Differentiated Products Demand Systems from a Combination of Micro and Macro Data: The New Car Market

Abstract

In this paper, we consider how rich sources of information on consumer choice can help to identify demand parameters in a widely used class of differentiated products demand models. Most important, we show how to use "second-choice" data on automotive purchases to obtain good estimates of substitution patterns in the automobile industry. We use our estimates to make out-of-sample predictions about important recent changes in industry structure.We thank numerous seminar participants, two referees, and the editors Lars Hansen and John Cochrane for helpful suggestions. We also thank the National Science Foundation for financial support, through grants 9122672, 9512106, and 9617887. We are particularly grateful to G.

The Welfare Effects of Coordinated Assignment: Evidence from the New York City High School Match

“…To exploit the richness of rank-ordered data, we do not restrict the correlation across the dimensions of γ i . Berry, Levinsohn, and Pakes (2004) show that data on top and second choices improves on estimates that only use first choice by revealing common characteristics between subsequent rankings for a given student. Rank-ordered data also allow us to relax the common assumption that random coefficients on choice characteristics are independently distributed.…”

“…It is worth noting that the difference in levels between our model and the data is small compared to the difference between the average distance to a high school in New York (12.7 miles from home) and the closest school (less than a mile from home). Berry, Levinsohn, and Pakes (2004) emphasize the importance of random coefficients models in the context of rank data for automobiles. In particular, they emphasize that when examining the within-consumer relationship between the attributes of alternatives ranked first and second, models without random coefficients do a poor job.…”

Product Differentiation and Mergers in the Carbonated Soft Drink Industry

“…The econometric procedure also corrects for the remaining within-panel unobserved serial dependence. The point- 4 In Dubé (2000), I present comparable evidence for other di¤erentiated products industries such as ready-to-eat cereals, canned soups and cookies. 5 One needs to estimate at least as many parameters as the square of the number of merged products.…”

“…For instance, a typical household shopping trip for CSDs may involve the purchase multiple units of an assortment of brands from a broad set of more than 1000 SKUs. 4 The traditional residual demand approach reduces analysis to the merged …rms' products without assuming speci…c consumer behavior (Baker and Bresnahan 1985), but its feasibility diminishes for mergers between …rms with extensive product lines 5 . The multi-level demand approach reduces the number of estimated parameters by grouping products a priori into segments and assuming consumers make sequential budgetting decisions (Hausman, Leonard and Zona 1994and Hausman 1996, Cotterill, Franklin and Ma 1996.…”

Practice Prize Winner—A Nested Logit Model of Product and Transaction-Type Choice for Planning Automakers' Pricing and Promotions

“…Unfortunately, their specification does not allow decision makers to plan future promotions or derive insights about how characteristics of each promotion program have driven the results or what the effects of competition might be. More recently, Berry et al (1995Berry et al ( , 2004, Sudhir (2001), Train and Winston (2007), quantify consumer response to price in the automobile market, but their price analyses are limited to MSRP and do not address the multiple instruments that are used for price customization. Bruce et al (2006) examine the logic of offering consumer rebates by automakers in a context in which consumers face a constraint in their "ability-to-pay" for a durable product (automobiles) with an alternative second-hand market.…”