Creative Industries Journal

ISSN: 1751-0694 (Print) 1751-0708 (Online) Journal homepage: https://www.tandfonline.com/loi/rcij20

‘Let's keep music special. F—Spotify’: on-demand

streaming and the controversy over artist royalties

Lee Marshall

To cite this article: Lee Marshall (2015) ‘Let's keep music special. F—Spotify’: on-demand
streaming and the controversy over artist royalties, Creative Industries Journal, 8:2, 177-189, DOI:
10.1080/17510694.2015.1096618

To link to this article: https://doi.org/10.1080/17510694.2015.1096618

Published online: 29 Oct 2015.

Submit your article to this journal

Article views: 13974

View related articles

View Crossmark data

Citing articles: 18 View citing articles

Full Terms & Conditions of access and use can be found at

https://www.tandfonline.com/action/journalInformation?journalCode=rcij20
Creative Industries Journal, 2015
Vol. 8, No. 2, 177 189, http://dx.doi.org/10.1080/17510694.2015.1096618

‘Let’s keep music special. F—Spotify’: on-demand streaming and the

controversy over artist royalties
Lee Marshall*

School of Sociology, Politics, and International Studies, University of Bristol, Senate House,
Tyndall Avenue, Bristol, BS8 1TY, UK

On-demand streaming music services have expanded significantly in recent years.

Services such as Spotify and Deezer are widely expected to become the dominant
means of mass music consumption in the future and have been significant factors in
helping recording industries in some countries arrest their long-term decline. The
emergence of such services (especially the largest of them, Spotify) has caused
controversy, however, with both major stars and smaller independent artists publicly
criticising the low levels of royalty payments such services have thus far generated.
This paper provides an overview of the controversy, concentrating on the criticisms of
Spotify from independent labels and musicians. In such a rapidly moving sector, it is
unwise to offer any definitive predictions and thus the purpose of this paper is to
outline as clearly as possible the parameters of the current debate. To achieve this, it
considers the business models associated with streaming music and their place within
the contemporary recorded music ecosystem, most notably in relation to piracy and
declining record purchases. It concludes with a consideration of streaming services in
relation to the broader power dynamics of the recording industry suggesting that while
services like Spotify may cause problems for independent musicians they rely upon a
logic of scaling that correlates with the existing practices of the major record labels.

Introduction
One thing that the history of popular music tells us is that the introduction of significant
new technologies is rarely uncontroversial. New inventions, such as the microphone
(Frith 1986) or the sampler (Theberge 2004) get challenged and ridiculed by parties
whose interests may be threatened by new social practices of music-making and music-
consuming. A recent example of this kind of conflict is the controversy over the amount
of payment being given to artists for allowing their music to be made available on stream-
ing services such as Deezer and, especially, Spotify. It is a controversy that periodically
appears in the mainstream media, such as Radiohead singer Thom Yorke’s proclamation
that Spotify is the ‘last gasp of the old industry… the last desperate fart of a dying corpse’
(Dredge 2013) or Taylor Swift’s decision to remove all of her back catalogue from the
service upon the release of her 2014 album, 1989 (Dickey 2014).
The controversy is not merely a few superstars grumbling, however, as the majority of
concerns have actually been raised by independent artists several rungs down the music
industry ladder. While such discontent could perhaps be assumed to be merely an eco-
nomic spat about how the pie should be divided, it actually reflects more deep-seated
uncertainties regarding the changing nature of the music commodity and its impact upon
the perceived value of music, as well as more long-standing issues regarding the

*Email: [email protected]

Ó 2015 Lee Marshall

178 Lee Marshall

dominant power relations within the music industry. This latter point is notable because it
contradicts a common perception that the disintermediating nature of the internet should
be a boon to musicians beyond the major label system (for example, Baym 2011;
Wikstrom 2009), with the emergence of ‘long tail’ economics (Anderson 2007) enabling
unsigned artists to sell their music to a wide audience. However, the seemingly inexorable
shift to streaming services has the potential to be a way through which old industry
structures reassert themselves. Yorke is thus correct to associate Spotify with the ‘old
industry’ but perhaps premature when describing it as a ‘dying corpse’.

