Last week, musician David Byrne penned an extensive op-ed critique of streaming music providers in The Guardian. Byrne argues that subscription services like Spotify and Rdio will ruin the creative sustainability of young musicians due to their low compensation rates. He writes:
The amounts these services pay per stream is miniscule – their idea being that if enough people use the service those tiny grains of sand will pile up. Domination and ubiquity are therefore to be encouraged. We should readjust our values because in the web-based world we are told that monopoly is good for us… In the future, if artists have to rely almost exclusively on the income from these services, they’ll be out of work within a year.
Byrne cites examples of other artists who have voiced criticism of Spotify’s business practice, including Thom Yorke and Nigel Godrich, the latter of whom released another argument against streaming music on the same day Byrne’s editorial was published. In addition to quoting other musicians unhappy with Spotify’s compensation rates, Byrne provides examples of artists who have been paid pittance for their music:
The major record labels usually siphon off most of this income, and then they dribble about 15-20% of what’s left down to their artists. Indie labels are often a lot fairer – sometimes sharing the income 50/50. Damon Krukowski (Galaxie 500, Damon & Naomi) has published abysmal data on payouts from Pandora and Spotify for his song “Tugboat” and [David] Lowery even wrote a piece entitled “My Song Got Played on Pandora 1 Million Times and All I Got Was $16.89, Less Than What I Make from a Single T-shirt Sale!”
That Byrne and other artists are voicing their displeasure with Spotify is understandable given the low royalty rates that Spotify pays out. According to The Cynical Musician, artists earn a mere $0.00029 per stream through Spotify, and other subscription services pay similar rates. This seems grossly unfair and borderline ludicrous for artists like Krukowski and Lowerey, especially when Spotify’s net worth is currently estimated at around $3 billion and it pays out $500 million to the major labels for use of their catalogs.
These numbers, however, belie a much more complicated financial reality that streaming services must negotiate. The arguments against Spotify that Byrne, Godrich, and others make also extend beyond propositions for fair financial compensation and into normative assumptions on what new music businesses should encourage or provide. Accordingly, the backlash against streaming music adheres to a music production model that has different base assumptions than those of music consumers. A more comprehensive look into how the producer understanding of the market differs from the consumer understanding of the market shows areas where all parties involved – artists, consumers, labels, and streaming services – can adjust their expectations and practices for the long-term sustainability of commercial success in music.
Bryne, Yorke, and Godrich argue that more effective revenue sources need to be tapped in order to fairly compensate artists for digital music purchases. They suggest that the growing demand for streaming music requires equitable increases in how much artists are paid for providing their music on these services. Both Byrne and Godrich also tilt this argument to particularly focus on new artists who don’t have income from merchandise or touring that more established musicians might. Byrne: “What’s at stake is not so much the survival of artists like me, but that of emerging artists and those who have only a few records under their belts (such as St Vincent, my current touring partner, who is not exactly an unknown). Many musicians like her, who seem to be well established, well known and very talented, will eventually have to find employment elsewhere or change what they do to make more money.”
While Byrne’s intentions are admirable, his singular focus on new musicians leads to a muddied conception of exactly how streaming services should compensate artists. Why should an “emerging” artist without other income sources be treated differently than any other established artist with low streams or digital sales? When Byrne talks about “musicians” in his piece, shouldn’t we assume that every artist on Spotify deserves to enjoy the ability to earn a living wage from making music? He quotes the Black Keys’ Patrick Carney: “For unknown bands and smaller bands, it’s a really good thing to get yourself out there. But for a band that makes a living selling music,” streaming royalties are “not at a point yet to be feasible for us.” Is Byrne arguing that even artists who generate almost no activity or interest deserve to earn money from streaming that, at a minimum, supports basic “feasibility” or subsistence?
The implications of this question are an extrapolation from Byrne’s thesis, but they do call into question how Byrne divides budding artists and the rest of the lower-profile musicians on Spotify. Byrne, Godrich, and Carney believe that promising artists require better support to compose new, creative music, but by their own logic, all artists on Spotify should be entitled to similar levels of fair compensation.
