Gambling activity in Australia

Findings from wave 15 of the Household, Income and Labour Dynamics in Australia (HILDA) Survey
Research Report – November 2017

Appendices

Appendix A: Comparisons with other Australian data sources

The statistics in this Appendix are intended to provide points of comparison between the HILDA Survey data and recent Australian gambling surveys.

Participation

Table 7.1 shows the past year participation rates reported in the most recent state and national gambling prevalence studies, alongside the monthly participation rates surveyed in the HILDA Survey.

Comparison with the two national gambling surveys suggests that the monthly gamblers identified in the HILDA Survey represented approximately 60% of Australians who gambled in the past year. Around two-thirds of past-year lottery participants were represented, indicating that most buy tickets on a regular basis. Casino table gaming was an occasional event for most, with only 15% of past-year participants represented in the HILDA Survey. Between 30-40% of past-year participants in all other activities appear likely to be regular gamblers.

Table 7.1: Past year participation rates in Australian gambling surveys compared to the monthly rate in the HILDA Survey
  HILDA National a National b NSW c Vic d Qld e SA f Tas g ACT h NT i
Year of data collection 2015 2012/
2013
2011 2011 2014 2011/
2012
2012 2013 2014 2015
Lottery 29.7 49.2 43.2 41 46.9 59 55.5 43.0 33.4 46.1
Instant scratch tickets 8.6 22.0 31.5 28 10.7 na 20.7 20.6 15.1 17.5
EGMs 8.2 20.7 19.4 27 16.7 30 26.5 18.6 30.2 22.9
Race betting 5.6 15.9 22.4 24 20.6 19 20.5 10.5 17.6 22.8
Sports 3.3 5.7 13.3 7 4.8 5 6.1 4.4 6.9 7.5
Keno 3.2 7.2 8.9 14 3.7 16 7.7 26.0 2.9 25.4
Casino games 1.1 5.9 8.7 6 5.1 6 6.1 6.3 5.8 13.4
Bingo 1.1 3.0 2.9 2 2.6 3 na 1.7 na 2.0
Any gambling 38.9 63.9 64.3 65 70.1 74 68.8 61.2 55.1 68.2

Notes: Excluding HILDA, all participation rates refer to past-year activity. Reference notations in table. Reference acronyms.

Sources: a Dowling et al., 2016; b Hing et al. 2014; c Sproston et al., 2012; d Hare, 2015; e Office of Economic and Statistical Research, 2012; f The Social Research Centre, 2013; g ACIL Allen Consulting et al., 2014; h Davidson et al., 2016; i Stevens et al., 2017.

Table 7.2 shows monthly participation rates reported by three Australian state and territory surveys. These are the only surveys to provide monthly participation statistics, and cover the smallest Australian jurisdictions. They nevertheless provide some points of comparison with the national monthly participation rates derived from the HILDA Survey. They illustrate the variability in activity participation between these jurisdictions and Australia as a whole.

Table 7.2: Monthly participation rates in Australian gambling surveys
  HILDA (National) Tasmania 2013 a ACT 2014 a NT 2015 a
Activity Adult population (%)
Lottery 29.7 23.1 16.3 15.2
Instant scratch tickets 8.6 5.4 3.5 2.7
EGMs 8.2 4.4 5.2 4.1
Race betting 5.6 3.4 2.8 4.6
Sports betting 3.3 1.5 2.6 2.7
Keno 3.2 7.8 0.2 5.3
Casino table games 1.1 0.7 0.7 0.8
Bingo 1.1 0.4 0.3 0.4
Private betting 0.9 0.5 na na
Poker 0.8 na na na
Any gambling 38.9 34.1 24.9 37.2
 Gambling population b (%)
Lottery 76.5 67.7 65.5 40.8
Instant scratch tickets 22.2 15.8 14.1 7.0
EGMs 21.1 12.9 20.9 11.1
Race betting 14.5 10.0 11.2 12.2
Sports betting 8.6 4.4 10.4 6.8
Keno 8.3 22.9 0.8 14.3
Casino table games 2.9 2.1 2.8 2.2
Bingo 2.8 1.2 1.2 1.0
Private betting 2.3 1.5 na na
Poker 2.0 na na na
Any gambling 100.0 100.0 100.0 100.0

Notes: na - not available. a calculated from data presented in report. b Any monthly gambling participation.

Sources: ACIL Allen Consulting et al., 2014; Davidson et al., 2016; Stevens, 2017.