An introduction to streaming services

Before outlining details of the controversy, it is important to differentiate streaming music
services. Often generically referred to as ‘music in the cloud’ (Morris 2011), streaming
services can be split into three types. The first is streaming radio (sometimes referred to
as webcasting). This can simply be an online version of existing radio services, or online-
only radio stations that specialise in particular genres or localities, but the more distinc-
tive form is personalised, in that the service selects songs to play for each individual lis-
tener based on previous listening preferences. This kind of streaming radio service thus
provides a more bespoke experience than broadcast radio but, in comparison to other
kinds of streaming service, the important characteristic is that it is non-interactive: the lis-
tener does not choose what songs will play. The most established streaming radio service
is Pandora, which announced in April 2013 that it had 200 million registered users in the
United States, 70 million of which were ‘active’ each month (Peoples 2013).1 The major-
ity of Pandora’s income comes from advertising but it also has 2.5 million subscribers
paying a monthly fee to avoid ads and listening limits. Another prominent player in this
field is iHeartRadio, a mixture of traditional radio services and Pandora-like tailored rec-
ommendations, created by terrestrial radio provider Clear Channel,2 and boasting 50 mil-
lion users (Solsman 2014).
The second kind of streaming music service are locker services, such as iTunes match
and Amazon’s Cloudplayer. In these services, users can listen to ‘their’ music directly
from the internet, either songs that they upload directly to the service’s servers or that are
automatically ‘matched’ (to, for example, a user’s Amazon CD purchases). Locker serv-
ices thus offer mobile access to a listener’s entire MP3 collection. However, they are a
relatively small part of the cloud-based music landscape and, in all likelihood, a tempo-
rary phenomenon, designed to ease consumer experience until the ‘purer’ forms of cloud-
based services are adopted more widely.
The third kind of streaming music services are ‘on-demand’. Such services are similar
to streaming radio services, such as Pandora, with one crucial difference: the listener
chooses what music is played. They are thus interactive rather than non-interactive. As
such, they provide a listening experience similar to locker services, except that choice is
not limited to an individual’s collection: instead the listener can choose from a vast cata-
logue of music (upwards of 30 million songs on some services). Spotify has established
itself as the most famous of these services but there are other significant providers. The
most important of them is Deezer, the second largest on-demand service with 16 million
active users and 6 million subscribers (Peoples 2014). It has garnered fewer headlines
than Spotify because it has strategically avoided entering the mainstream US market but
is the most global service, available in 180 countries. Other services include Rhapsody
(until the emergence of Spotify, the largest subscription service in the United States, with
2 million subscribers), US-based Rdio and WiMP, based in Norway.3
 Creative Industries Journal 179

On-demand services often include radio functions so there is some overlap between
services but the most important distinction is the one between non-interactive and interac-
tive services. Within the music industry, it is widely accepted that on-demand services are
heralding a significant reconfiguration of recorded music consumption, from being based
on ownership of music to being based on access, renting rather than buying records. This
acceptance reflects the manner in which streaming music has begun to enter the main-
stream in recent years. This can be seen in the global expansion and accelerating growth in
user/subscriber figures among the aforementioned services but, probably more significantly,
in the way in which the major internet companies are now becoming involved in the on-
demand streaming market. For example, Amazon offers a limited streaming service as a
benefit to subscribers of its Amazon Prime service, while Google actually provides two
streaming subscription services: Google Play Music All Access and YouTube’s Music Key
service.4 On top of this, Apple, a long-time rejecter of streaming music, purchased Beats
Music (primarily a headphone manufacturer but with a small on-demand streaming music
service) in May 2014 and launched its own on-demand streaming service in June 2015.

The backlash against Spotify

The broader issues discussed in this paper, concerning the changing nature of the
musical commodity and the economic reward for recorded music, are applicable to all
of the on-demand streaming services. This paper will focus on Spotify, however, given
that it is the most prominent of such services and the one which has been the centre of
public debate about the levels of remuneration received by artists for streaming.
The service was launched from its native Sweden in 2008 and it currently offers a cat-
alogue of ‘over 30 million tracks’ in 58 countries (Spotify 2014). In January 2015,
Spotify announced that the service had achieved 60 million active users and 15 million
paying subscribers (Spotify 2015). Its growth in subscriber numbers has been impres-
sive: according to public announcements, it took almost 2 years to get its first 500,000
subscribers but then added 2 million in approximately a year, a further 3.5 million in
the following 18 months, 4 million more in the next 14 months and 5 million more in
just 7 months. It is growing exponentially and is the poster boy of the new, optimistic,
digital record industry the one that has stopped bleating about piracy and is doing
something new and funky. Yet despite, or perhaps because of, its success, Spotify has
also received a large amount of criticism, most notably from artists about the levels of
remuneration they receive from the service.
The public discussion of Spotify payments first began in 2009, when it was widely
reported that Lady Gaga had received a payment of just $167 for her song ‘Poker Face’,
which had been streamed over a million times (for example, Smith 2009).5 The statistic was
misleading6 but the mud stuck it is easy to find recent stories repeating the bogus fact (for
example, Marsden 2012) and the idea that Spotify did not adequately reward artists
gained public credence. It was around the time that Spotify launched in the United States in
2011 that the controversy surrounding payments became more consistent as a number of
smaller US record labels withdrew their catalogues from the service. Seemingly, the first to
do so (though not the first to announce it) was Projekt Records, a Brooklyn-based gothic
rock label, formed in 1983, which withdrew its catalogue from all on-demand streaming
services in July. Label founder Sam Rosenthal stated that these services ‘are not a viable
way forward for the music industry’ and was vocal in his criticism of Spotify:
In the world I want to live in, I envision artists fairly compensated for their creations…
Spotify is NOT a service that does this. Projekt will not be part of this unprincipled concept.
180 Lee Marshall

