Why Spotify Pays So Little

No, it's completely irrelevant. This is a discussion about a pricing model, not about an artists' lifestyle. Unless you want to think of the choices of artists whether to distribute on Spotify or not, depending on the pricing model, to be a 'lifestyle'.

As for what pjc50 wrote, no, he didn't quite get the point of the article.

Yes, perhaps the most equitable way to charge is for artists to set a price for 1 stream, or 1 permanent download (iTunes).

But the Copyright holders are NOT being paid 'per-play'. Sure, they're paid more when the plays go up, but that's not necessarily indicative or a 'per play' model. Just like your gas station doesn't sell gas 'per mile' you drive, yet if you drive an extra mile you obviously pay more at the gas station than if you didn't drive that mile.

Spotify is not a pay per play model for copyright holders. It's a pay-by-share model. This means that if 10 people are subscribed, and they all play two songs 10 times, the copyright holder gets as much money as when these 10 people played both songs 100 times each.

Both songs get 50% of the artist revenue each, in both cases, which is 70% of 10x $10 subscriptions worth, which is $35 each per song, whether both are played once, or 10 times, or 100 times, no difference.

Obviously NOT pay-per-play but pay-by-share.

If it was pay-per-play then there'd be an actual difference in the songs' revenue if both songs were played 10 times or 100 times. Not the case with Spotify.

Now, this is what the article is about:

There are multiple (here just 2) ways to implement pay-by-share.

1) You total up ALL plays on Spotify, and ALL revenue. So say you have $1m of subscription revenue, and 1 million song plays in total. You then pay out artists the fraction of their plays. Got 10 plays? 10 plays / 10m total plays * 1m revenue = $1, that's your revenue.

2) PER USER, you total all the user plays, and take the $10 subscription, and then pay out the artists as a fraction of how much the artist listened to that particular user. Artist got 30 plays, user played songs 100 times? You take 30 user-plays / 100 user-plays * $10 = $3, that's the artist's revenue.

Is there a difference? The assumption is yes.

The assumption is that some music, like classical music, gets fewer average plays by classical music lovers, than an average pop song. i.e. a classic music lover might listen intently to a 15 minute song, and then another, and have 30 minutes of music, and generate 2 plays.

And a pop song that's 3 minutes long, will get played 5x in that 15 minute period. And instead of 30 minutes of conscious listening, it'll be on in the background for hours and hours as it's made for radio. So you get 4 hours of 3 minute songs, that's 80 plays.

Now say there's 1 pop lover, and 1 classic music lover. And say both categories have 2 artists that get listened equally.

In model 1, you'd have 1 classic music lover listening to 2 classic artists and generate 2x 2 plays per day = 4 plays. And the pop lover generates 2x 80 plays is 160 plays, for a total of 164 plays. The two music lovers both pay $10, so you get $20, and the two pop artists both take home their 80 play share out of a total of 164 plays on $20, or: $9.75. The classic music artists take home 2 plays / 164 total plays * $20 each = 24c.

In this model 1, we see that the classic music lover that pay $10, is subsidizing the pop artists with a huge fraction of his $10, despite the fact he only listens to classical music.

But in pricing model 2, the two classic artists would each get 50% of the share of the classic music lover: $5. And the pop artists would both get $5, too. Only there'd still be more pop song lovers, so they'd still make a ton more money because which is fine, as again there are more pop song lovers. But then there'd also be more pop artists, so there'd be more competition.

Neither models are pay-per-play. They're both pay-per-share. But in model two, the classic music lover who plays 0 pop music, is paying 0% of his money to pop songs. While in model 1, over 90% of his money goes to pop songs which is being called a subsidy for songs that are shorter, more culturally fit to play as background music/noise.

In model 2, obscure, less popular, longer songs, songs with fewer average plays etc, get a decent bit of revenue, because they compete for revenue shares of users who listen to other like minded music, and thus they compete against other obscure, less popular (in absolute numbers), longer songs, songs with fewer average plays etc.

I think it's clearly a more fair approach. And I think it's clear it has nothing to do with 'preferred lifestyle', and the argument is certainly no straw hat.

It is not moving goal posts its considering whether the 'solution' will actually work given circumstances outside the technical feasibility.

I think it depends on local copyright laws, but I think in most places you buy a license to play music publicly and with that license you are allowed to play any music legally acquired.