On the way to and from a meeting at the OU's main campus in Milton Keynes yesterday, I heard BBC news programmes repeatedly dole out the British Phonographic Industry's latest unsubstantiated statistics about how much they wish people to believe they've lost in music sales due to illegal file sharing on the Net.
The calculation they do goes something like this.
Assume there are 10 million teenagers illegally sharing music.
Assume the average teenager swaps 40 songs (through choosing a relatively small number they can claim they are underestimating the problem, though in reality no one knows how many copyright infringing files are swapped).
Assume the average album has 15 songs and costs £15.
So the average teenager is swapping £40 worth of songs per year.
Assume they would otherwise have bought £40 worth of music in CDs or legal downloads.
That means we've lost 10 million X £40 worth of sales per year.
That means we've lost £400 million worth of sales per year.
So that means we've lost £1.2 billion sales over the past three years.
You don't have to be a statistician to see there are huge unsubstantiated assumptions here and that these "statistics" have no basis in reality. What's remarkable is the degree to which we, and the good folks at influencial news outlets like the BBC, will accept that doing some simple (and often not even mathematically sound) arithmetic will magically impart meaning to some numbers picked randomly out of thin air. The BPIs numbers therefore are entirely meaningless. Are they right however to be worried about illegal file sharing? Possibly. Various detailed academic studies have suggested that file sharing contributes to an increase in sales, a decrease in sales and some on balance have detected no significant effect. What is clear is that the effect of file sharing on music sales is not a simple linear relationship. We can identify various categories. For example,
1. The cash strapped file sharer (typical teenager, say) who does indeed use the file sharing networks to get music they would otherwise not be able to afford.
2. Those who use p2p to get music they would otherwise have bought - these are the folks the BPI have arguably got a legitimate beef about.
3. Those who use p2p to get music that is not otherwise commercially available or is rare etc.
4. Those who use p2p to get music that is out of copyright or is released eg under a creative commons or other less restrictive licence
5. Those who use p2p to sample music in anticipation buying a legitimate copy if they like the songs.
6. And you could go on describing a whole range of other dynamics and motivations that encourage folks to use p2p
At the weekend on a morning BBC news programme there was a spokesman from the BPI on with a member of the Drifters bemoaning the short 50 year term of copyright for music recordings. It's not fair said the Drifters man - I want to be able to pass on to my grandkids that I wrote that song 50 years ago. Its not fair said the BPI man, we invest 15 to 18% of our earnings in new talent. We must be given a longer copyright term, like they have in the US. There's not another industry in the world outside of copyright that have a minimum 50 year monopoly on sales of their product but for the poor BPI it really isn't enough. It's a pity there couldn't have been someone to put the case against extension of the copyright term, especially since the ministers in the UK government responsible for looking at this are so keen on the prospect. But when it comes to IP you really do have to suspend your disbelief.
I see Ian's been getting irritated at the latest statistical sophistry too.