For a stream on Spotify… on average $0.0013 is paid to Projekt’s Digital Distributor. 5000
plays generates around $6.50. In comparison, 5000 track downloads at iTunes generates
$3487. … To earn the U.S. monthly minimum wage $1160 892,307 plays a month are
needed at Spotify. This is not a viable number for artists. (Rosenthal 2011)

The next label to pull its catalogue out of Spotify, and the first to be publicly reported,
was Century Media, a metal label based in Dortmund and California. In August, they
announced the withdrawal of its repertoire, along with its associated labels, ‘to protect
the interests of its artists’. While praising Spotify as ‘a great tool to discover new music’,
Century countered that:

Physical sales are dropping drastically in all countries where Spotify is active… . Since the
artists need to sell their music to continue their creativity, Spotify is a problem for them. This
is about survival, nothing less and it is time that fans and consumers realize that for artists it
is essential to sell music to keep their heads above water.

Century concluded that:

Spotify in its present shape and form isn’t the way forward . … Ultimately, in some cases, it
will completely kill a lot of smaller bands that are already struggling to make ends meet.
(Century Media 2011)

More label withdrawals followed. Curiously, metal labels, especially LA-based

metal labels, seemed the most keen to remove their catalogues from Spotify. In Sep-
tember 2011, Napalm Records withdrew their repertoire, stating that ‘the income
generated through streaming is so insignificant that neither we, as a record company,
nor our artists are able to further support this method of distribution. We… object to
price models that endanger our existence and that of our bands’ (see Metal Under-
ground 2011). Also withdrawing in September 2011 were Prosthetic Records and
Metal Blade, while Sumerian Records withdrew most of its catalogue in November.
Beyond the LA metal scene, a more widely publicised story emerged in November
when ST Holdings, a UK digital distributor for independent electronic and dance
labels, withdrew from on-demand streaming services the catalogues for 234 of the
238 labels it distributed. In announcing their withdrawal, the company highlighted
that, in the third quarter of 2011, their digital revenues dropped for the first time in
their history (by 14%) and that iTunes income in particular was badly affected, fall-
ing by 24%. During the same period, on-demand streaming services accounted for
82% of ‘plays’ of their music, but just 2.6% of revenues, with Spotify returning
£2500 for 750,000 streams. This low rate led the company to voice concern that the
value of music was being undermined, stating that ‘music loses its specialness by its
exploitation as a low value/free commodity. Quoting one of our labels, “Let’s keep
the music special, fuck Spotify’” (in Resnikoff 2011).

The royalties debate and the changing nature of the musical commodity
It is important to keep these examples in context. The withdrawal of a handful of small
independent labels is not going to cause too much of a dent in a catalogue of 30 million
tracks.7 It is also worth noting that some of these companies (such as Century) have sub-
sequently returned their catalogues to Spotify. However, the controversy about the
amount of money paid to artists by streaming services has continued, especially among
 Creative Industries Journal 181

