The Scottish Government

« Previous | Contents | Next »

Listen

CHAPTER EIGHT: GAMBLING IN SCOTLAND

8.1 This chapter presents an analysis of new data from the British Gambling Prevalence Survey, conducted in 1999, to outline patterns of gambling participation in Scotland. It also uses information about patterns of deprivation to comment on likely geographic patterns of problem gambling in Scotland.

Participation in Gambling in Scotland

8.2 In 1999, the National Centre for Social Research conducted the first nationally representative survey of gambling in Britain. The aim of the survey was to provide baseline and prevalence data on adult gambling behaviour. In total, 7,680 people aged 16 and over participated in the survey, 84% of whom lived in England, 10% in Scotland and 6% in Wales. ¹⁵ Respondents were selected at random, using the Postcode Address File ( PAF), and data was gathered through face-to-face interview and self-completion questionnaires. Population estimates from the Office of National Statistics were used to weight the data to reflect the age and sex profile of the British population.

8.3 In this section, we look at the number of people living in Scotland who took part in gambling in the year prior to interview, and the number and type of gambling activities they engaged in. At the outset, however, two caveats should be noted. First, the data are over six years old and should not, therefore, be seen as a current picture of gambling participation in Scotland. Second, the Scottish sub-sample, although representative of the adult Scottish population, consists of just 745 cases.

8.4 While this is large enough to afford a reasonable degree of precision for results at the level of Scotland as a whole, the possibilities for more detailed sub-group analyses are limited. For example, analysis was run on participation in gambling by: martial status; employment status; social class; income; and highest educational qualification, but when statistical testing was performed the bases were too small to show any significant differences. Despite these caveats the 1999 survey remains the only data source of its kind (though a new Britain-wide prevalence study is to be carried out in 2006) and its broad findings are likely to remain relevant. Moreover, detailed sub-group analysis based on the sample as a whole (i.e. including respondents in Scotland, England and Wales) is likely to provide a reasonable basis for predicating patterns of gambling behaviour in Scotland.

How Many People Gamble in Scotland?

8.5 One of the main aims of the 1999 survey was to provide robust baseline information levels of participation in gambling. In order to measure this, respondents were shown a list of 11 gambling activities and asked whether they had participated in each activity in the preceding week and in the 12 months prior to interview. Respondents were not asked about their frequency of gambling - simply whether they had participated in the last year or week - in other words, the survey provides a measure of prevalence rather than incidence of gambling activity. It did, however, record the number of different forms of gambling engaged in and it is likely that this is a reasonable proxy for extent of involvement.

8.6 Looking at the sample overall (table 8.1), seven out ten (72%) respondents said they had participated in one or more of these activities in the twelve months prior to interview. Table 8.1 also suggests that there was no significant difference in levels of participation in gambling (70% in Scotland compared with 73% in England and Wales.

What do People Gamble On?

8.7 Taking part in the National Lottery Draw was by far the most popular gambling activity, with 65% of all respondents saying they had purchased a National Lottery ticket within the past twelve months. Respondents were three times as likely to participate in the National Lottery Draw as in the next most popular activity, which was the purchase of scratchcards (including those sold by Camelot, the current organisers of the National Lottery) by 22% of the population. Again, there was no significant variation between Scotland and the rest of Britain.

8.8 The only significant differences - where Scottish respondents were less likely to bet than were those from England and Wales - were betting on horse or dog racing and private bets.

Table 8.1 Gambling activities in the year and week prior to interview

8.9 Although these data show no significant differences between Great Britain and Scotland as a whole, other research has found that gambling participation and expenditure in Scotland tends to be somewhat higher than the national average. For example, one survey found that one in ten of the population in Scotland - twice the national average - are frequent players of the lottery, football pools, bingo and betting shops (Mintel 2004). Others have found that Scots spend more on the Lottery than do players anywhere else in the country (Fitzherbert 1995; FES 1999).

Number of Activities Engaged in During Previous 12 Months

8.10 Turning now to the number of gambling activities people engaged in, the picture in Scotland is again very similar to the rest of Britain.

