In this paper we will investigate the question of who gambles in the UK. This should be a fairly simplistic question as for the last forty to fifty years, the British public has been able to freely gamble in betting shops, bingo halls and casinos and more recently online. Prior to this the majority of gambling was illegal and had been so for about a century. However this did not stop the British public’s desire for a ‘flutter’. Illegal street betting was endemic, pubs resounded to the sound of money being won and lost on all sorts of sports, games and competitions, casino gaming happened at private parties and slot machines could be found in all sorts of establishments. The law was widely disregarded to say the least. So estimates on gambling before the 1960’s are understandably sketchy and inaccurate. The contentious issue is that the same could be said to apply even now.
Britain’s history of gambling over the 20th and now 21st Century is one of a nation who gambles vigorously and this gambling is done by all social classes and gender. Gambling was and is a part of British culture, society and language and continues to be so. The Victorian prohibitions blatantly failed to stamp it out (much like their other attempts to banish ‘vice’), but we had to wait until the 1960’s before the government of the day felt they could begin to embrace the reality of people’s leisure choices rather than be dictated to by the puritanical ideologues of the minority in Parliament and pulpit. This paper will show that throughout Britain’s recent history, the nation has gambled happily and continues to do so. But, as we shall see, getting an exact estimate of who gambles, how and how much is a very complicated matter.
As stated above, all forms of gambling (except betting at a racecourse or on credit) were illegal at the beginning of the 20th Century. The most prevalent form of illegal gambling was betting on horse races and other sports. This had been driven by the growth in mass media, as the newspapers carried the details of the ‘runners and riders’ to all parts of the country and with the invention of the first internet, the electric telegraph, the results of races could be relayed to anxious punters almost instantaneously. As can be seen from the report to the Select Committee of the House of Lords on Betting (1902), illegal street betting was ubiquitous;
“The Committee reached the conclusion that “betting is generally prevalent in the United Kingdom, and that the practice of betting has increased considerably of late years especially amongst the working classes; whilst on the other hand, the habit of making large bets, which used at one time to be the fashion among owners and breeders of horses, has greatly diminished”. The Committee were of the opinion that “even when due allowance has been made, both for the increase in population of towns, and the rise in wages, betting is undoubtedly more widespread and general than it used to be.” They considered that “the increased prevalence of betting throughout the country is largely due to the great facilities afforded by the Press, and the inducements to bet offered by means of bookmakers’ circulars and tipsters’ advertisements”. It has been proved conclusively that the practice of betting in the streets has increased very much of late years, and is the cause of most of the evils arising from betting amongst the working classes”.[i]
Betting (or indeed any form of gambling) by the working classes was seen in an extraordinarily derogatory light by today’s standards (with the exception of some parts of the Labour Party). Considering that the above statement came from a Committee of the House of Lords who were probably all born around the time of the Battle of Waterloo, it is hardly surprising that they viewed what was a blatant disregard for the law by a mass of the public in a bad light. Furthermore views about the working class at this time tended to range from those who saw them as poor due to their own fault and that gambling was just another example of their lack of moral fibre to those who saw gambling as an evil that kept the working class away from more high minded pursuits such as studying the Bible or Karl Marx. It would take decades before the views of the establishment would appreciate that the working class had just as much a right to their leisure as any other. Arguably, it was also the scale and the blatant nature with which illegal betting operated throughout the country which raised the ire of those in power; the Select Committee on Betting Duty (1923) also makes reference in their report to the prevalence of street betting:
“the appalling hold betting has got of the large majority of the community” and again amazement is expressed at “the extent to which betting exists at the present time. The evidence of all the witnesses agrees that practically every class in the community now bet, and that the habit has taken hold of women, both old and young, and of boys and girls in many cases even under sixteen years of age”…. It was estimated that “a yearly turnover [turnover is amount staked and not the amount lost by customers] of £200,000,000 [approx. £5.9BN at 2005 prices[ii]] could be safely assumed” “All the evidence goes to show that though the average stakes are now diminishing, due no doubt to the decrease and lack of wages, there is now, five years after the war [World War I], no decrease in the number of people betting, but an increase”.[iii]
The interwar period has been considered a highpoint in working class mass leisure. Social legislation partly prompted by the sacrifices made in World War I (the majority made much earlier) meant that the working class was getting paid more and with paid holidays and days off had more leisure time than ever before. It is here that we see the mass crowds going to football matches, greyhound stadiums and horseracing (though horseracing has always been a festival of gambling for the masses). Not surprisingly they spent some of their increased wages on sports betting. It should be said though that much of the history of organised sport is down to gambling, with sporting associations in the 18th and 19th Century organising themselves and unifying their rules to accommodate the needs of gamblers. So by this time there was already a lengthy social history of mass betting on sport. In house entertainment until the 1950’s was rather limited (the piano and then the radio), and much more so than today were people spending time out of their homes (which for many were small and squalid by today’s standards) in social environments (e.g. pubs, music halls, sporting venues) and betting and gambling were part of the social melee. The rich had for centuries been gambling, mostly at casino games and also on horses of which they were owners. The difference for them was that their activities were not considered a social problem, even though many examples showed great instances of harm by today’s standards.
