| Abstract: | The paper discusses and illustrates estimation methods for identifying and modelling the influence of producers on the prices for a differentiated product. A two‐step fixed effects approach is contrasted to the random effects specification. The techniques are employed to analyse prices for over 260 Australian premium wine producers. The fixed effects specification is preferred given significant correlation between wine attributes and random producer effects. The estimation of fixed producer price effects identifies statistically significant price premiums and discounts, which average 15%. Fixed producer price effects are estimated to depend on the quality reputation of the producer, its level of experience, producer size and the use of multibrands by conglomerates. In part, results indicate that price discounts are associated with producers who have low‐quality reputations, are small and recently established. |