This model, of course, is unworkable, and it’s why Spotify compensates artists based on stream counts. While it might be ideal for an innovative, lower-profile band to earn a healthy income despite only generating a slim number of streams, it’s not a sustainable business model. There are 80 million songs and tens of thousands of artists on Spotify, and the vast majority of traffic focuses on the same top-40 tracks that populate the Billboard and iTunes charts. To guarantee every single artist a basic guaranteed income (to guarantee fairness across the board) requires fixing a high base royalty rate to the lowest common denominator- say, a band that generates only 1,000 streams of their album in a year. Spotify is already operating at a quarterly loss with their current royalty rates; this increase would put the company out of business faster than the quashing of a Smiths reunion rumor.
This isn’t to say that the rate Spotify pays is fair. “Daft Punk’s song of the summer, “Get Lucky”, reached 104,760,000 Spotify streams by the end of August: the two Daft Punk guys stand to make somewhere around $13,000 each,” Byrne reports. That’s not a terrible number, but as it’s one of the top songs on Spotify, it doesn’t speak well for the vast majority of artists who don’t generate that kind of reception.
Perhaps an even better way to quantify the unfairness of Spotify’s rate is to compare the “per-stream” profit of Spotify vs. a paid music download. Most songs retail for $0.99 as individual downloads, and we might think of this retail price as the right to stream a given song in perpetuity or for life. Let’s assume a person buys a song at age 20 and will play that song an average 20 times each year for the next 60 years. That’s effectively 1200 streams, which theoretically generate revenue of $0.00083 / stream. Even if the artist only gets a fraction of that total as a royalty (the label probably takes a pretty good chunk of the $0.99 retail price), this amount is still substantially higher than Spotify’s $0.00029 / stream- and most people will only play a select few tracks this frequently over the course of their life. Based on this conception of what “purchasing music” entails, Byrne and co. are correct that Spotify’s royalty rate should be higher.
The problem is that roundly criticizing streaming services for low rates alone is not enough; a rising base royalty tide inherently raises all artistic ships, but too high of a tide washes everything away. The missing piece in Byrne’s argument is the presence of a discretionary mechanism for more effectively compensating artists who need that extra funding to support otherwise-unworkable projects. Without this mechanism, there isn’t a good way to fairly distinguish who deserves an even higher rate.
Godrich alluded to this issue in one of his earlier posts about Spotify:
Catalogue and new music cannot be lumped in together. The model massively favours the larger companies with big catalogues. They need the new artists to be on the system to guarantee new subscribers and lock down the “new landscape.” This is how they figure they’ll make money in the future. But the model pays pittance to the new artist right now. An inconvenient fact which will keep coming up. I feel a responsibility to speak up when I see something going on which I think is unfair. I’m not bitching about not getting paid. It’s about standing up for other artists’ rights. It’s up to streaming providers to come back with a better way of supporting new music producers. It’s not for us to think up how it could work. That’s your department.
In addition to the aforementioned issue of qualifying how to meter higher royalty rates, there are two problems with the argument Byrne and Godrich are making here. First: why are they placing agency on Spotify for low royalty rates and not music labels for trying to negotiate better deals? Godrich: “The big labels did secret deals with Spotify and the like in return for favourable royalty rates. The massive amount of catalogue being streamed guarantees that they get the big massive slice of the pie (that $500 million [which Spotify pays out to labels]) and the smaller producers and labels get pittance for their comparatively few streams.” This isn’t a just a problem on Spotify’s end; it sounds like oligopolistic behavior by major labels has resulted in a failure to consider their clients’ best interests (or, conversely, it’s creating crowding-out effects that give less leverage to smaller independent labels). Byrne and Godrich should continue to advocate for higher rates from Spotify, but to an equal or even greater degree, the target of their discontent should be labels for mortgaging artistic development for pure profit.
