Consumer Surplus of Maritime and Amateur Aviation Radio Users

Discussion - John Swanson

Discussions focused on SP design issues raised by the attributes which were chosen to present in pairwise comparison, and the difference between the valuations obtained from quality and cost increases and decreases.

Comparisons of equipment lifetime costs were chosen to prevent policy bias, with the question order randomised and decoy costs also added. Some thought a shorter time frame, such as a year would give a truer estimate of WTP. But the questions framed needed to be realistic questions, costs per year would be hard for people to compare.

There was a need to ensure the respondents were thinking and choosing with the same time frame in mind. The question was raised as to whether all individuals were considering the same estimated equipment lifetime. In the survey the lifetime estimate of each individual was obtained and used.

The running costs of a boat or light aircraft, are significantly higher than the radio equipment replacement costs. Therefore the questionnaire may be asking more about the willingness to pay for the boat than the radio. This too is a potential problem.

Samples for particular sectors were small, as low as 30 in some cases, but this was representative of the population and analysis was not performed at that level. Sampling bias may have been present if the consumer surplus of radio to overseas registered boats differed from UK registrations, as none were included in the survey sample.

Design abstraction may have resulted in unconstrained response bias, ie because people were not obliged to act according to their response. (Standard practice is to model Random Utility as a function of product characteristics for non-market goods).

Analysis of cost reductions and cost increases were performed separately. However, respondents did not respond to cost reductions to the same degree. This 'inertia' to change, can be incorporated by the use of non- constant coefficients in the model. Work in this area is currently being done by Daly, Brownstone, Train (Berkeley), McFadden (Berkeley) and Ben-Akiva (MIT) among others.