The crack cocaine of gambling?

More than fifty years have passed since bookmakers and casinos were legalised in Britain, but gambling remains at least partially taboo in some quarters. Few now remember the concerns raised about the National Lottery when it was launched in 1994, but more recent panics about Internet gambling and 'super-casinos' showed that gambling remains a sinful vice in the eyes of some.

The current gambling bogeyman is the ‘fixed-odds betting terminal' (FOBT), a device that offers virtual casino games in licensed betting shops. These machines are routinely dubbed ‘the crack cocaine of gambling’ and it is claimed that players can lose £18,000 in an hour. They are blamed for a rise in problem gambling and it is said that Britons spend £42 billion on the machines every year. It is also claimed that betting shops have proliferated as they scramble to cash in on the popularity of the machines.

These accusations are repeated so frequently in the media that they have come to be taken as truisms and, as is often the way in 21st century Britain, a campaign is underway to ban them. A new IEA paper - The Crack Cocaine of Gambling? - assesses the claims made against this new variant of gambling machine and finds a familiar pattern of scare-mongering that owes more to folklore than evidence.

Tellingly, the ‘crack cocaine of gambling’ label has no verifiable source and has been used by anti-gambling activists to describe virtually every new gambling product since the 1980s. It is a memorable, but ultimately empty, phrase designed to attract attention to whichever form of gambling is currently regarded as suspect. Existing evidence does not support the suggestion that fixed odds betting terminals have led to a nationwide rise in problem gambling, nor do the data suggest that these machines are uniquely ‘addictive’ or seductive. Indeed, it is far from certain that there has been a rise in problem gambling at all, despite wider availability and greater advertising.

The claim that Britons spend £42 billion on FOBTs each year is based on a misleading definition of the word ‘spend’. It is true that £42 billion was put into these machines last year, but campaigners neglect to mention that 97 per cent of what is put in is returned in prizes. The actual spend (i.e. stakes minus prizes) is around £1.5 billion, which is similar to the amount spent on traditional over-the-counter betting.

As for the ‘dramatic proliferation’ of betting shops, their numbers began to decline in the late 1960s and reached an all-time low at the turn of the century. Since then, there has been only a slight resurgence, with numbers rising by 4.5 per cent between 2000 and 2012. This is not a dramatic proliferation by any stretch of the imagination, and, contrary to popular belief, the bookmaking industry’s gross gambling yield has fallen slightly in recent years. The popular perception of a booming bookmaking industry is probably due to betting shops moving out of the side streets and onto the high streets - thanks to lower rents in credit-crunch Britain - and therefore becoming more visible.

Finally, the idea that people can lose £18,000 an hour on these machines is based on the improbable scenario of a player feeding £100 into the machine every 20 seconds and losing every time. Leaving aside the practical obstacles involved in this feat, the chances of losing so many games in succession are billions to one. This kind of loss, or anything close to it, would simply never happen in reality.

None of this myth-busting proves that fixed-odds betting terminals are ‘safe’ or ‘harmless’ - like all forms of gambling, they are likely to attract some compulsive gamblers - but it does suggest that campaigners are resorting to exaggerated and misleading rhetoric in the absence of solid data. The reliance on anecdotal evidence and unsubstantiated claims is characteristic of similar previous panics which were subsequently abandoned when it became clear that the new activity was neither especially pernicious nor particularly contagious.

The existing evidence base is incomplete in the case of fixed-odds betting terminals, but we can say with confidence that the prohibition or over-regulation of these machines will close off an avenue of pleasure to a significant number of people who enjoy them, leading to players turning to the less regulated online market instead. It will also damage Britain's bookmaking industry, which has worked hard to keep pace with changing tastes. It would have a detrimental impact on employment and would significantly reduce tax revenue paid by the gambling sector. Instead of clamping down on popular pastimes on the basis of hearsay, regulators should be encouraging innovation by British companies in a competitive global industry.

