Apparent Consumption of Alcohol, Australia methodology

1 This publication provides estimates of apparent consumption of alcohol based on the availability of alcoholic beverages in Australia. It provides estimates of the quantity of pure alcohol available for consumption in the form of beer, wine, spirits, and Ready to Drink (pre-mixed) beverages (RTDs), plus estimates of the total volume of beer and wine available for consumption. Estimates of the quantity of pure alcohol available for consumption from cider are included from 2004-05 onwards.

2 Also provided are estimates of apparent per capita consumption for each of the above alcoholic beverages, for the population aged 15 years and over. This population is consistent with international standards for measuring trends in apparent consumption over time. See paragraphs 51 and 52 for information regarding population estimates.

3 Given the assumptions underlying this data, the data is most useful as a guide to trends in alcohol consumption in Australia.

4 Data are only available on an annual basis, at the national level. Data are not available by particular demographic characteristics (such as state/territory, region, age, sex or country of birth). As such, it is not possible to account for any effect that changes in the age structure of the population over time may have had on the apparent consumption of alcohol in Australia. Data derived from individuals' self-reported consumption are available from ABS National Health Surveys, and can be disaggregated by a number of demographic variables such as age, sex and geography. In this publication, ABS National Health Survey data has been used to derive estimates of apparent consumption of cider (see paragraphs 38 to 50 for more information).

5 The 2017-18 domestic wine consumption estimates are derived using transaction data and Wine Australia projections (see paragraphs 21 to 33 for more information).

6 The assumed alcohol content of wine has been revised in this publication resulting in revisions to the consumption of pure alcohol in wine and total pure alcohol consumed series from 2008-09 onwards. These revisions have resulted in an increase in the alcohol content for white and red table wine (see paragraph 30 for more information).

7 The assumed proportion of total apparent alcohol used to calculate cider consumption for 2015-16 and 2016-17 were revised following the release of the 2017-18 National Health Survey (see paragraphs 38 to 50 for information about how cider is derived indirectly from ABS National Health Surveys).

8 Estimated Resident Population used to calculate per capita consumption figures were revised for 2011-12 to 2016-17 following the rebasing which aligns all ABS population data to the new Census base (see feature article). This update will impact all per capita consumption estimates from 2011-12 onwards.

9 Some estimates for beer, wine, spirits and RTDs for 2011-12 onwards have been revised in line with revisions to the Estimated Resident Population. As cider estimates are calculated using total alcohol consumption, these estimates have also been revised in some cases. Revisions have caused only marginal differences in the data.

10 The scope of this collection is beer, wine, spirits, RTDs and cider (from 2004-05 onwards) available for consumption in Australia.

11 Data for beer, wine, spirits and RTDs are collected from import clearances via the Australian Government Department of Home Affairs (DHA), excise tariff data from the Australian Taxation Office (ATO) (which only applies to alcohol sold in Australia and is collected for beer and spirits only), domestic sales of Australian produced wine from winemakers and an indicator based on domestic sales data to derive a 2017-18 wine estimate. For more information refer to paragraph 21. Data for beer and wine also contains an estimated component for home production. Data for cider has been derived indirectly from ABS National Health Surveys.

12 It should be noted that estimates of 'apparent consumption' of alcoholic beverages (excluding cider) are derived using information related to supply (that is, data on domestic sales of Australian produced wine, excise data on alcohol produced for domestic consumption, data on imports and an estimated component for home production). No adjustments are made for:

• changes in stocks;
• duty-free alcohol imported by individual overseas travellers;
• alcohol which is imported into Australia, cleared through a bonded warehouse and then subsequently re-exported; or
• alcohol that has been stored or cellared, used in the preparation of food or discarded as waste.
   

13 All alcohol available for consumption in a particular year is assumed to have been consumed in that year.

14 Import clearance data are used in this publication to measure the quantity of alcohol imported into Australia. Import clearances relate to goods which are brought into Australia directly for home consumption, plus goods cleared from a bonded warehouse (that is, goods cleared into the Australian market for home consumption following payment of duty). Refer to International Merchandise Trade, Australia: Concepts, Sources and Methods, 2018 (cat. no. 5489.0) for more information.

