8501.0 - Retail Trade, Australia, 2002  
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 04/02/2003   
   Page tools: Print Print Page Print all pages in this productPrint All

Feature Article - Contribution of Gambling to Retail Estimates


This article was published in Retail Trade, Australia (Cat. No. 8501.0), December 2002


INTRODUCTION

The October 1996 issue of this publication contained an article on the potential impact of gambling on short-term movements in the Retail Trade series. The article concluded that for the period January 1995 to June 1996 the effect was minimal.

Since 1996, the contribution of net proceeds received from licensed gambling activities to retail turnover has increased in different degrees in each state and territory and a number of users have expressed concern that growth in gambling could be significantly impacting movements in the Retail Trade series. To address these concerns, this article examines the impact of net proceeds received from licensed gambling activities in in-scope industries on the Retail Trade series from July 1996 to June 2002.


BACKGROUND

Any interpretation of the Retail Trade series requires a clear understanding of the industries included and the definition of turnover used. The industries included in the Retail Trade series, as defined in the Australian and New Zealand Standard Industrial Classification, 1993 (cat. no. 1292.0) (ANZSIC), are:

  • the Retail industry (ANZSIC Division G) except for Milk vending, Marine equipment retailing, Household equipment repair services, Motor vehicle retailing and Motor vehicle services
  • Non-petrol sales of convenience stores of selected petrol stations (ANZSIC Class 5321)
  • Pubs, taverns and bars (ANZSIC Class 5720)
  • Cafes and restaurants (ANZSIC Class 5730)
  • Clubs (hospitality) (ANZSIC Class 5740)
  • Video hire outlets (ANZSIC Class 9511)
  • Hairdressing and beauty salons (ANZSIC Class 9526).

Businesses which fall within the ANZSIC classes of Lotteries (9321), Casinos (9322) and Gambling services not elsewhere classified (9329) are not included in the Retail Trade series. As TABs are included in 'Gambling services not elsewhere classified' (ANZSIC class 9329), they are excluded from the Retail Trade series. However, the commission from TAB agencies operated by hotels and clubs is included.

Net proceeds from gambling activities are included in the estimates of total turnover for businesses in-scope of the Retail Trade series. Prior to 1994, businesses reported total turnover as one figure and, although net takings from gambling was included, it could not be separately quantified. Since the July 1994 reference month, businesses in the Pubs, taverns and bars and the Clubs (hospitality) classes1 have reported three components which sum to total turnover. This allows the contribution of net proceeds received from licensed gambling activities to be measured. The three components of turnover reported are:
  • Membership fees and subscriptions received
  • Net proceeds received from licensed gambling activities
  • General turnover (e.g. takings from bar and bottle shop sales, meals and accommodation, and hiring and booking fees).

Net proceds received from licensed gambling activities (called net proceeds from gambling in the rest of this article) includes:
  • commission from TAB and Keno transactions
  • net takings from Bingo
  • net revenue received from poker machines and other electronic gaming machines.

'Net revenue received from poker machines and other electronic gaming machines' means the commission earned or the amount remaining after payouts and government gambling taxes are removed. Takings from non-gaming coin-operated amusement machines are included as General turnover.

Businesses in the Newspaper, book and stationery retailing industry may receive commissions from selling Lotto and lottery tickets, but this information is not reported separately and is therefore not included in this article.


METHODOLOGY

The three components of turnover collected from Hotels and licensed clubs have primarily been used to assist businesses to correctly report total turnover and were not intended to be produced as separate estimates. The current investigation involved creating quarterly estimates of net proceeds from gambling for the Hotels and licensed clubs industry for the period from September quarter 1996. Procedures are being developed which will enable these data to be produced more readily in the future.

Estimates of total turnover have been subject to detailed imputation, validation and outlier treatment but this has not necessarily occurred for the turnover components. Previously when Hotels and licensed clubs did not report their monthly turnover, only their total turnover was imputed. To create the new net proceeds from gambling series, values were imputed for missing components of turnover. To ensure additivity, the components of turnover received the same outlier treatment as total turnover. The new series has been edited but, due to the elapsed time since the data were collected, the editing has not been as detailed as that for the total turnover series.

Over the last few years, the ABS has introduced both methodological improvements and enhanced statistical infrastructure. These improvements required revisions to the historical Retail Trade series to ensure it remained continuous. These revisions have been taken into account in producing estimates of net proceeds from gambling.

For estimates from September quarter 1998, there is an allowance for business provisions in the net proceeds from gambling series which means that this series and total turnover are on the same conceptual basis. Prior to this business provisions were included in estimates of total turnover but they could not be included in the estimates of net proceeds from gambling. This means estimates of net proceeds from gambling will be understated for periods prior to the September quarter 1998. However, this understatement is unlikely to be significant. It is estimated that for June month 1998 the understatement of net proceeds from gambling is about $8m (about 2% of net proceeds from gambling).

Data for the period July 1995 to June 1996 were included in the previous article but have not been included in this study as the data could not be accurately adjusted for the improvements mentioned above. For similar reasons, data by state are only available from July 1998.

