Estimating consumer preferences for online music services

Abstract

This article analyses consumer preferences with regard to important attributes of online music services. Conjoint analysis and a random coefficient discrete choice model using Bayesian approach with Gibbs sampling are used to estimate the preferences. Based on the quantitative results, we use simulation to look at how a new pricing strategy and the threat of legal penalty for file sharing would influence the online music market. Findings include these: estimated willingness to pay for downloading one music file is significantly less than the actual price of the file; consumers are sensitive to longer search and download times for music files and very sensitive to the threat of legal action; and consumers are not sensitive to online music services broadening their catalogues. Finally, the simulation shows that a combination of increased transaction costs for illegal file sharing and lower-priced digital music files would inhibit illegal file sharing and bolster the number of people purchasing music legally from the online services.