Does piracy of creative goods such as movies, books, or songs reduce paid demand for those goods? This seemingly straightforward question has proven surprisingly difficult to answer in the real world.
Piracy may draw away customers who might otherwise have paid. But it’s also possible that consumers of pirated copies are, by and large, not people who would have paid to consume if they couldn’t get access for free. Piracy may also help spread the word about a good movie, book, or song. This sort of informal advertising might drive up paid consumption, even if some people who would have paid are lost to piracy. It’s also possible that some combination of all these things might happen, with uncertain net results.
In a new empirical paper, titled Avengers Assemble! When Digital Piracy Increases Box Office Demand, Klaus Ackermann, Wendy A. Bradley, and Jack Francis Cameron offer a nuanced and interesting study of the effects of piracy on the movie industry. The effects of piracy, as it turns out, have a lot to do with what kind of movies we’re talking about.
The authors built a novel dataset that identified the existence and the timing of the earliest upload of a high-quality pirated copy for every U.S. movie release. The authors did this with data on the appearance of movie piracy “torrents” between January 2004 and January 2020 from online piracy site The Pirate Bay (TPB). The authors then matched this data with movie release information during the same period from the well-known IMDb database.
Merging these two streams of data allowed the authors to match up release dates and “piracy dates.” They measured changes in box-office revenue for pirated movies in the first 48 days after their releases in theaters, relative to the preceding period following releases before the movies were pirated. They then used a formula to adjust for the general fall-off in movie box-office revenues over time. If piracy was substituting for paid demand, the authors could pick up that effect by comparing (time-adjusted) pre-piracy vs. post-piracy box office revenues across many films.
The authors hypothesized that piracy has different effects on different types of movies. Specifically, they theorized that “spectacle” movies—the kind of movies that people want to see in the movie theater—may be less affected by piracy than “story” movies that people are more content to watch on their computers. In other words, spectacle movies may benefit more from word-of-mouth advertising that piracy may provide while losing fewer customers to demand substitution, compared to story movies.
To aid this assessment, the authors constructed two measures of movie “spectacleness.” One used a movie’s release in 3D or IMAX formats as a proxy for that quality (because “spectacle” movies are the kinds of movies that people want to see in these especially immersive formats). A second categorized movies into genres associated with spectacleness and story-focus by measuring the number of movies in various genres nominated for the “best visual effects” Oscar (associated with spectacleness) as opposed to the “best original screenplay” Oscar (associated with story-focus).
Based on data for more than 400 movies, about half of which have been pirated within the first 48 days of release, the authors concluded that piracy had the mixed effects they predicted. For films for which in-theater viewing adds value (“spectacle” films), there is a 13% increase in average daily box office revenue after the appearance of a high-quality pirated version of the film online. For story-focused films, on the other hand, there is as much as a 30% decline in average daily box-office returns after the appearance of a high-quality pirated version of the film online. This is consistent with the idea that piracy acts as a substitute to films focused on story, where the full value of the film can be consumed at home.
The authors’ findings shouldn’t be too surprising. Think for a moment about the music industry. Recorded music is more vulnerable to piracy than live music because a big part of the appeal of live music is the immediacy and communal experience of the concert. Such experiences cannot be replicated in a pirated recording.
So we might expect that during the post-Napster but pre-Spotify/Apple Music era when online piracy was driving down revenues for recorded music, there would be an industry shift toward more focus on live music. There was indeed a very rapid growth in that period of big live music firms such as Ticketmaster and Live Nation. Moreover, during that period the rise of live music revenue very closely mirrored the decline of recorded music.
In 2000 (just after Napster’s debut), recorded music represented 53% of the global music industry. By 2017 (when paid streaming started to restore lost recorded music revenues), recorded music’s share of total music industry revenues had dropped to 38%, while live music went from 33% to 43% of the industry.
Something analogous is happening in the motion picture industry, although the effect is probably not as pronounced. That is, the industry’s product mix may have shifted toward “spectacle” films because these sorts of film tend to be more resistant to piracy. Indeed, the authors gesture in this direction, stating that their findings suggest because the value of a film is linked to its “spectacleness,” the industry would be wise to adjust its creative output on the margin—i.e., to produce more spectacle films—to blunt piracy’s effect rather than investing in the law enforcement efforts that would be required to reduce piracy by any substantial amount.
Alternatively, movie studios may seek to insulate story films against piracy by, for example, releasing them to streaming channels simultaneously with theatrical release. Or, maybe movie studios could invest directly in upgrading theaters for these story-focused films to enhance the in-theater viewing experience in other ways, such as by making the theater a place for fun and social interaction. (Theaters such as the Alamo Drafthouse are already offering this kind of experience).
If so, then the principal effect of movie piracy may not be to lower the overall demand for movies or the number of movies produced. It may be to shift the kind of movies produced, or, more subtly, to shift the way that movies are presented to the public. Unlike the relatively simple framework in which piracy leads to fewer movies, the real effect of piracy may be more subtle, and the case for investing significant resources (especially public resources) in anti-piracy efforts less clear.
