Participation and level of play in the UK National Lottery and correlation with spending on other modes of gambling

Abstract

The paper analyses UK data, collected from the Family Expenditure Survey, for over 6000 households. The focus is on participation in, and level of expenditure on, the national lotto game. Controlling for standard demographic, socio-economic and income variables, and some non-standard lifestyle indicators, it examines correlation between lotto play and expenditure on six other forms of gambling as well as on alcohol and tobacco. Correlations with lotto participation and expenditure are found to be positive and strong in most cases. Implications for the regulation of the gambling sector and for taxation policy are outlined and discussed.

Keywords:

• lotteries
• gambling participation
• drinking
• smoking
• Family Expenditure
• Survey

Acknowledgements

David Forrest is grateful for support through a Fellowship from the Leverhulme Trust. David Gulley was supported by the Gibbons Research Professorship. Efficient research assistance in assembling data was provided by Olga Otborkina at Bentley College. FES data were obtained from the National Statistics website, www.statistics.gov.uk. FES data are Crown Copyright material, reproduced with the permission of the Controller of HMSO, London.

Notes

 1. Information on grants awarded is provided at www.lottery.culture.gov.uk

 2. For example, there is only a 15% tax on bookmaker takeout.

 3. A breakdown of ‘where the money goes’ is presented at www.national-lottery.co.uk

 4. ‘Weak’ regressivity means that lower income households pay a greater proportion of their income in tax. ‘Strong’ regressivity implies they spend a larger absolute amount on the tax.

 5. Smith shows the proportion of household income accounted for by stakes on the lottery falling almost monotonically across the income deciles, from 1.67% for the poorest 10% to 0.13% for the richest 10%.

 6. A number of studies, related to the USA and Australia, have employed panels of states to assess impacts on lottery revenue from the introduction and growth of casino gaming. Elliott and Navan (2002) and Fink and Rork (2003) found, for America, that lottery revenue had been cannibalised by riverboat and commercial (non-tribal) casinos respectively; these findings are consistent with those from a study by Walker (2007) on the generality of casinos. The Australian case, analysed in Farrell and Forrest (2008), is possibly more relevant to that of Britain since states were introducing casino style gaming to an environment where there was already high availability of betting and other modes of gambling as well as lotto and scratchcards. The richness of Australian data permitted them to distinguish between two types of casino. Destination casinos, of which there is typically one per state, had no impact on the lotto market. However, in states that allowed the emergence of dense networks of casino type facilities located within hotels and clubs, where the product offering was high prize electronic gaming machines (sometimes hundreds in number), the consequent decline in lotto sales typically reached 20–30%. The implication drawn was that lottery sales were highly vulnerable to modes of gambling that competed with it in terms of accessibility and high prize levels. A similar finding in the USA was that lottery sales in border counties typically fell substantially when neighbouring states introduced large scale slot machine gaming facilities at racetracks (Tosun & Skidmore, 2004).

 7. For a general discussion of the reliability of the FES data, see FES 2000-01User Guide, v.1.

 8. Dummy variables for other months and regions proved insignificant in preliminary estimation.

 9. Thus, a household's spending decisions could of course be modelled with a system of simultaneous equations. Doing so would add great complexity to our analysis, while adding little in the way of improved results because our variable of interest, lotto expenditures is a very small fraction of total spending.

10. This is a long-odds lotto-style game, based on soccer results, that is provided by the private sector and offers jackpots of the order of magnitude of £1 million.