By the time of the Royal Commission on Lotteries and Betting 1932-33, chaired by Sir Sidney Rowlatt, we see attempts to estimate the size of the betting industry which are hampered by the lack of hard evidence due to the illegal nature of the activity and the common mistake of mixing up turnover (the amount staked) with what we now call Gross Gam(bl)ing Yield, which is the amount customers actually spend (customer stakes minus customer winnings), as this is the amount they lose, both of these factors hinder forecasts of gambling spend until gambling is legalised in the 1960’s.
“The volume of taxed betting during the years 1927-8 and 1928-9 was about (or a little over) £90,000,000 [approx. £3BN at 2005 prices] a year. This figure was fairly equally divided between course and office betting [betting was only legal on course or by credit and therefore predominantly the activity of the rich].
We understand that in general tax was not paid on illegal betting [there had been an attempt to tax illegal off course betting but not make it legal, unsurprisingly it failed] , and it has been stated that there was some evasion in regard to legal betting. The figures of the volume of taxed betting cannot therefore be taken as disproving the accuracy of the estimate given in the draft report of the Chairman of the 1923 Committee. The first annual report of the Racecourse Betting Control Board (for the year 1929) contained a section dealing with the total volume of betting, which indicated a figure of £230,000,000 [approx. £7.6BN at 2005 prices] as the probable annual turnover on betting.”[iv]
Betting obviously wasn’t the only gambling activity of the masses. Some of the reasons for setting up the 1932-33 Royal Commission had been the growth in gambling on the Irish Hospitals Sweepstake (a lottery product linked to horseracing) and other gambling related competitions being run by newspapers:
“In Appendix IV we give particulars of the total money subscribed from this country to the Irish sweepstakes, which was nearly £10,000,000 [approx. £369.8M at 2005 prices] in the year 1932.
Account must also be taken of the increasingly large sums expended as entrance fees for newspaper competitions of various kinds which are a disguised form of gambling. The Secretary to the Post Office informed us that approximately £3,000,000 [approx. £110.9M at 2005 prices] was believed to have been expended in one year on entrance fees to newspaper competitions in the form of postal orders and stamps”[v]
Although these sums may seem small by today’s standards, when added to the forecasts for betting shown above, it shows that, although illegal in the majority, gambling was a mass market pursuit involving the majority of the population gambling, in total, a significant amount of money. The Rowlatt Commission would achieve little in its attempt to deal with the issue of gambling and the interruption of the Second World War would only prolong the need for further consideration of the issue. In 1949 another Royal Commission would be set up, The Royal Commission on Betting Gaming and Lotteries chaired by Henry Willink, which would utilise the new methods of social research that were developed prior to the War and during it. When the Willink Commission’s Report was published in 1951, it was able to give some of the first scientifically based estimates of gambling participation:
“The results of the Social Survey [Government Social Survey conducted in 1950] suggest that at least 4 out of 5 adults in all walks of life have taken part at some time in their lives in some form of gambling, including such relatively trivial forms of gambling as raffles, whist drives, private sweepstakes and an annual “flutter” on the Derby or Grand National”[vi]. The Commission went on to look at the Survey’s results regarding participation in the football pools, dog racing and horse racing. For the football pools; “the results of the Social Survey all indicate that the number of persons sending in football pool coupons during recent years has been about 10 million, of whom about 7 ¾ million sent in a coupon during each week of the season [UK population in 1950 was approx 50 million]. The number of persons who take part in this form of betting is rather higher, as many coupons represent the combined efforts of a number of individuals. The results of the Social Survey indicate that during the 1949-50 season the total number of persons who take part in football pool betting was about 14 million or rather more than 1 in 3 of the adult population”[vii] For dog racing; “The evidence, which is considered fully in Appendix II, indicates that the annual attendances (i.e. individual admissions) have been about 32 million in recent years, but that the number of regular attenders does not exceed about 500,000. Over 80% of the attendances are accounted for by some 350,000 persons who attend once a week or more frequently. There is no indication of widespread betting on dog-racing off the course.”[viii] For horseracing; “The results of the Social Survey indicate that the number of persons who bet on horse-racecourses each year is approximately 2 ½ million, of whom the great majority attend only once a year or slightly more frequently. The number of people betting fairly regularly on horse-racing off the course appears to be about 4 million though probably as many as 1 in 2 of the adult population have a bet on such races as the Derby and the Grand National.”[ix]
If we replace the football pools with the National Lottery (both being the mass market gambling product with a small stake and the potential for a life changing jackpot of their day) then there are great similarities with 1950 the state of gambling today as in the industry is dominated by one big weekly product (the pools and now the National lottery) and there are a large number of people who regularly indulge in betting, but an even larger amount of the population have a bet on the Grand National and the Derby. Greyhound racing, like the football pools has sadly declined in their popularity due to changing social habits and alternative pastimes and gambling products.
