on alcohol that is consumed in such a way that it leads to alcohol-related harm.²⁵ They concluded that the external costs of alcohol roughly match the amount currently collected by the alcohol excises.