Table 8.2 Number of gambling activities engaged in during year prior to interview

Who is Most Likely to Gamble in Scotland?

8.11 As noted earlier, the possibilities for detailed sub-group analyses are limited by the sample size in Scotland. Nevertheless, it is possible to offer some observations about variations in gambling behaviour by age group and sex. Table 8.3 looks at: the proportion of men and women who participated in each type of gambling activity in the twelve months prior to interview; the proportion who took part in any gambling activity; the number of activities; and the mean (average) number of activities.

8.12 A number of points are worth noting. In Scotland, there was no significant difference between the proportion of men (70%) and women (69%) who said they had taken part in at least one gambling activity during the year prior to interview. Interestingly, men in England and Wales were more likely than women to have taken part in at least one gambling activity (76% and 68% respectively). Scottish men were also below the British figure for taking part in any gambling activity in the year prior to interview (70% compared with 76%).

8.13 Looking at the individual types of gambling activities, some gender differences are also apparent. For example, men were significantly more likely to gamble on football pools, fruit machines, horse or dog racing or take part in private bets than women, whereas women were more likely to play bingo than men. These patterns are very similar to those for the whole British sample.

8.14 But while, overall, men were no more likely than women to have gambled at all during the previous 12 months, men were more likely to have engaged in a greater number of gambling activities (1.7 compared with 1.3), and were twice as likely to have gambled on four or more activities (13% compared with 6%). The same pattern was found for the British sample as a whole.

Table 8.3 Type and number of gambling activities in the year prior to interview by sex (Scotland only)

8.15 Looking at table 8.4 we can see that participation in gambling was also linked to age with those aged 35-54 being most likely to have taken part in any gambling activity in the year prior to interview (78% compared with 67% of 16-34 year-olds and 63% of those aged 55 or over). However when the average number of activities are examined there is no distinction between the 16-34 and 35-54 age groups who both had a mean participation figure of 1.7.

8.16 There was also some evidence of variation by age in the type of gambling activity, with 35-54 years old more likely to participate in the National Lottery Draw (74% compared with 58% of 55+ age group and 55% of the 16-34 year olds). The younger age group were also more likely to gamble on scratch cards and fruit machines. These patterns are again consistent with those for Britain as a whole.

8.17 It can be seen that table games in a casino were played by only 3% of the population, in line with figures for the rest of the country, and that they were played disproportionately by males and those aged 16-34.

Table 8.4 Type and number of gambling activities in the year prior to interview by age (Scotland only)

Note these percentages have not been compared with the British sample because the age categories have been combined to enable analysis.

Problem Gambling Prevalence

8.18 One of the key aims of the 1999 survey was to provide an estimate of the prevalence of problem gambling in Britain. In this section, we examine prevalence rates for Scotland and compare them with the British sample as a whole. The survey team used two screening tools to identify problem players: the South Oaks Gambling Screen ( SOGS) and DSM-IV¹⁶. Both screens can be used to measure lifetime and current problem gambling behaviour.

8.19 Using the SOGS, 0.8% (63 cases) of the British sample as a whole were classified as problem gamblers. Similar findings were obtained using the DSM with 0.6% (44 cases) of the British sample identified as problem gamblers. For Scotland, the comparable figures were 1% (8 cases) and 0.7% (5 cases). These do not represent statistically significant differences from the figures for Britain as a whole.

8.20 Because of the very small number of cases, it is not possible to extrapolate further on the basis of the Scottish data. Given the very strong similarities between Scotland and the rest of the sample in terms of participation rates, however, it seems reasonable to assume that the key drivers of problem gambling in Scotland will be similar to those for Britain as a whole. Findings from the 1999 prevalence study concluded that, once others factors were controlled for, 'problem gambling' was associated with: being single and male; having a parent who was or had been a problem gambler; and being in the lowest income category.

8.21 As we saw in Chapter Five, international research has also suggested a link between problem gambling and living in areas of deprivation.

