MITSUKUNI NISHIDA, University of Chicago
Email: nishida@uchicago.edu

Competition among multi-store chains is common in the retail industry. This paper proposes a method for estimating a model of strategic store network choices by two chains. In contrast to previous studies, I allow chains to not only choose which markets to enter but also how many stores to open in each of those markets. To deal with the huge number of possible network choices, I use lattice-theoretical results. I show that a chain's net trade-off between costs and benefits from clustering their stores in a market can be either positive or negative while still ensuring the existence of an equilibrium, thereby providing a way to freely estimate this within-market effect from the data. By integrating the model with revenue data, I show that one is able to decompose the within-market effect into cost savings from economies of density and lost revenues from competition with one's own stores. I apply the technique to a unique cross-sectional data set from the convenience store industry in Okinawa, Japan, to obtain cost and demand parameters. The estimates imply a strong business stealing effect among own stores. I then use the estimated structural model to perform two counterfactual analyses. First, I evaluate the impacts of a hypothetical merger of two chains. I find that post-merger total surplus is expected to decrease due to reduction in the number of stores and total sales. Second, I consider how significantly the zoning regulation introduced in Japan in 1968 affects store network choices.

ROBERT SEAMANS, University of California, Berkeley - Haas School of Business
Email: seamans@haas.berkeley.edu

This paper links empirical literature on the use of price as an entry deterring mechanism with literature on the effect of multi-market contact on competition. The analysis uses a dataset of cable TV system prices to provide evidence that incumbent cable TV firms use price to deter entry by telecom overbuilders as well as cities with municipal utilities. There is also some evidence that multi-market contact with telecom overbuilders results in lower prices. However, there is no evidence that incumbents use price to deter cable overbuilders. In addition to linking entry deterrence with multi-market contact, this study has two other unique features. First, it establishes entry deterrence using two techniques, one of which relies on theory by Ellison and Ellison (2008) on non-monotonic price decreases in response to entry probability. Second, it uses detailed price and channel data at the service tier level.

JIN-HYUK KIM, Cornell University
Email: jk384@cornell.edu

This paper analyzes DRM-based technological tying, where the content and hardware form a system. A closed DRM system makes the legal content incompatible with a rival's hardware, whose users must then obtain illegal copies. The main finding is that the tying firm gains market power in a competitive hardware market and invests in product upgrades at a later stage. Welfare implications of the policy that requires an open DRM system are also discussed.

CURTIS TAYLOR, Duke University
Email: curtis.taylor@duke.edu
LIAD WAGMAN, Duke University
Email: liad.wagman@duke.edu

When firms can identify their past customers, they may use information about purchase histories in order to price discriminate. We present a model with a monopolist and a continuum of heterogeneous consumers, where consumers can opt out from being identified, possibly at a cost. We find that when consumers can costlessly opt out, they all individually choose privacy, which results in the highest profit for the monopolist. In fact, all consumers are better off when opting out is costly. When valuations are uniformly distributed, social surplus is non-monotonic in the cost of opting out and is highest when opting out is prohibitively costly. We introduce the notion of a privacy gatekeeper - a third party that is able to act as a privacy conduit and set the cost of opting out. We prove that the privacy gatekeeper only charges the firm in equilibrium, making privacy costless to consumers.

IVAN MARYANCHYK, University of Arizona - Department of Economics
Email: ivanm@email.arizona.edu

The aim of this research is to analyze whether and when ratings are informative signals about the quality of movies. The ratings data of Netflix is used to fit a structural Bayesian learning model. This model links revealed experience utilities of raters, previous consumers, to the product choice of the future consumers of the same good. I postulate that movies are chosen based on the prior beliefs' and signals' precisions. The extent of signals' use depends on their informativeness, that is on how many consumers revealed their preferences before. The results demonstrate that consumers learn about the quality using ratings as signals. The signal produced by one rating is very noisy and might not be taken into account. The more people rate, the better are signals' quality. Consumers are not considerably dispersed in how they value quality.

HSING K. CHENG, University of Florida - Warrington College of Business Administration
Email: hkcheng@ufl.edu
HONG GUO, University of Florida - Warrington College of Business Administration
Email: guohong@ufl.edu

This paper proposes a holistic view of a network organization's computing environment to examine computer virus propagation patterns. We empirically examine a large-scale organizational network consisting of both social network and technological network. By applying information retrieval techniques, we map nodes in the social network to nodes in the technological network to construct the composite network of the organization. We apply social network analysis to study the topologies of social and technological networks in this organization. We statistically test the impact of the interplay between social and technological network on computer virus propagation using a susceptible-infective-recovered epidemic process. We find that computer viruses propagate faster but reach lower level of infection through technological network than through social network, and viruses propagate the fastest and reach the highest level of infection through the composite network. Overlooking the interplay of social network and technological network underestimates the virus propagation speed and the scale of infection.