As the net proceeds from gambling, the Hotels and licensed clubs and total retail turnover series all display strong seasonality, the data presented in this article have been seasonally adjusted on a quarterly basis. The seasonal adjustment process for the quarterly series is similar to that used for the main Retail Trade series. However, the time span available is fairly short (especially for the state series) and seasonal patterns may not be fully established until more data are available, and trading day patterns cannot be detected with this small time span.

The net proceeds from gambling series is subject to the same sampling and non-sampling errors associated with the total turnover series. As with the total turnover series, every effort is made to minimise non-sampling error. The following table provides an indication of the sampling error associated with these estimates in original terms. The standard error associated with the seasonally adjusted series will be similar.


STANDARD ERRORS

Relative
Standard
Standard
Estimate
error
error
Data series
$m
$m
%

Total retail turnover
March quarter 1999
33,455.8
314.7
0.9
March quarter 2002
40,811.3
408.1
1.0
Net proceeds from gambling
March quarter 1999
1,043.3
77.2
7.4
March quarter 2002
1,425.4
97.0
6.8



The ABS recommends that users exercise some caution in interpreting movements for series with a relative standard error higher than 5%.


GAMBLING IN AUSTRALIA

As mentioned above, not all gambling industries are in-scope of the Retail Trade series. In 2002 the ABS released some data which places this in perspective. Gambling Industries, Australia, 2000-01 (cat. no. 8684.0) provides an overview of gambling industries and Clubs, Pubs, Taverns and Bars, Australia, 2000-01 (cat. no. 8687.0) provides more detail on these industries.

These sources estimated that in 2000-01 total net income from gambling was $13,839m, of which the Hotels and licensed clubs industry received $5,957m or 43% of the total net income.

The estimate of net income from gambling in 2000-01 of $5,957m for Hotels and licensed clubs is prior to gambling taxes and levies of $1,361m being deducted. If these taxes and levies are excluded, the resulting estimate of net proceeds from gambling of $4,596m is comparable to an annualised Retail Trade estimate for 2000-01 of $4,682m. The two estimates differ slightly as the two series are compiled for different purposes using different samples and slightly different methodologies and concepts.


FINDINGS

From September quarter 1996 to June quarter 2002, the Australian Retail and Hospitality/Services series grew by 33% (in seasonally adjusted terms) while Hotels and licensed clubs grew by 25%. However, net proceeds from gambling grew by 86% over this period.

Over the six-year period to June quarter 2002, net proceeds from gambling has increased from about 2.5% of the Australian Retail and Hospitality/Services turnover series to about 3.5%. Over this same period, net proceeds from gambling has increased from about 25% of the turnover for Hotels and licensed clubs to about 36%. In both cases the increase has occurred gradually with no one particular period displaying a significant increase in contribution.


GAMBLING IN RETAIL TRADE, AUSTRALIA, SEASONALLY ADJUSTED

HOTELS AND
GAMBLING
TOTAL RETAIL
LICENSED CLUBS



Net
Proportion of
Proportion of
proceeds
Turnover
gambling
Turnover
gambling
Quarter
$m
$m
%
$m
%

1996
September
788.6
31,939.5
2.5
3,196.9
24.7
December
808.3
31,920.7
2.5
3,218.5
25.1
1997
March
782.8
32,286.9
2.4
3,236.7
24.2
June
842.0
32,724.6
2.6
3,343.6
25.2
September
865.2
33,304.6
2.6
3,327.1
26.0
December
885.4
33,590.7
2.6
3,370.3
26.3
1998
March
935.5
33,524.9
2.8
3,355.3
27.9
June
1,015.6
34,163.9
3.0
3,532.1
28.8
September
958.1
33,342.9
2.9
3,054.3
31.4
December
975.4
33,661.8
2.9
3,107.1
31.4
1999
March
1,014.6
34,429.8
2.9
3,211.2
31.6
June
1,060.8
34,742.4
3.1
3,248.6
32.7
September
1,101.9
35,719.7
3.1
3,333.8
33.1
December
1,106.9
36,194.3
3.1
3,387.4
32.7
2000
March
1,126.2
35,768.8
3.1
3,332.5
33.8
June
1,101.6
36,740.9
3.0
3,294.8
33.4
September
1,139.3
36,766.1
3.1
3,438.1
33.1
December
1,166.5
37,553.3
3.1
3,496.4
33.4
2001
March
1,170.9
38,400.5
3.0
3,545.3
33.0
June
1,212.7
39,255.3
3.1
3,678.8
33.0
September
1,290.4
39,959.8
3.2
3,817.4
33.8
December
1,327.8
40,808.9
3.3
3,743.6
35.5
2002
March
1,348.3
41,317.5
3.3
3,778.8
35.7
June
1,469.7
42,438.8
3.5
3,995.8
36.8



The next table shows net proceeds from gambling as a proportion of Retail and Hospitality/Services turnover for each state and territory. The states/territories with the largest proportion are New South Wales and the Australian Capital Territory while Western Australia and Tasmania have the smallest proportion.

