When digital music became available online, through various file-sharing websites, the music industry witnessed a substantial decrease in sales. The big recording companies – EMI, Sony, Universal and Warner – blamed the decline on piracy and intellectual property rights infringements. They responded by suing thousands of individuals who had shared digital music, resulting in several landmark court cases. The industry also responded by adopting technology – digital rights management (DRM) technology – which was embedded in music files purchased on-line, that made copying all but impossible. However, in April 2007, EMI unexpectedly removed such DRM technology from its music files, which then enabled individuals to share the music files they had purchased with others.
EMI’s move came as a surprise; so much, that many thought it was an April fool’s joke. This would obviously harm their sales, wouldn’t it? Individuals could not only share the music files they purchased with friends and family, but technically they could potentially also upload them on sharing sites. Surely, enabling this form of piracy would depress EMI’s sales? Yet, 2 years later, the other three record companies followed suit.
The question how it affected record companies’ sales is pretty much impossible to answer by just looking at the numbers, because there is no way of telling how sales would have developed had the record companies not removed DRM technology. Then, however, a clever PhD student from the University of Toronto – Laurina Zhang – spotted a golden opportunity to do some truly illuminating research. The fact that EMI removed DRM some time before the other three record companies, created what researchers excitedly call “a natural experiment”. She realised that by exploiting the time difference between these events and running a so-called difference-in-differences statistical model, she would be able to draw firm conclusions about how lifting DRM had affected sales. Eagerly, she got to work.
She painstakingly collected data on 5,864 albums released by 634 artists before 2007 – the date EMI dropped the DRM technology – and traced both their on-line and physical sales. She then put all this in a statistical model and calculated whether EMI’s album sales went up or down as a result of removing DRM technology from all its records (in contrast to the other record companies, which had not yet removed DRM). And the answer was unambiguous and clear: EMI’s sales went up.
In fact, EMI’s sales went up by no less than 10 percent. But why was that? How on earth could lifting the copy-restrictions actually stimulate sales?! To answer this question, Laurina (again) did something clever. She measured how the lifting of DRM influenced the sales of different types of albums: ranging from best-selling albums (more than a million sold) to relatively tiny niche records (less than 25,000 copies sold) and the results suggested a compelling story.
Removing the DRM restrictions enabled search and sharing among consumers, which caused many people to be exposed to music they otherwise would never have heard of. This left the sales of top-selling albums unaffected (people would hear those for instance on the radio anyway), but significantly stimulated the sales of lower-selling albums, by no less than 30 percent. Enabling sharing made people discover new fringe artists or rediscover old ones, and subsequently they went on to buy (other) albums of those artists on-line, through iTunes or Amazon. Consequently, EMI benefited significantly from lifting the DRM restrictions.
To me Laurina’s research offers a fascinating turn in our understanding of intellectual property rights (IP) and how they sometimes might have unexpected consequences. More specifically, with on-going digitisation in various industries – including games, books and television – it might make producers and rights-holders take notice and think twice.
But it also tells a wider story of sharing and search. It shows that for firms offering specialty products in their portfolios, enabling search and giving consumers access to sample products becomes key. Tapping into people’s network and desire to share with others could be a valuable opportunity to exploit for this purpose, rather than one to fight and restrict. Managers are often inclined to fiercely protect what they see as their competitive advantage and shield it from imitation and dispersion. However, sometimes it might be worth opening up, and reap the benefits in turn.
Yet, what I – as an academic – also much value in Laurina’s work is that it shows the sheer purpose (if not necessity) of academic research, to understand the world around us. Sometimes things – especially in the long-run – are different than meets the eye. What seems obvious at first (“of course removing copy protection is going to hamper sales!”) can be shown to work very differently when examined in a thorough and systematic way by someone who knows what she is doing (“in fact, if you dig a bit deeper, removing copy protection appears to increase sales”). And understanding the world around us clearly is a necessary first step for us to be able to improve it.