IFPI struggles to retain relevance

International wing of the RIAA, the IFPI, has issued one of its periodic reports calling for more to be done to stop piracy, topped with an eye-catching 'fact':

It also revealed that despite a crackdown in some countries on individuals who share large amounts of music, for every track sold legitimately there are still 20 unlicensed tracks downloaded.

Fascinating, eh?

But is it a fact?

Let's take a quick look at the full report to see, shall we?

First, we know the figure for legal downloads is 1.7 billion. This appears in a little box, complete with a source - albeit the IFPI itself. But they should know.

So, this means there are 21.4 billion unlicensed tracks, then. And that figure comes from... um, nowhere. It seems to be made up, as there's not even a source, much less a methodology.

In fact, if you look at this figure too hard, it starts to undermine the claims made on a country-by-country basis in the report. So, we're told:

2.6 billion illegal music files are downloaded in Mexico and another 1.8 billion in Brazil per year

- according to IPSOS research. That's 4.4 billion tracks in just two countries, which means about a fifth of all unlicensed tracks are downloaded in those two nations. Japan, we're told, accounts for 400,000,000 illegal downloads, bringing the total up to 22 and a half per cent of all global downloads supposedly being accounted for by three nations.

Let's add in China, shall we? The IFPI quotes a figure (again, unsourced) that the Chinese music business is worth $74million, with a digital piracy rate of 99%. In other words, for every single legitimate download, 99 are "stolen", supposedly. But what are the actual figures for this? The IFPI doesn't say, so we'll turn to an article from Knowledge at Wharton which is probably the source of this 99% figure and explains:

"That market has shrunk to several billion yuan a year because of piracies and the rise of Internet music. China's online music market has grown to between 10 billion and 20 billion yuan, and less than 1% of that is legal.

Now, elsewhere in the same piece the unit price of a download is mentioned:

A8.com charges 0.5 yuan to 2 yuan to download a song and offers a monthly plan for unlimited downloads for 20 yuan (15 yuan during special promotional periods). Top100.cn charges 1 yuan, Aigo 0.99 yuan, and Taile 2 yuan to 5 yuan for a single track download.

So, an average 1 yuan per track seems fair, and we can extrapolate that if the total sales of the industry are rougnly 15bn yuan, of which one per cent is legitimate, then than means the industry is selling 150,000,000 legal downloads a year. Given that the IFPI insists that there are 99 illegal downloads for each legitimate one, this gives us a rough but fair figure of 14.85 billion alleged illegal downloads in China.

So, four countries, and we're up to 91% of all illegal downloads accounted for by four nations. Factor in the stuff done in Russia and the number of illegal downloads that must take place in America and it's starting to look like Western Europeans are actually illegally. Only they can't be, as the IFPI also tartly claims:

In Europe, Spain and the Netherlands have
a huge online piracy problem resulting in
underperformance of their legitimate market sector. According to Jupiter Research, over a third (35%) of all internet users are now regularly filesharing infringing music in Spain and 28 per cent in the Netherlands.

Basically, then, the statistics in the IFPI handbook don't actually add up. Now, it could be argued that the IFPI's claim about a 1:20 legal:illegal download rate is merely an attempt to be cautious, but since the rest of the document is quite screechy, it seems unlikely. The varying forms the piracy data is given in - a number of downloads here, a percentage of the entire market there, the number of internet users who might be downloading over there - suggests that the IFPI knows that their numbers don't stand up to ten minutes oversight from a schoolboy with a slide-rule and so choose to try and smear them in the interests of a good headline. It's dishonest, and means that we should approach all their data with a very long pair of tongs.

As, of course, should their conclusions - for example, this, talking about Mexico:

Online piracy has hit the core music buying population in the region – research in Mexico shows that 64 per cent of music downloading is carried out by consumers in the wealthier ABC economic categories

Now, let's just resist the temptation to 'well, duh' the observation that the poorer Mexican consumers aren't downloading music tracks in a country where IT penetration amongst the lower social classes is so low the One Laptop Per Child initiative is distributing computers. Instead, just admire the totally unsubstantiated claim that the ABC social classes represent "the core music buying population", and the equally unproven suggestion that those 64% of downloaders are responsible for the 25% in Mexican physical sales. The IFPI has lots of claims, and many, many numbers: it's a pity that they so seldom match up.

They do announce an amusing new approach to fighting peer-to-peer networking - just saying it's rubbish:

Research by IFPI debunks a myth about illegal P2P services: in fact, fans get better choice on legal sites. IFPI conducted research with a sample of 70 acts on the legal site iTunes and on the copyright infringing service Limewire. In 95 per cent of searches the artists requested had more songs available on iTunes than on the leading P2P service.

Which prompts the question, if the supply is so poor on Limewire, what are you bothered about?

Still, fair play to the IFPI, with their rigorous research - although they don't say if it's a random sample; they don't even list the seventy acts they searched on: it could be that they were just looking for rubbish acts nobody's much interested in. Or perhaps the IFPI forgot that a lot of filesharers mung the names of bands and songs to avoid bots finding them.

Anyway - something, say the IFPI, must be done, and what they're pushing for is for ISPs to be forced to scan your internet communications. In effect, then - despite admitting in its own report that

only 20 per cent of internet users worldwide use P2P services

the IFPI thinks it's acceptable for all of us - including the 80% who don't use peer-to-peer - to have our privacy invaded merely to chase the chimera of illegal downloads.

If the IFPI wants governments to hand over our privacy, the least it could do is get its figures straight before asking.

John Kennedy is a man in a hurry, though:

FPI Chairman and CEO John Kennedy says: “A turning tide of opinion is one thing – a concrete programme of action is another. There is only one acceptable moment for ISPs to start taking responsibility for protecting content – and that moment is now.

Of course he wants action now. With EMI pulling out of the organisation, it's looking less and likely they'll be around to demand action next year.