Keynote: Chuck Weinberg (British Columbia) An Empirical Study of Uniform and Differential Pricing in the Movie Theatrical Market (Paper: Jing Yan , Jason Ho  and Charles B. Weinberg)
Although movies vary widely in appeal, quality, and cost, a multiplex in most markets charges the same ticket price for all non-3D movies it shows (during the same time period) and another price for all 3D movies. Both industry executives and academics researchers have suggested multiple reasons for the uniform pricing puzzle. However, minimal empirical work directly addresses this issue, likely because there is no observed price variation across different movie titles within a theater. In Hong Kong, however, prices vary both within and across multiplexes. Building a dataset consisting of daily ticket prices and sales by theater and title, we study this pricing puzzle empirically. To estimate the effects of different pricing on demand, our structural demand model extends the well-known Principles of Differentiation model by nesting the original PD model under the choice of whether or not to see a movie. This allows for a fuller study of primary demand effects from pricing.
Among the various rationales advanced on the pricing puzzle (for example, consumers’ preference for uniform prices, the influence from the upstream channel members, and the costs of implementing differential pricing), a critical factor, rarely considered, is the effect of competition among movie theaters. This paper thus focuses on how uniform ticket prices can soften the competition, which would otherwise be intense under differential pricing. Specifically, we expect the differential ticket pricing would create two opposing effects: (1) an increase in profits due to the additional surplus the differential ticket prices can extract from consumers; (2) lower profits due to the competition intensified by each multiplex having a larger set of price alternatives.
Based on the empirical results, which show that demand is sensitive to price after controlling for a variety of other effects, we conduct counterfactual simulations to compare the outcomes of differential and uniform pricing. Assuming that competing theaters are playing a Bertrand pricing game, we find movie theaters realize a higher profit (on average, because not all theaters are better off) when using a differential as compared to a uniform pricing strategy. However, the profit improvement is relatively small, consistent with the view that the competition-softening effect of uniform pricing may offset the surplus-extracting effect of differential pricing in practice. The paper concludes with a discussion of managerial implications.

Yuanfang Lin (Toronto) The Blessing of Targeted Innovations in a Competitive Market (Paper: Yuanfang Lin, Amit Pazgal and David A. Soberman)
Innovation is a key component of competitive strategy but a …firm must introduce as well as create innovations to gain advantage. These introductions are anything but simple. Rapid advances in technology and product evolution can result in innovations not being fully understood by consumers. As a result, choosing the optimal level of product innovation in competitive markets is challenging. To examine this question, we study a …firm’s decision in terms of both the nature of innovation and pricing when it enters a market where an incumbent provides a basic product with technology that everyone understands. In a market with two consumers, an entrant chooses between introducing a new product that represents a) a drastic innovation targeted to one consumer or b) a general improvement of smaller magnitude that both consumers value. Drastic innovation implies advanced functions over the basic product but these new features may not be appreciated by all consumers. This implies that the willingness to pay for an innovation classifed as drastic is often heterogeneous. Our analysis shows that, when an entrant introduces a drastically innovative product that is perceived heterogeneously, price competition with the incumbent entails mixed-strategy equilibrium where the entrant maximizes pro…fit from the consumer who pays extra for the new features. In contrast, an entrant with a new product class…ed as a general improvement sets price to capture business from both consumers. Here, sales for the incumbent’s basic product tail off. The ability to substantially reduce the incumbent’s sales suggests that general improvements should be globally preferred to drastic innovations with narrower appeal. This reasoning is ‡flawed because product innovations have indirect effects on market competition. In particular, to create equivalent pro…fit for an entrant, a general improvement needs to create signi…cantly more value in the market than a drastic innovation for a sub-segment. The reason is that a locally drastic innovation relaxes price competition with the incumbent. This …finding has managerial implications for fi…rms that use new product development and pricing to compete for consumers who are heterogeneous in their appreciation of new technology.
For paper click Targeted Innovation

Paul Messinger (Alberta) A New Model of Asymmetric Competitive Structure Using Store-level Scanner Data (Paper: Fang Wu, Paul R. Messinger and Terry Elrod)
This article develops a model of competitive structure that can be applied to widely available store-level brand sales and price data. We estimate maps of the competitive brand relationships that are assumed to jointly underlie cross-price elasticities, own-price elasticities, and brandspecific intercepts. Our methodology uses an adaptive Bayesian approach that shares information across different brands and different terms in a set of demand equations. Drawing upon recent psychometric research, we express the asymmetries present in cross-price elasticities as the difference between what we refer to as brand power parameters, and we identify relationships between a focal brand’s power parameter, clout, vulnerability, own-price elasticity, and spatial density. We apply the model separately for two datasets that consist of weekly sales and prices for beer and soft drinks. Our estimated maps of competitive brand relationships have implications for both brand management and antitrust policy.
For paper click Asymmetric Competitive Structure

