File sharing: Guilty as charged?

New numbers on declining music sales could mean that MP3 trading really is hurting CD sales. But that still doesn't mean we should lock up the pirates.

By Damien Cave

Aug. 23, 2002 | Does MP3 file trading hurt the music industry?

It's a question that has caused heated debate ever since Napster exploded on the scene in 1999. And as sales of recorded music have declined over the past two years, it's a question that has taken on ever-greater importance -- for the music business, Congress and music fans.

Up until recently, there has been little hard data to support anyone's claims that file trading is hurting -- or helping -- music sales. But at least one researcher, University of Texas (at Dallas) economist Stan Liebowitz, author of an upcoming book (set for publication Sept. 7) titled "Rethinking the Network Economy," is digging hard for quantitative answers.

In May, Liebowitz published a paper suggesting that the record industry would soon be seriously harmed by MP3s. But in June, by the time Salon caught up with him, he was questioning his own conclusions after having examined the numbers and finding little solid proof that file sharing was hurting CD sales.

Two months later, he's changed his mind again. Sort of. In an insightful, yet-to-be published paper that analyzes 30 years of record sales figures, Liebowitz argues that MP3s are in fact having a significant negative effect on the CD market. He acknowledges that new data could once again lead to new conclusions, but for now, Liebowitz says, "I've moved somewhat closer to the record company position."

Salon called Liebowitz at his home in Dallas to discuss his findings.

When we last spoke, you said you had yet to find proof of harm from MP3s. What's changed?

The one big piece of evidence that I didn't have when we talked before was a half-year year 2002 number [that appears to indicate a 9.8 percent decrease in album sales.] There has to be a caveat in here, which is that I don't know if this number is correct. It's a half-year number that I saw in USA Today, from SoundScan.

If it were the case that there was a 9.8 percent drop on albums, when you look at the historical record of the ups and downs of the CD industry, [that's] a bigger decline than we've seen in 30 years. It starts to look unusual.

How much bigger is the decline? Is it a significant drop or a slight depression?

I haven't figured out the percentage but it's definitely bigger than the other ones. Now, let me add another point. When we last spoke, we were talking about a 5 percent decline in sales and I said, Look, if this is a recession then a 5 percent decline isn't so unusual. At that time, I had assumed that record sales moved with income; during a recession, you could expect fewer records to be sold. When I actually ran the numbers, with income as a variable, it had a very small impact. It was what is known as statistically significant but it was so small that you could ignore it. So in fact, you couldn't conclude that because we're having a recession, you might expect a 5 percent reduction in record sales. That's the other prop I was leaning on [to show that MP3s are having an effect.] There's evidence that something different is going on.

But assuming SoundScan's figures are correct, your paper seems to hang on a matter of degree. If 2002 sales are down 9.8 percent, you argue, and if this continues, the decline will be the biggest in 30 years. But aren't there other possible reasons for the decline?

I mention that there are these supposed instances of doldrums in musical creativity and you read about them from time to time. But it's a hard thing, at the moment, to measure that.

But isn't it possible that the intersection of several other unprecedented factors wholly independent of MP3s could be causing the decline in sales?

That's right. It is certainly not conclusive, by any means, that there's real damage going on from MP3s. It could be that we're having a bit of doldrums in terms of taste; it could be that we're all using CDs now and nothing else so since they're a little more durable than other formats that could be part of it. But it is at least beginning to look like there is damage being caused. But remember, the original story was that there's so much MP3 downloading going on so we should see a really big impact fairly easy. And now we're seeing a medium impact, which still could be explained by other things -- but we can't discount the MP3 possibility.

If the record industry is somehow not able to stop the downloading, I think we'll know by 2003. We'll get the end of the 2002 year numbers when December's over. We'll see what actually happened. And I expect that by 2003, whatever's going to happen will have happened. This is a great experiment for people who are curious about issues like this. We'll eventually find out. So while it's premature to say this is the smoking gun that shows that harm is there, it is certainly more indicative of harm than what had been there with just the 2001 numbers.

Next page: So, should the studios be allowed to hack into the computers of file traders?