PREVISIONES A CINCO AÑOS

One hundred and fifty-five undergraduate students from a large university were recruited to participate in a 20-minute within-subject computer study in return for $5 cash17. At the

beginning of the study, these respondents were asked to indicate their favorite men’s college basketball team (over 90% of respondents listed the home university as their favorite team). Subsequent questions were then automatically adjusted so that they were pertinent to that particular team. To be eligible for inclusion in the study, respondents had to answer a set of questions designed to test their understanding of the options concept; four students were dropped from the sample because they did not pass this screen, resulting in a final sample size of one hundred and fifty-one fans.

Fans indicated their WTP for tickets to the final game for three different pricing conditions: (1) advance selling, (2) consumer options, and (3) full information pricing. Across the pricing conditions, I offered fans the same broad range of prices ($0 to $400, in increments of $20). In the advance sellingcondition, fans indicated their WTP for a “regular ticket” given three uncertainty levels (i.e., probability of the favorite team making it to the final game; γ=0.25, 0.50,

16

Of the fans in our study, 63% volunteered to participate in the drawing. Participants were informed that if their name was drawn, they would be required to buy the ticket by paying the average price they had indicated in our study. The winner of the drawing paid $200 for an upper-level ticket to the 2007 Championship game at the RCA Dome in Indianapolis.

17

We use a within-subject design because the consumer options concept is new. To help ensure that this concept was understood, fans had to think about the option ticket as well as the more familiar regular ticket. Pretests further indicated that fans were not overwhelmed with the three pricing conditions because their responses simply involved clicking on their preferred dollar amount.

42

and 0.75) of their favorite team making it through the tournament. Based on extensive pre- testing, in the consumer options condition I fixed the levels of the option price (P_o) for these three probabilities at $20, $40 and $60, respectively18. Given a certain option price, fans indicated the exercise price they were willing to pay (

∧

e

P ) at the three different uncertainty levels

(thus, expected WTP for the option ticket is )

∧ + _e

o P

P γ 19. In the full informationcondition, fans indicated their WTP for a “regular” ticket after they were told whether or not their favorite team actually made it to the final game. Importantly, fans had the choice of not buying a ticket under all pricing conditions—they could simply indicate a WTP of $0.

Because this is a within-subject design, the presentation order of tickets was counter- balanced to control for possible order effects20. To control for differences in payment timing

across the pricing conditions (e.g., fans pay an exercise price in the future, whereas they pay in the present under advance selling), I asked fans to indicate their WTP the exercise price in the present should their favorite team make it through the tournament. To control for the possibility that WTP may be influenced by the mere fact that choices become unavailable in the future (e.g., Ariely and Shin 2004), I asked fans to assume that both ticket types (options and regular) were

18

Fans were explicitly instructed that a WTP of $0 for the exercise prices across all the uncertainty levels would be treated as if they did not want to purchase the option ticket.

19

We do not allow fans to choose both the option and the exercise price because we want to obtain their WTP without confounding our measures with individual differences in risk preferences. For example, suppose we allow subjects to state their willingness to pay for both the option and the exercise price, and one subject was willing to pay an option price of $50 and an exercise price of $100 given a 0.25 probability of her favorite team making it to the big game. Suppose another subject was willing to pay $25 for the option price and $100 as the exercise price given a 0.50 probability of her favorite team making it through the tournament. In this case, the expected value of both subjects’ willingness to pay is the same ($75). However, it can be argued that the first fan who was willing to pay more for the option when there was a lower probability of the team playing in the final game is more risk-averse than the latter.

20

Because we find no significant order effects in any of our analyses, we ignore this factor in the remainder of this chapter.

43

only available at the start of the season (before the uncertainty of the final team matchups has been resolved).

Finally, because I also want to better understand the behavioral determinants of consumer willingness to pay under various pricing approaches, I follow Padmanabhan and Rao (1993) and measure fans’ risk preferences based on their responses to a lottery question (fans who are willing to pay less for the lottery than its expected value are considered to be risk-averse; these fans represent 81% of the sample).