Empirical evidence about music parodies

Kris Erickson, Evaluating the Impact of Parody on the Exploitation of Copyright Works: An Empirical Study of Music Video Content on YouTube (2013):

The discussion section provides evidence that the presence of parody is not likely to cause any economic damage to rightsholders; in fact, the data suggests that the presence of parody is related to increased audiences for original works and therefore increased revenue. A small but growing market for skilled amateur parody production is identified and described.

It's a bit hard to tell given the way the data are broken down, but the definition of "parody" used in the study seems to include vids--instances in which a song is used to tell a new story with existing video clips.  The study doesn't explain whether it classified vids as "weapon" parody, where a song is used to comment on something else, but it seems likely.