some smaller independents and DIY artists.8 Broadly speaking, the complaints can be
characterised in two ways. First, musicians argue that on-demand streaming services
undermine sales of digital files and physical media and, therefore, they are not a financial
model that can sustain their musical careers. Underpinning this, however, is a broader
aesthetic and moral argument about what music should be worth, with the micro-pay-
ments generated by each stream being seen as devaluing music itself.
I will return to these issues shortly but, first, I want to outline the two ways in which
Spotify has responded to such criticisms. The first is to argue that, for many artists, the
amount that they receive from Spotify reflects their contract with their record label and
not the amount that Spotify pays out (e.g., in Rosenthal 2011).9 There is some truth to
this and it is important to recognise that the question ‘how much does Spotify pay artists?’
is somewhat misleading given that Spotify does not ‘pay artists’ any more than a physical
record store ‘pays artists’. Instead, Spotify pays labels, or distributors handling labels
and, as such, what proportion of that money is returned to artists is a matter of individual
(though often standardised see Stahl 2013, 165 170) contracts. It is necessary to be
cautious, therefore, of information circulated online about the amount a particular artist
received from streaming. However, there is obfuscation in Spotify’s response also.10 Not
all of the artists raising concerns have record label contracts, with many instead using dig-
ital distributors that charge a flat rate commission which makes it possible to ascertain
what Spotify actually paid out. A large number of these independent artists have made
significant efforts to provide reliable data on streaming rates (for example, the cellist Zoe
Keating and the aforementioned David Lowery, who maintains a ‘streaming price bible’
on his Trichordist blog). Furthermore, the fact that independent labels, and not just indi-
vidual artists, have been complaining about the revenue received from on-demand
streaming suggests that the low level of income received by artists is not just a matter of
exploitative contracts.
Aside from claiming contractual confidentiality, Spotify’s main public response to the
issue of royalties has been to provide some really big figures the global amount that it
has returned to rights holders. In 2012, Spotify founder, Daniel Ek, stated in an interview
that ‘the whole debate about artists payments is slightly misconstrued. We have paid out
over half a billion dollars to rights holders now, and that’s doubled in the last nine months.
That’s a significant amount of money’ (in Dredge 2012).11 The word ‘misconstrued’ is
quite deliberate because the argument being made is that artists and the media are wrong to
treat on-demand streaming services as comparable to sales. Spotify has repeatedly argued
that it ‘does not sell streams, but access to music. … [It] is not a unit based business and it
does not make sense to look at revenues from Spotify from a per stream or other music
unit-based point of view. Instead, one must look at the overall revenues that Spotify is gen-
erating’ (in Houghton 2011). Thus, the really really big or the really really small numbers
get to the heart of those different ways of understanding streaming services.
To be clear, this is not merely an argument that it is impossible to equate a certain
number of streams with one purchase. Rather, it is an argument that the nature of what is
being sold is qualitatively different than in the past and, therefore, artists need to recon-
sider how they are rewarded not merely how much they are paid but the basis on which
they are paid: in the ‘old’ model of music-buying/owning, consumers paid (and artists
therefore received) a one-off payment; in the new model of music-renting, consumers
continually pay for access to music. The theory is that continual small payments will
compensate (and more) for the withdrawal of the one-off payment. That is why Spotify
argues that ‘people need to transition from unit-based thinking to consumption-based
thinking’ (in King 2012).
182 Lee Marshall