Yupin Yang (Simon Fraser) Who Sponsors Whom and Why? An Empirical Investigation of Sports Sponsorships (Paper: Yupin Yang and Avi Goldfarb)
This paper applies a two-sided matching model to investigate the formations of sports sponsorships using a dataset containing the shirt sponsorships from 43 English football clubs during the period from 1990 to 2010. We find that sponsorships become less valuable as the distance between the club and the sponsor’s head office grows and that better-performing clubs can attract more distant sponsors. In addition, there is an assortative matching between a club’s attendance and a sponsor’s revenue. Based on the estimates from the two-sided matching model, we simulate the counterfactual matching outcomes if sponsorships on alcohol and gambling are banned. Our estimates suggest that such bans will not have the biggest impact on the (relatively successful) clubs that currently have alcohol and gambling sponsors. Instead, it is clubs with low attendance and clubs in low income, less populated areas will be most affected.
For paper click Who Sponsors Whom.

Tanya Mark (Guelph) A Dynamic Segmentation Framework: Assessing Omnichannel Behavior of Customers (Paper: Tanya Mark, Jan Bulla, Rakesh Niraj and Ingo Bulla)
Retailers are experiencing a dramatic shift in the buying habits of their customers as more customers buy across different channels over the duration of their relationship with a firm. From a firm’s perspective, an omnichannel approach to managing customers is of utmost importance because their priority is to capture as many of a customer’s purchase occasions as possible. While it certainly is better for retailers to get customers to buy from one of their own channels compared with their buying from a competitor, it will be even better to direct customers to the channel that is the most effective and efficient to serve their needs. Consequently, we argue that as more information on cross-channel purchasing behavior becomes available, firms need models that utilize these data in order to make marketing decisions that nudge the consumers to the “best” channels to serve them over time. We develop a dynamic model of channel choice and purchase frequency while simultaneously accounting for customer heterogeneity, the various stages of a customer-firm relationship, and responses to a marketing activity. We extend the classic hurdle model by introducing a hidden Markov process to accommodate unobserved customer heterogeneity in patterns of both purchase frequency and channel choice. Our findings suggest direct mail increases the likelihood of a purchase in the telephone channel buy only for the first purchase incidence and for customers with infrequent buying patterns. When we account for the frequency of purchases across channels, we find differential effects of this marketing activity on purchasing behavior. For active customers, this instrument is effective for increasing the likelihood of additional purchases across various channels. Our research contributes to the growing literature on how firms can influence purchasing behavior through different channels beyond a customer’s first purchase.
For paper click Omnichannel Behavior.

Kersi Antia (Western) All for One, One for All: Governance and Bankruptcy in Franchisor-Franchisee Relationships (Paper: Kersi Antia, Sudha Mani and Kenneth Wathne)
Despite much being made of franchisor-franchisee interdependence and the importance of relationship governance, little is known about the actual extent to which both parties’ fortunes are intertwined and about the costs – in particular, the catastrophic cost of bankruptcy – accompanying the governance mechanisms relied on by franchisors to ensure performance by their franchise partners. The present study addresses both these limitations. Integrating data from multiple archival sources, we undertake a census of all bankruptcy filings by 1,130 franchisors and their franchisees over a thirteen-year observation window; we then relate the incidence of franchisor and franchisee bankruptcy to the governance mechanisms commonly relied on by franchisors. We not only document the actual interrelationship – both contemporaneous as well as lagged – between the occurrence of franchisor and franchisee bankruptcies, but also provide evidence of network externalities among franchisees. Our findings point to important conflicting governance effects, in that monitoring and incentives, while reducing franchisee bankruptcy incidence, are actually associated with a higher likelihood of franchisor bankruptcy. We also find prior estimates of franchise system failure rates to be significantly overstated.

William Allender (McMaster) Consumer Search and the Choice Overload Hypothesis (Paper: William Allender, Tim Richards and Sungho Park)
We use an experimental approach to study the relationship between consumer’s preference for variety and subsequent search and purchase behavior. By imposing a search cost, and a preference for variety, we induce search behavior that leads the participants to make both single, and multiple-purchase decisions. This paper shows that the ability to shop more efficiently allows consumers to focus on …finding products that meet their exact speci…cations in differentiated product categories. With uncertainty over product attributes, no single choice stands out as a clear favorite a priori, so utility rises in the number of choices that are available, but at a decreasing rate. Our results suggest that retailers can broaden their assortments to persuade consumers to patronize their store, but not without bound. Because search is costly consumers will eventually become overwhelmed by the number of products o¤ered and search a competing store as well, or possibly instead. The degree to which consumers are overburdened by the variety offered exhibits signi…cant heterogeneity among consumers.
For paper click Consumer Search.