Popular music is popular because it is listened to by more people, and that includes more people who are not such great music fans. The music industry needs to take this kind of information on board as we plan for the future. We have enough perverse incentives already embedded in the way we do business without wilfully amplifying or adding new ways to make fools of ourselves.
It took a bit of research to clarify how and why music becomes popular. We like to know what other people are listening to, so popularity is in itself a driver of additional listening by more listeners. It turns out we think we like music better if we think it is more popular too. The hyper curious can chase down work done by Salganik, Dodd, and Watts, who studied this phenomenon, even to the extent of faking popularity, with depressing results for those who think that cream rises despite our biases.
Some of us like a wider range of music than others. But it turns out that the model of an eclectic listener who dives deep and never surfaces is mostly a myth. Pop seems to be pop with eclecticists, as well as the casual listener; Anita Elberse has studied this in detail with Felix Oberholzer-Gee, in response to the influential Long Tail ideas expounded by Chris Anderson.
As her online bio says, “Professor Elberse is one of the youngest female professors to have been promoted to full professor with tenure in Harvard Business School’s history”. Looking at some large data sets Elberse found that nearly all of us start at the top of the pop pyramid; most of us stay there; a some explore a bit deeper and further. Elberse put this much more succinctly than I can:
consumers of the most obscure content are also buying the hits
The people who only like popular music turn out not really to like much music in their lives. They are the classic light listeners. Again, that should be no surprise as they don’t demonstrate much deep interest with their music choices. And conversely, the eclectics are the heaviest users of content services. Of course they are. They are passionate, and adventurous.
Many people wish all these awkward findings were not so evident, and despite what the world is showing them are campaigning for a change in the way that subscription fees are divided up and paid to copyright owners and musicians. Their big idea is that the money from each subscriber should be paid out only for those tracks that particular subscriber listened to. It seems to be fair, as it treats the monthly sub as a kind of pre-payment for actual consumption.
This was proposed by Pedersen in a study of a small data set from a Danish streaming service. His conclusion was that the popular artists did better with per user distribution (although some Danish artists also benefit), but that it was an open question how consumers would respond if they knew about the method. I shall quote his own conclusions:
Switching from the current pro rata distribution model to a per user distribution model would primarily benefit the most popular artists.
[and]
There are two primary benefits of shifting to a per user distribution model. First, it re-establishes the economic connection between the consumer and the artist, where the fees paid by the subscriber is distributed among the artists she actually listens to. Secondly, it benefits local artists, which could be interesting from a cultural policy perspective.
I would argue that we should be cautious about relying on any conscious and deliberate economic purpose from music listeners when we are thinking about how to develop our industry. It seems too much to ask that consumers should choose between royalty structures when they are wondering how to get their music, even where we might expect some to prefer suppliers that try to be fairer.
But a small group of commentators have jumped on the per user distribution idea and turned it into a campaign. Some clearly don’t understand the data, and claim that it benefits indie and niche artists. Others simply consider it fairer. It is an odd kind of fairness however, that decides in retrospect that one listen is worth more than another listen, and that listens by light listeners are worth more than listens by committed and engaged music fans. And it would have the very obviously unfair result of rewarding artists more to the extent that their listeners could be persuaded to spend less time with music as a whole, and with other artists.
We have to decide what kind of music fans we want. I have argued elsewhere for diversity and discovery as the signs of a healthy market. It seems to me that we get there by making sure that all the incentives are lined up so that we in the industry want our audiences to love more music from more artists, rather than less from fewer.
In subscription that means each listen should be paid out at the same rate, or at a rate that the copyright owners and musicians have negotiated. We have a huge range of choices for people who want music, and for the lighter listeners perhaps a few good compilation albums bought as CDs or downloads are better value. But there is no justification at all on the grounds of either fairness or outcomes to hypothecate subscription fees.