When journalists encounter statistics, music-downloading edition

By Felix Salmon
September 6, 2011

The NYT’s Janet Morrissey, this weekend, had a long profile of a music-download service which as far as I can else no one else has even so much as thought about for about three years. The story itself is not particularly noteworthy, but there’s one paragraph near the beginning which encapsulates very neatly a lot of what can go wrong when journalists encounter statistics. Here it is:

About 95 percent of music downloads in 2010 were unlicensed and illegal, with no money flowing back to artists, songwriters or record producers, according to Alex Jacob, a spokesman for the International Federation of the Phonographic Industry. So riches could await a company that persuades some of these Internet scofflaws to change their ways.

This paragraph is structured in a standard journalistic manner: start with a fact, then say what your source was for that fact, and then draw a conclusion from that fact. Except that in this case, the standard journalistic formula is twisted inside out.

First of all, the central assertion is almost certainly wrong. I feel comfortable in saying that it’s simply not true that 95% of music downloads in 2010 were illegal. That number was, literally, incredible in January 2009, when the IFPI first trotted it out as a number averaged over a three-year period. And since then, the iTunes music store and other legal services have grown very fast. What’s more, even the IFPI seems to have backed away from the figure: it appears nowhere in its 2011 Digital Music Report. The highest such number there is the claim that “in the UK, for example, 76 per cent of the music obtained online in 2010 was unlicensed”.

In any case, the last people you’d trust to come up with a reliable figure for such things would be the industry group devoted to demonizing and exaggerating the threat of illegal downloading. Morrissey doesn’t seem to understand, here, that you can’t just quote the IFPI and be done: the fact that the number is coming from the IFPI, far from being a reason to believe it, is a reason not to believe it. At the very least Morrissey should have asked whether the number was based on any kind of empirical research; instead, she felt comfortable simply parroting the nonsense being put out by the paid flack for a pressure group, and reporting it as fact.

And what’s more, the statistic isn’t even relevant. Let’s say I download 361 albums one night over Bittorrent, and I have 19 friends who each download and pay for a single album on iTunes. Then it would be true to say that among me and my friends, 95% of the music downloads were unlicensed and illegal. But it would also be true to say that 95% of us are downloading music legally and paying for it. Would riches await a company that persuaded the scofflaws to change their ways? No, because there’s only one of me. And if I start paying for the music I download, you can be sure I won’t be downloading 361 albums in one night.

Before you start quoting statistics, then, it’s always worth (a) knowing where exactly they come from; (b) verifying them independently if you were fed them by some pressure group; and (c) making sure that they say what you say that they say. Otherwise, you just end up looking credulous and silly.

Update: IFPI’s Alex Jacob responds in the comments, saying that his 95% figure is “based on independent research in a series of markets around the world”.


We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see http://blogs.reuters.com/fulldisclosure/2010/09/27/toward-a-more-thoughtful-conversation-on-stories/

Another one of my journalistic favorites goes something like this:
1. Assert there is a problem.
2. Find a single example illustrating that problem.
3. Imply that single example conclusively demonstrates a systemic problem.
4. QED.

Regrettably, I’ve seen you do this on more than one occasion. You’re better than that.

Posted by Curmudgeon | Report as abusive

QTrax is pursuing an already-dated model, so the issue of the rate of illegal downloads is pointless.

In the nascent digital media years, people were desperate for content any which way they could get it, and content owners were fairly adamant about the issue: illegally download it, or you get nothing.

Times have changed. Today forward, it is all about cheap or free cloud-based services, not free, legal downloads.

My library offers three completely free (and without DRM) MP3 downloads a week. However, once those Google Music Beta and Spotify invites came, I never bothered to download another track from my library.

Same thing goes for videos. Once you have legal access to streaming content (and either free or at a cheap rate) you start to care less about owning content, and more about the breadth of titles that are available.

Stop obsessing with torrents already…even before MP3 digital media, we had the mixed cassette tape and the warehouses of mom and pops sellers of illegally copied CDs, cassette tapes, VHS, Beta and DVDs.

Posted by GRRR | Report as abusive

This myth is for some reason trotted out by politicians when they have no clue that it comes from some tiny sample size from one night that was extrapolated out across the world. There are also studies that show that downloaders of illegally obtained music are also the biggest purchasers of legal music, so just like listening to the radio to see what they like, maybe they download to see what’s worth getting?

Microsoft used to trot out (and still does if you ask) the same line about how the world was stealing its software and how many billions it was costing them. This from a company with 94% market share with Windows and over 90% with Office. The argument there was the same lousy extrapolation, multiplied by their highest retail price presumably on a per program, not per package basis, yet they will happily sell their Office package software at less than $10 to employees of large organisations just to maintain the high market share that drives the purchase of legal copies.

