Nordisk Copyright Bureau (NCB) administers the royalties of authors concerning records and CDs...The authors were divided into four groups in the analysis: Group 1, which receives DKK 1-10,000 from NCB annually, Group 2, which receives DKK 10,001-50,000, Group 3, which receives DKK 50,001-150,000, and finally Group 4, which receives more than DKK 150,000...
First and foremost, it turns out that the number of NCB recipients in the period after 2000 is consistently higher than at any point in time during the heydays of the recording industry from 1995-2000. Thus more authors were successful in releasing text and music in the period of declining sales than in the period of massive increase in sales. Furthermore, we may also conclude that the number of NCB recipients increases slightly in the three "poorest" groups. The number of authors who receive more than 150,000 is decreasing.
Less money shared by more people necessarily entails that someone is getting poorer. Since the number of NCB recipients in the high end (more than DKK 150,000) has declined and the other groups have more members now, a loss of income for the poorest artists would be expected. However, this is not the case.
NCB's own figures show that the average payment for Group 1 (less than DKK 10,000) has increased by 16.7% from 2001 to 2005. The income of Group 2 has decreased by 2.2%, the income of Group 3 has increased by 2.9, and the average income of Group 4 has decreased by 18.2%.
So, there are fewer artists in the high income Group 4 and they make less money than they used to. The low income Group 1 now contains more artists and they make more money than before.
I should clarify this somewhat confusing discussion about the distribution of payments. Less total money is being distributed to more people. Suppose that this took the form of a pro-rata decrease in income - so that the lowest 10% earn 10% less, the highest 10% earn 10% less and so forth. This would have the following effect: The number of people in the high end as measured by those earning more than a fixed cut-off would decline, and (if as is almost certainly the case, the density function is convex in the upper range) the amount of money they would earn per person would go down. At the low end the number of people would increase and if the density function is concave in the lower range, the amount that they would earn per person would actually go up. So basically, all the complicated distributional information Pedersen reports is basically saying that earnings per person went down.
Another salient fact is that while the sale of recorded music during the period has declined by a factor of about two, live performances have almost doubled. This may mean that income at the high end decreased less than in the recorded sales data if high end performers get a higher share of live performance revenue.
What does this mean for copyright? Copyright primarily serves to enhance the incomes of the top artists. So the fact that their income from recorded sales has dropped is pretty much what you might expect if file-sharing has weakened copyright. The rationale as to why it might be a good idea to enhance the income of top artists (as has been discussed in some previous posts and comments) is that it may provide more incentive to enter the industry in the first place. The key point is that the data here isn't consistent with that argument: the total number of people producing music has gone up rather than down.