Casinos and Patterns of Deprivation in Scotland

8.22 Given this link between problem gambling and deprivation, it is advisable to look at levels of deprivation and other risk factors associated with problem gambling in Scotland in order to assess the likely impact of the introduction of new casinos. To recap: problem gambling is associated with particular indices of social deprivation, such as unemployment, low income and low levels of education and skills. In addition, individuals of Chinese ethnicity also have higher rates of problematic casino play than the general population.

8.23 Using data from the Scottish Index of Multiple Deprivation 2004 ( SIMD) ¹⁷, Labour Market Statistics (2006) and the 2001 Census, this section outlines the profile of the regions which have recently bid for casino licenses, and considers the likely social impact of casinos on the residents of these regions.

Table 8.5 SIMD Data

Proposed Casino Locations

8.24 Glasgow City Council, West Dunbartonshire and Midlothian have bid for regional Casino licenses, while Renfrewshire, Falkirk and Dumfries and Galloway have bid for one of the eight large casino licenses that will be awarded.

8.25 The potential sites for the Glasgow casino are within the city itself. Two are along the Clyde waterfront: one at the Scottish Exhibition and Conference Centre ( SECC) and one at the Glasgow Harbour development near Partick. The third is at Ibrox. The Midlothian site is south of Edinburgh, on the city bypass at Straiton. The West Dunbartonshire site is at Bowling, outside Glasgow, on a major motorway route to Loch Lomond.

8.26 The proposed site of the large casino in Renfrewshire is Paisley; in Dumfries and Galloway it is Stranraer; and in Falkirk it is the Falkirk Gateway site, an area earmarked for leisure and retail use.

8.27 At present, there are already twelve casinos in Scotland: one in Dundee, two in Aberdeen, four in Edinburgh and five in Glasgow. All are small by international standards.

Regional Characteristics

8.28 SIMD data are based on 6,505 Data Zones ( DZs), which contain an average of 750 people. Each DZ is ranked according to its level of deprivation - the DZ ranked 1 is the most deprived and the DZ ranked 6505 is the least deprived. ¹⁸ Table 8.5 presents SIMD data from the 15% of DZs ranked most deprived in Scotland. Councils applying for licenses are highlighted in bold. It should be noted that the City of Edinburgh has been included in this, as the Regional casino site proposed for Midlothian is within the catchment area of Edinburgh itself.

8.29 The most obvious point to note is that Glasgow City has by far the highest score of deprivation with over half (54%) of its population living in the 15% of most deprived data zones. Glasgow City also accounts for 40% of the population of Scotland who live in the worst 15% of DZs. West Dunbartonshire also has high deprivation scores, with 27% of its population living in the worst 15% DZs, while only 0.9% of the local authority population of Midlothian do so. However, the city of Edinburgh, which would be the catchment area for the Straiton casino, has 11% of its population living in the most deprived zones. Of the contenders for large casinos, Renfrewshire has the highest proportion of population in the most deprived DZs (19%), while Dumfries and Galloway has 5% and Falkirk has 7%.

8.30 It is worth noting that eight out of ten of the local authorities with the highest proportion of population in the 15% of most deprived DZs are in the West of Scotland. Of the 'city' authorities, Aberdeen and Edinburgh have the lowest proportion of population in the most deprived DZs.

8.31 In terms of income and employment, Glasgow has higher rates of unemployment than Scotland as a whole (5% compared with 3.3% at the time of writing), and a higher proportion of these are long term unemployed (17% compared with 13.4% ¹⁹). Long term unemployment appears to be falling faster in the city than in the country as a whole however. Overall, a significantly lower proportion of people of working age is economically active than in Scotland as a whole (70% compared with 79%). Median earnings are 4% lower than the Scottish average (Labour Market Statistics April 2006).

8.32 The picture is similar in West Dunbartonshire, which has rates of unemployment comparable to Glasgow. Long term unemployment is slightly lower than in Glasgow (14%), although it is falling more slowly than the country as a whole. There are slightly lower numbers of people of working age who are economically active in the region than in Scotland as a whole (76%). Median earnings are 12% lower than elsewhere in Scotland.