The other, more underlying problem with Godrich and Byrne’s arguments is their fundamentally different conception of individual artists’ value versus that of most music consumers. Byrne and Godrich recognize the unique creative potential of each individual artist and advocate financial support for these artists to encourage their development. In contrast, I hypothesize that most music consumers, influenced by the ongoing devaluation of digital music, view artists as producers of interchangeable products who have minimal unique value.
Let’s break this hypothesis down by first looking at the consumption trends of digital music purveyors. The Top 100 tracks on iTunes, Spotify, and other online music providers generally reflect the same broad consumption patterns, in that the same songs (usually pop radio hits) account for the highest of sales and traffic. This suggests the majority of music consumers use streaming music providers to primarily access the songs of the moment. Beyond hot tracks, consumers then find value in the back-catalog that contains their favorite artists- a small group of bands or singers that they follow closely and would support beyond casual streaming (going to concerts, buying albums, etc). The remaining value of streaming services comes from the long tail, or the extensive back-catalog that allows listeners to quickly and easily access a huge bank of tracks that fit a specific moment or immediate need. Soundtracks for parties, seeking out new artists, and other “one-off” listens fall into this category.
Herein lies a substantial difference from Byrne’s arguments. “A culture of blockbusters is sad, and ultimately it’s bad for business,” he says, but based on the listening patterns exhibited on major music services, this culture is currently the main driver of industry profit, with “new” artists often acting as simply a bonus for a sizable number of people.
In theory, many music listeners would agree Bryne’s assertion that “a culture of blockbusters is sad.” The problem, however, is that it’s not a culture of blockbusters that is crowding out new musicians, but a larger-than-ever supply of these new musicians. In Byrne’s excellent book How Music Works, he describes a number of changes in how record companies have operated since the mid-twentieth century. Two of those changes stand out in terms of how artists now make music:
1) Recording costs currently approach zero for many projects. “Now an album can be made on the same laptop you use to check email,” Byrne writes.
2) Manufacturing and distribution costs are also approaching zero. “Digital distribution is pretty close to being free. Digitally, it’s no more expensive to distribute a million copies than a hundred,” he says.
If there are minimal costs to both create music and disseminate it to potential listeners, basic economics dictates that the supply of music is going to increase since more people now have the ability to make and share it. And supply has increased, greased by improving technological standards that allow people to download and stream music in almost no time at all. We may be living in a “culture of blockbusters” as Byrne asserts, but we’re simultaneously living in an era of unprecedented musical flourishing spurred by the collapse of a hierarchical structure for getting music to the public. Anyone can make music and earn a following with greater ease than ever before.
Ironically, though, the increasingly horizontal nature of music creation is spurring the deleterious effects that Byrne discusses in his op-ed. When the supply of artists increases, I argue that consumers begin to place less value in a given individual artist because there are so many alternate songs or albums that can act as effective supplements. This problem is exacerbated by a service like Spotify, which people use primarily to find hits and see the long-tail artist catalog as a cherry on top. If there is a fixed percentage of long-tail streams (versus hot 100 or radio hit streams), and more artists continue to enter this fixed pool, the average individual attention for each artist is going to drop. More artists results in a devaluation of individual musicians since there’s simply more potential music to choose from.
The technology that has allowed more people to create music is also shaping how we conceive of the value of music and individual songs. That music is now so easily accessible on YouTube and via piracy suggests it is being assigned an inherently lower value than other forms of art. I believe that our modern conception of media value is increasingly a positive function of two factors: file size and length. The larger an object’s file size and the longer its length, the higher the price we’re willing to pay for it (notwithstanding the actual unit costs that are associated with each object). I’d maintain this is partly why people are willing to pay $50-60 for video games (large file size, 40+ hours) and $15 for films (medium file size, 2-3 hours). This isn’t to say every piece of media is inherently valued at these levels across the board, but it suggests a trend for what people expect a given media product should be worth.