Expenditure

Table 7.3 shows mean participant past-year expenditure by activity, derived from the HILDA Survey and the Tasmanian and Australian Capital Territory (ACT) prevalence surveys. These were the only Australian surveys to provide mean gambling expenditure statistics at the time of writing. As would be expected, the HILDA Survey estimates, which reflect the mean expenditure of regular gamblers, are much higher than the estimates from these two studies, which reflect the mean expenditure of those who gambled at least once in the past year.

Table 7.3: Mean past-year gambling expenditure in Australian gambling surveys
  HILDA (National) 2015 Tasmania 2013 a ACT 2014 a
  $ $ $
Lottery 695 448 337
Instant scratch tickets 248 94 72
EGMs 1,292 682 633
Race betting 1,308 1,186 309
Sports betting 1,032 551 200
Keno 425 286 na
Casino table games 1,369 395 225
Bingo 863 211 na
Any gambling 1,272 964 605

Notes: na - not available. a Mean spend per gambler (any participation in past year). HILDA Survey values based on weighted data and capped expenditure. All values expressed in 2015 dollars.

Sources: ACIL Allen Consulting et al., 2014; Davidson et al., 2016.

Table 7.4 shows total past-year gambling expenditure statistics supplied by the Australian gambling industry (Queensland Government Statistician's Office 2016), alongside totals derived from the HILDA, Tasmanian and ACT self-report surveys.

The total expenditure figures reported by industry are much higher overall than the estimates derived from the self-report surveys. A contributing factor is that the industry figures reflect the total past-year expenditure of all gamblers in the respective jurisdictions including tourists, whereas the Tasmanian and ACT survey figures reflect the sum of resident gamblers' typical expenditure in the past year. The surveys therefore exclude amounts from non-residents as well as untypically high spend events or periods on each activity. The HILDA Survey figure is even further limited to the past-year expenditure of regular resident gamblers on activities that they spent money on in a typical month. The amount gambled using overseas operators is also unknown, further limiting comparisons between Australian industry and gambler expenditure.

At the activity level, all three survey-based expenditure estimates for lotteries and instant scratch tickets are much higher than the figures reported by industry, whereas the estimates for race betting, EGMs and casino table games are much lower. In the case of lotteries and instant scratch tickets, it is clear that people over-estimate their expenditure. In the case of race betting, EGMs and casino table games, the difference is likely explained by a combination of "untypical" or unplanned over-expenditure, the expenditure of infrequent gamblers, and underestimations of expenditure by survey participants.

Table 7.4: Past year gambling expenditure reported in Australian gambling surveys and Australian Gambling Statistics industry survey
  Industry 14/15 HILDA Industry 12/13 TAS 13 Industry 13/14 ACT 14
Activity $M $M $M $M $M $M
Lottery 1,801 3,577 41.7 71.9 20.9 34.0
Instant scratch tickets 192.5 368 4.3 7.3 2.0 3.2
EGMs a 11,589 1,820 118.3 45.7 173.5 38.1
Race betting 2,815 1,265 41.1 47.2 23.4 16.4
Sports betting 815 579 1.9 8.4 - 4.2
Keno 330 226 30.4 27.6 0.7 0.4
Total gambling expenditure b 22,734 8,609 334.0 222.0 238.0 101.0

Notes: na - not available. HILDA values based on weighted data and capped expenditure. All values expressed in 2015 dollars. a Industry EGM expenditure data includes hotel and club but not casino expenditure. b Includes gambling activities not presented separately, such as casinos and bingo.

Sources: Queensland Government Statistician's Office, 2016; ACIL Allen Consulting et al., 2014; Davidson et al., 2016.

Gambling problems

Table 7.5 shows rates of gambling problems reported in recent Australian studies.

The HILDA Survey shows that 1.1% of Australian adults can be classified as problem gamblers. This is around twice the rate reported in most recent gambling studies. A major contributing factor is the HILDA Survey sampling frame. The HILDA Survey administers the PGSI to a population representative sample, whereas the gambling studies only administer it to people who gambled in the past year. The PGSI rates derived from the HILDA Survey therefore include people who may not have gambled in 2015, but nevertheless reported experiencing harms in 2015 associated with their prior gambling behaviour. For example, problem gambling in 2013 may have caused financial problems that stretched into 2015.

The HILDA Survey further shows that 2.1% of typical monthly gamblers could be classified as problem gamblers. This is around twice the rate among past-year gamblers reported in recent Australian surveys. This is because people with gambling problems participate more regularly than people without problems.

