Having a Bubble Bath or Why the “Buried” Piracy Report Got It All Backwards

DISCLAIMER: The following text may contain sarcasm

Conventional wisdom says if you pour cold water into a hot bath, the temperature of the bath water will fall. New research, however, challenges this outdated view. That’s right: I was having a nice, hot bubble bath and decided to do a scientific experiment. So I opened the cold water tap and let it run into the tub. After ten minutes, I measured the temperature and again after fifteen minutes. To my surprise, the temperature was the same both times! It felt strange, because I was freezing, but you can’t argue with research. My experiment shows that pouring cold water into a hot tub does not decrease the water temperature. I thought about writing a 300-page report about it, but the government would only bury it.

You guessed it, I’m not really talking about bathing but the supposedly buried report that says piracy does not hurt legal sales. This idea is one of the pirates’ favourite daydreams. The recently leaked report adds to the daydreaming. I read the 307 pages so you wouldn’t have to. The mistake is on page 74. The claims go against established research, empirical evidence and common sense. The reason for the misleading conclusion is method problems (intentional or not, your guess is as good as mine).

Established research:

The available peer-reviewed research on the topic is as clear as academic research ever gets that piracy does hurt legal sales. Here is a good research overview. (Sidenote: When pirate-MEP Julia Reda says that the report is “consistent with other studies”, she means the wealth of studies that demonstrate that active pirates are also active legal consumers. From that, pirates like to draw the conclusion that pirate consumption is also a driver of legal demand. But the fact that the same music-lovers and film-buffs get content from both pirate and legal soruces, does not say anything about how one causes the other. Days with high ice-cream sales also have higher drowning incidents, but we don’t believe banning ice-cream would save anybody. Correspondence does not imply causality. Cum hoc ergo propter hoc is a fallacy so ancient, it has a name in Latin. End sidenote.)

The authors of the “buried” study seem to agree (page 83):

The academic debate whether file sharing reduces or increases legal demand for music is not settled. Although an increase appears to be counterintuitive, it might be achieved through so called ‘sampling’ or ‘exploring’, were consumers use downloading to sample a song from a particular album or artist before purchasing the music legally. Although various surveys found some evidence of sampling (3 out of 14), the net result of file sharing on music sales is considered negative in most papers (6 studies found a negative effect on purchases and only 1 discerned a positive effect). If the studies are restricted to peer-reviewed papers, only those with negative or insignificant estimated displacement rates remain.

Empirical evidence:

Global sales of recorded music were 27,8 Billion US-dollars in 1999. 17 years later, the figure was 16 Billion US$ or almost 50% less. 1999 was the year Napster started. Of course, there may be other reasons. Perhaps the interest for recorded music decreased significantly? Not very convincing

Common sense:

If I have a pirate copy of a movie, record, game, book, map or whatever, I will only rarely get a legal copy too. It might happen occasionally, but it’s the exception and not the rule.

Method problems:

The main problem is that the study only looks at how sales changed from 2009 to 2013 (this is page 74 and following). This is like checking the temperature ten minutes after I start adding cold water to my bath! The large-scale impact of internet piracy of music started in 1999. Taking the first data-point ten years later is in the best case stupid (and in all other cases intentionally misleading). Since this study was uncovered, I suggest the Unit 0.1 at DG MARKT who commissioned the report send it back and tell the researchers to do it again with a correct data set.

This data fault is so severe, it is hard to say anything about the rest of the method problems, but let me add one thing anyway. I must admit that I don’t follow the long discussion on various countries’ anti-piracy measures. If the conclusion that piracy does not harm legal sales would have been accurate, the anti-piracy measures would have no impact at all. But, as that conclusion is wrong, those measures would serve to decrease illegal use and to some extent increase legal use, thus offsetting the decrease brought by the impact of piracy on legal sales.

All told, this report is a very complex way of arguing an impossible position. I get the impression that it was not subject to peer-review, so the academic quality cannot be confirmed. The fact that the report was not released does not necessarily mean that it delivers some kind of “inconvenient truth” that somebody wants to keep away from the public. It could also be that the report simply is lousy.


The whole conversation about whether piracy hurts legal sales is actually pointless. Nothing stops a creator from giving away their work for free. That decision should be for the creator and not the pirate to make.

Don’t believe everything “buried” reports say. And don’t pour cold water into your bath.


EDIT: Updated more accurate sources and figures for music sales