Pre-Release Movie Piracy and Box Office Sales: Estimates and Policy Implications

In this paper we analyze a new dataset collected from an Internet file-sharing site to examine the impact of pre-release theatrical piracy on box office revenues. Using a variety of empirical models and propensity score matching controls for endogeneity, we find that the presence of pre-release piracy reduces a movie's box office revenue by approximately 15%, that the vast majority of this reduction occurs in the movie's opening weekend, and that pirated copies of higher quality have a less severe impact on box office sales than lower quality pirated copies do. Theory and Model Development To test the impact of pre-release piracy across a movie's lifecycle we need to build a model to explain both the box office sales of movies and the distribution of sales to different time periods. Most movies see the highest level of sales in the opening weekend, with sales declining in subsequent weeks. Consistent with existing literature in marketing and information systems (e.g. Sawhney and Eliashberg 1996, Krider and Weinberg 1998), we model the box office revenue of movies using an exponentially declining function: it i i it i t m t i it e e m