As far as I'm aware (which admittedly might not be that far), all independent research on music piracy carried out to date has concluded that piracy either has no discernible effect on music sales, or does in fact have a positive effect. To a lesser extent this also applies to software piracy, where such has been lumped in with music piracy in some studies.
While there are certainly studies which conclude differently, all of them (or all which I have read/skimmed/glanced at/heard of) have been paid for by the MPAA, IFPI and/or other industry organisations. To my mind research paid for by an interested party should be automatically ignored as inherently untrustworthy, regardless of subject, which means that all empirical knowledge currently available concludes that piracy is either harmless or mildly beneficial.
Probably the first ever independent study was carried out by
University of North Carolina, by Oberholzer-Gee and Strumpf. The best known study was carried out by the
University of London, Industry Canada, and Decima Research in 2006.
A 2009 study funded by Virgin Media again found that pirates buy more music than non-pirates, echoing
a similar study from the same year by the BI Norwegian School of Management which found that pirates bought up to ten times the amount of music compared to people who never used file sharing, torrent services or similar.
Finally a very interesting study titled
"Media Piracy in Emerging Economies" was released earlier this year. This is a 440-page collaboration by social scientists from six countries and takes a much broader view, delving into causes, enforcement, policies and much more. Very well worth perusing if one is interested in the subject.
There are several more studies for which I've not bothered to supply links, but which ought to be readily available for the google-savvy. Notably a Swedish and a Spanish study, both independent.
Of course, whether music piracy is directly comparable to ebook piracy is an open question, as I'm not aware of any research at all dealing specifically with ebooks.
In any case it seems to me to be a consistent trend here, in that every new piece of independent research on the subject of piracy continues to come up with data negating the message spread by the MPAA, IFPI et al. While I don't think that necessarily makes piracy A-OK or that there's nothing more to debate, I do wish that the debate was grounded in the empirical knowledge that's available, rather than gut feeling and self interest.