Beyond Plain Vanilla: Modeling Joint Product Assortment and Pricing Decisions

Abstract

This paper investigates empirically the product assortment strategies of oligopolistic firms. We develop a framework that integrates product choice and price competition in a differentiated product market. The present model significantly improves upon the reduced-form profit functions typically used in the entry and location choice literature, because the variable profits that enter the product-choice decision are derived from a structural model of demand and price competition. Given the heterogeneity in consumers' product valuations and responses to price changes, this is a critical element in the analysis of product assortment decisions. Relative to the literature on structural demand models, our results show that incorporating endogenous product choice is essential for policy simulations and may entail very different conclusions from settings where product assortment choices are held fixed. Electronic copy available at: http://ssrn.com/abstract=1077441 AbstractIn this paper, we take a first step toward exploring empirically the product assortment strategies of oligopolistic firms. Our starting point is a discretechoice demand model for differentiated products. We incorporate the demand model into an equilibrium supply model, in which firms compete by first choosing which products to offer and then by setting prices. We show how modeling joint product assortment and pricing decisions enriches standard product choice models by allowing insights into how demand characteristics affect firms' product offerings in a competitive environment. We furthermore demonstrate that incorporating endogenous product choice into demand models is essential for policy simulations (e.g., mergers) as it entails at times dramatically different welfare assessments than the common assumption that product assortments are exogenous.

Enriching interactions: Incorporating outcome data into static discrete games

“…Finally, the emphasis on inequality conditions derived from revealed preference arguments and the speci…c decomposition of the errors into structural and non-structural components draws upon the recent moment inequality literature 2 There is a wide and growing literature on discrete games in both economics and marketing. Notable examples include Aguirregabiria and Mira (2007), Bajari, Benkard, and Levin (2007), Berry (1992), Pakes, Ostrovsky, and Berry (2007), Draganska, Mazzeo, andSeim (2009), Hartmann (2010), Ho (2009), Orhun (2006, Pesendorfer and Schmidt-Dengler (2007), Sweeting (2009), Vitorino (2007, and . provide an overview of this rapidly expanding …eld.…”

Endogenous product choice: A progress report

“…25 Based on their results, Draganska et al (2009) simulate the effect of a merger between the two largest ice cream manufacturers and find that incorporating endogenous product choice into the analysis reduces consumer surplus due to decreased variety, but increases consumer surplus due to lower prices at the new varieties. The difference in consumer surplus allowing or not endogenous varieties is slight.…”

“…Papers in the former literature therefore add profitability conditions familiar from entry models to the standard (demand + pricing) framework to recover the fixed costs of such product introductions. Examples in this vein include Draganska, Mazzeo and Seim (2009), Eizenberg (2011), Nosko (2010), and Sweeting (2011. 22 In other settings, the set of products cannot easily be changed (or cannot easily be modeled) and the analysis focuses on the continuous choice of one or more product characteristics.…”

Integrating assortment selection, pricing and mixed-bundling problems for multiple retail categories under cross-selling

“…Also in the literature, assortment planning problem may be considered dynamically for a multi-period horizon (e.g., Hall et al, 2010;Talebian et al, 2014, Rabbani et al, 2014 or applied in an oligopolistic environment instead of a monopolistic one (e.g. Draganska et al, 2009;Zhang, 2015); also, the price may be changed over planning horizon (Saha & Basu, 2010). As Ghoniem et al (2013) mentioned, there are many significant computational challenges in the literature of product line optimization problems, and this problem have been mostly tackled via heuristics (McElreath et al, 2010).…”