The table below shows one of the first official estimates of gambling expenditure (provided in the Commission’s Report), which one would almost automatically query due to the illegality of much of gambling at the time (football pools were considered legal as a form of credit betting). However, as the quote from the Report below provides an estimate of the amount of personal expenditure on gambling as compared to total personal expenditure, giving a result of 0.8%, as we will see later on in the paper, this is not far from that found in the 1970’s or the current ratio that we’ve seen in the first decade of the 21st Century. This suggests that the figures provided are quite credible.
Variations in personal expenditure on gambling £ millions[x]
“Two general conclusions may be drawn from these figures. The first is that personal expenditure on gambling immediately after the war and until taxation was imposed was about two and a half times what it was before the war. If, however, account is taken of the decrease in the value of the £, the post-war expenditure would be rather less than one and a half times the pre-war expenditure. The second conclusion is that, apart from the effects of the imposition of taxation, there has been a noticeable decrease in gambling during the last three years, particularly in dog-racing.
The relative importance of expenditure on gambling. It is interesting to compare these estimates with the estimates of personal expenditure relating to other forms of entertainment and personal indulgence as given in the White Paper on National Income and Expenditure of the United Kingdom, 1946 to 1949.
Personal expenditure on gambling and other items compared
Gambling in betting shops was to be legalised in the Betting and Gaming Act 1960. Due to the Parliamentary focus on betting shops and consequent lack of scrutiny on other areas of gambling, the Act also brought about the emergence of the commercial gaming industry (bingo and casinos) which it had specifically tried not to do. Due to bad draughtsmanship, a loophole existed in the law which allowed commercial gambling to happen by operators setting themselves up as private clubs and by altering the rules of how the games were played and who acted as the ‘bank’. Not only did this mean that there was an explosion in the number of gaming clubs (there were c.1,200 in 1968), with gaming found in cabaret clubs, dinner clubs, night clubs, but as the law was unclear it was also un-regulated, and this meant the infiltration of criminal gangs. The newspapers of the time trumpeted stories of customers being threatened for payment of their gambling losses and publicans being forced to take rental of gaming machines at disadvantageous rates. The media played a great a part in whipping up moral outrage at the country becoming plagued by pathological gamblers even though there was not much hard evidence of the fact (a similar story would happen with the Gambling Act 2005). It did however, prompt the government to introduce new regulation, the Gambling Act 1968, that would reduce the number of casinos down to 120, ensure that bingo could only be for soft gambling and that criminality was driven out by a system of personal and premises licensing that was enforced by a regulator, the Gaming Board for Great Britain, whose powers have been described as ‘draconian’.
Even with the legalising and regulating of gambling that was to happen in the 1960’s, there was still a paucity of reliable data about gambling activity, evidence-based policy making was yet to be invented. In the debate on the Gaming Bill on 13th February 1968, William Deedes MP (Con, Ashford) would state;
“Our main source of this information in the last few years has been the Churches Council on Gambling, and all hon. Members will wish to pay tribute to the work done by that Council. However, its evidence is, alas, not accepted everywhere as being reliable or authentic.
It is not clear from the Bill or from what the right hon. Gentleman said how we are to supplement our knowledge, and I think it is important that we should. We should, from now on, be able to get continuous information about the extent of gambling, the amount being spent and some of the social evil consequences that may ensue. If we are to act sensibly and, as the Home Secretary observed, flexibly, we must have more base information, preferably over the whole sphere.
