When Napster introduced the concept of sharing songs, it was not so much the idea of sharing that was wreaking havoc on the music industry; it was the fact that albums were now being sliced and diced.
Traditionally, the music industry has been organizing around the concept of album sales. When developing a new music album (which is only a compilation of several music tracks,) the music industry decided to bundle some good songs, along with some so-so and sometimes some bad ones. The idea is that they would promote a few songs in the media and use those as a way to sell albums. Where the economic breakdown happens is that not all songs have the same value. As a result, the idea that a hit song is worth the same amount as a B-side falls apart. So if you take the current album-related economics model, you end up with a product of 10-15 tracks, which retails for somewhere between 15 and 20 dollars. Based on that concept, one can argue that all the songs on an album are worth about 1 dollar (15 dollars divided by 15 tracks.)
Some songs are going to be promoted more heavily than others so what happens is that the statistical distribution of prices across tracks has much more variance. A hit track could be worth as much as 2-3 dollars while a B-side track could be worth as little as 10 cents. Online music stores, however, do not work along those lines; they still price all tracks at the same level, no matter what their popularity is. What this does is create an imbalance in selling music as popular tracks are underpriced while unpopular ones are overpriced. Since there are generally only about 2-3 hits on an album, the value of that album now falls to 2-3 dollars, leaving the music industry to try to figure out how to make up the difference.
The economics of downloaded music need to also take into account the fact that the only thing being delivered is a digital asset. This means that the music producer actually reduces their cost since they do not have to deliver a medium like a CD or tape in order to make that sale.
The solution to this problem is to carefully analyze how all the tracks on an album will perform and price each accordingly. In a perfect world, songs should be priced based on an algorithm that would take into account their popularity. This would reward early adopters for discovering a song before it became popular by offering them a lower price (since we can assume that word of mouth on a good product will only influence later sales.)
In 2000, I looked at a number of possible business models for the music industry. While many of those have been implemented, there is still room for innovation in that space.
Tomorrow, we will look at how some of the digital music stores are now fighting for new supremacy in the space and how their different models are impacted by the modular by design approach.