Entry

Paper Session

Sunday, Jan. 8, 2017 6:00 PM – 8:00 PM

Hyatt Regency Chicago, New Orleans
Hosted By: American Economic Association
  • Chair: Joseph Mazur, Purdue University

Estimation of Discrete Games With Weak Assumptions on Information

Lorenzo Magnolfi
,
Yale University
Camilla Roncoroni
,
Yale University

Abstract

We propose a method to estimate static discrete games with weak assumptions on the information available to players. In contrast to the existing literature, we do not fully specify the information structure of the game. Instead, we allow for all information structures consistent with the assumptions that players know their own payoffs and the distribution of opponents’ payoffs. We make this approach tractable by adopting a weaker solution concept: Bayes Correlated Equilibrium (BCE), proposed by Bergemann and Morris (2013, 2015). We characterize the sharp identified set obtained under the assumption of BCE behavior. In simple games with modest levels of variation in observable covariates, identified sets are narrow enough to be informative, while avoiding the misspecification resulting from strong assumptions on information. In an application, we estimate a game theoretic model of entry in the Italian supermarket industry, and quantify the effect of the presence of large malls on competition. Our model yields parameter estimates and counterfactual predictions that differ from those obtained under the restrictive assumption of complete information.

Foreclosure, Entry, and Competition in Platform Markets With Cloud Storage

Mark Tremblay
,
McMaster University

Abstract

Platform providers in two-sided markets compete over time by introducing new generations of their platforms. New generations are often backward compatible through some form of cloud storage so that consumers' content, preferences, apps, software, and games from the previous generation can still be used on the current generation. When platform providers compete over time, consumers that switch platforms are unable to use their previous content. With two-sidedness, the incumbent platform provider is able to use its price to the other side of the market, the price to content providers, to endogenously determine the strength of the consumer carryover utility that its consumers face when considering switching platforms. Thus, two-sidedness with cloud storage results in endogenously determined switching costs. I find that cloud storage is not used to foreclose platform entry but instead is used to soften competition between platform providers. Furthermore, the resulting equilibria comport with anecdotal evidence of attempted entry across several platform industries, such as smartphones, video game consoles, personal computers, eReaders, and online streaming subscriptions.

Market Entry With Frictional Matching and Bargaining: An Experimental Investigation

Knut-Eric Joslin
,
BI Norwegian Business School

Abstract

We investigate market entry decisions when firms have a productive opportunity but must recruit a worker through a frictional process and then negotiate a wage. Specifically, we experimentally test a modified version of the ``market entry game'' in which labor market institutions are introduced in a simple fashion. We modify the basic market entry game in the following fashion: Prior to the start of the game, players are assigned either the role of a ``firm'' or of a ``worker.'' In the first stage of the game, firm players decide to either participate in a ``matching market’’ or to take a certain payment. Next, players who participate in the matching market are matched according to a constant returns to scale matching technology; this means that when more firms enter, competition for workers increases. Finally, in the last stage of the game, players ultimatum bargain over a surplus. Despite the complications introduced in our game, results from the entry portion of our experiment are largely consistent with the stylized facts from the existing literature: Entry tends to equalizes expected profits from market entry with the opportunity costs and aggregate entry probabilities conform with the predictions of the symmetric mixed strategy equilibrium. Moreover, we identify the same biases found in other experiments. We explain the bias in the pattern of entry using a novel equilibrium with noise. In the bargaining portion of the experiment, we find that the size of the surplus affects the bargaining outcome: Larger surpluses generate lower offer shares. This pattern can not be accounted for by standard behavioral preferences. Notably, however, the modal offer is optimal in terms of expected payoffs: After weighting offers by the probability that they are accepted, the most common offer maximizes payoffs.

Political Connections and Market Structure

Camilla Roncoroni
,
Yale University
Lorenzo Magnolfi
,
Yale University

Abstract

This paper empirically investigates how political connections affect supermarket entry in the Italian grocery retail industry, then quantifies the welfare cost of political influence. We focus on the largest grocery retailer in Italy, a network of consumer cooperatives that has historical links to political parties. We estimate a game-theoretic model that accounts for both the interdependence among firms’ entry decisions and the effect of market-level variables, among which we include a measure of political connections. The informational environment is affected by the presence of political connections, and the connected player might be better informed than its competitors. To take this into account, we adopt a new method to estimate the entry game under weak assumptions on the informational environment. We find a positive effect of political connections on cooperatives’ profits, and a negative effect on some competitors. In a counterfactual, we examine the effect of removing political connections and find that consumers' can benefit substantially in markets where political connections act mainly as barriers to entry, but can also stand to lose where connections are necessary to overcome the burden of entry regulation.

Discretely Innovating: Barriers to Entry, Contestablility and Innovation

Steven Bond-Smith
,
Curtin University

Abstract

This paper considers the effect of a discrete entry barrier (i.e. only an integer number of firms is permitted) on innovation in an endogenous growth model to draw conclusions about the relationship between contestability and innovation under Cournot and Betrand oligopoly. Sector-specific workers become a measure of contestability and provide a tool for calibration. The paper finds that sectors with low contestability have lower innovation. In particular innovation under Cournot oligopoly is always less than under Bertrand with particularly pronounced effects when contestability is low. With increased contestability, innovation increases towards the limit in the continuous entry model as space for less than one firm becomes a smaller proportion of the sector. Wage inequality varies depending on the extent that the barrier to entry is binding upon a marginal entrant.
JEL Classifications
  • L1 - Market Structure, Firm Strategy, and Market Performance