On Streaming and Commodity Pricing

UPDATED Aug 1, 2025:
The following, which reads partly as an apology for Spotify, has me feeling icky, now that I’ve learned about Spotify’s CEO’s investments in AI-driven weapons tech. I am much less “of two minds about Spotify” now, and actively exploring alternative ways of finding and managing digital music.

I got thinking about this after reading a number of pieces lately about Spotify – people critizing Spotify for being exploitative and underhanded and monopolist. I don’t really disagree with the substance of these reviews, but the whole discourse brought to mind for me an old blogpost from a decade or more ago, in which I worked through some of these thoughts for the first time. I reposted that last week: Cats and Commodity Pricing.

I am very much of two minds about Spotify. Obviously, they are a big ugly corporation playing the winner-take-all Internet capitalism game, where gathering the largest user base means you gain untrammeled power. At the same time, I can’t help feeling that Spotify delivers a straightforward and pretty hefty amount of value to me, and does so for a pretty low price (more about that below). Indeed, Spotify’s ‘contract’ with me as a consumer appears pretty uncomplicated: I pay them the monthly subscription fee and they let me listen to just about anything I want, whenever I want.^[1] They’re obviously collecting data about me, but if that stupid “Wrapped” thing they do every year is any indication, they’re not learning much. Compared to other big Internet companies (Google, Microsoft, Apple, Meta…) Spotify appears relatively benign, as I don’t see a lot of evidence of weird ulterior agendas, dark patterns, and all the other enshittified nonsense that’s so common on the Internet these days.

But appearances are not the whole story, are they? That Spotify gives me all the music I want for roughly $13CAN per month is surprisingly close to the deal I was getting from record and CD stores back in the 90s, when I last paid for music the old fashioned way. Back then, I spent probably something in the order of $200 a year on CDs or vinyl: not so different from my outlay today. Admittedly, if we correct for inflation, I should probably be paying something more like $30 a month today, but that difference is a detail; it doesn’t speak to the structural issues of a service like Spotify.

The huge difference between shopping at the CD store in 1994 and subbing to Spotify today is the vastly larger catalogue of music available via streaming: orders of magnitude larger. And here is where I think a lot of the complaints about Spotify miss something key.

People like to complain that only Taylor Swift and her echelon can make any money on Spotify. It’s not strictly true, but the bare fact is that the music business is and always has been a “long tail” business, where a small number of hits make the vast majority of money and then there’s a much larger “long tail” of other stuff that makes much less money. This was laid out in a tidy little article by Chris Anderson in Wired magazine back in 2004. Back then, Anderson was breathlessly heralding what he saw as a new era of retail, in which the physical and economic limitations of inventory would give way to an electronic market in which we would “Make everything available,” “Cut the price in half; now lower it,” and “Help me find it.” Anderson was kinda half-right about all this. Amazon, Netflix, and Spotify have done a world-changing job at making (almost) everything available, and cutting the price in half or more, but Anderson’s techno-optimism about the filtering that would help us make sense of all this turned out to be another thing altogether. In fact, “helping people find things” turned out to be another game entirely – another business, and one ripe for enshittification, or at least payola. We definitely do not live in the filter-driven utopia Anderson imagined. But undoubtedly, a lot of stuff is pretty accessible now.

Long-tail theory explains a fair bit of Spotify’s success. It also accounts for a big chunk of the criticism, too. Spotify, we’re told, is bad for musicians; the average musician makes peanuts from streaming revenues; musicians have to make their money from touring and merchandise nowadays; and the algorithm works against art and artists. All these things are arguably true. But, as an old guy, can I please ask when were any of these things different?

The Music “Industry”

The history of the music industry goes back to the 19th century, when people bought mass-produced sheet music, and morphed into something recognizable to us early in the 20th century when recording technology became possible and then, by degrees, ubiquitous. Back in the day, to get recordings out to market, you needed a recording studio, tons of specialized gear to make a master recording, and then the capital to work with a pressing plant to manufacture and distribute records. Industrial-scale sales and promotion was also part of that process. This kind of capital-intensive heavy industry is not so different from lots of mass media: publishing, journalism, film, radio, etc. And we know from endless biographies of musicians that this system was really not friendly to artists. The idea that there was some pre-Spotify world in which musicians could easily make a living is a fantasy. Back to the “long tail” again: if it didn’t look like you could sell zillions of records, no record company would sign you in the first place, leading to the situation I recalled from my childhood: the number of possibilities in the record-store universe was vastly smaller than the universe of music available on streaming services today.

There was a time in the 1980s and 1990s when “indie” labels and DIY bands had a moment, as recording technologies got way cheaper and more accessible (Tascam Portastudio, anyone?) and subcultures flourished in which indie acts could circulate and find audiences. But realistically, this was never very big, compared to what the organized networks of major labels, radio stations, and promoters were capable of. Could musicians make a living? In a punk-rock sense, maybe. But that moment did set us up for the massively complex plurality of music production and consumption today, on top of which Spotify sits.

By my telling of this story (which I acknowledge is just mine and could probably be contradicted by others’ accounts), I’m not sure the Spotify-dominated landscape is all that different from what preceeded it in the 20th century: a handful of big stars get all the attention, make all the money, and everyone else labours in relative obscurity. The part about big stars is pretty much the same; perhaps the “relative obscurity” part is arguably different.

