Willingness Toupee

David M. McEvoy, O. Ashton Morgan and John C. Whitehead

No 19-01, Working Papers from Department of Economics, Appalachian State University

Abstract: In this paper we tackle the hairy problem of male pattern baldness. We survey balding men and elicit their willingness to pay to move from their current sad situation to a more plentiful one. Then we comb-over the results. What’s the average willingness to pay to move from a glistening cue ball to a luscious mane? About $30,000.

Key Words: mullet, skullet, comb-over, ducktail, Beatlemania, buzz cut, whiffle, pageboy, attribute non-attendance

The sole reference: Carilli, Anthony M., “Scarcity, Specialization, and Squishees,” Chapter 1 in Homer economicus: The Simpsons and economics. Joshua Hall, ed., Stanford University Press, 2014.

Some sample footnotes:

1. As is standard in the discipline, author order is determined by reverse Norwood Baldness Scale.
2. The “stone piece” was a block of dark slate tied around the head to achieve the appearance of a full head of hair. While there are no sources of any such thing actually taking place, the authors imagine that it must have happened.
3. “In ‘Simpson and Delilah,’ Homer attempts to pursue an executive position in which he doesn’t have a comparative advantage. Mr. Burns confuses Homer with a young go-getter and promotes him to an executive position after Homer has managed to scam himself some Dimoxinil–a miracle cure for baldness–and grow some hair.” (Carilli 2014, p. 11)
4. It is important to note that the authors did not even bother looking for other studies.

7. Both of these models can be found in the NLogit manual (www.limdep.com) or via Google Scholar. They’re legit but we really don’t want to add any references besides the Simpsons book.

9.Referee #2 may try to claim that you cannot estimate WTP from a mixed logit model with a price parameter distribution that includes negative values because these respondents’ WTP will be undefined. Since distributions that constrain WTP to the positive realm do not perform as well statistically as the normal (we didn’t really check this) and (likely) generate goofy WTP estimates, we choose to present WTP estimated with the mean coefficients. The gullible, er, reasonable, reader will just go along with it since the MXL WTP number is so close to the ECLC WTP estimate and this lends reliability to our data.