In the light of the Royal Commission's thinking, we must also ask the question: what is excessive gambling? Certainly in the collective sense, the 1960 Act manifestly failed there. There was an estimate—I tend to mistrust large estimates of this kind—that the total extent of gambling was in 1960 £750 million [approx. £11.4BN at 2005 prices]. A corresponding estimate now puts it in the region of £2,000 million [approx. £20.5BN at 2005 prices], of which horse betting probably accounts for about a half. Even allowing for a change in money values, that is a considerable increase over the last few years. I accept from the Home Secretary that this is not only excessive but is, in some cases, inspired or assisted by a criminal element. I will return to this subject when dealing with some of the proposals for cities outside London.”[xii]
One needs to question whether these estimates were relating to turnover or to gambling spend, the latter being the much smaller more relevant number. It is also quite ironic that for much of the 20th Century, the British government has relied on data about gambling from the work of groups that were set up to get gambling banned. While the Churches Council on Gambling had over the years softened its approach, from prohibition to restriction, it was still their view (and arguably still is), being mostly of the non-conformist persuasion, that gambling was bad and that it should be restricted far more than it ever has been. The influence of the non-conformists continues to do this day and has, arguably, far more influence than their membership or current social mores would suggest.
In 1968 we also see the survey work that would culminate in the 1974 publication of ‘Gambling, Work and Leisure: A Study Across Three Areas’ Edited by D.M Downes. Although funded in the majority by the Churches Council on Gambling, this study can be seen as one of the first, independent, comprehensive attempts to look at the prevalence of gambling in the UK (although Otto Newman’s ‘Gambling: Hazard and Reward’ had been published two years earlier, it focussed solely on betting shops). In 1968 the research team interviewed a random sample of adults (aged 18+) from households in the county boroughs of Sheffield and Swansea and the former Municipal Borough of Wanstead and Woodford (now part of the London Borough of Redbridge) about their gambling behaviour in “the context of general spare-time and specific leisure pursuits”[xiii].One of their ambitions was to test some sociological hypotheses about why people gamble. They considered eight theories; anomie, alienation, working class culture, structural-functional imperatives, decision making, risk taking, home-centredness and work centred leisure. Space and the topic of this paper does not permit any detail and explanation of these theories or of the results, apart from saying that their hypotheses were all disproved apart from a few definitive statements;
“The sequence of refuted propositions which unfolded as we tried theory after theory with at best only limited success, and at worst total failure of the data to match the theories, was itself predicted in view of the difficulties involved in constructing theories which go beyond singular propositions. As things stand, we can at least claim to have cleared the ground of a great deal of presupposition and piecemeal theory stitching, and from the few relationships which have emerged, are better placed to get on with building a better theory.”[xix] Unfortunately, half a century later, we are still no nearer having a definitive answer to why people gamble.
In their 1968 survey, they found the following about the way people gambled:
The football pools (as stated above should be considered the National Lottery of their day) were really a lottery product. Although the ambition is to predict as many score draws as possible from the fixture list, this was never really considered a skill based product and many customers would keep the same choices on the coupon each week regardless of which matches they related to. Downes et Al found that participation rates were high, 52.5% of men and 26.2% of women occasionally did the Pools[xxx]. Of these the number of men was lowest amongst the self-employed (38.7%), then the upper middle class (43.8%) with the highest being the upper working class (57.9%). For women, it was the upper middle class who had the lowest participation rate (18.3%), followed by the lower middle class (20.1%) with the lower working class having the highest (23.8%)[xxxi]. Men spent more than women with a mean gross annual expenditure of £10.9 [approx. £107.43 at 2005 prices] compared to £4.5 [approx. £43.69 at 2005 prices][xxxii]. Though it should be stated that these data relate to those who filled in the football pools coupon and that men were more likely to share their pools coupons with work mates, while women were more likely to either fill in the coupon alone or share it with their husband.
As stated above, betting shops had been legalised with the Betting & Gaming Act 1960 which came into force in 1961. Prior to this there had been widespread illegal betting that permeated every part of society. While initially the betting industry were reticent to grasp the opportunity of licensed betting offices, by the time of the survey they were almost at the high point in terms of numbers of betting shops (c.14,000). Downes cites Newman in stating that betting shops, at the time’ ‘are by far the principal arenas of gambling activity in Britain – which is certainly so for commercialized gambling”[xxxiii].The survey found that betting shops were (and to some extent still are) male preserves, with 17% of men regularly using them compared to just 2% of women who regularly use them. When considering those who just occasionally use them; the difference was equally stark; with 27% of men compared to 7% women[xxxiv]. This is reflected in their expenditure in betting shops with men having a gross yearly spend of £111[approx. £1,141.08 at 2005 prices] compared to £5 8s [approx. £55.51 at 2005 prices] for women[xxxv]. Average stake for men (none is provided for women) was £1 1s 6d[xxxvi] or approx. £11 at 2005 prices (higher than today which is approx. £9) and the betting was predominantly on horses (91.1% for men, 99.8% for women), while today less than half of all bets are on horses[xxxvii].