Another thing that needs looking into more seriously is the ‘divide and rule’ of disallowing cross-border selling. This is anti-competitive and probably drives much software pirating. Most software products – even downloadable ones – are cheaper if you buy them in the US or from the US, and companies like Adobe, Microsoft and to a lesser extent Apple all charge more in other places. It’s no wonder some people take the actions they do. Even if in reality the damage really isn’t that severe.

Posted by FifthDecade | Report as abusive

Nice analysis and example.
I especially liked how you pointed out that if you were downloading illegally, if you went straight how insignificant the affect on the profitability to the music industry would be. Most of the commenters get this. And businesses that mint money (e.g., Microsoft) get this.

I think it is very, very tough for some people to accept the fact that their “creations” are not worth as much as they want them to be. I very much enjoy your blog, but how much would I pay for it? Hmmmm…the thing of it is, there is so much competition. But even without any other choices, I doubt I would pay even a buck a month.

Posted by fresnodan | Report as abusive

geez! “the thing of it is, there is so much competition. But even without any other choices, I doubt I would pay even a buck a month.”
That looks terrible, makes me feel bad, and makes it sound that you are worthless.
The fact of the matter is that I read dozens of blogs daily, and sample dozens as well. So if you look at it from the standpoint that out of tens of thousands of people spouting off, I find you one of the few worth coming back to…well, that should make you feel much better!

Posted by fresnodan | Report as abusive

I thought it would be useful to clarify a couple of questions raised by Felix Salmon’s article.

Janet Morrissey asked what IFPI’s current estimate is for the percentage of music downloads worldwide that are unlicensed and illegal. That figure is 95 per cent.

While legal download stores have increased sales since the figure was first calculated in 2009, illegal music downloading has largely kept pace with that growth and continued to grow as broadband networks have rolled out worldwide. This growth in illegal downloading is verified by numerous independent studies. Felix draws attention to the UK estimated digital piracy rate of 76 per cent which comes from Harris Research. There is also a New Zealand estimate of 80 per cent provided by IPSOS. Both are consistent with the global 95 per cent estimate as these lower figures are counterbalanced by markets such as China, which has an estimated 99 per cent digital piracy rate.

These estimates are based on independent research in a series of markets around the world. IFPI always attempts to present an accurate picture of developments in both the legitimate digital music market and piracy trends.

IFPI does not present every illegal music download as a lost sale. We point out that the explosion in digital piracy in the last decade has mirrored and been a major cause of declining industry trade revenues. It is self-evident that if the digital piracy rate could be reduced and more users migrate to legal services then more money would flow into the industry and be available for discovering, nurturing and promoting artists.

Legitimate music services, as well as artists and producers, stand to benefit from users’ migration from illegal websites to legal forms of accessing music. Therefore Janet Morrissey’s assertion that “riches could await a company that persuades some of these internet scofflaws to change their ways” is a fair and reasonable one.

Posted by ifpialex | Report as abusive

There’s another problem with Morrisey’s assertion. The idea that folks who are illegally downloading things would cough up significant amounts of money if they couldn’t seems unlikely to me. I’d think that the majority of those people would simply do without or listen to internet radio. In general, claims of all the money being lost to pirates has always seem overblown by orders of magnitudes; people are a lot more careful about purchasing things when real money changes hands.

Speaking of not spending money on digital content, the last three titles I thought about downloading to my Kindle from Amazon were more expensive in Kindle format than paperback. They didn’t get downloaded. Sheesh, whatever happened to the brave new world of digital media that was supposed to be better for the user?

Posted by David239 | Report as abusive

If I want to buy a legal download I am restricted to the choice available from the country in which I am resident. Is it any wonder that people go for illegal downloads just to avoid a record company telling them what they can and cannot buy?

Posted by FifthDecade | Report as abusive

Great catch Felix! There is one other journalistic trait that should be pilloried here. It goes like this:
1) There are two types of people A and B
2) Type A people have this behavior (on average) that Type B people don’t
3) Therefore, if we turn these Type A people into Type B, or breed more Type B people, or somehow eliminate some Type A people, we will get more of the (desirable) Type B behavior.

This may be true in an engineering setting in which you can move the columns of a building back and forth to change the center of gravity. However, this is almost always false in social science problems, that you can cause Type A people to become Type B people. In this example, it is implausible that these illegal downloaders can be made to change their ways; if Bittorrent disappeared tomorrow, I bet that a lot of these people would not be buying tons of music legally. This is exactly David239′s point above.

Posted by junkcharts | Report as abusive