We can illustrate this with a graph:

What is the source of this graph? It looks suspiciously like something you’ve just made up. The edge cases you describe do in fact skew the way you describe, but there are other edge cases that will skew the other way. What statistics do you have that a “subscriber division” model will result in a substantially different distribution? You may be surprised. A subscriber division model means the most casual listener distributes the largest “cents per stream” and the casual listener may be the one most likely to select a major label filled playlist.