News Releases

Debunking the damage of illegal downloads

Reactions to professor's research shows the power behind academic research.

Wednesday, March 10, 2010

It is a rare day to walk by Koleman Strumpf's office without hearing music flowing from his doorway. Strumpf, Koch Professor of Economics, readily confesses his love of music. Indeed, he has such an admiration for music that he even found a way to weave it into his research.

Back in 2000 when the Napster trial was in full swing, file sharing began to pique Strumpf's interest. His interest was so great that he decided to read a few of the briefs filed in the Napster case. What he found astonished him.

"The expert analysis in the Napster trial was something I probably would not give a passing grade in one of my classes," Strumpf states emphatically.

Music executives' battle cry was that illegal downloading was hurting record sales. In Strumpf's view, though, the evidence in support of this phenomenon was weak. He began contemplating how to test the legitimacy of the music executives' claim.

It wasn't until he spoke with Felix Oberholzer, a business professor at Harvard, that Strumpf formulated a method to proceed.

Oberholzer and Strumpf realized they had to overcome two barriers: 1) how to capture the data (both illegal downloads and retail sales), and 2) how to analyze the data once they had it.

As it turned out, the two researchers were easily able to obtain data on illegal downloading; all they had to do was cull massive log files. Oddly enough, it was the retail sales that proved the most troublesome. Aside from the top-ten list, not much information is published regarding retail sales. Fortunately, Oberholzer found a way around this dilemma.

With the first barrier whipped, the only question was how to conduct the analysis. The two decided to investigate illegal downloading during school holidays in Germany. Though seemingly abstract, this plan was actually quite masterful because, at that time, Germany had a high level of broadband, which meant people from around the world could quickly download music files held by German students. Furthermore, because the students were on vacation, they had all-day access to the Internet, providing a greater number of music files for longer periods of time.

If the music industry's claim was correct, retail sales should be displaced during periods when file sharing increased. Therefore, there should be lower sales of albums released in weeks with German school holidays compared to albums released in weeks with no German school holidays.

However, after crunching data on millions of observations, Oberholzer and Strumpf found nothing of the sort; their data revealed no indication that illegal downloading affected retail sales. According to Strumpf the results were polarizing.

When they first circulated their paper, the two received at least 1,000 emails, most of which praised their work. However, the people who were unhappy with the results were very unhappy. One music executive sent a letter to Strumpf that simply said: "You guys are in academic fantasyland. Get a life."

Strumpf said it was empowering to witnesses the feedback and to know so many individuals were interested in his research.

Oberholzer and Strumpf's research was published in the February 14 issue of the Journal of Political Economy.