8.33 Midlothian appears to have a more buoyant economy. The unemployment rate is lower than the Scottish average (2.4%), with low rates of long term unemployment (11%) which are also falling faster than the Scottish average. Significantly higher numbers of working age people are economically active than in Scotland as a whole (84%), although median earnings are still 11% lower than for the country overall. However, characteristics of the city of Edinburgh, which would be within the catchment area of the proposed casino, should also be taken into account. Unemployment in Edinburgh is around the Scottish national average, and though long term unemployment is similar (14%) it is rising considerably faster than in the country as a whole. A slightly higher proportion of the working age population is economically active and median earnings are 12% higher than the national average.

8.34 In Renfrewshire, unemployment is around the Scottish average, while long term unemployment is lower (11%) and is falling faster than in Scotland as a whole. The economic activity rate is only slightly lower than the Scottish average (78%), and median earnings are 6% higher than in the rest of the country.

8.35 In Falkirk, unemployment and long term unemployment (13%) are similar to the national average, although the latter is rising more quickly than in the rest of the country. Proportionately more people are economically active than in the country as a whole (80%), although median earnings are 4% lower.

8.36 In Dumfries and Galloway, although the unemployment rate is lower than the national average (3%). Though long term unemployment is similar (14%) it is rising faster. A higher proportion of people is economically active in the region (82%), and median earnings are 8% lower.

8.37 Although they make up a small proportion of the Scottish population overall, numbers of Chinese residents are highest in Glasgow and Edinburgh, where they comprise 0.7% and 0.8% of the cities' populations respectively. Total numbers are greater in Glasgow however (3876 compared with 3532). They make up 0.3% of the population of Renfrewshire, 0.2% of West Dunbartonshire and Dumfries and Galloway, and 0.1% of both Midlothian and Falkirk (Census 2001).

Potential Impacts of Casinos

8.38 From what we have seen of the international evidence, it appears that problem gambling is associated with individuals who experience multiple forms of deprivation, and particularly who are unemployed and/or on low incomes. Although it should be stressed that these groups are not the ones most likely to visit a casino, when they do, they are most risk from developing problems with their playing.

8.39 From the socio-economic profile of the proposed sites for casino development outlined above, it seems likely that some regions will have higher proportions of their population who may be at risk from developing gambling problems than others.

8.40 On the other hand, it is also the case that potential benefits - in terms of job creation and economic regeneration - are likely to be greatest in the areas of greatest deprivation. Research has shown that casinos can create full time, entry level jobs (unlike many other branches of the gambling industry) which are badly needed in communities suffering from unemployment and underemployment. Economic benefits tend to be concentrated within the immediate vicinity of gambling facilities themselves. In this sense, there is a case to be made for casinos providing local employment for people in deprived areas, especially in locations where individuals may be unable or unwilling to travel outwith the immediate area for work. However, the case for employment is not straightforward, as it is not certain how many of these new jobs would actually go to the local people who need them the most. The long term unemployed in particular may lack the skills required by the casino, especially in terms of the better paid 'front of house' jobs.

8.41 The following discussion will review the potential impacts of casinos, and highlight areas where the negative impacts of casinos could be greatest, first in terms of regional casinos, followed by large ones. Throughout, it will also outline the limitations of this exercise.

Proposed Regional Casinos

8.42 Regional casinos will be designed as resort or 'destination' venues providing a range of facilities such as restaurants, bars, cinemas, shopping and entertainment, which create an entire leisure experience, and of which gambling may be only one aspect. They will intend to attract the majority of their customers from outwith the local population, and will especially target large numbers of tourists. They will be allowed up to 1250 Category A machines, including FOBTs, offering unlimited prizes and accepting large stakes.

8.43 Table 8.6 presents some of characteristics known to be associated with problem gambling for the regional authorities that have applied for Regional casino licenses. It should be noted that again, details for Edinburgh have been included in this, as the Regional casino site proposed for Midlothian is within the catchment area of Edinburgh itself.

