Alcohol Price Policies as an Instrument of Health Equity: Differential Effects of Tax and Minimum Price Measures.

Increasing the price of alcohol has been shown to reduce alcoholrelated harm (Babor et al., 2010). Thus the price of alcohol is a relevant consideration in improving the health of the population. Existing regulatory strategies to manipulate price include changing tax rates or specific price controls such as minimum unit pricing (MUP). The effects of these strategies on overall consumption are significant (Stockwell et al., 2012; Holmes et al., 2014), as well as their effects on consumption by sub-groups such as people on low incomes. Indeed, modelling in the UK using the so-called ‘Sheffield’ model (Brennan et al., 2014) suggests that the introduction of MUP would impact most on the consumption of low-income harmful drinkers (Holmes et al., 2014). Leaving aside issues of fiscal equity, in terms of hitting the pockets of poor drinkers harder, the modelling predicted two beneficial effects of MUP on the reductions of purchasing by this group of drinkers: it would result in overall decreased spending on alcohol and produce better health outcomes as a by-product of the reductions in purchasing and consumption. Although much of the recent discussion surrounding price policies has been focused on MUP, the broader question is how price changes resulting from any strategy differentially impact on different drinkers, be they grouped by socioeconomic status, drinker category or by the beverage that they choose to drink. Such differential effects on poorer and richer drinkers are often the subject of public policy debates, but usually the emphasis is on the regressivity of the price measures, meaning that alcohol price increases tend to absorb a greater proportion of the income of poorer drinkers than of more affluent drinkers (not necessarily of the poor rather than the affluent in general, since the poor are more often abstainers—Ashton et al., 1989). Six papers (five revised from papers presented at the Kettil Bruun Society Thematic Meeting on Alcohol Policy Research in Melbourne in September 2014) presented in this special section tackle this important issue. They are focused specifically on price and/or taxation policy and examine how these policies impact on affordability. The paper from Sornpaisarn et al. (2015) sets the scene, presenting simulations of different combinations of taxmethods, including ad valorem and specific taxation and different combinations of the two, to examine their impact on tax per unit of ethanol while taking into account the perceived quality per unit—i.e. what value the consumer assigns to each beverage. They demonstrate that while taxation according to value (ad valorem taxation) resulted in a lower tax burden for cheaper products and the specific-tax method resulted in a lower tax burden for higher quality beverages, combination methods such as mixed specific and ad valorem taxation can result in unintended outcomes. Depending on the type of combination method applied, these could include reducing price differentials between mediumand low-quality beverages or removing incentives to keep the alcohol content of cheaper beverages down. The authors stress the importance of ensuring that the net result of a particular combinationmethod does not provide an incentive to increase the ethanol or decrease the quality of beverages in order to avoid a tax burden. In a similar vein, Vandenberg and Sharma (2015) use an annual compilation of highly-accurate barcode scanning data of off-licence purchases in Victoria, Australia to contrast the effects of MUP with specific taxation. Adapting and extending the Sheffield model (Brennan et al., 2014), they estimate slightly greater effects of these policy changes on poorer drinkers, primarily through changes to the price of wine (taxed the least in Australia). The resultant potential health gains from likely consumption changes are estimated to be greater for MUP than for a specific taxation increase, particularly for low-income drinkers, while the increased costs are only mildly regressive by income, particularly for moderate drinkers. Jiang and Livingston (2015) provide the first study on the relationship between alcohol consumption, price and affordability over time. Using Australian data, their model shows that increases in price resulted in initial short-term decreases in consumption, but that these levelled off over time. However, their findings suggest that increases in affordability have longer-lasting impacts, with increases in consumption still found in the long term. This finding picks up on the long neglected suggestion (Bruun et al., 1975) that alcohol price elasticities may well not be symmetrical; the concept of alcohol dependence predicts, for those who are dependent, that price increases would have less or shorter effects than price decreases. Of the beverage categories they consider, beer and spirits appeared more susceptible than wine to price-based changes in consumption, but all trends in beverage-specific consumption were at least somewhat affected by changes in price over time. Kerr et al. (2015) examine the consequences of the marked increase in liquor sales outlets that accompanied the end of a state monopoly on spirits sales in Washington State in 2011. Using measurements taken before and after the end of the monopoly they find that many items increased in price, primarily as a reflection of the increased taxes implemented to make up for the loss of revenue from the monopoly’s end. Bulk spirits at large-scale stores did not increase in price, instead showing a non-significant decrease, despite being one of the items possibly related to harmful consumption. This finding highlights how competition in the alcohol market can alter prices differentially in ways that may not always be uniform and result in undesired outcomes from a public health perspective. Callinan et al. (2015) examine who it is that purchases cheap alcohol, offering evidence on which drinker groups would most likely be affected by a MUP strategy. They find that effects on groups of drinkers differed depending on whether the outcome variable was the Alcohol and Alcoholism, 2015, 50(6) 629–630 doi: 10.1093/alcalc/agv061 Advance Access Publication Date: 15 June 2015 Editorial