The author proposes a strategic model of entry that allows for positive and negative spillovers among firms. The model is applied to a novel data set containing information about the store configurations of all U.S. regional shopping centers and is used to quantify the magnitude of interstore spillovers. The author addresses the estimation difficulties that arise due to the presence of multiple equilibria by formulating the entry game as a mathematical problem with equilibrium constraints (MPEC). Although this study constitutes the first attempt to use this direct optimization approach to address a specific empirical problem, the method can be used in a wide range of structural estimation problems. The empirical results support the agglomeration and clustering theories that predict that firms may have incentives to colocate despite potential business stealing effects. The author shows that the firms' negative and positive strategic effects help predict both how many firms can operate profitably in a given market and the firm-type configurations. The relative magnitude of such effects varies substantially across store types.
- Direct optimization approach
- Incomplete information game
- Mathematical programming with equilibrium constraints (MPEC)
- Shopping centers