8.45 Of the regions considered for these developments, Glasgow has the highest concentration of risk factors associated with problem gambling. It has the highest deprivation scores in the country, with over half of the population of the city living in the 15% of most deprived data zones. In addition, unemployment is high and economic activity and median income relatively low.

Table 8.6 Characteristics of authorities applying for Regional casino licenses

8.46 West Dunbartonshire also has high deprivation scores (27%), high levels of unemployment and low median incomes. As mentioned earlier, it is noteworthy that eight out of ten of the local authorities with the highest proportion of population in the 15% of most deprived DZs are clustered in the West of Scotland.

8.47 In terms of the three proposed sites for the Regional development, Midlothian has the lowest deprivation scores (0.9%) and low unemployment but also low median incomes. However, it must be remembered that the Midlothian casino proposal would serve the population of the city of Edinburgh which has a deprivation score of 11%.

8.48 The first thing to note is that all of the proposed casino sites, with the exception perhaps of Dumfries and Galloway, are within easy reach of areas of considerable deprivation, with Glasgow in particular possessing particularly high levels of deprivation. However, given that all of the proposed sites are close to pockets of high deprivation, and all are within 50 miles of each other, it is likely that the impacts of any one casino would be felt in surrounding regions.

8.49 In terms of the risk posed to ethnic groups, a casino development would probably have the greatest negative impact in Glasgow city and Midlothian. Compared to the country as a whole, there are relatively high numbers of Chinese people living in Glasgow, as well as in Edinburgh, the latter of which would be the local catchment for the Midlothian casino at Straiton. It should be stressed, however, that overall numbers are extremely small.

8.50 It should be noted that this 'mapping exercise' is an instance of applying a blunt tool to some complex issues, and the overall picture is complicated by a number of factors which need to be taken into account:

8.51 Local market:

• The real question is the extent to which these casinos actually attract local residents. The international evidence has suggested that those which draw on large non-local markets tend to have low social costs, and those who rely on locals as their primary customers have high ones relative to economic benefits. Urban or suburban casinos tend to draw large numbers of local residents to them, and so social costs remain within the community where the gambling facilities themselves are located. All of the proposed developments (with perhaps the exception of Dumfries and Galloway) could be considered urban or suburban casinos, although all will attempt to attract large numbers of visitors.
• There is evidence that the proposals for regional casinos are more likely to be successful than the ones for smaller developments in this: research shows that people are willing to travel to larger, destination or resort casinos. In fact, surveys have shown that around 90% of visitors had travelled more than 50 miles to the largest casinos in the U.S. However, far smaller numbers travelled to smaller ones, and the pattern that has emerged from overseas is of smaller casinos being patronised in greater numbers by locals, with larger ones attracting more from outwith the immediate area. It is the latter that is the most desirable proposition, as large casinos located away from population centres have the greatest potential to deter impulsive playing by locals, and so have the potential to avoid high levels of negative social impacts.
• Whether or to what extent any of these proposals will actually manage to attract large numbers of non-locals is simply not known.

8.52 Accessibility:

• It was noted earlier that research shows the location of a casino within 50 miles of an individual's home can double the risk of problem gambling behaviour. To begin with, it should be noted here that all of these proposed casino sites are within around 50 miles of each other (again, with the exception of Dumfries and Galloway), and so the impacts of one could conceivably be felt in surrounding regions, regardless of where any single one was actually built.
• Issues of accessibility also combine with questions of how attractive casinos might be to those who are most at risk. From what we know about casino participation already, we can see that casinos are not particularly popular among those on low incomes, the unemployed and those with low levels of education. For deprived groups living in the Central Belt of Scotland, casinos may not present a sufficiently attractive proposition to merit the time and effort required to actually visit one, especially if this involved a journey of up to 50 miles.
• However, for any to whom a casino is easily accessible (for example, the city centre proposals in Glasgow at Ibrox, Partick and the SECC), the situation might be quite different. The location of casinos in these could potentially encourage impulsive playing. Then again, it should be noted that Glasgow already has five casinos which are located in the heart of the city centre, and so are already more accessible than the regional development would be.
• An additional factor whose impact is difficult to assess concerns the recent smoking ban in public places. Problem gamblers, as well as individuals from deprived social groups, have higher levels of smoking than the general population, and evidence from Australia showed a dramatic decrease in gambling among these groups when smoking was banned from gaming venues. Although a smoking ban may not prevent regulars from visiting casinos in Scotland, it is possible that it might deter new visitors.
• Overall, it is simply not possible to predict what the uptake of casino gambling will be among these groups of individuals. However, existing knowledge of current patterns of gambling participation suggests it might be small.