Each state and territory places some restrictions on the number of gaming machines, whether it be a total cap and/or a cap per hotel or club. The low contribution of net proceeds from gambling in Western Australia reflects poker machines in that state being restricted only to Casinos. In addition, Hotels and licensed clubs in Victoria and Tasmania only receive a commission or venue share for poker machines located on their premises and this amount is proportionally less than the net takings received by Hotels and licensed clubs in other states and territories.


PERCENTAGE OF NET PROCEEDS FROM GAMBLING BY STATE, SEASONALLY ADJUSTED

New
Australian
South
South
Western
Northern
Capital
Wales
Victoria
Queensland
Australia
Australia
Tasmania
Territory
Territory
Quarter
%
%
%
%
%
%
%
%

1998
September
5.7
1.1
1.7
2.2
-
0.2
1.0
4.1
December
5.9
1.1
1.7
2.4
0.1
0.1
1.1
4.2
1999
March
6.1
1.2
1.7
2.4
0.1
0.3
1.0
4.2
June
6.1
1.2
2.1
2.5
-
0.3
0.9
4.2
September
6.1
1.2
1.8
2.3
0.1
0.4
0.9
4.2
December
6.0
1.3
2.1
2.3
-
0.6
0.8
4.1
2000
March
6.1
1.3
2.3
2.3
-
0.8
1.0
4.2
June
5.8
1.3
2.3
2.6
-
0.7
1.1
3.9
September
5.7
1.5
2.3
2.6
-
0.6
0.9
4.2
December
5.8
1.3
2.5
2.9
0.1
0.6
0.9
4.3
2001
March
5.8
1.2
2.3
3.0
-
0.6
0.9
4.4
June
5.9
1.3
2.3
2.8
0.1
0.9
0.9
4.4
September
6.0
1.3
2.4
3.1
0.1
0.9
1.5
4.2
December
6.1
1.6
2.3
2.9
0.1
0.7
1.6
4.2
2002
March
6.0
1.6
2.6
3.1
0.1
0.6
1.7
4.4
June
6.5
1.6
2.7
3.2
0.1
0.5
1.5
4.4

- nil or rounded to zero (including null cells)


From September quarter 1998 to June quarter 2002, the proportion of net proceeds from gambling to total turnover increased by 1.0 percentage point (pp) in Queensland and South Australia, 0.8 pp in New South Wales, 0.5 pp in Victoria and Northern Territory and 0.3 pp in Tasmania and the Australian Capital Territory. There was no significant change in the proportion for Western Australian.

Although the increase in proportion for each state and territory generally occurred gradually, the increase in the proportion for New South Wales in the June quarter 2002 (0.5 pp) was considerably higher than the increase in previous quarters. Victoria, Queensland and Northern Territory also showed an increased proportion in the last few quarters of the series. The contribution in Tasmania was higher in 2000 and 2001 (between 0.6 and 0.9%) than it was in June quarter 2002 (0.5%).

The following chart compares quarterly, current price growth rates of the Australian Retail and Hospitality/Services turnover series and the same series excluding net proceeds from gambling. The latter series includes non-gambling revenue from Hotels and licensed clubs. A chart on page five of this publication shows the seasonally adjusted and trend monthly levels of the Australian Retail and Hospitality/Services turnover series excluding all turnover for the Hospitality and services industry.

Graph - Comparison of total and total minus gambling



The chart shows that over the last two years, the two series have had the same quarterly growth rate for five of the eight quarters but the series excluding net proceeds from gambling had weaker growth for the remaining three quarters. Although only two years have been graphed for space reasons, the previous two years show a similar pattern, but in the earliest years, 1996 to 1998, the growth in the total series is mostly slightly higher than the series without gambling. However, the growth rates were not significantly different on any occasion.


CONCLUSION AND FUTURE RELEASE OF DATA

This investigation has confirmed that net proceeds received from licensed gambling activities has increased as a proportion of the Australian Retail and Hospitality/Services turnover series but that the contribution has not had a significant impact on quarterly movements for the series. However, the contribution from net proceeds from gambling has increased over time and this growth is particularly evident in the Hotels and licensed clubs series. Users should be aware of this when interpreting the Australian Retail and Hospitality/Services turnover series. In future issues of this publication, the ABS will provide additional information to assist users in monitoring the proportion of net proceeds from gambling to total turnover.

In the short-term an extra paragraph will be included in the explanatory notes of this publication mentioning the contribution of gambling to the Australian Retail and Hospitality/Services series and the contribution to the latest estimates will be updated quarterly. In the medium-term, the series for net proceeds received from licensed gambling activities will be made available electronically.


RELATED IN FORMATION

The ABS produces a more comprehensive range of data on gambling via its service industries program. Detailed surveys of the Pubs, taverns and bars industry, Clubs industry and Casino industry are usually conducted every three years. For more information about the service industries program please contact Graham Boxsell on (02) 6252 5633.


FURTHER ENQUIRIES

For more information about this article please contact Kate Bradbury on (02) 6252 5730.