Alcohol industry and governmental revenue from young Australians

Additional keywords: adolescents, alcohol consumption, policy, taxation, young people.

The consumption of alcohol is commonplace across various countries, cultures and settings.¹ Because of its pervasive reach, a significant amount of research has investigated the consequences of alcohol consumption. There is a substantial literature examining the social and health costs of alcohol. Alcohol is responsible for a large amount of hospitalisations and deaths in Australia: 3.2% of Australia’s burden of disease and 4.6% of the global burden of disease.^2–4

Alcohol use by young people continues to be an area of particular research interest for several reasons. Alcohol consumption, particularly heavy alcohol consumption, adversely affects brain development in young people.^5,6 Hence, alcohol consumption may have adverse effects while the brain is developing into the early to mid-20s.⁶ Large proportions of young people consume alcohol in ways that place themselves and others around them at risk of acute harm because of falls, assaults, road crashes and other injuries.⁷ Therefore, research into the patterns of alcohol consumption, as well as revenue generated, is of interest given the substantial health and social impacts associated with alcohol use by young people.

Substantial proportions of young people consume alcohol.⁸ This extends even to young people who are below the legal alcohol purchase age (18 years) in Australia. For example, Doran et al. estimated that approximately 30% of Australian adolescents aged 12–17 years have consumed alcohol.⁹ In addition, substantial revenue is generated from alcohol consumption by young people in Australia in two forms: industry revenue and taxation. Further, studies by Doran et al. suggest that revenue generated by the consumption of alcohol by Australian adolescents increased over the period 2002–05.^9,10

The purpose of the present study was to update and extend estimates of alcohol consumption by young people in Australia, as well as the alcohol industry and taxation revenue generated. For the remainder of this paper, the following terms are used to refer to Australians in the various age groups examined: (1) those aged 12–17 years old are termed ‘adolescents’; (2) those aged 18–25 years are termed ‘young adults’; and (3) when referring to the entire sample age groups (12–25 years), the term ‘young people’ is used.

Some underlying assumptions regarding the alcohol purchasing habits of young people were made because of limitations in the data available for the present study. First, all alcohol was assumed to be accessed from ‘off-trade’ outlets (e.g. bottle shops), which are generally lower priced than ‘on-trade’ premises (e.g. bars and restaurants).¹¹ Sharma et al. distinguished between alcohol purchased from ‘off-trade’ and ‘on-trade’ sites;¹² however, this was not possible in the present study because of data limitations. Second, it was assumed that young people purchased in bulk to minimise the cost per unit. This may not hold for all young people who have limited disposable income because bulk purchases, despite having the cheapest unit price, generally cost more in absolute terms. However, to make a meaningful generalisation it was assumed that all young people were able to buy at the cheapest unit price. Third, it was assumed young people do not prefer to purchase premium alcohol products that are more expensive. Evidence for this has been found in the UK.^13–15 These assumptions are consistent with providing conservative estimates of revenue generated from alcohol consumption. All price and revenue estimates in this paper were adjusted for inflation and are expressed in 2014 Australian dollars.

The study sample consisted of 22 814 Australians (including 3601 young people) from the 2010 National Drug Strategy Household Survey (NDSHS). That survey collected information on the attitudes and behaviours towards alcohol consumption of individuals aged 12 years and above, across 26 648 households in Australia.¹⁶ Drinkers in the present study were defined as individuals who stated that: (1) they had consumed alcohol within the 12 months before the survey; and (2) continued to consume alcohol. The prevalence of alcohol consumption by age and gender was then estimated. Two-tailed Chi-squared tests were used to test for significant differences in drinker status across demographic groups, with P < 0.05 considered significant.