Downes et al considered different aspects of gaming; gaming in pubs which encompassed betting on darts matches to card games and dominoes and gaming at home which was usually card games such as bridge, whist, rummy and poker. Neither of these is of interest to this paper as we are concerned with commercial gambling and for this Downes et al was fortunate that the Gaming Act 1968 had just been passed and so could investigate legal commercial gaming for the first time. What they found was that only a very small number participated in casino gaming; only 0.1% of men and 0.3% of women did so regularly and 3.5% of men and 1.5% of women did so occasionally[xxxviii]. The found that casino attendance was skewed to the more affluent and the better educated and, interestingly, there was a higher proportion of casino gamblers who went to church sporadically than went less frequently[xxxix]. This affluent bias is borne out by the casino gaming expenditure figures (turnover weighted), which Downes suggests may be underestimates; men £56 1s [approx. £576.19 at 2005 prices] and women £20 1s[approx. £206.11 at 2005 prices][xl].
When it came to bingo, Downes is rather condemning when he states that; “Bingo has come to occupy rather the status of a joke in British social life. In song, parody and social comment, it has come to epitomize the ‘dead end’ use of leisure to which the newly ‘affluent’ working class have resorted in the absence of any ability to do something better with their time. Since its sudden rise in popularity in the immediate aftermath of its legalization in the 1960 Betting and Gaming Act (though its status before the Act was not so much illegal as anomalous), bingo has also come to be associated with middle-aged women in particular, so much so that its attraction has been held to have something (usually unspecified) to do with the ‘crisis’ of the menopause.”[xli] Downes was also adamant that this stereotype is not confirmed by the data, which shows that 11% of men and 10% of women regularly played bingo and 17% of men and 16% of women occasionally played. But then goes on to suggest that this slight male majority may be due to the inclusion of working men’s clubs in the sample [bingo was played as a form of fundraising]. The stereotype does hold though as it was almost exclusively a working class activity and ‘its appeal does seem inversely proportionate to degree of success in terms of formal education’[xlii]. Mean annual weighted turnover for sporadic frequency men was £49 [approx. £503.72 at 2005 prices] and for women was £65 [approx. £668.20 at 2005 prices]. There was a high correlation with high bingo spend and church attendance, which may be linked with the social nature of the activity (half the men and four-fifths of the women never play alone).
Prior to the Betting and Gaming Act 1960, (legal) gaming machines were to be found only in amusement arcades and a limited range of social clubs. The explosion in commercial gaming which the poor drafting of the 1960 Act brought with it a growth in gaming machines as they were allowed in a bigger number of premises. The survey found that 9% of men and 2% of women played on them regularly, while 29% of men and 17% of women played on them occasionally[xliii]. What the Downes et al survey found most striking was that; “the attractions of fruit machine playing, whatever they are, are inversely associated with growing older’[xliv]. One in five of the youngest (18-25) men played regularly compared with one in ten of the younger (26-45) and one in twenty five of the older (46 – 65) group. Downes suggests that the appeal could be to do with the speed of the reels, the sounds and the lights which appeals to a younger audience and which have subsequently become the mainstay of a number of academic works (and careers) into gambling behaviour. Customer spend figures for regular or sporadic play (annual turnover) are odd as Downes gives a figure of £40 [approx. £411.20 at 2005 prices] for men and £125 [approx. £1,285 at 2005 prices] for women[xlv]. Since Downes only suggests that these may be under-reported figures and does not question the significant gender difference, one may consider these figures to be possibly the wrong way round.
The next attempt to consider the gambling public was conducted for the Royal Commission on Gambling in 1978. Chaired by Lord Rothschild, it had been set up to look at the way sports could be funded from betting and concerns about certain lotteries and prize competitions. It ended up promoting the cause of a setting up a national lottery, very much along the lines of what would actually happen sixteen years later. As part of the Commission’s deliberations about a national lottery, they conducted some extensive survey work into gambling behaviour:
“In a sample of adults aged 18 and over, 94%, representing some 39 million, said they engaged in some form of gambling. But for large numbers this involvement was confined to an occasional bet – 63% of the sample bet on major races – or to buying a raffle or lottery ticket. With the important exception of the football pools, few people claimed to be regular participants in the various forms of gambling[xlvi].
