Release windows and piracy

It takes a long time for movies to be released in New Zealand. How long? Mike Dickison's been trying to figure it out, while plotting his escape from Christchurch*. So far, it looks like about a 50 day wait on average. What's that like? I didn't know there was a remake of Girl With a Dragon Tattoo so figured the new posters up meant that the film we'd seen a couple years ago was finally having a broader theatrical release. Perhaps that's just me being dense, but it didn't seem crazy.

So while we're hearing all our friends tweet about the latest releases, the only way of staying in the conversation is piracy. Two-month delays in getting films here might have made sense when they had to be carried by steamship; it's perhaps a bit ridiculous when theatres with digital projection equipment can have the movie emailed to them.

I'd wondered whether these release windows explained New Zealand's robust file-sharing culture. Stealing content just doesn't seem that wrong where content providers refuse to provide a legal way of accessing it. So I started casting about for ways of getting data on NZ filesharing to test whether length of release delay explains filesharing volume.

Looks like I won't have to bother. Danaher and Waldfogel have an excellent paper up at SSRN. Here's TorrentFreak's summary. Long story short, the introduction of BitTorrent depressed box-office revenues outside of the US in a way that correlates with the length of the window between US and foreign release. Within the US, the introduction of BitTorrent hasn't sharpened the gradient for revenue drop off with weeks since first release, which we would have expected if greater online availability increased with time (which it does) and depressed box office revenues.

If studios delay the release of films that they expect to be less popular in a particular foreign market and rush in films they expect will have more viewers, then we'd see lower revenues for those longer-window films even in the absence of piracy. But, as they use a difference-in-difference approach with films released before and after BitTorrent, we'd need that studios suddenly became better at this post 2003 for the results to be contaminated. They also run a robustness check using a triple difference-in-difference approach exploiting that sci-fi and action movies are more heavily downloaded than, for example, romance films. Longer international release windows depress box office earnings, with the effect likely due to BitTorrent.

The paper also provides an very nice explanation for studios' use international windowing:

• Cost of physical prints is high, so it's cheaper to send films abroad after a first run in the US. This also goes some way toward explaining our sometimes grainy film quality at theatre: every theatre is effectively a second-run theatre.
• Most foreign markets are theatre-poor relative to the US, so they have to be more selective in which films to screen. Waiting to see what's successful in the States before committing to a run here can make sense. 
• Promotions centred around film star appearances require international windowing.

Those three things have to dominate the piracy cost of their windowing strategy, unless we want to assume the studios are idiots. The authors estimate that piracy reduced weekend box office returns by $240 million in 2005. I'd be a bit surprised if the gains coming from the three bullet points above outweigh that, but I'm not the one with money on the line in this game. Or it could be that the studios are pushing hard on legal solutions as a way of keeping the benefits of international windowing while avoiding piracy losses. I'm pretty sceptical that that's going to work.

Update: The OatMeal is on point.

* While we're thinking about things that aren't working, here's Mike's piece in the Herald. Depressing but realistic.

After the first quake shock had worn off, there was a unexpected elation in the air. People were itching to reclaim the rubble and turn destruction into a fresh start. The Gap Filler project screened outdoor movies in an empty lot, and made a book exchange out of an old fridge; Greening The Rubble built parks where there used to be buildings. The urge to help-to do something -filled community meetings and swamped the City Council with suggestions for the rebuild, giving rise to that utopian document the Central City Plan, which painted a picture of tree-lined cycleways, green markets, and inner-city apartments.

Not only would the quake damage be fixed, so would decades of urban sprawl and central city neglect. Ponies for everybody. Ponies with free wireless.
...
Apart from a fortnight when I was barred from my apartment by a police cordon, I've been living in the central city since the February quake, watching earthquake tourists circle the Red Zone on sunny weekends, and seeing buildings gradually disappear week by week. I've watched the crack in my wall get slowly wider, and energy and optimism leak away, replaced by frustration, cynicism, and a dawning realisation that bringing a heart and life back to the city will take a decade or longer. And that the only people who can speed that up are politicians and insurance companies, not the citizens.

I'm a bit more optimistic than Mike. New life is popping up all over the suburbs. EPIC will break ground downtown and provide an anchor for new development. So long as we can keep Council from stuffing it up...