I find it very hard to believe that £42 billion is put int these machines every year. That is getting on for £1000 on average for every adult in the country. Can this really be so?
Notwithstanding the mathematical polemic presented above, the issue of gambling is a moral one and hopefully it will remain so. Its reduction to a purely utilitarian - economic issue should cause some concern. That £97 are returned as prizes for every £100 put in is meaningless in that context. So what? So what if it returned £99? There are age-old, timeless questions that human beings have placed in the moral arena. Why shouldn't they stay there? Gambling is questioned by every culture almost, the most in the Indo-European cultures. In the Indian semi-religious mythological text Mahabharat, one of things that led to crisis and war was gambling / wagering and its adverse moral outcome keeps coming up to haunt the heroes and villains right through the epic (its the longest epic known to mankind).
As a Labour Party member I find myself in deep shock finding an IEA paper I agree with. It is a shame that you claim that the Gambling Commission scrapped the Gambling Prevalence Survey. That was funded directly by the UK government and was cut by the coalition in those early days of pretend red tape burning. The Association of British Bookmakers has since made clear that the industry is willing to fund it (submission to the triennial review of gaming machine limits). You also fail to explain just how dodgy "Fairer Gambling" are with their figures, reproduced by the Guardian, Mail and in most of the articles you quote. The data they used has been published by the Guardian in spreadsheet format which gives it some increased credibility, indeed Caroline Lucas MP when writing to the minister sites the “Guardian” figures. They are here http://www.guardian.co.uk/news/datablog/2013/jan/04/5bn-gambled-britian-... The data is misleading for a host of reasons but the worst is that it is all based on counting the number of FOBTs and then assuming that each takes an industry average in wagers. They use total bets not gross yield to make the numbers seem as big as possible. The bigger deception is pretending that the data is local, it is not, it is a national average applied to the FOBT count. for an area. Now the bookmakers are in high streets and these tend to be “poorer” areas but the central claim that they are targeting the poor is false. Bookmakers go for footfall to decide where to set up. This January Guardian data is easy to spot as a national very basic model not local data as many constituencies have the exact same figures. Amusingly if you download the full spreadsheet you can see that the total bet claim for the constituency looks precise – ie Birmingham Ladywood £163,067,753 but they do themselves an injustice, the figure in the spreadsheet is far more precise £163,067,753.3333. They are claiming local betting data to the fraction of a penny! it gets worse though. The Guardian did a follow up article and this time “Fairer Gambling” give us the FOBT betting and yield figures for every single constituency. http://www.guardian.co.uk/news/datablog/2013/feb/11/cost-of-gambling-bri... Now this time non of the constituencies have identical figures, indeed when we download the data we find that the model has been tweaked to give us regional differences. Interestingly the regions match the Euro constituencies, shown only in the downloadable spreadsheet not the article, naturally. From the spreadsheet you can add a column to work out that the model used is again based on an assumed yield per machine – this time the average weekly yield per FOBT has three assumed figures – £657.77 or £967.31 or £1000.31. Every London region FOBT yields £1000.66 a week whether it is in Barking or Chelsea. The £967.31 yielding machines are in: Eastern Euro Region, South East Euro Region, South West Euro Region, The £657.77 yielding machines are in: East Midlands Euro Region, North East Euro Region, North West Euro Region, Scotland Euro Region, Wales Euro Region, West Midlands Euro Region, Yorkshire and Humber Euro Region. Now not explaining this crude model is misleading enough but Fairer Gambling still have the problem of identical constituencies revealing their model. Taking just the first two constituencies listed alphabetically -Aberavon and Aberconwy. Both have 9 bookies and 33 machines. They are both in Wales and so the assumed yield per week per machine is the same at £657.77. That is the model and it should give us £657.77*52 weeks*33 machines to give us the gross annual yield of £1,128,733.32 for BOTH constituencies. Instead we have £1,130,990.78664 and £1,131,298.623 listed (though shown to the penny not to the millionth of a penny). “Fairer Gambling” have inserted a fudge factor. Aberavon is said to have 33.066 FOBTs in the spreadsheet (though shown as 33) and Aberconwy 33.075 FOBTs (again shown as 33). Each constituency has been individualised by fractionally altering the number of FOBTs in the constituency – but never by enough to alter the model which is always a fixed yield per FOBT depending upon region. The only reason to have this false number for the FOBTs in the calculation is to fudge the figures a little to produce individual constituency figures that are different and so do not reveal how crude the model is. The “fudge factor” is deliberately altering the yield per constituency to hide the simple nature of the model and make it look like a calculation for each constituency. The whole “report” is deceptive – it pretends to give constituency level data but is really a fairly crude model. Very deceptive but it has delivered them lots more press coverage at a local level and even let them get the story rehashed for Wales and Scotland. It is clear from the spreadsheet that they are using this false data to lobby individual MPs and MEPs – they even list who they are in the spreadsheet to help them lobby. Judging by Caroline Lucas’s letter some MPs have been entirely taken in, they believe this “data” is a real report not a crude model that deliberately misleads regarding what is being spent on these machines by conflating yield with total bets and essentially inventing local data from crude assumptions and a deliberate effort to hide the model used.

Post new comment

The content of this field is kept private and will not be shown publicly.
Type the characters you see in this picture. (verify using audio)
Type the characters you see in the picture above; if you can't read them, submit the form and a new image will be generated. Not case sensitive.