Participation in various types of gambling[xlvii]
People who regularly participate in gambling 1977[xlviii]
If we look at the table below, we can see the equivalent of a Gross Gaming Yield figure of £793.5M for 1976 (approximately £4.2BN at 2009 prices) which can be compared to the personal expenditure on gambling (which should equate to Gross Gaming Yield) of £70.2M (approximately £1.6BN at 2005 prices) in 1949, implying that gambling spend increased 260% over 27 years (or 9.62% per annum). This would be an amazing growth in any element of expenditure and must surely be due to the legalisation of gambling that happened half way through the period. In the Gambling Commission’s Industry Statistics 2009/10 they estimate the Gross Gambling Yield for gambling generated in the UK (thus ignoring much of online gambling spend) to be £5.7BN[xlix] or a 36% increase since 1976. This would imply an annual growth of just over 1% in gambling spend year on year, however much of this growth can be put down to the introduction of the National Lottery in 1994. While we have no data to show how the spend on National Lottery products has impacted on total gambling spend, we have a good proxy in terms of total gambling taxes (duty), which by 1995/96 (the first full year of the Lottery’s operation) had increased by 42%[l] due to the National Lottery.
Gambling in England, Scotland and Wales: Calendar 1976[li]
In the above table, the gross revenue before tax figure of £793.5M for 1976 is considered too low by the Royal Commission on Gambling and as a note to the table they state “Miscellaneous items such as fixed odds coupon betting, prize bingo competitions, lotteries and raffles increase column (4) by £80 million to at least £873 million”[liii]. The Report then goes on to state that “this figure of £873 million may be compared with the Gross National product of about £109 billion for calendar 1975, it is 0.8% of the latter”[liv] The 1949 Royal Commission provided a figure of 0.8% of total personal expenditure. As we will show later, when we look at Family Expenditure Survey data for the first decade of the 21st Century this proportion of personal expenditure has only increased by a small amount, even with a major deregulation of gambling and the invention of internet gambling.
Money spent in various ways in calendar 1976, £ million[lv]
Average weekly household expenditure on some items in 1976, out of a total of £61.70, £[lvi]
According to the 1978 Royal Commission:
“This table suggests that, at least in terms of money, betting is a small component of the average household’s weekly expenditure, about one ninth of what is spent on alcoholic drinks, one sixth, of expenditure on tobacco, about the same as on animals and pets and a third of that on entertainments such as theatres, sporting events and television and radio. About the same amount was spent on butter, eggs, sweets and chocolates and soft drinks. This small weekly expenditure per household provides such a large aggregate figure for gambling per year because there are 19 million households in the country and 52 weeks in the year. By multiplying, one arrives at a figure of £395 million. This figure may be compared with the more accurate figure of £873 million [above] which itself covers only the main forms of commercial gambling. The errors in sampling procedures inherent in all surveys may account for part of the difference between the two figures. But the discrepancy which other gambling surveys have also encountered is so large as to call for another explanation; it is because of ambiguous attitudes towards gambling, many householders are reluctant to admit, perhaps even to themselves, the full extent of their gambling expenditure”[lvii]
The difference between industry statistics (usually based on tax receipts) and personal expenditure (based on the family expenditure survey conducted annually by the Office for National Statistics) exists even today. Although the Gambling Commission took over regulating the whole of the British gambling industry (apart from the National Lottery) in 2007 and has had every operator submit financial returns annually (and the National Lottery’s data is publicly available), it has unfortunately been unable to come up with a Gross Gaming Yield figure for the whole industry until it’s 2009/10 Industry Statistics publication. Until the Gambling Commission came up with its figure of £5.7BN for 2009/10[lviii] they had been working off an estimate based on HMRC figures for 2006/07 of £9.9BN[lix]. Family expenditure survey data for 2006/07 gives an annual gambling expenditure figure of £4.6BN[lx], less than half of the Commission’s figure for the same year while its figure for 2009/10 is much closer with a family expenditure survey figure of £5.5BN for 2009/10[lxi], only £0.2BN adrift. Considering how much political time and newspaper column inches have been spent deliberating over the extent of gambling it is remarkable that we are still to find any figures that can be considered completely definitive.
