The British betting agency, William Hill, is known for accepting unusual wagers. They have taken bets that a man would set foot on the moon before January 1, 1970 (at odds of 1,000:1), that Elvis will be found alive (1,000:1), and that Tony Blair will prove to be an alien (10,000:1). In addition, they have accepted bets from people who are trying to lose weight, often at odds of up to 50:1 and with payoffs of over $6,000. The person placing the bet, whom we refer to as the “bettor”, wins the bet if he loses a specific amount of weight by a certain date. If successful, he receives a payoff based on the odds and the size of the wager.
The first recorded weight loss bet was placed in 1993, when Mr. Graham Trow wagered $68 that he could lose 28 pounds in 56 days, with odds of 28:1. Mr. Trow managed to win his bet and received $1,900 for his efforts. Since then, William Hill has received many requests for bets from individuals seeking to incentivize their dieting efforts. Using a data set comprising bets from 1993 to 2006, we provide insight into the decision making behind this unusual activity.
At first it seems puzzling that a bookmaker would accept a high-stakes bet from someone who arguably has control over the outcome. While gambling is a common activity, it is rare that the bettor has direct influence over the probability of winning. In most professional and collegiate sports, for example, persons with an influence over the outcome of an event are prevented by law from betting on it. In many of these weight loss bets, the bettor could win the bet by reducing his daily food consumption by the equivalent of a Starbucks hot chocolate. William Hill is willing to accept these types of bets for two reasons: the low success rates and the large amounts of free publicity that the bets generate. Since 80 percent of bettors lose their bets, William Hill only pays out an average of $500 per year on weight bets. In return, they receive the equivalent of tens All amounts listed in dollars have been converted from nominal British Pounds (GBP) to year 2004 US dollars. (sometimes hundreds) of thousands of dollars of free publicity in the British media.
Initially, weight bets do not appear to be much more successful than other weight loss programs. According to a comprehensive review paper by Ayyad and Andersen (2000), the median long-term success rate for dietary treatment of obesity is 15 percent, very close to the overall success rate of weight loss betting. Granted that traditional dieting has a high failure rate, one might still expect that providing a substantial monetary reward for losing weight would be more effective. In Section 4, we will explore how bets could actually be much more effective than the overall success rate suggests.
We use correspondence with the bettors to propose an explanation for this market based on the time-inconsistency literature. Bettors describe their motivation for making weight loss bets in a manner that is consistent with the model of partial naivete in O’Donoghue and Rabin (2001). Bettors are aware that they have self-control problems so they actively seek out commitment mechanisms (such as personal gambles) to alter their own behavior. However, they underestimate the magnitude of their self-control problems and sign up for bets that they can’t possibly win. The commitment mechanism may succeed in encouraging some weight loss but rarely, if ever, is it sufficient to prevent loss of the bet.
The betting market we study is small and idiosyncratic, but it contains insights into the growing problem of obesity. If a potential payoff of thousands of dollars is not sufficient to motivate people to achieve a desirable goal, then reducing obesity levels with direct economic incentives (an in creasingly popular policy proposal) is unlikely to be effective. We do not claim that the individuals we study are representative of the general population; it is very likely that people with higher than average costs of losing weight self select into weight loss bets. Nevertheless, the high failure rates we observe suggest that the merits of monetary incentives to effect behavioral change should not be overestimated. Instead, we evidence that suggests the temporal structure of incentives should receive equal, if not greater, attention.
Our paper is organized as follows. In Section 2 we begin with a brief discussion of the challenges posed by obesity and of attempts to use incentives to motivate weight loss. Section 3 describes our data set and outlines the process of placing a weight loss bet. In Section 4 we present a potential explanation of bettor behavior and generate some testable hypotheses. Section 5 considers the bookmaker’s decision-making process and uses the odds offered by the bookmaker to test the predictions of the model. Finally, in Section 6 we discuss our findings and implications for future research.
Download
PDF Ebook Betting on Weight Loss ... and Losing: Personal Gambles as Commitment Mechanisms
