In August, Professor Michael D. Smith and Rahul Telang from the Carnegie Mellon University in Pittsburgh published a paper about how piracy impacts sales of media products. Their study is based on review in the field of academic literature in economics, marketing and information systems. Smith and Telang conclude that almost all papers published in major peer-reviewed academic journals find evidence of statistically significant harm to sales of recently released content as a result of illegal file sharing. 

Comparative analysis in this field is tricky. For instance, not all the literature seems to endorse this evidence. For example, a survey in the Netherlands by Helberger, Huygen and Van Eijk showed that media pirates purchase as many CDs as non-pirates do. However, according to Smith and Telang, the problem with this survey is that it ignores a potential endogeneity problem. "If there are unobserved characteristics of consumers (say their interest in music) that might influence both a consumer's propensity to pirate and their propensity to purchase, then one cannot conclude anything about a survey that finds that pirates purchase as much as non-pirates do. Put another way: while pirates might purchase as much as non-pirates do, we have no way of knowing how much these pirates would have purchased if piracy weren't available." they write in their report. 

All in all, it's hard to draw firm conclusions about the relationship between piracy and the sales media products. The papers described in the study of Smith and Telang span a variety of methods, time periods and contexts. Still, the overall line is that piracy influences the sales of media products in a negative way. 

Source: Smith, Michael D. and Telang, Rahul, Assessing the Academic Literature Regarding the Impact of Media Piracy on Sales (August 19, 2012). 

For additional scientific analysis in this field, see the study Feiten om te Delen (Dutch only).