Every time somebody listens to a song on Spotify it generates payments, but Spotify does not calculate royalties based upon a fixed “per play” rate.
Your buffet analogy isn’t really apt here because streams aren’t limited resources, and they all have the same effective wholesale cost to Spotify (again, assuming no secret deals).
You're totally right on the merits, but it's worth noting that the big record labels are composed entirely of and built their empire solely upon unjust-seeming secret deals, and we have no reason to believe that they've stopped that modus operandi now merely because they're dealing with Spotify.
I think the real problem that isn't being discussed is the fact that on a service such as Spotify it is impossible for an artist to be independent, his/her efforts WILL support majors via their equity in streaming services.
I listen to Spotify and Pandora almost exclusively. Mainly Spotify. I probably listened to 300 tracks of Emancipator a month on Spotify. At 0.00786 per play, that works out to 2.85. For the last 2 years: 2.85 * 24. That's $56.59. I seriously doubt I would have spent that much on iTunes. In fact, from their site, I can only spend $46 on their music.
http://emancipator.shop.redstarmerch.com/Dept.aspx?cp=69283_...
In the long run, it seems to me that Emancipator is winning on Spotify, even if I overestimated by a factor of 3 and only listen to 100 tracks a month. Presumably, I will be listening to Emancipator for a long time.
I don't think most people listen to 300 tracks of any artist per month consecutively. Say you are doing 30 tracks per month on average by an artist, that's $5.66, less than buying three 10 track albums.
I have hundreds of albums on CD (I'd guess ~500 or 600). There are many that I probably only listened to five times.
My main gripe is that because of this stupid model, it pushes a lot of the artists I would listen to away from spotify because it doesn't connect them directly with the revenue stream they should be getting from their true fans. Artists have a hard time surviving without that.
But there's a separate listener: the explorer. The value of any given track is exceedingly low, but these are the people you depend on for the discovery function. Personally, I prefer that role, but the cost before Spotify was exorbitant unless you were an industry insider.
If you think on principles, Spotify should in the long term aim for maximum customer satisfaction. This means optimizing the satisfaction per customer instead of per music play, seems to be currently done. By concentrating money on bands each user listens to each of those bands (getting more money) should theoretically improve in quality (and give Spotify more attention), increasing overall user satisfaction. It's a simple reasoning.
Or split Terry's money between the artists they actually listen to.
This scheme is similar to the proposal to calculate artist pay independently for each user, but I think it's a little simpler, since you still only calculate the artist pay out of the global pot, you're just weighing each play by the user's total monthly plays. And you can put a minimum on it too, so the per-play weighting doesn't take effect until the user listens to more than a given number of tracks per month (this way the user who has a subscription but just occasionally listens to a song here and there isn't paying the artist $1 per play, which is to say, until you reach a certain point, listening to other songs won't "devalue" your previous tracks).
This is a bit like being a vegetarian at a buffet and complaining about your admission fee paying for steak which all those other people are eating. It's true, but you came to the buffet because it was much cheaper than a la carte with better service.