One of my students years ago pointed out that the distinction between being an artist and being a celebrity: one is vastly more lucrative than the other. We should not be fooled into conflating these two identities. Don’t get me wrong: I am not making a judgment here about aesthetics or culture. Taylor Swift may be a great artist. But that’s not what makes her a billionaire. If anything, it is her ability (and her team’s ability) to discern and manage the difference that probably leads to her gigagntic success.

So Taylor Swift can make billions – not just via streaming but especially through touring. Elton John, U2, and Bruce Springsteen, Swift’s nearest competitors in recent years, get this as well. People love to get mad and blame Ticketmaster and Live Nation – and I’m sure sure we can point to some nastiness done at that level – but, in the words of ex-Ticketmaster CEO Fred Rosen, “It’s called capitalism!” The Rolling Stones made more money in this past decade as old, old men on the road than they did in their entire, monstrously successful career beforehand. That’s celebrity and the business model that has emerged around it. It ain’t new.

So let’s bring this back to the structural. I suggest there has never been a business model that has actually worked for artists. Capitalism doesn’t give a fig about artists, or art. But it does very well in trading on celebrity, and to the extent that artists can become “blockbuster” stars, then capitalism rewards them. But let us not fool ourselves that capitalism makes a place for artists – or that it is some particular unfairness in the business practices of specific corporations (Spotify, Ticketmaster, etc.) that somehow needs to be be fixed in order to make it fair. The game is not fair, never has been.

The Value of Things

So we are left pondering the “tension between culture and commerce” as we like to say in Publishing Studies. Publishing, like Music, is an industry, a commercial practice, that trades in cultural ‘goods.’ It’s an awkward fit, and it has always been an awkward fit. What are cultural goods worth? There are a number of different ways to answer that question. One way is to look to the market to deliver a value (à la ECON 100): the dynamics of supply and demand in the marketplace will provide a commodity value for anything. This is the pain point for musicians and writers, and this is what I was trying to illuminate in my post on Cats and Commodity Pricing. What’s a book really worth? What’s a song worth? What’s a performance worth? What’s a cat worth?

Capitalism can only ever offer the commodity value: your book is worth what people will pay for it in the marketplace; your music’s worth is tied to how famous you are; your cat is, well, worthless (sorry, of course I mean priceless). Capitalism is not going to solve these fundamental aporias; it can’t be fixed or tinkered with to make it more ‘fair.’ If we actually care about culture (and cats), we have to look beyond capitalism. In Publishing, we talk about the role of “cultural capital” and its ability to stand in for, and be convertable to economical capital, but that only goes so far, as it still relies on a marketplace based on scarcity, novelty, and status.

Scale plays an important role in how we think about these things. I remember realizing this years ago when attending a poetry reading event put on by Coach House Books on a sweaty summer evening in Toronto: the event was at a pub, and it was jam-packed with people! I was amazed so many people would come out for poets! But then I realized that the assembled audience were probably a good chunk of the entire audience for Coach House’s poetry books: here were a couple of hundred people who were paying attention to the publisher, to the authors, to the verse, and who could probably be counted on to buy enough copies of the books to make their profit-and-loss statements almost make sense. That is, at the scale at which Coach House was operating – publishing A-list Canadian poets in a big city with a well-established network of people, discourse, and events – poetry publishing could make sense, as it could in similar scenes in New York, London, Berlin, and so on. Could it make sense in a smaller centre? Could it make sense on a larger scale – nationally or internationally? Probably not. There was a known sweet spot for the scale of poetry publishing, evolved over hundreds of years and well trodden, probably up until the point Rupi Kaur became an Instagram celebrity and sold more poetry books than anyone would have ever imagined. But that wasn’t a repeatable model – just celebrity strikes again.

My optimistic conclusion is that, within the overall “long tail” model of a market – the kind of thing that describes what Spotify, Netflix, Amazon do – there can be spaces where smaller publics exist, within which ideas, expressions, and artworks can circulate. Within those spaces there is room for minor celebrity, relative economic success, and the sustenance of a cultural sphere. It is fragile, though.

It’s fragile because it can be destroyed easily from a couple of different directions. First and most obvious to everyone who has ever worked in the “cultural sector,” if the network of economic resouces (including market revenue, but also grants, audiences showing up, and probably also a layer of personal wealth splashed around) gets disrupted, then the scene dies and the network fragments – and needs to be painfully rebuilt again over a long time.

But the other threat is that we mistake the big market – Spotify, Amazon, etc. – as the real cultural centre, and we begin to expect that the same rules apply for everyone. This is tempting of course because the marketing messages from the big Internet platform companies are relentlessly trying to convince us that they are the solution to all our problems. It ain’t true though… these things can’t sustain a cultural plurality, and to whatever extent they purport to serve “the long tail” the fundamental rules of capitalism will always favour blockbusters, hits, and celebrities.

So, I think, let Spotify be Spotify. It does what it does; it’s hard to imagine a different system delivering the sheer mass of stuff that it does, at least while copyright exists. But the conclusion that goes along with this is that Spotify can’t possibly be the only game in town. It has to exist alongside smaller, more niche, more local, more specialized spaces where music and audiences can circulate. Spotify itself relies on those, since Spotify certainly isn’t doing anything to nurture new artists. So it pays to keep awareness of the plurality: of artists, of systems, of scales at which things can operate. As always, we only get the systems we deserve.