Table 7.5: Gambling problem rates in Australian gambling surveys
Report HILDA National a National b NSW c Vic d Qld e SA f Tas g ACT h NT i
Year of data collection 2015 2012/
2013
2011 2011 2014 2011/
2012
2012 2013 2014 2015
  Population (%) reporting gambling problems in the past year
Non-gambler & non-problem gambler 92.1 94.7 88.0 87.9 87.5 92.5 89.7 93.7 94.3 88.3
Low risk gambler 4.2 3.0 7.7 8.4 8.9 5.2 7.1 3.9 4.2 8.1
Moderate risk gambler 2.6 1.9 3.7 2.9 2.8 1.9 2.5 1.8 1.1 2.9
Problem gambler 1.1 0.4 0.6 0.8 0.8 0.5 0.6 0.5 0.4 0.7
  Gambling population (%)
Non-problem gambler 83.3 j 91.7 81.4 81.3 82.2 89.7 85.2 89.9 89.5 84.6
Low risk gambler 8.7 j 4.7 11.9 13.0 12.7 7.0 10.3 6.4 7.7 10.7
Moderate risk gambler 5.9 j 3.0 5.8 4.5 4.0 2.6 3.6 2.9 2.1 3.8
Problem gambler 2.1 j 0.6 1.0 1.2 1.2 0.7 0.9 0.8 0.8 0.9

Notes: Victoria, South Australia, and Queensland utilised a modified five response PGSI (Never=0, Rarely=1, Sometimes=1, Often=2, Always=3).

Sources: a Dowling et al., 2016; b Hing et al., 2014; c Sproston et al., 2012; d Hare, 2015; e Office of Economic and Statistical Research, 2012; f The Social Research Centre, 2013; g ACIL Allen Consulting et al., 2014; h Davidson et al., 2016; i Stevens et al., 2017; j Monthly gamblers only. 

 

Appendix B: Supplementary tables

Table 8.1: HILDA sample size by gambling activity
Activity Any expenditure on a typical month? How much per month? 
(On average)
Lottery 4,293 4,263
Instant scratch tickets 1,243 1,231
Electronic gaming machines 1,250 1,243
Race betting 818 810
Sports betting 474 461
Keno 496 484
Casino table games 125 121
Bingo 164 159
Private betting 115 110
Poker 102 97
Any gambling 5,742 5,709

Notes: only participants aged 18 and over are included

Table 8.2: HILDA sample size by risk group
  HILDA 
Self-Completion Questionnaire respondents
Regular gamblers
Activity N N
Non-gambler &/or non-problem gambler 13,398 4,776
Low risk gambler 598 502
Moderate risk gambler 376 340
Problem gambler 157 115
Respondents 14,529 5,733

Notes: Only participants aged 18 and over are included

 