15 Data provided by the ATO are administrative by-product data collected for the levying of excise tariffs.

16 Data sources relating to wine consumption is provided in paragraphs 21 to 29 for more information.

17 Estimates of the volume of beer available for consumption (in terms of total volume and volume of pure alcohol) are obtained from excise data on Australian production and import clearance data, as well as an estimated amount for home production.

18 Changes were made to excise data for beer provided by the ATO due to excise tariff reform in July 2006. Since then, only data on the dutiable quantity of alcohol in beer is provided to the ABS by the ATO. Data on the first 1.15% of alcohol in beer (which does not attract excise) and data on the total volume of beer is no longer available, therefore this data is estimated using separate strength estimates for packaged and bulk beer for each of the three beer strengths. This means the total quantity of alcohol and total volume of beer available for consumption, and apparent per capita consumption for beer, may not be directly comparable with data before 2006-07. The table below shows the average alcohol strengths of domestically produced beers for 2012-13 to 2017-18.

19 As a result of excise tariff reform in July 2006, beer brewed on commercial premises for non-commercial purposes not separately identified previously was introduced to the ATO collection. Beer less than 3% volume of alcohol is included in low strength beer, while beer greater than 3% volume of alcohol is included in full strength beer (as the amount of mid strength beer brewed on commercial premises for non-commercial purposes is negligible).

20 Prior to 2008-09, figures for beer included an estimated component for home production which was based on the survey Home Production of Selected Foodstuffs, Australia (cat. no. 7110.0), conducted in 1992. After a review into the estimated component for home production, incorporating advice from industry, the estimate for the home production of beer was marginally increased from 2.1% of total domestic beer available for consumption to 2.2%.

21 This issue incorporates a new methodology for estimating wine consumption for the 2017-18 reference year. Unlike the years prior to 2016-17 where wine estimates were collected using ABS surveys and combined with imports data from customs, the 2017-18 estimates are derived by applying a projection to the 2016-17 estimates. This projection is based on indicator assumptions derived from historical transaction data in combination with current transaction data and estimates from Wine Australia’s domestic production and sales report (see report) in addition to imports data from customs. This method takes account of all available information on the sales of wine in the domestic market, while remaining consistent with the historical series.

22 The 2016-17 estimates are a projection derived by applying an estimate of annual growth to the Domestic Wine Sales estimates from the ABS 2015-16 survey. The growth rate was developed using Wine Australia domestic sales estimates as an indicator. The domestic sales estimates can be found in Wine Australia's annual publication Australian wine: Production, sales and inventory 2016-17. Because this indicator has been produced annually for a number of years, ABS has been able to confirm the validity of the annual indicator by applying it to the 2013-14 ABS wine sales estimates (see Shipments of Wine and Brandy in Australia by Australian Winemakers and Importers, June 2014 (cat.no 8504.0) to derive a parallel annual series to compare with the published apparent consumption of alcohol time series estimates. As the difference in apparent consumption estimates between these series was found to be less than 2% in any year, they are considered within acceptable limits and fit for the purpose of contributing the wine component to the Apparent Consumption of Alcohol series.

23 The 2015-16 wine consumption estimates were collected in the Sales of Australian Wine by Winemakers survey. This collection was a sample survey of businesses from the ABS business register.

24 The 2015-16 Sales of Australian Wine by Winemakers survey surveyed wine making business entities that were classified to ANZSIC 0131 or ANZSIC 1214 or were considered to be significant contributors to the wine industry. Businesses with less than 5 employees were excluded from the sample frame and the survey aimed to completely enumerate businesses with 50 or more employees or those who were considered to be significant contributors to the wine industry. Some 149 businesses were selected in the 2015-16 wine survey (68 in the completely enumerated stratum) and the survey achieved a response rate of 85.9%.

25 Prior to the 2015-16 collection, the survey was a partial coverage census of wine making businesses with sales of at least 250,000 litres or more in either of the previous two financial years. For more information on the collections prior to 2015-16 see Shipments of Wine and Brandy in Australia by Australian Winemakers and Importers, June 2014 (cat.no 8504.0).

