Distribution Intensity and New Car Choice
Published 8/1/2008
Author: Randolph E. Bucklin, S. Siddarth, and Jorge M. Silva-Risso
View this content
Executive Summary
For much of the marketing mix, empirical evidence on effect sizes is widespread, and consensus has emerged on issues such as average magnitudes of price and advertising elasticities. Conversely, empirical studies for the distribution element of the marketing mix have been scant, with only a few studies on the relationship between distribution intensity and market share. In addition to academic interest in this issue, practitioners also need to be able to assess how changes in distribution (e.g., the size and structure of a dealer network) might affect demand for their products and services. Consider the challenge facing the U.S. automakers General Motors, Ford, and Chrysler, all of which have publicly stated their desire to reduce the size of their dealer networks but confront high costs in doing so. To target their scarce resources, firms need to gauge the effect on demand, if any, from proposed dealer consolidation plans (e.g., consolidating in one city versus another). To select which dealerships to target for consolidation, managers also need to assess what may happen if one location is shut versus another. Conversely, companies such as Toyota, whose sales in the Untied States have been growing, may be interested in adding new dealers to their network. Decisions regarding where to do so might also benefit from new insight into the relationship between distribution and sales.
The authors develop a new method to assess how changes in the intensity of mature distribution networks—specifically those in the U.S. automotive industry—may affect consumer choice. They capture distribution intensity by car make (e.g., Honda, Toyota) at the disaggregate level using the exact geographic locations of individual buyers and new car dealers. They develop three buyer-centric measures for the intensity level of each competing make from this information: (1) dealer accessibility (the buyer’s distance to the nearest outlet for each make), (2) dealer concentration (the extent to which multiple dealers for each make are located near a given buyer), and (3) dealer spread (dispersion of the multiple dealers for each make relative to the buyer’s location). The authors propose a logit choice model, estimated with Bayesian methods, to study the association of these measures with new car choices. The model was applied to buyer records in the midsize premium sedan category, drawn from an anonymous sample provided by the Power Information Network. All three buyer-centric measures of intensity were significantly related to new car choice. Buyers were more likely to select cars whose dealer networks had shorter distances to the closest outlet (accessibility), more dealers within a given radius from the buyer (concentration), and locations that skewed toward the buyer (spread). Based on the modeling results, the market share elasticity of distribution intensity averages approximately .6 across the new car models in the study. The approach should help firms evaluate the potential effects of expanding or contracting distribution networks for mature products.
Biography
Randolph E. Bucklin is Peter W. Mullin Professor in the Anderson School of Management at the University of California, Los Angeles, where he has taught since 1989. He holds a PhD in Marketing and an MS in Statistics from Stanford University and an AB in Economics from Harvard University. His research interests are in the quantitative analysis of customer shopping behavior, market segmentation, and interbrand/interstore competition. His work emphasizes the development of models for use with disaggregate-level data, such as Universal Product Code scanner panel and Internet clickstreams. He is the coeditor of Marketing Letters and serves on the editorial boards of Journal of Marketing Research, Marketing Science, and International Journal of Research in Marketing.
S. Siddarth is Associate Professor of Marketing in the Marshall School of Business at the University of Southern California. He holds a BTech from the Indian Institute of Technology, Delhi; a PGDM from the Indian Institute of Management, Calcutta; and a PhD in Marketing from the Anderson School of Management at the University of California, Los Angeles. His research interests lie in the development and use of response models to understand consumer and market behavior. He currently serves on the editorial board of Marketing Letters and International Journal of Research in Marketing.
Jorge Silva-Risso is Associate Professor of Marketing in the A. Gary Anderson Graduate School of Management at the University of California, Riverside. His research interests include econometric models of consumer response, marketing productivity, and the effects of the Internet on consumer behavior. His work has been published in American Economic Review, Journal of Marketing Research, Marketing Science, Journal of Econometrics, Journal of Marketing, and other journals in business and economics. Jorge won the 1995 Clayton Award from the Marketing Science Institute, the 2006 Marketing Science Practice Prize, and the 2007 Paul Green best-paper award and was a finalist for the 2007 INFORMS Edelman Award. Currently, he is the ISMS Vice President of Practice. He holds a PhD in Management (1996) and an MBA (1991) from the University of California, Los Angeles.
Journal of Marketing Research, Volume 45, Number 4, August 2008
View Table of Contents.