UCL News


Few really want to be millionaires

24 August 2005

Analysis of contestants on the game show, ' Who Wants to be a Millionaire? ' has shown few people really want to be millionaires, the 2005 World Congress of the Econometric Society, which is hosted by UCL (University College London), will hear today.

The paper, which will be presented by Professor Gauthier Lanot of Queen's University Belfast found that few people are willing to take the necessary risk to win the jackpot. Many contestants assess the risk of failing to answer the question they are presented and quit while they are ahead.

Working with Professor Ian Walker of Warwick University and Professor Roger Hartley of Manchester University , Professor Lanot used an economic model to analyse the 515 contestants for the first eleven series of the show (it started in June 2003). Overall, three players won the £1 million prize. Two-thirds of players quit while they were ahead and one-third didn't answer their final question correctly. 'Failures' left the studio with an average of £17,438 while 'quitters' went away with an average of £72,247.

"The data provides a real insight into how people in the UK and elsewhere take risks in life especially when they're facing choices where uncertainty plays a big role," explains Professor Lanot.

"The format of gameshows like Who Wants to be a Millionaire? is as good as it gets when it comes to measuring the sensitivity of individuals to risk. It provides us with data which is akin to data from controlled experiments in other fields. Economists are intrigued by how people take risks in life but no research council or university would be prepared to give us many millions of pounds to reproduce such a game show environment."

Who Wants to be a Millionaire? has a number of features that makes it well-suited to analysis. The format is straightforward, it involves no strategic decision-making, and the prizes are cash, and paid immediately. The range of prizes is large, up to £1 million.

Each player is faced with a sequence of 15 multiple-choice questions. At each stage the contestant can answer the current question and stands to, approximately, double their current winnings but at the risk of losing a specific amount if their answer is incorrect. Alternatively the contestant can quit and leave the game with the winnings to date. At each stage of the game contestants are reminded that their winnings so far belong to them - to risk, or walk away with.

The data was transcribed from the original videotapes of the first 11 series. To establish how representative of the UK population the contestants on the tapes were, they were surveyed to obtain information about their characteristics, which was compared with selected information from population surveys such as the Labour Force Survey, the Family Expenditure Survey, the British Household Panel Study and the Gambling Prevalence Survey.

The comparison showed that more men took part in the game show, which is consistent with previous studies that have found men are more likely to take risks. But overall the analysis indicated that contestants weren't more likely to take risks than the average member of the population.

"The proportion of individuals who said their household contents was not insured was similar to the average population and the proportion that report being regular lottery ticket purchasers is also quite similar. This suggests contestants on Who Wants to be a Millionaire are a good model for the rest of the population in the kind of risks we all take," explains Professor Lanot.

"Consider a hypothetical situation where you have to take part in a simple game of heads or tails. If Heads appears, in half of the cases, you are asked to pay, say, £1000. If on the other hand Tails appears, again in half of the cases, you will be paid £1000.

"The question of how much would you be prepared to pay to avoid playing the game relates to your coefficient of risk aversion, in such a way that if you are very risk averse you will be prepared to pay a relatively large amount, while if you are not affected by the presence of risk you are not prepared to pay anything.

"Knowing a precise value for this coefficient will allow economists to understand individual behaviour better. In particular in relation to the take of individual insurance, pensions plans or explaining saving and investment behaviour in general. Furthermore, it may be useful to the government when it decides whether or not to undertake large public projects with uncertain consequences such as the construction of new hospitals, the funding of the education system, or even the decision to go to war."

Professor Gauthier Lanot of Queen's University Belfast will present the paper, 'Who really want to be a millionaire? Evidence of risk aversion from gameshow data' on Wednesday 24 August between 14.15 and 16.15 BST.


For further information, please contact:

Professor Gauthier Lanot,
Queen's University Belfast,
Mobile: +44 (0) 7947581011,
Email: g.lanot@qub.ac.uk

Judith H Moore,
UCL Media Relations,
Tel: +44 (0)20 7679 7678,
Mobile: +44 (0) 77 333 075 96,
Email: Judith.moore@ucl.ac.uk

Notes to editors

A copy of the paper is available on request from the UCL press office

About the 2005 World Congress of the Econometrics Society

The Econometric Society is the leading international learned society in the field of economics, and its quinquennial world congress is recognised as the most prestigious in economics. UCL is hosting the ninth Econometric Society World Congress from 18-24 August 2005, which is the first time the Congress has been held in London and has not been hosted by a UK institute for 35 years. A full copy of the programme can be accessed on the 2005 Econometric Society World Congress website: http://www.eswc2005.com/