The next great spurt of gambling research is prompted by the build up to the new Gambling Act that would be passed in 2005 and come into force in 2007. As the new law was about deregulating the industry and removing many of the now archaic controls that had been put in place in the 1960’s, there was for the first time in gambling legislation, a real emphasis of consumer protection and responsible gambling. The political deal was that deregulation could only go ahead if the social externalities could be minimised and some of the strictest regulations in the world were brought in to protect customers who may be susceptible to problem gambling. As part of this process it was decided that regular prevalence studies into gambling would take place, the first before the Act to set a base line, and then there would be regular surveys after its enforcement to monitor how the industry was doing. As is the way of politics, the time for the Act to come into force took longer than expected and the pre-Act survey was conducted in 1999 (published in 2000) and then not updated until 2007 and again in 2010 (and then discontinued due to budget cuts).
Since we now are privileged to have three survey’s asking basically the same questions, basically in the same way and in a reasonable timeframe, we can start having a serious look at who gambles and how. Rather than look at the results individually, we have used the comparative data of the British Gambling Prevalence Survey 2000, 2007 & 2010 as tabulated in the 2010 edition and, where necessary, used data from the 2010 edition.
Comparison of participation in gambling activities for 1999, 2007 and 2010: Gambling activity in the past year[lxii]
Comparison of participation in gambling activities for 1999, 2007 and 2010: Gambling activity in the past week[lxiii]
For those who have gambled in the last week:
Profile of gamblers
Unfortunately it is, yet again, expenditure that proves a most elusive figure when it comes to research into gambling. The British Gambling Prevalence Studies of 1999 and 2007 both make an attempt at researching it, but by 2010 they would appear to have given up as the task was too difficult and the academics were travelling in a different direction. The 2010 study states:
“For each activity undertaken monthly or more often, respondents were asked how much money they usually spent on that activity in a month. Measuring expenditure on gambling is extremely difficult. The approach used in the 2007 survey, by which respondents were asked to report the total amount won and total amount lost for each activity, revealed that respondents were likely to have had a number of cognitive and heuristic biases when reporting these data (see Chapter 2 of the 2007 report for a fuller discussion). Therefore, this approach was not recommended for re-administration in the present survey. Furthermore, the focus of the present survey was slightly different in that it was not our aim to accurately capture data on net expenditure but to administer questions that would allow us to divide gamblers into a range of sub-groups based on higher to lower involvement. Work by Wood and Williams comparing different question wording against diary data suggested that broad information about gambling ‘spend’ could serve as a proxy measure of gambling volume. Therefore, the questions used in the present survey, by which respondents were asked to report how much money they usually spent on an activity in a month, were a refinement of the approach recommended by Wood and Williams.”[lxxiv]
They go on to provide a modicum of data:
“The amount of money spent in an average month was asked for each activity undertaken monthly or more often. The total amount of money spent gambling on all activities was calculated by summing together the figure for all 16 activities. As with the time spent measure, the distribution was skewed towards the minimum: 44% of regular gamblers (49% of men; 38% of women) spent an estimated £5.50 per month on gambling. That is, they regularly gambled on one activity and spent between £1- £10 on this activity. Further attempts to identify ‘medium-spend’ and ‘high-spend’ groups resulted in somewhat unsatisfactory cut-points (for example, simply dividing the remaining sample into two groups gave a high-spend group who reported spending between £21 a month and £3952 a month). Therefore, a distinction was made between the top 10% of regular gamblers, who spent an estimated £61.50 or more per month on gambling (mean expenditure of £209.92 per month) and the remaining 90% of regular gamblers who spent less than this amount (mean expenditure of £14.82 per month). The former is called the ‘high-spend’ group and the latter the ‘non high-spend’ group.”
From the discussion in the 2007 Prevalence Study, the main issues for those involved with the Prevalence Study appear to be that net expenditure – i.e. the amount of money people actually spend on gambling (i.e. stakes minus loses plus winnings) is the preferred measure of gambling spend (much as Gross Gambling Yield is a preferred industry statistic than turnover) and this has not always been surveyed (gross expenditure has), thus we haven’t had an accurate estimate on true spend. Furthermore, when respondents were asked what they have spent and won on their gambling, they have had a tendency to overestimate winnings and underestimate losses. Respondents also seem to be unsure of remembering how much was won or lost over a given period and men may have a tendency to exaggerate.
The discussion section of the 2007 Study considers its results when considering net expenditure which show that most gamblers have responded that they have ‘broken even. The Study team state: “Given that all sectors of the gaming industry make ‘considerable profits’, the results in this study clearly show that many gamblers do not appear to be making a realistic assessment of their previous week’s spending”[lxxv]. The unreliability of the data from their chosen methodology would seem to be the reason for not continuing asking how much is spent on gambling, one of the critical questions that researcher and policy makers need answering. It would appear that for the Study team, solving this problem would be too complicated.
