As the Digital Audio Insider suspects, Amazon is conducting a large scale experiment with prices for digital music downloads. For a single day, Opeth’s Still Life was only $1.99 and immediately boosted to position #3 on amazon’s best seller list. Today it’s the Greatest Hits album from Creed (also $1.99, also #3). Last.fm numbers indicate that both bands are fairy popular (345,852 listeners for Opeth, 523,844 listeners for Creed), but both albums are from the back catalog, what makes this even more astounding.

There is the suggestion that an album should cost $5, my guess is that in mid term the prices will even drop to around $3. This seems to be the amount a consumer is willing to pay on a voluntary basis. (Remember the Radiohead In Rainbows numbers. Divide $2,736,000 by 1,200,000 downloads, then you get $2.28 in average.)

Why so low? It’s not only random customer behaviour. The value of an mp3 file itself, as is with any digital information, tends to zero. It’s an economic fact, because the marginal costs for the production of a digital copy are nearly zero.

Despite this a digital download has a value for the customer, because of two reasons:

First, the service that provides the download is a value. If you pay, you save yourself hunting in file sharing networks and ending up with mistagged mp3s in bad quality. The better the service, the higher the value. iTunes is more convenient than amazon’s mp3 store, so they might be able to realize a higher price.

Second, people will pay because it’s fair. But you are not paying for the file, your want to support the artist. People (mostly) want to be fair. In this view, you are paying for the consumer-artist relationship. This means that probably a direct sale from the band’s website has a higher value for the customer – if they know that the money goes directly to the artist – that could equal much more than $3.