26 Other than the change in scope, the Sales of Australian wine by Winemakers 2015-16 survey and 2013-14 survey are largely comparable, with two differences to note. The reference period for the 2015-16 survey was one financial year (2015-16), while the 2013-14 and previous years data were based on the sum of four quarterly surveys. Respondents to the survey in 2013-14 and prior years were asked to report "Other wine products" including wine cocktails, marsala, apertif, tonic wines, de-alcoholised, low and reduced alcoholic wines, flavoured wine and vermouth at the final question in the survey. In the 2015-16 survey this question was changed to specifically only ask for "Other alcoholic wine products" including all of the above and specifically asking respondents to exclude Non-alcoholic wine. This may be contributing to the reduced rates of "other" wine within this publication (which also includes sparkling, carbonated wine and fortified wine).

27 As a sample survey the 2015-16 Sales of Australian Wine by Winemakers survey is subject to a small sampling error. This small error has not been published as the error decreases to marginal when the data is combined with imports data (and then further when combined with beer, spirits and cider data to derive total alcohol). Further, the surveys conducted prior to 2013-14 to collect domestic wine data were subject to non-sampling error due to the lack of coverage of smaller wine producers.

28 The Sales of Australian Wine by Winemakers survey was not conducted in 2014-15. In order to facilitate a complete 2014-15 dataset (i.e. to be able to publish total alcohol), estimates for domestic wine in 2014-15 are interpolated using the average change of each type of wine over 5 previous years' data. Interpolated 2014-15 domestic wine data were combined with 2014-15 imports clearances from the DHA to produce the 2014-15 wine data used in this publication.

29 Whilst the change in methodology between 2013-14 and 2017-18 in addition to the interpolation method applied to calculate the 2014-15 data is not expected to have a significant impact on the final results, comparisons between these years should be interpreted with caution.

30 In preparing the 2017-18 issue of the Apparent Consumption of Alcohol, ABS in consultation with the Australian Wine Research Institute (AWRI), updated the assumed alcohol content of white and red table wine. This update is based on analytical samples of Australian wine vintages from 1989 to 2017 made by AWRI using scientific analyses to measure the alcohol content using an internationally accredited laboratory facility. Changes in alcohol strength in wine over time reflects the impact changing environmental conditions, industry practices and consumer preferences have had on wine. The alcohol strength estimates were revised back to 2008-09, and these revisions result in an increase on average of 0.4 percentage points for white wine and 0.7 percentage points for red wine. The assumed alcohol strength of sparkling and carbonated, vermouth, fortified and other wines have not changed. This update results in revisions to the pure alcohol consumed from wine and subsequently total consumption of pure alcohol from 2008-09 onwards.

31 Earlier research by AWRI (Godden and Gishen, 2005) indicated that, overall, the average alcohol content of wines in Australia have increased since the mid 1980s. As a result, for the publication Apparent Consumption of Alcohol: Extended Time Series, 1944-45 to 2008-09 (cat. no. 4307.0.55.002), assumptions used in the calculation of alcohol in table wine were revised back to 1980-81. As volumes of red and white wine are available from 2000-01 onwards, separate assumptions were made for red and white wine for these years. Similarly, the alcohol content of sparkling wines was assumed to increase linearly between 1979-80 (10.6%) and 2008-09 (11.2%). The same assumptions for wine are used in this publication. Assumed concentrations of alcohol in wine are listed in the following table.

32 From 2009-10 onwards, data on domestic sales of vermouth are not available separately, but are included in 'Other wine not elsewhere included'. While alcohol content for these two categories differ, the very small amount of vermouth produced in recent years (less than 0.2% of total domestic wine volume in 2008-09) means that the inclusion of vermouth in 'Other wine not elsewhere included' has a negligible effect on the total volume of pure alcohol in wine.