Not surprisingly, music comes off pretty poorly in this equation. Low file sizes make it the easiest form of media to pirate and upload to YouTube, and short track lengths (usually 3-4 minutes) suggest a commodity that is exchangeable and transient. This is especially the case when there is no corresponding physical packaging.
The increase in music supply further affects the purchase of full albums beyond individual songs. The greater the quantity of available music to listen to, the less likely it is that a consumer will go all-in and splurge on only one album, especially if he/she is on a limited budget. The purchase of an album reflects a greater willingness to put more time into the music. It’s a deeper investment in the artist’s work. For most people, though, there is very little reason to pick up a random album on a whim. Even a $7 sticker price is asking too much for an unknown value. We put money into things that last longer because we have reason to think this length provides justifiable return. Most people only buy a select few games or watch a limited number of movies because they pick and choose based on subject material that’s likely to appeal to them. The other element is that the supply of available movies and games is much more limited than music today. Byrne’s rationale for cheaper music production extends to why people don’t value individual albums as much – more artists are around to make more music. When so many supplementary items are available, it’s unlikely a consumer will put his financial and temporal eggs in one basket to purchase a single album when others could fill a similar need.
All of this is to say that a fundamental shift has already occurred in how people consume music, and there’s no going back to the status quo of twenty or even ten years ago. We’re at a nexus where musicians fear they won’t be able to make sufficient profit to sustain their work and where consumers are placing less value on individual units of music, be they songs or albums. What can be done to have these trends coexist?
I’m no expert in music making or the music business, and many people smarter than me are trying to tackle this question. But a few potential avenues stand out:
1) People are more risk-averse when buying full albums due to the high cost associated with them. Record prices need to be lowered or additional physical content needs to be added to CD sets to further entice buyers to move beyond digital streaming. LPs that throw in extra photos and a free digital download of the record are a step in the right direction, for example.
2) Streaming rates need to be higher, as discussed above. Though services like Spotify are operating at a loss, discontent will continue to mount if rates are not increased. Artists should try to negotiate more equitable deals with record labels, too. Label revenue per stream on Spotify is $0.0016, according to The Cynical Musician, which is nearly five times as much as the average artist rate.
3) Creative marketing is imperative; bands are now, more than ever, responsible for their own financial destiny. Godrich’s admonishment to record labels of how finding new profit avenues is “your department” for finding a solution to low streaming rates doesn’t hold any water at all; this isn’t some problem that artists can offload and expect to be solved. While labels might bear the brunt of trying to find more profitable avenues, artists now exert greater control in getting their music to people in new ways.
I would imagine that Byrne would be pessimistic about these three suggestions, and I don’t think he’d be incorrect- they’re not enough to spark a sea change in how musicians are compensated. “Musicians might, for now, challenge the major labels and get a fairer deal than 15% of a pittance, but it seems to me that the whole model is unsustainable as a means of supporting creative work of any kind. Not just music,” he says.
But Byrne’s pessimism about the future of artistic innovation in music and other spheres of culture seems, on the whole, too pervasive. He suggests in How Music Works that many, if not most, musicians would make music without any incentive for profit because it’s something they love to do. The ubiquity of great music today suggests that even if making music becomes increasingly less of a career-sustaining activity, the impulse to create something completely new will still exist. And the technology is there to get people to hear it. Perhaps record labels will become an institution that, in addition to marketing the biggest pop starts, fund projects that require a prohibitive up-front investment that couldn’t be accomplished on a DIY level.
So yes, streaming music is having negative effects for a number of musicians, and that needs to change. But the problem isn’t so cut-and-dry as to say that providers like Spotify are the problem and a culture of hits is an inevitable outcome. The underlying changes in how people create and value music need to be more thoroughly considered in order to find new best practices to solve this problem. Rectifying these differing consumer and producer views of the market is the first step to a more nuanced understanding of digital music consumption and finding ways to equitably satisfy artists, consumers, labels, and providers.