Case of the Day: Measuring Utility
Consider the Freakonomics feature on page 115 of the Goolsbie, Levitt, and Syverson text about the valuation of the continued presence of the Minnesota Vikings (American) football team by residents of Minnesota. (The quite complicated paper by Crooker and Fenn upon which this feature is based is available here, but you are not expected to read it unless you want to.)
1. Sports teams with large fan bases may be "public goods" in that many people can enjoy the benefits of the team without paying; they can watch the team on television, read about it in the media, and derive vicarious enjoyment from its success (when the team is successful) without making any measurable expenditure. Why does this public-good aspect of the product make it difficult to assess its value?
2. As noted in footnote 2 of the feature, this "contingent valuation method" study uses surveys to try to assess the value of a hypothetical option to people. What are the inherent difficulties of this approach? How much faith do you have in the results? Given the public-good nature of the product, can you think of any other alternative methods for valuing the team?
3. The authors of the study sent out 1,200 surveys to randomly selected residents of Minnesota, of which 42% were returned and used in the study. Based on common sense, what kinds of recipients would be more or less likely to return the survey? In other words, would you expect that the 42% of residents returning the surveys would be typical of the entire population or do you think that those choosing to respond would differ in their responses (and, in particular, in their valuation of the team) from those choosing not to respond? How might this "response bias" affect the results of the study?
4. The authors asked respondents how many Vikings games they had attended. The average response to this question was 0.33. Minnesota had about 5 million people in the late 2000s; Viking attendance was about 500,000 for the season. Respondents also reported that they watched an average of 8.2 (out of 16 possible) games on television; the Minneapolis Star Tribune reports that the Vikings' TV rating (the share of houses with a television watching an average game) in 2009 was 38.7%. What does this information about the respondents imply about the possibility of response bias (as considered in the previous question)?
5. Suppose that you were an intern for a Minnesota legislator and were asked to assess the implications of this study for a bill to use state money to subsidize a new stadium. In a short paragraph, what would you have told her?