This author feels that the people responsible for the British Gambling Prevalence Study have seriously let down the gambling industry and those interested in researching it by ducking the question of gambling expenditure. Considering the cost of the survey (reputedly between £500K and over £1M), it seems farcical that a better attempt could not have been made to identify one of the more important and relevant variables when looking at gambling, whether it be from a business, public policy or social research perspective. There are many aspects of social science research that examine personal behaviour where respondents have difficulties in accurately recalling events and they are of a much more personal/ contentious nature than gambling (e.g. sexual behaviour, illegal drug use, other forms of criminality). At the very least, a methodology could have been chosen which was imperfect, but with caveats and a consistency through the different Reports would have at least given an indication as opposed to what appears to be a misplaced attempt to categorise gamblers into abstract groupings for no particular purpose.
Fortunately for this paper, we have the data of a consistent, longitudinal survey, that may well suffer from similar reporting difficulties that proved insurmountable for the British Prevalence Study, but has kept with the same methodology (but not always in the reporting of the data) to allow us to get a good general idea of the British public’s spend on gambling. This is the previously mentioned, family expenditure survey that has been conducted for many years by the Office for National Statistics.
The table below shows that gambling payments have remained fairly static over the last decade, even with the implementation of the most deregulating gambling legislation ever. There does however seem to be a very recent increase in gambling gross spend in 2009-2010 in betting.
Source: Family Expenditure Surveys 1999/2000 – 2009/2010, Office for National Statistics
Source: Family Expenditure Surveys 1999/2000 – 2009/2010, Office for National Statistics
Source: Family Expenditure Surveys 1999/2000 – 2009/2010, Office for National Statistics
Source: Family Expenditure Surveys 1999/2000 – 2009/2010, Office for National Statistics
This obviously has the effect of at least halving (and in some years far more) the proportion of gambling spend as a proportion of total family expenditure. If we take our weekly net expenditure figure for 2009-2010 of £1.80, multiply by the number of households (25,980,000 according to the 2010 Family Expenditure Survey) and then by 52 weeks in the year, we get a figure for net expenditure of £2,431,728,000, less than half of the £5.5BN the survey gives for gambling payments (gross spend) or the £5.7BN that the Gambling Commission gives for Gross Gambling Yield (the same as Net Expenditure) for the same year. It would appear that the inconsistencies in the data identified throughout the last century remain prevalent. However if we compare this data with the family expenditure data for 1976, we see that in 1976 net expenditure was 0.65% of total weekly expenditure and in 2009/10 it was 0.47%, so although the data is dubious, it would appear that spend is dropping in the long term even though it might be increasing in the short.
The only conclusion one can draw from the above is that the gambling industry and government is crying out for reliable statistics to be produced by a neutral body that would look at who gambles, on what, how often and with how much. This should not solely be for the purposes of academia, constructing their own theories on problem gambling but for the industry and policy makers who need the raw data to look at how the industry grows and wanes. The Gambling Commission would seem the obvious choice to conduct this research, but given the paucity of data they have supplied so far and the criticism of its accuracy (even for such simple data as the number of gambling outlets and gaming machines); maybe a new independent body needs setting up. Recommendations number 1 and 2 of the Royal Commission on Gambling 1978 were:
Regardless of how and why such a unit is set up, the critical issue is that fifty years after the gambling industry was legalised, we still do not have an accurate picture of how the British public interacts with it. All we can say are generalisations; that the majority of the public gamble at some time, that a big proportion of the public gamble regularly, men gamble more than women, parents gambling influences their children to become gamblers in later life, gambling seems to reduce when people settle down to have children and then increase when elderly, that the very poor gamble more than others is a misconception and that average weekly spend on gambling appears to be low compared to other items such as alcohol and tobacco.
The industry and the British public deserve to be served better by those academics and regulators who consider gambling and deserve to be given comprehensive data. Why? Because gambling has always been a contentious issue and every time it is debated in Parliament, it is done so in an environment of hype and frenzy stirred up by the press (looking for scandal and moral outrage) and the faith groups (who are venting their moral outrage) and sometimes the truth, as in evidence based policy making, is forgotten. If there was an independent research unit considered credible by all sides of the debate than hopefully out politicians could consider the issues on their merits rather than the headlines or guesstimates that have so far played such a big part in deciding the industry’s future and that of the majority of the British population who have always enjoyed a bit of a ‘flutter’.