Research Note—Testing File Sharing's Impact on Music Album Sales in Cities

What is it about?

This article examines how sales of sound recordings changed after the birth of Napster and its progeny. It finds that cities with higher levels of internet use also had the largest decline in record sales, consistent with the possibility that piracy is harmful to sales. It also finds that genres that are least heavily downloaded, such as classical or jazz, have the smallest declines relative to heavily downloaded genres. The specific empirical result implies that sales of sound recordings would have grown during the early years of Napster except for the fact that piracy became important during those years, leading to a decline in sales.

Featured Image

Why is it important?

This study has several advantages over other studies in the literature. First, it uses cities as the unit of analysis, as opposed to, say, individual sound recordings. This avoids the simultaneity problem found in studies examining downloads of individual songs where it is very difficult to properly control for the fact that popular songs are both the most heavily purchased and also the most downloaded songs. It also has an advantage of using entire populations as opposed to small samples of survey respondents who are asked to remember how many songs they downloaded or purchased. Instead, actual recording sales data are used. And it is able to control for socioeconomic characteristics across cities by using a "fixed effects model" that examines changes over a period of 5 years to control for differences in cities that are largely fixed for those 5 years. The results are economically important and robust to various changes in specification.