Annual consumption was determined using the graduated frequency (GF) method, based on the reported consumption of various quantities of alcohol in a day, for the past 12 months.¹⁷ Alternative consumption measures include the simpler ‘quantity frequency’ and the ‘yesterday’ methods. The former often fails to capture periods of heavy drinking outside of ‘usual’ routine, whereas the latter is prone to missing infrequent drinkers.^17,18

In the GF method, quantity responses were given in decreasing ranges down the page (20+ standard drinks, 11–19 drinks etc.), with frequency responses across the page starting with the most frequent (everyday, 5–6 days a week etc.). The midpoints of ranges were taken, except for the top quantity ‘20+’ where a value of 24.81 standard drinks was assigned based on the equally weighted day-of-the-week mean of ‘yesterday’ responses that were 20 drinks or more (n = 64). This allows for an empirically based estimate to what otherwise would be an arbitrary value for the top unbounded category. Midpoints for lower ranges have been shown to have little bias.¹⁹ A recurring issue of the GF method is respondents reporting more than 365 drinking days in a year, which we observed in 30% of our sample. This was adjusted by downward capping; that is, only the first 365 days starting from the highest consumption level were counted.¹⁹ Multiplying these quantity and frequency responses gave the annual standard drink consumption by gender and age, which were then extrapolated to the respective Australian population statistics in 2010.²⁰ The significance of differences in the quantity of drinks consumed by age and gender was tested using two-tailed t-tests, with P < 0.05 considered significant.

Respondents were asked to identify their main drink from 13 categories. This allows the consumption share of the various beverages by age and gender to be derived.

Price and product data, including alcohol content and container volume, were sourced from the Dan Murphy’s online alcohol retail store (August 2014; http://danmurphys.com.au/, accessed 6 August 2014), which provides more comprehensive data across a greater number of products than other online liquor retailers. After removing identical products with different packaging and bundled packs with non-alcoholic items, the initial extracted sample of 8143 products was reduced to 5234 unique alcoholic beverages.

Standard drinks were calculated according to guidelines from the Australian Tax Office.²¹ In Australia, one standard drink is equivalent to 10 g alcohol. This can be calculated using the beverage container volume and alcohol content data from the Dan Murphy’s website (http://danmurphys.com.au/, accessed 6 August 2014).

The Australian Government imposes a two-step tax on alcohol beverages in Australia. First is an alcohol tax, followed by a 10% Goods and Services Tax. Alcohol tax in Australia falls into two broad categories: (1) a volumetric ‘excise duty’ that is based on the volume of alcohol in the beverage; and (2) the value-based ad valorem Wine Equalisation Tax. The tax rate that applies depends on the type of alcoholic beverage. Because prices in the present study were taken from June 2014, tax rates as of June 2014 from the Australian Taxation Office were used.²¹ Import duties and excises were not considered in the present study.

Because alcoholic beverages are often sold in non-standardised containers (e.g. 750 mL wine bottles, 500 or 375 mL beer cans etc.), the price and tax of each product were divided by the number of standard drinks the beverage contains to obtain a price and tax per standard drink. This allows for standard units that can be compared across beverages.

The price per standard drink is equivalent to the unit total revenue generated, which was distributed to two main parties: (1) government via taxation; and (2) the alcohol industry. Total revenue for each beverage was calculated by multiplying the total number of standard drinks consumed annually (Table 1) with the price per standard drink for each beverage and its share of total consumption for each gender (Table 2). Aggregating these across sex and age groups yields total revenue from alcohol consumption (Table 3). Repeating the calculations with the tax per standard drinks yielded government tax revenue from alcohol consumption by young people (Table 3). The difference between total revenue and government tax revenue was industry revenue (Table 3).

