And Now, for Something Special

Peeps! I am very excited! I got a guest contribution! Granted, this was the result of months of cajoling, pleading, and holding forth on whatever wiles I’ve got, but I think the piece is more than worth it! A., as I’ve said before, is a smart man.

So with a great deal of fanfare and drum rolls, here’s A.’s piece


File sharing is one of the most common online activities. I don’t know the current numbers, but even as early as 2004 – ancient history in the file sharing world, in the U.S. alone, households shared about 4 billion files a year (Karagiannis et al. 2004). That’s not a typo. Billion – with a “b”. Four billion, every year, in the U.S. alone. There is little doubt that the number today is much larger. These are truly fantastic numbers.

As books are increasingly consumed as ebooks – Amazon recently announced that Kindle books sales outnumbered the sales of printed books, the role of file sharing in the publishing industry is likely to become a major issue, and one well worth studying. While I am not aware of any research that looked carefully into file sharing and the publishing industry, previous work about the music industry can perhaps teach us something. More importantly, it can give us some ideas about how to approach this important question.

So, what do we know about the effect of file sharing on the music industry?

The remarkable increase in the numbers of files shared got the music industry really worried. A number that is often cited is that between 2000 and 2005, the number of CDs shipped in the U.S. fell by 25% (RIAA 2006). Since even just standing in place over 5 years is usually thought of as a failure – losing so much ground was perceived more or less as the end of the world as we know it.

The music industry sprang into action. They sued thousands of people, and even made it all the way to the Supreme Court to rule on the legality of file sharing, though it is not clear what effect these efforts had. In April 2003, in what was probably the most important move, the iTunes store was opened, and within 5 years they became the largest music vendor in the U.S., changing (dare I say – revolutionize?) the way the market for music operates.

However, one thing remained unclear – was it really the rise of file sharing that led to the decline in CD sales? How can we test that? As often is the case, there’s the easy but wrong answer, and there’s the hard, possibly correct one. The obviously wrong way to approach the question is to look whether songs that are shared more in file sharing websites sell less CDs. That is, not surprisingly, not the case. Big hits both sell more CDs and are shared more. Clearly, that doesn’t mean that file sharing increases CD sales.

The possibly correct way to approach this question is to use what social scientists call an instrumental variable. I will spare you the long boring details, but here’s the gist of it: if you can find something that is affecting the number of times that a song is shared online (that is, the number of times someone downloaded it from a file sharing website), but that is not correlated directly[1] with the number of CDs of this song that were sold – you are golden. It is not easy to think of something that does affect a song popularity on the net, but that we can confidently assume doesn’t affect its popularity in the stores. Still, in a paper from 2007, Felix Oberholzer-Gee from Harvard and Koleman Strumpf from the University of Kensas had an idea.

School vacations in Germany!

Say what?

Well, it turns out that in the P2P file sharing website called OpenNap, Germany is the single biggest foreign contributor of files that Americans download. In fact, 16.5% of the files that Americans downloaded originated in Germany. Canada is a distant second with 6.9% – talk about globalization… Now, during German school vacations, German kids spend more time with their computers uploading files. Better yet – they stay up late, and so they upload files when Americans are busy downloading files. The result is that during German school vacations, it is significantly faster for an American to download music from OpenNap. That in fact can be shown to have an effect on the number of downloads in the U.S.  Moreover, it seems safe to assume that it shouldn’t affect CD sales in the U.S directly. So, Oberholzer-Gee  and Strumpf ran all the numbers, used vast amounts of data, did a lot of robustness checks – you name it. No matter how you looks at the numbers, the conclusion is pretty strong – file sharing does not cause a drop in CD sales.

It is an excellent piece of research, and the paper was very successful. It was the leading article in the Journal of Political Economy, one of the most prestigious economics journals. That said, as is almost always the case, some people strongly disagree (check out this guy for example).

Now, back to books and publishing. I am not an expert, so maybe I missed some stuff, but I am not familiar with any estimation of the size of the file sharing phenomenon in ebooks. However –  it definitely exists. Some of my best friends… amm… borrow many books in these file sharing websites. But what is the effect it’s going have on book sales? ebooks sales? What can we learn from the research about music file sharing to the threat posed to the publishing industry?

For one, it gives us an idea of how to go about testing the effect file sharing already has on book sales. If someone can find a good instrument, that is – something that affects file sharing activity, but doesn’t affect ebook sales directly, she will be half way to making an important contribution to our knowledge of the subject.

Second, I would argue that if in the music industry there is no evidence for an effect of file sharing on CD sales, it is unlikely that it will have an effect on ebooks sales. There are three important differences between CDs and books when it comes to file sharing:

1. Most people read most books only once, so they are much more likely to lend books than they are to lend CDs. That is, even without file sharing, there is probably a lot more book sharing than CD sharing. Therefore, file sharing technology is likely to have a smaller effect in books than in music.

2. In the music industry, it was often said that CD buying was driven by heavy users – people that regularly purchased three or more CDs a week. These people were the first to move to file sharing, many of them, it is argued, just stopped buying CDs. In books, I suspect that heavy users are already using library services, rather than buying 3 books a week. In other words, the people likely to be the first to move to file sharing, unlike in music, were not big book buyers anyway. Clearly, that is not to say that no one will buy less books when they are available for free online. It only suggests that the effect is likely to be smaller than in music.

3.  iTunes store only came about as a response to the threat of file sharing. Amazon and others are well under way to improve the way they sell ebooks long before file sharing starts posing a real threat.

To conclude, speculation is fun, and if I had to guess, I’d say file sharing is only going to increase the buzz around popular books, thus possibly increase sales, and also allows more no-name writers to make it big time. But this is all just guessing. What we really need are some hard facts. Any ideas for an instrument anyone?

[1] Note that it is ok if through its effect on file sharing, your instrument will affect CD sales. The requirement is that it doesn’t affect CD sales directly.

This entry was posted in Books and tagged , . Bookmark the permalink.

I think I'm getting addicted to comments. Please feed the addict & leave a reply.

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s