Deal Or No Deal

How much do you love the new American version of the international hit “Deal or No Deal?” Do you watch it as I do, with a calculator in hand, computing the expected value and predicting bank offers on the fly? Do you marvel how the program showcases behavioral economics in action, demonstrating that the statistically “best” choice—rejecting any bank offer that does not meet or exceed the expected value of the player’s case—is not generally the correct choice in light of risk aversion and the diminishing marginal utility of money?

No? Well, you may think that makes me cooler than you, and in a way it does, but neither of us is as cool as the team of Dutch economists who have analyzed the show in depth and published a paper entitled Deal or No Deal? Decision Making Under Risk in a Large-Payoff Game Show.

Here is the abstract of their paper:

The popular television game show “Deal or No Deal” offers a unique opportunity for analyzing decision making under risk: it involves very large and wide-ranging stakes, simple stop-go decisions that require minimal skill, knowledge or strategy and near-certainty about the probability distribution. Based on a panel data set of the choices of contestants in all game rounds of 53 episodes from Australia and the Netherlands, we find an average Arrow-Pratt coefficient of relative risk aversion (RRA) between roughly 1 and 2 for initial wealth levels between €0 and €50,000 and assuming full rationality. Risk aversion is lower if we allow for myopic or hyperopic framing. The RRA differs substantially across the contestants and some even exhibit risk seeking behavior. The cross-sectional differences in RRA can be explained in large part by the previous outcomes experienced by the contestants during the game. Most notably, consistent with the “break-even effect”, the RRA strongly decreases following earlier losses and risk seeking arises after large losses.