33 Changes made to customs tariff codes from 1 January 2012 have resulted in the combining of previously discrete wine codes for some wine types. See Information Paper: Changes to AHECC and Customs Tariff, 2012 (cat. no. 5368.0.55.017) for more information. Sensitivity analysis performed by the ABS indicates that assumptions made about the alcohol content of these combined codes have only a negligible effect on the total volume of pure alcohol in wine.

34 For spirits and RTDs the amount of alcohol available for consumption is only available as the quantity of pure alcohol. Data are obtained from import clearance data from the DHA and excise data on domestic production of spirits from the ATO, with an adjustment to account for the excise paid on imported spirits which are commercially mixed with locally manufactured soft drinks after importation. Since 2003-04 the excise data used in these estimates have been obtained from the ATO. In previous years, excise data was obtained from the DHA.

35 Prior to 2011-12, estimates of the volume of pure alcohol of imported flavoured cider obtained from customs had been included in estimates of spirits. However, as cider was reported separately for the first time in the 2011-12 publication for 2004-05 onwards, imported flavoured cider has been removed from spirits estimates for 2004-05 onwards. This has a negligible effect on spirits, as the volumes of imported flavoured cider during this period were only small.

36 RTDs can include spirit based, wine based and other unspecified based products.

37 Import clearance data used to estimate RTDs are 'distilled alcoholic beverages not elsewhere specified' and 'spirituous beverages not elsewhere specified' (both having an alcoholic strength by volume exceeding 1.15% but not exceeding 10%).

38 Historically, it has not been possible to include estimates of the amount of alcohol available for consumption in the form of cider in this publication, as, unlike beer, wine, spirits and RTDs, no suitable source of data on either volume of cider or volume of pure alcohol in cider has been available. As cider has made up only a relatively small proportion of all alcohol consumed in Australia, this absence of data has not negatively affected the overall accuracy of estimates of the total volume of alcohol available for consumption.

39 Recent years, however, have seen an increase in the popularity of cider. Given this increase, some estimate of apparent consumption of cider is required.

40 Unlike estimates of apparent consumption of beer, which are calculated according to the dutiable quantity of alcohol in beer and obtained from the DHA and ATO, cider is in general taxed according to the price at which the cider is sold (under the Wine equalisation tax). The total monetary value of this tax cannot be used to derive estimates of the volume of cider available for consumption, as prices of cider vary by brand and by volume of container. Taxation data is therefore not a source of data for cider.

41 In the absence of a direct source of information, in the 2011-12 and subsequent issues of this publication the volume of pure alcohol available for consumption in the form of cider has been derived indirectly from the 2004-05, 2007-08, 2011-12, 2014-15 and 2017-18 ABS National Health Surveys using self-reported information on individuals' consumption of alcohol in conjunction with total apparent consumption of alcohol.

42 The National Health Survey (NHS) is a population survey conducted by the Australian Bureau of Statistics on a regular basis (roughly every 3 years) that collects a range of health-related information. Included in the survey are a number of questions on individuals' consumption of alcohol in the week prior to the survey interview.

43 Each survey respondent who had consumed alcohol in the week prior to interview was asked questions about the type, brand, number and serving sizes of the beverages they had consumed. This information was used to calculate, for the main beverage types, both the total volume consumed and the total volume of pure alcohol consumed, for all Australia.

44 It should be noted that this information is not sufficient to determine total annual consumption of cider, or indeed other alcoholic beverages, in Australia. This is because data is collected for individuals' consumption over a single week only, and also due to under-reporting of consumption of alcohol in population surveys (both in terms of persons identifying as having consumed alcohol in a specific period, and in the quantities reported). For a discussion of potential data quality issues associated with the collection of alcohol data in the National Health Survey, see Alcohol Consumption in Australia: A Snapshot, 2007-08 (cat. no. 4832.0.55.001).

45 Instead, proportions of each beverage type relative to all beverages were calculated from the NHS (see table below). While some variation exists, the proportions for each beverage are relatively consistent between the NHS and Apparent Consumption of Alcohol data. For example, beer comprised 45.1% of total pure alcohol consumed according to the 2014-15 NHS, similar to the proportion recorded in Apparent Consumption of Alcohol for the same year (38.8%). This indicates that in proportional terms, the pattern of alcohol consumption in the NHS is similar to overall apparent consumption of alcohol, and can therefore be used to indirectly calculate estimates of the volume of pure alcohol from cider for inclusion in apparent consumption of alcohol data.