8.53 Problem gambling prevention strategies:

• As we saw in Chapters Five and Six, the provision of specialist prevention and treatment strategies can counter the negative effects of increased availability of gambling on the local population. Some international studies found no increase in problem gambling among local residents after the introduction of casinos in Washington State and Oregon in the U.S.; Windsor in Canada; and in cities in Australia and New Zealand. These patterns were largely attributed to the development of comprehensive prevention strategies and specialist service provision to prevent and treat problem gambling.
• In conclusion, the strategies that regional bodies might adopt to combat the development of gambling problems could have a crucial impact on the overall social impact of a casino.

Proposals for Large Casinos

8.54 Table 8.7 presents some of characteristics known to be associated with problem gambling for the regional authorities that have applied for large casino licenses.

Table 8.7 Characteristics of regions applying for large casinos

8.55 While regional casinos have the potential to attract large numbers of visitors, from whom it is expected they will draw the majority of their revenue, smaller ones tend not to. Evidence has shown that people are less likely to travel longer distances to visit a smaller casino. In this sense, it can be expected that the casinos proposed for Renfrewshire, Falkirk and Dumfries and Galloway would draw the majority of their customer base from their local populations. Given that it has the highest score of social deprivation of the proposed sites, it can be suggested that Renfrewshire would experience the greatest negative social impact relative to economic benefit here, and Dumfries and Galloway the least negative. Again, this also implies that Renfrewshire could have the most to gain from the potential economic benefits of a casino development

8.56 However, accessibility is also an issue here and of the three sites, Dumfries and Galloway is probably the least accessible to non-locals. This could mean that even higher proportions of casino visitors were recruited from the surrounding community. On the other hand, it can be noted that these large casinos would house fewer EGMs than regional ones (150 compared with 1250), with smaller jackpots. As larger jackpots are associated with greater potential to encourage problematic play, the type of risk posed by these machines could be considered to be less than that posed by the high jackpots machines in a regional casino.

8.57 The numbers of Chinese people living in all three regions is relatively small, although far larger numbers are within the fifty mile radius. Both the Renfrewshire and Falkirk casinos would be within reach of the populations of Glasgow and Edinburgh which have larger Chinese populations. However, whether or not they would actually travel to the casino is uncertain. Given the fact that casinos already exist in both cities, it is likely they would not.

Conclusions

8.58 It should be stressed that this 'mapping exercise' is an instance of applying a blunt tool to some very complex issues. Although they may create employment opportunities in deprived areas, the potential economic benefits of a casino development is difficult to quantify, since the extent to which new jobs would be filled by those who need them most is not known.

8.59 In terms of social impact, the real question is the extent to which any of these casinos would draw on their local populations. The attractiveness of a casino to the local population and the numbers of individuals from vulnerable groups who might actually visit it are all unknowns. In addition, the steps taken by public authorities to prevent the development of gambling problems could have a significant impact on the eventual social impact of a casino.

8.60 In a very broad sense, it can be stated that in Scotland, small casinos and those located in urban or suburban areas would have the greatest potential for negative social impacts, particularly if those areas are already characterised by high levels of social deprivation. However, such impacts can be ameliorated by the provision of specialist prevention services. Given this, it is important that public authorities adopt a proactive approach to ensure that the appropriate resources to prevent and treat problem gambling are specifically targeted within the areas that might host or be proximate to casinos.

« Previous | Contents | Next »