New Downloading Study

I love RSS. The University of Chicago Press journals now do RSS, or at least the ones that I read (JLE, JLS and JPE). Now amidst the Steve Jobs Thoughts on Music and reports that EMI may move to unDRMed MP3s, this from the most recent issue of the Journal of Political Economy. Oberholzer-Gee & Strumpf, The Effect of File Sharing on Record Sales: An Empirical Analysis:

For industries ranging from software to pharmaceuticals and entertainment, there is an intense debate about the appropriate level of protection for intellectual property. The Internet provides a natural crucible to assess the implications of reduced protection because it drastically lowers the cost of copying information. In this paper, we analyze whether file sharing has reduced the legal sales of music. While this question is receiving considerable attention in academia, industry, and Congress, we are the first to study the phenomenon employing data on actual downloads of music files. We match an extensive sample of downloads to U.S. sales data for a large number of albums. To establish causality, we instrument for downloads using data on international school holidays. Downloads have an effect on sales that is statistically indistinguishable from zero. Our estimates are inconsistent with claims that file sharing is the primary reason for the decline in music sales during our study period.

Comments

[Erasmussimo]

Gee, there's nothing like a little data to alter the terms of a debate. I find the result counterintuitive, but I suppose that this is another case of an elegant intuition being brought low by an ugly little fact. Still, I wonder if this fact can be extrapolated to the future. Can we be certain that, in the long run, the availability of a product for free will not reduce its overall sales?

[Bruce Boyden]

This sounds like the same paper released, as a working draft, in 2004; Stan Liebowitz had some comments on that earlier draft at http://www.utdallas.edu/~liebowit/ .

[Doug Lay]

I don't take Prof. Liebowitz very seriously. His Institute has the look of one of those think tanks whose studies are designed to produce results favorable to their friends and sponsors in the business world.

Nonetheless I agree with Erasmussino that the results from the Oberholzer-Strumpf study look pretty counter-intuitive, and I also wonder how much validity a study conducted several years ago has today.

Truly, however, I wonder if this study is even asking the right question, if the goal is to determine the efficacy of laws like the DMCA.

Prof. Picker postulates in a comment on the thread below that a long-standing property right (copyright) is being threatened by a new technology (peer-to-peer file sharing). Then he proposes (I am paraphrasing; I hope I have this right) that the law should evolve to help copyright holders maintain the ability to enforce their long-held property right, in the face of this technological change.

I don't agree with all of the above. (Is intellectual property really property? Is legally-protected DRM really about enforcing existing rights, or does it create new and fundamentally different rights?) However, even if one agrees wholly with Prof. Picker's postulates, it seems that new laws designed to protect old rights in the face of new technology should be evaluated based on whether they successfully address the threat posed by the new technology.

So, the question to ask when evaluating the DMCA anti-circumvention provision is not whether file sharing cuts into legitimate record sales, but rather whether the DMCA anti-circumvention provision (and the DRM protected by the DMCA) has any impact on the amount of file sharing going on.

The study Prof. Picker cites above certainly should be of enormous interest to record companies trying to decide on digital music distribution strategies. But I don't really see why it should be of much importance from a legal point of view. If p2p file sharing makes a property right unenforceable, then, if I follow Picker's logic right, the law should change to make enforcement possible again, even if studies were to show that file sharing actually HELPS CD sales.

The real question should be "Does the new law actually work to make enforcement of the property right possible again?" Does the DMCA anti-circumvention provision have any impact on p2p file sharing? Here there is plenty of anecdotal evidence that says no, along with a powerful analyic argument presented in the Darknet paper. It would be interesting to see if any social science professors (preferably not Mr. Liebowitz!) could come up with an empirical study to test the claims in the Darknet paper.

The questions I would have for Prof. Picker, then, would be (a) what level of evidence would you require before conceding that the DMCA anti-circumvention has failed in the goal of protecting copyright against the challenge of p2p technology, and (b) what should the law do next?

[Erasmussimo]

The problem faced here is primarily a technological one: the owners of the IP are unable to secure it. At present, there is no practical means of securing IP in MP3 format. No law can correct that problem. The copyright owners of music are unable to enforce their copyright.

We acknowledge that this is a Bad Thing, which we would like to fix. We can certainly imagine a number of technological fixes. Suppose that we built a technology that was absolutely secure. Then there would be no need for additional law; existing law would address any problems.

Where it gets tricky is the scenario in which the IP is mostly secure, but can still be cracked by some complicated scheme. Would it then be appropriate to enact special law to address that complicated scheme? I don't think that's a good idea, as technology moves so quickly that any such law would be obsolete in a few years.

