The release of the latest NPD estimates of music consumption by internet uses causes some gloom over at Paid Content:
Demand for music down in Q3; paid downloads grow a mere 2%
But is that actually supported by NPD's report? (Their figures, it should be noted, are all drawn from the US market). Not quite, as it turns out.
15% of internet users bought a download in the last quarter; in the same period in 2007, 13% of internet users did so. This is not, you'll note, a growth of 2%, but of two percentage points - the growth is actually more like 15%. A market growing by that size, against the current backdrop, is surely something to be applauded, rather than miscalculated and dismissed as a "mere" figure.
More importantly, as PaidContent notes:
In general legal music download volumes grew by 29% in Q3.
That's pretty impressive growth.
And there's some even better news for people who sell music:
Teens purchased 34% more paid digital downloads compared with year-ago
Not bad, considering the prevailing glum-wisdom that young people won't pay money for music.
With all this purchasing and growth, it's hard to see how the gloomy 'demand for music' stands up. Let's look at the original press release:
According to The NPD Group, the leader in market research for the entertainment industry, year-over-year consumer demand for music among Internet users in the U.S. fell 2 percent in the third quarter (Q3) of 2008. This overall decline in music acquisition includes purchased CDs, purchased digital music downloads, files obtained on peer-to-peer (P2P) file sharing Web sites, and borrowing music files to rip to a computer or burn to a CD.
So, that's not quite 'demand for music' as 'number of discrete tracks obtained' which has fallen. Could it be that part, or maybe all, of that 2% would be wiped out by the number of people streaming music from Last FM, Imeem or YouTube - or simply listening to online radio? It's a fatal flaw to assume that demand for music can only be satisfied by the acquisition of a thing; it suggests that NPD's surveys might be counting the wrong things in the first place.