« Neuroscience and Law | Main | Privacy's Other Path »
March 12, 2007
What's the Analog Hole worth? Twenty-Four Cents
I've overstayed my welcome, so I'll be signing off with this post. Thanks to Dan and the other permabloggers for letting me participate.
Point a video camera at a television screen, aim a microphone at a speaker, or run a cable from the "line out" to the "line in" ports on the back of your computer, and you're ready to exploit the so-called analog hole. Just press "play" on one device and "record" on the other, and you can copy a movie, television show, or song, even if the original is supposedly protected by digital rights management technology designed to prevent copying.
The analog hole--which arises from the fact that relatively-easy-to-protect digital content must be converted into harder-to-protect analog signals if we humans are to see or hear them--has given Hollywood and the recording industry a fair amount of heartache, has led them to displays of public consternation, and has even resulted in some proposed legislation.
Despite its frequent appearance in DRM debates, the analog hole is suprisingly unexplored in legal scholarship. Westlaw's JLR database contains a mere thirty-seven articles that use the phrase, most in passing, and SSRN returns only three hits. Most of the commentary relies on an empirical assumption that has never before been rigorously tested: Exploiting the analog hole creates copies of such low quality as not to be good substitutes for the originals.
Doug Sicker, an Assistant Professor of Computer Science at my University, together with Shannon Gunaji, a grad student, have tried empirically to test this assumption by conducting a series of surveys assessing, among other things, what the analog hole means for the typical music consumer. Doug asked me to help bring the early results to the legal academy, and our little article, entitled The Analog Hole and the Price of Music: an Empirical Study, has been posted to SSRN and will appear soon in the Journal of Telecommunications & High Technology Law.
Our results after the jump.
We came to three primary conclusions, although none of our survey sizes were large enough to support statistically significant conclusions (we hope to replicate the study on a larger sample):
First, some of our survey respondents could detect the differences between a digital original and an analog hole copy of the same song. The results, however, weren't dramatic, and many respondents couldn't tell the two apart. Of course, there is a danger to generalizing too much from this conclusion. There are many ways to create an analog hole copy, and the amount of signal degradation varies widely. Our results apply only to two specific methods.
Second, using an econometric model, we came to a tantalizingly specific conclusion: the analog hole is worth twenty-four cents. We mean by this that our respondents were willing to buy signal-degraded analog hole copies of music, but only if priced twenty-four cents less than a digital original. This builds nicely on Chris Sprigman's article which attempts to explain why nearly every digital track sold online costs ninety-nine cents.
Third, we ran a second survey of people (all college students) whose music collections contain more non-purchased than purchased music. These, putative "pirates," surprised us by expressing a willingness to pay for music, but only if it was sold for twenty to forty cents per song. Although this is well below the market rate, it suggests that a shift in pricing might bring some pirates back into the fold.
Please read the paper if you're interested in the conclusions we draw from these results or to see how this fits into the other writing on the subject.
Posted by Paul Ohm at March 12, 2007 01:42 AM
Trackback Pings
TrackBack URL for this entry:
http://www.concurringopinions.com/movabletype/mt-tb.cgi/1816.
Comments
One comment I'd make is that your third finding is that this would seem to exactly mirror the business model of allofmp3.com.
I'd also note that part of the issue with the analog hole is that it requires a good deal more time to implement, and more skill, since the analog files need to be played back in real time, rather than a substantially faster digital read. Track breaks also must be manually set, which can be a serious pain.
Posted by: Ubertrout at March 12, 2007 11:06 AM
Paul -- Thanks for an excellent visit!
Posted by: Daniel J. Solove at March 12, 2007 09:57 PM
One thing that you mention in passing but don't really focus on is that the analog hole only has to be passed through one time, after which the DRM-free copy of the product will be digitally transferable. In a world with perfect DRM and peer-to-peer networks, this suggests that consumers will be choosing between paying for a perfect copy or downloading a copy produced with extremely high-quality analog components (such copies would presumably be more widely propagated than inferior analog copies).
I would be curious to see the results of your survey if you included an "Amateur Musician Copy" that was produced using ~$10,000 worth of home audio equipment (or a rented music studio with similar levels of equipment).
Posted by: Jake at March 13, 2007 10:47 AM