Table 1.  Drinkers and standard drinks, Australia 2010
Note, because the data for those aged 12–15 years are sparse, they have been aggregated into one group. Percentage figures refer to the prevalence rate of alcohol for each age group, and gender

Table 2.  Price and tax per standard drink, Australia 2010
RTD, ready-to-drink

Table 3.  Industry and government tax revenue, Australia 2010
The 10 drink categories have been condensed to six broad groups. Low-, mid-and full-strength beers have been combined into ‘Beers’, whereas bottled, cask and fortified wines have been combined into ‘Wines’. RTD, ready-to-drink

In 2010, there were approximately 2.7 million young drinkers (aged 12–25 years), of whom approximately 518 000 were adolescent drinkers (Table 1). The prevalence of alcohol consumption increased with age, with much larger proportions (>50%) of those aged 16 and 17 year being drinkers than adolescents aged 12–15 years (P < 0.01). Among young adults aged 18–25 years, the prevalence rate was higher again compared with younger age groups, at approximately 85% (P < 0.01). Sixty-three per cent of young people were drinkers compared with 78% across the population (P < 0.01). Prevalence rates were similar between males and females (P = 0.086).

Table 1 also presents standard drink consumption by age and sex. Young adults consumed more alcohol at around 17 standard drinks per week per drinker compared with 8 standard drinks per week for adolescents (P < 0.01). The weekly consumption by young people of 16 drinks was also higher than the 15 standard drinks by the general population (P < 0.01). Male young adults consumed more standard drinks per week per drinker than female young adults (P < 0.01).

Ready-to-drink (RTD) beverages were the preferred main drink for male (41%) and female (65%) adolescent drinkers (Table 2). Other alcoholic drinks that were mainly consumed were spirits and beer. For young adults, there was a difference in alcoholic beverage preferences. First, although a substantial proportion of young adults indicated RTD beverages were their main drink (17% of male and 31% of female young adults), this was much lower than for their adolescent counterparts. Second, the preference towards beer almost doubled to 44% for young adult males, whereas for young adult females bottled wines more than tripled to 26%. Spirits consumption remained high and consistent at approximately 20%, regardless of age group and sex.

Price and tax per standard drink are presented in Table 2. Bottled wines, ciders, RTD beverages and spirits were the most expensive per standard drink. However, the taxes per standard drink collected were disproportionate to their prices, a result of the different tax rates applicable. RTD beverages were the most expensive and taxed the highest rate on a per standard drink basis.

Table 3 presents disaggregated revenue estimates from alcohol consumption in Australia. Young people generated $4.8 billion in total revenue in 2010 (note, results are presented in 2014 Australian dollars throughout), comprised of $2.8 billion (59%) in industry revenue and $2.0 billion (41%) in government tax revenue. Beers provided the most industry revenue (39% of total industry revenue) whereas RTD beverages generated the most tax revenue (33% of total tax revenue), followed closely by spirits (29% of total tax revenue).

Total revenue from adolescent drinkers was estimated to be $517 million, comprising $281 million (55%) in industry revenue and $235 million (45%) in government tax revenue. On a per drinker basis, each adolescent drinker generated a revenue of $997 per annum, whereas drinking by young adults generated $2000 per annum. RTD beverages were the main source of revenue to both industry and government from adolescent drinkers because of their higher tax rate and a strong consumption preference by this age group. For young adults, beer was the largest source of industry revenue ($1.05 billion) given the large shift in preference towards beer.

Although the focus of the present study was on young people, estimates for the Australian population aged 12–75 years old are included in Tables 1 and 3. Some useful observations are as follows: the 2.7 million young drinkers made up 20% of all drinkers in Australia and consumed 20% (2.1 billion) of total standard drinks. Tax revenue from young people accounted for 25% of population tax revenue and contributed to 21% of total revenue. Overall, young people generated $1805 in revenue per drinker per annum compared with $1663 for the general population.

Substantial proportions of adolescents (approximately one-third) and young adults (~85%) are drinkers. The average consumption for young people was approximately 16 standard drinks per week, whereas the Australian National Health and Medical Research Council’s recommendation is that young people below 18 years of age should not consume alcohol.²² Furthermore, substantial amounts of industry and governmental revenue were generated from alcohol sales to young people.