46 According to the 2017-18 NHS, cider comprised 2.5% of the total volume of pure alcohol consumed. It was therefore assumed that cider would make up 2.5% of the total volume of pure alcohol of beer, wine, spirits and RTDs combined in the apparent consumption of alcohol collection in 2017-18 (186 566 million litres; see table below). Applying 2.5% to this figure results in an estimate of 4781 million litres of pure alcohol of cider available for consumption in 2017-18, and therefore 191 247 million litres of pure alcohol available for consumption from all beverages.

47 Estimates of cider have not been produced for years prior to 2004-05 as information is not available from earlier National Health Surveys.

48 For years in between the National Health Surveys, the trend observed between the previous two surveys is applied to the proportion used to calculate the contribution of cider to pure alcohol available for consumption. So 3.1% and 2.8% have been applied to the 2015-16 and 2016-17 data respectively to reflect the trend between the 2014-15 and 2017-18 National Health Surveys.

49 It should be noted that the method described above is an approximation, but allows an assessment of the relative impact of recent patterns of cider consumption on the total level of apparent consumption of alcohol in Australia.

50 Updating the assumed alcohol content of wine, results in revisions to consumption of pure alcohol in wine but also the total pure alcohol consumed time series starting from 2008-09 onwards. As cider is derived as a proportion of total alcohol consumption, any changes to this series impact cider estimates. However, the revisions are considered minor with negligible impact on the overall apparent consumption estimates.

51 Apparent per capita consumption data included in this publication are calculated by dividing the quantity of beverage or pure alcohol available for consumption by the estimated resident population of Australia of persons aged 15 years and over in Australia at 31 December each year. Please note that the adjusted estimates by beverage type are not available in the data cube.

52 For more information on population estimates see Australian Demographic Statistics (cat. no. 3101.0).

53 Adjustments are based on the NHS, which can provide information on the frequency of alcohol consumption among the population and the proportion of drinkers who consumed each type of alcoholic beverage based on consumption in the last week, while accounting for the population aged 15 years and over who consumed alcohol in the last 12 months.

The most recent NHS was conducted in 2017-18 and included a sample of approximately 21,300 people in 16,400 private dwellings across Australia.

54 The Census and Statistics Act 1905 provides the authority for the ABS to collect statistical information, and requires that statistical output shall not be published or disseminated in a manner that is likely to enable the identification of a particular person or organisation. This requirement means that the ABS must take care and make assurances that any statistical information about individual respondents cannot be derived from published data.

55 Percentages and percentage movements have been calculated using un-rounded numbers, and may differ from figures obtained from rounded numbers presented in tables.

56 ABS publications draw extensively on information provided freely by individuals, business, governments and other organisations. Their continued cooperation is very much appreciated; without it, the wide range of statistics published by the ABS would not be available. Information received by the ABS is treated in strict confidence as required by the Census and Statistics Act 1905.

57 Other ABS products which may be of interest to users include:

• Alcohol Consumption in Australia: A Snapshot, 2007-08 (cat. no. 4832.0.55.001)
• National Health Survey: First Results, 2017-18 (cat. no. 4364.0.55.001)
• Apparent Consumption of Alcohol: Extended Time Series, 1944-45 to 2008-09 (cat. no. 4307.0.55.002)
• Shipments of Wine and Brandy in Australia by Australian Winemakers and Importers (cat. no. 8504.0)
• Australian Aboriginal and Torres Strait Islander Health Survey: First Results, 2012-13 (cat. no. 4727.0.55.001)
   

58 ABS products and publications are available free of charge from the ABS website http://www.abs.gov.au. Click on Statistics to gain access to the full range of ABS statistical or reference information.

59 As well as the statistics included in this and related publications, the ABS may have other relevant data available on request. Enquiries should be made to the National Information and Referral Service on 1300 135 070 (email: client.services@abs.gov.au).