Appendix C: Variable definitions

Subpopulation categories HILDA Survey variable Definition
Sex OHGSEX Male or female
Age group OHGAGE Grouped into age categories reflecting life stages, Young adult 18-29, early-middle age 30-49, later middle age 50-64, Older adults 65+
Indigenous status OANASTI Not of Indigenous back ground or Indigenous (combined Aboriginal, Torres Strait Islander, and Aboriginal and Torres Strait Islander categories)
Region of birth a OANCOB Grouped based on Australian Standard Classification of Countries into those born in Australia, Europe (North-west Europe and Southern and Eastern Europe), Asia (South-East Asia, North-East Asia, South and Central Asia), Other (Oceania and Antarctica excluding Australia, North Africa and the Middle East, Americas, Sub-Saharan Africa)
First language spoken OANENGF, OANLOTE Modified "Is English the first language you learned to speak as a child?" to code those reporting they did not speak a language other than English as "yes"
Highest education level OEDHISTS, OEDHIGH1 Classified based on number of years of schooling completed and level of post-school education obtained. Categories reflect standard levels of education. Below year 10, Completed year 10/junior secondary, completed year 12/senior secondary, certificate or diploma (cert. III or IV, adv. diploma or diploma), bachelors or higher
Employment OHGES Employed full-time (35+ hours per week), employed part-time (<35 hours per week) unemployed but looking for work, retired, non-working student, not employed and not looking for work (includes home duties)
Relationship status OMRCMS Considered married or in a de facto relationship if they reported being married or living with someone in a relationship, otherwise single
Household composition Adapted from OHHTYPE Single adult household (one adult aged 15 or more), Couple only household (2 persons aged 15+ who identify as a couple), household with children (one or more adults aged 15+ with one or more children aged less than 15) multiple adult household (2 or more persons aged 15+, excluding couple only households)
Housing tenure OHSTENR, OHSMGPD Own outright, own with a mortgage, rent (or pay board). A small number of participants reported other living arrangements (either living rent free or in rent-buy scheme). While these participants were included in calculating percentages their data is not presented due to small numbers
Remoteness OHHSRA Using ASGS 2011 Remoteness Area. Outer regional, remote and very remote combined due to low numbers
SEIFA quintile OHHSAD10 Collapsed from SEIFA 2011 decile of index of relative socio-economic advantage/disadvantage
Equivalised disposable household income quintile OHIFDITP, OHIFDITN Total household disposable income was equivalised for household size using the "modified OECD" scale, the first adult in the household as having a weight of 1 point, each additional person who is 15 years or older allocated 0.5 points, and each child under the age of 15 allocated 0.3 points. Equivalised household disposable income was then divided into 5 categories of roughly equal size. (<$29,500, $29,500-41,499, $41,500-53,999, $54,000-73,499, $73,500+)
Main source of household income OHIFISI, OHIFNISI, OHIWSFEI, OHIBIFIP, OHIBIFIN, OHIFPPI, OHIFINIP, OHIFININ, OHIFWFLF Main source of household income was determined by the largest contributor to total household income from either (a) salaries and wages, and business income; (b) government pensions, allowances or benefits (includes parenting payments and non-income support payments); or (c) superannuation, annuities or investments (including private pensions). A very small number of households received no income, or income from other sources. These were retained when calculating percentages but not presented due to small numbers
Grocery spend OHXYGRCI Total household expenditure on groceries. Includes food, cleaning products, pet food and personal care products. Does not include alcohol or tobacco
Utility spend OHXYUTLI Total household expenditure on electricity bills, gas bills and other heating fuel such as firewood and heating oil. Does not include water, telephone or internet bills
Household income quintile OHIFDITP, OHIFDITN Total (unequivalised) disposable household income was divided into 5 roughly equal groups (<$38,000, $38,000-63,749, $63,750-92,249, $92,500-131,999, $132,000+)
Problem Gambling Severity Index OGAPROB OGAMORE OGALARGE OGAHEALT OGAGUIL OGAFIN OGACRIT OGABORR OGAANDAY Responses to 9 individual items scored from 0 to 4 were summed and categorised as per standard PGSI thresholds, 0: non-problem gambler, 1-2: Low risk gambler, 3-7: Moderate risk gambler, 8 or above: Problem gambler. The PGSI was administered to all participants, and unless otherwise noted figures represent the total population, regardless of gambling expenditure.
Lottery participation OGALOTU Answered yes to expenditure on lotto or lottery games a typical month
Instant scratch tickets participation OGASCRU Answered yes to expenditure on instant scratch tickets in a typical month
Electronic gaming machines participation OGAPMU Answered yes to expenditure on poker machines or slot machines in a typical month
Race betting participation  OGABETHU Answered yes to expenditure betting on horse or dog racing (excluding sweeps) in a typical month
Sports betting participation OGABETSU Answered yes to expenditure betting on sports tickets in a typical month
Keno participation OGAKENU Answered yes to expenditure on keno in a typical month
Casino table games participation OGACASU Answered yes to expenditure on casino table (e.g., blackjack, roulette) games in a typical month
Bingo participation OGABINU Answered yes to expenditure on bingo in a typical month
Private betting participation OGAPBETU Answered yes to expenditure on private betting (e.g., playing cards or mah-jong with friends and family) in a typical month
Poker participation OGAPOKU Answered yes to expenditure on poker in a typical month
Any gambling participation OGALOTU, OGASCRU, OGAPMU, OGABETHU, OGABETSU, OGAKENU, OGACASU, OGABINU, OGAPBETU, OGAPOKU Answered yes to expenditure on at least one of the above gambling activities in a typical month
Annual lottery spends OGALOTA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Instant scratch tickets spend OGASCRA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Electronic gaming machines spend OGAPMA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Race betting spend  OGABETHA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Sports betting spend OGABETSA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Keno spend OGAKENA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Casino table games spend OGACASA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Bingo spend OGABINA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Private betting spends OGAPBETA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Poker spend OGAPOKA Average dollar spends per month multiplied by 12. Expenditure figures presented in this report have been capped at the top and bottom 1% using a Winsorising technique
Total gambling spends OGALOTA, OGASCRA, OGAPMA, OGABETHA, OGABETSA, OGAKENA, OGACASA, OGABINA, OGAPBETA, OGAPOKA Sum of all annual capped expenditure on gambling activities
% household income spends OGALOTA, OGASCRA, OGAPMA, OGABETHA, OGABETSA, OGAKENA, OGACASA, OGABINA, OGAPBETA, OGAPOKA, OHIFDITP, OHIFDITN Total capped annual gambling expenditure divided by household disposable income