It seems to me that we need to take a larger view and develop an entire body of law predicated on the principle that ALL information has an owner, and that the owner can specify the manner in which the information is used. What's interesting here is the possibility of revising the Internet so that every packet is marked with its ownership information, and the servers themselves enforce the law. In other words, you can't send information over the Internet unless you have an established right to it. This would also require that all information storage and processing devices be designed to enforce the law as well. In other words, you couldn't burn a CD unless you have the right to do so. This is certainly technically feasible, although it would entail an enormous effort and would require licensing of all wafer fab facilities. Still, wafer fabs are big and expensive enough that we can nail them down legally.

Of course, someday we'll have home wafer fab printers, and we'll have to migrate to something new...

[Michael Risch]

As I read this study, it seems like the claim is that because record sales do not go down when the supply of German based P2P downloads increases (with various time lags), then there is no effect on record sales.

There seem to be two primary problems with the study design:

1. The sample seems exceedingly small and non-representative(one-sixth of .01 of one percent of all downloads)

2. The study's entire design falls apart if we weaken the assumption that the supply of downloads in period x will have a definite effect on sales in period y.

What if, for example, just knowing that downloads will be available means you won't buy, even if you don't download? I can get song a this week and song b next week from the same album - do we attribute the lack of record sales to that week only? You could correct for this problem, but I don't see it in this study, and I don't know that the sample is robust enough to do it effectively in any event.

[LAK]

Eras,

Why so shocked? I still buy albums because I want the whole thing right away, including album art. This is on iTunes and CDs in stores. I download certain songs off of file sharing networks as well, but only ones I would not otherwise buy, where I would otherwise be dead weight loss due to anticompetitive conduct and market inefficiencies.

I have no moral qualms with this for a number of reasons. The major music labels all conspired for years to fix the price of music when their distribution channels were the only ones readily available to consumers. They settled a class action for fixing the price of CDs a few years ago and basically got away with murder.

Ideally, as the economic theory goes, in a competitive envirnoment a commodity's price is supposed to approach its marginal cost, which in this case is nearly zero. If I value a shitty pop song at 3 cents, but can't get it out of my head and want to listen to it once or twice, why should I pay $1 for it (let alone $15 for it a 10 other songs I don't want, which is how it was for years - tying anyone?)? I shouldn't, and wouldn't otherwise. The fact is I could record it legally at no cost myself, and ironically in much better quality off of FM compared to a crappy 128kbps compressed mp3 file. As someone who cherishes his records and analog sound, I find it funny that just because some crappy compressed file is digital, it is somehow more worthy of protection than a uncompressed analog signal.

And surely you see the inevitablily of all this? The major music labels served their purpose in a pre-digital era - they facilitated the recording and distribution of music when costs of doing both was high, and by virtue of this a few huge companies pretty much dominated the whole music business from the picking of the artists who they allowed to reach the mass market, to the promotion, to the tours and sales of the music down to the wrapper. And they profited and profited. And exploited artists along the way (especially in the early days and easpecially if you were black), and extracted monoply profits out of consumers, and were responsible for some of the worst forms of mass market manipulation. They brainwashed teenagers with sex and marketing, and still do in fact(though Brittney just isn't as compelling as Elvis - with modern studio effects and production, how you look is more important to these companies than the actual amount of musical talent the pop star has).

The digital age is fantastic. It makes recording and distributing music extremely low cost. It renders major music labels somehat obsolite, and they hate it. What does this mean?

When Brittney can't make all that much money from CDs, she's going to have to go perform to make her money. At least she'll get most of it and not the music company. Uh oh, she better get some talent quick. Imagine that, people making music because they love to, and not for just for profit as a creation of a pop cultured machine. People having to perform their music to make a buck! What a wonderful world it will be when the pop garbage these companies shove down childrens throats is rendered unprofitable by virtue of progress. A Hegelian moment.

Also, even if there are fewer sales due to the ease of pirating, musicians now have the opportunity to bypass the major labels and should get together and start a collective itunes where the artists profits from the sale with no record company middle man.

Besides, the msic labels all still sell tons of music and make tons of money. They released stats about how album sales are down, yet they rarely talk about how much single song sales have increased when before they didn't even exist.

And F them should I download a song I purchased last year on CD but which subsequently became scratched and skips.

Music is love. Greed sucks. Down with Britney Spears. Up with talent and live performance.

[Pablo]

A study should also be done as a way to prove something. In many instances, studies are done when the person sponsoring the study already have an idea where the whole thing will go, like the old movie cliche, who stands to profit?