There are no reliable, publicly available estimates of government expenditure on measures to address harm from alcohol in Australia, including measures such as prevention strategies, education and treatment services. Expenditure on prevention of alcohol harm in particular is likely to be modest on the basis that it is estimated that only 1.5% of governmental health expenditure goes to public health.²³

Researchers and health groups have suggested that a proportion of alcohol taxation revenue could be hypothecated to prevent and reduce harm from alcohol.^24,25 Given that surveys have identified high levels of community concern about alcohol use by young people and associated harms, a useful area of investigation may be the potential to hypothecate alcohol taxation revenue generated by young people for the purpose of addressing harm from alcohol among young people.²⁶ For example, a proportion of alcohol taxation revenue could be directed to funding components of a comprehensive strategy to reduce harm from alcohol.²⁷

The substantial industry revenue generated by young people’s drinking raises questions about the appropriateness of the alcohol industry profiting from drinking by those under the legal alcohol purchase age. Although alcohol industry groups publicly state their commitment to reducing underage drinking,²⁸ researchers and health experts have expressed concern about the industry’s apparent conflicts of interest and questioned the effectiveness of industry-led initiatives that purport to address harm associated with alcohol.²⁹ For example, a study by Pettigrew et al. identified that a youth-targeted ‘responsible drinking’ campaign developed by an alcohol industry social aspects/public relations organisation may be counterproductive.³⁰

Furthermore, previous studies have shown that alcohol consumption is associated with large economic costs. To provide some perspective on the costs of alcohol consumption, Collins and Lapsley estimated alcohol-related social costs in Australia to be in excess of A$15 billion in 2004–05, with approximately two-thirds and one-third of these costs being attributed to tangible and intangible social costs, respectively.³¹ Manning et al. estimated direct costs of alcohol consumption to be approximately A$14.4 billion in Australia in 2010.³² These costs are substantial. Furthermore, because one of the main drivers of alcohol costs is loss of productivity, the adverse per capita lifetime cost of alcohol consumption for young people is expected to be even starker. It would be in the public interest to divert some revenue towards prevention policies in order to encourage less harmful levels of alcohol consumption.

The revenue and quantity estimates in the present study are larger than reported for 2005 by Doran et al. even after accounting for inflation.⁹ This is due, in part, to the use of the GF method, which has been shown in other studies to estimate larger quantities of alcohol compared with the weekly recall method used by Doran et al.⁹ The choice of the GF method was dictated by the data available from the NDSHS. Nevertheless, both methods have their respective merits.³³ Another potential reason for the higher estimates in the present study is that the sample also includes adolescents who do not attend school, and it is possible that their alcohol consumption may be higher. Larger revenues in the present study could also be the result of higher taxation and excise duties levied on alcohol products, as well as higher retail prices of alcohol beverages, since the study of Doran et al.⁹

The limitations of the study are largely associated with the assumptions outlined in the Methods section. First, it was assumed that drinkers only consumed one type of alcoholic beverage, namely the one they specified as their main drink in the NDSHS. Secondary drink responses are not considered because it was not possible to objectively determine the proportions of the main and secondary drinks consumed on the basis of the available data. Second, the prices from Dan Murphy’s are among the cheapest compared with other competitors.³⁴ Therefore, the revenue estimates reported herein are likely to be conservative.

The present study has provided updated estimates of drinking and its associated revenue for Australian adolescents, and extended the estimates to include young adults. The use of actual alcohol prices and product information from the unique dataset collected for the present study provides holistic and accurate price and standard drink estimates for the different alcoholic beverage categories.

A substantial number of young people were drinkers, and the average number of standard drinks consumed by young alcohol drinkers was high. The revenue estimates from the present study show there is significant revenue from alcohol sales in Australia. Given the high costs of alcohol consumption and the substantial revenue generated from young people, some revenue should be diverted towards education and prevention policies.

The authors acknowledge funding for this research from the McCusker Centre for Action on Alcohol and Youth.