Recent evidence from aggregate models of automobile demand indicates that, when not corrected for quality differences, market price elasticity of demand is substantially biased downwards. This note presents new information on market price and income elasticities derived from a disaggregate demand model that controls for cost, household income, vehicle attributes and perceived quality, consumer search, and manufacturer. Based on an extensive household survey of new vehicle purchasers in 1989, market price and income elasticities are estimated to be -0.87 and 1.70, respectively. Moreover, excluding vehicle quality from a well-specified model is found to have little effect upon the estimated market elasticities. Copyright 1996 by MIT Press.

A brand switching model that considers the choices: previous choice, current choice, and substitute choice, if the current choice were not available, is developed and estimated. An important assumption of the model is that the market consists of two types of consumers: “Loyals” and “Shoppers.” The model provides estimates for: (1) the proportion of Loyals in a submarket, and (2) the success of each submarket in capturing Shoppers. We illustrate how the model can be used to characterize the competitive structure of a market. We also show how the estimated parameters of the model can be validated using simulation and a theoretically derived covariance matrix. Our empirical application to automobile data from J. D. Power and Associates provides insights into several interesting marketing phenomena including: (1) the competitiveness of domestic versus foreign makes of automobiles, (2) competition between submarkets defined by form, and (3) the effect of previous dealer experience on loyalty and shopping.