The economics of streaming music

While Spotify’s position reflects a broader cultural shift toward ‘renting’ music, its argu-
ment that musicians should move away from ‘unit-based thinking’ is seemingly under-
mined by the fact that musicians are paid on the basis of how many times their songs are
streamed by the service. In fact, Spotify royalty statements list each stream and its associ-
ated rate (though Spotify refers to calculating per-stream rates as ‘reverse engineering’).
Understanding Spotify’s position that it is wrong to focus on individual streams requires
an understanding of the broader economic logic of streaming, which depends upon the
totality of streaming rather than the aggregation of individual streams.
Before explaining Spotify’s payment system, it is important to recognise that Spotify
does not seem to be ‘cheating’ by paying a lower royalty rate than other digital suppliers.
At SXSW in 2012, Spotify’s chief content officer, Ken Parks, reported that the company
pays out ‘in the neighbourhood of 65 70% of revenue to rights holders’ (in Peoples
2012a) and it has repeatedly affirmed that it pays ‘nearing 70%’ or ‘approximately 70%’.
The 70% figure is important because it is what Apple returns to rights holders for iTunes
downloads and it broadly matches historically conventional allocation in the record
industry.12 So while not confirmed precisely, Spotify is at least paying rights holders a
broadly conventional rate for the right to use music. It distributes money by dividing its
total revenue by the number of streams played on the service. However, the amount of
revenue Spotify generates relies mainly upon the number of subscribers it has (plus a little
extra from advertising), whereas the number of times music can be streamed is effectively
limitless. Thus, when this amount paid is compared to the overall number of songs
streamed, the numbers look embarrassingly small – the broad average from the informa-
tion posted online is about $0.004 per stream (0.4 or 0.5 cents per stream, although
streams get different rates of return according to a variety of factors, most notably
whether it is played by a subscriber or someone using the free service). Given the tiny fig-
ures quoted from musicians on per-stream basis, or the ‘30,000 streams only earned me
$120’ stories, it is understandable that musicians might be concerned about the amount of
money they are receiving. How can Spotify suggest that streaming is a viable option for
artists?
Spotify’s answer to the question is scale. Given the company pays a fixed proportion
of revenue back to rights holders then, as Spotify’s overall revenue increases so, too, does
the amount of money returned to rights holders (70% of a billion dollars is more than
70% of 10 million dollars). This means that scale is key to the viability of Spotify and
similar services: the more free users it can convert into paying subscribers the greater the
level of return to artists and labels, not because more people are playing their music but
because there is a bigger pie to share out and the effective per-stream rate thus increases.
This can actually be seen in the royalties data that has been shared online: whereas the
average rate per stream was initially in the region of $0.002 to $0.003 per stream (as can
be seen in Sam Rosenthal’s statement quoted earlier), today it is $0.004 to $0.005. Ek
announced in November 2014 that ‘500 thousand listens on Spotify would pay out
between three and four thousand dollars’ which is an effective per-stream rate of $0.006
to $0.008 (Ek 2014). The logic of the Spotify model has thus shown signs of working,
though it remains to be seen whether services can scale sufficiently to provide a viable
means of income for musicians. According to Spotify’s artist-in-residence, D.A. Wallach,
that particular tipping point is achievable, stating that ‘if we can get to the scale of Netflix
which has 20 million subscribers we estimate we’d be paying out to artists what
iTunes is paying out on a year to year basis’ (King 2012).13
 Creative Industries Journal 183

Artists are thus being told that they are wrong to treat streams like sales and that they
should adopt a patient, long run view. Yet while large labels with cash reserves can poten-
tially ride out the storm, artists who need money to live on now argue that streaming is
undermining their current income by cutting into their digital sales. Both Century Media
and ST Holdings made explicit reference to the fact that income from digital sales had
fallen dramatically since the emergence of on-demand streaming. Spotify’s position, per-
haps unsurprisingly, is that streaming does not have a detrimental effect on digital sales.
For example, Katie Schlosser, a director of label relations at Spotify, has stated that they
have been able to provide labels with ‘data that’s proving and demonstrating the fact that
streaming revenue is additional to actual unit download consumption or physical music
sales’ (Resnikoff 2013). Perhaps more surprisingly, the argument has been supported by
the major record labels. For example, at an IFPI event in London in 2012, then president
of Universal Music Group’s global digital business, Rob Wells, described the argument
that on-demand streaming ‘cannibalises’ sales as ‘absolutely bogus’ (Peoples 2012b).
The year 2012 was a good year to claim that streaming does not undermine sales, as
overall revenue for the recording industry increased for the first time this century, if only
by 0.2% (IFPI 2013). There were also individual instances where streaming and selling
seem to have complemented each other well. For example, one of the biggest albums of
2012 Mumford and Sons’ Babel became the most streamed album on Spotify in its
first week of release (‘around eight million listens’) while simultaneously recording the
highest first week sales for a US album that year (600,000 units) (Peoples 2012b). Fur-
thermore, there are specific territories where streaming has played a significant role in the
local recording industry returning to growth after more than a decade of decline. For
example, in Norway (where streaming accounted for 65.3% of recorded music revenues
in 2013) the industry grew by 7.2% in 2012 and 10.6% in 2013 (IFPI Norge 2014) while
in Sweden (where streaming contributed 71.2% of recorded music revenues in 2013) the
recording industry has seen 3 years of consecutive growth (IFPI Sverige 2014).
Given their distinctive characteristics (small markets with high broadband penetration
and the home of two of the major streaming services), it is unclear whether these
Scandinavian examples are reliable indicators of future trends in much bigger music
markets. Most notably, neither country had well-established paid download services prior
to streaming services emerging (Mulligan 2014b). This makes them very different from
the United States and the United Kingdom, where download services were contributing
between 40% and 50% of recorded music income by the early 2010s. And, revenues from
download sales declined in the United States for the first time in 2013 (Pham 2014) and
the United Kingdom in 2014 (Titcomb 2015). Critics of streaming services were quick to
attribute the blame to Spotify (for example, Titcomb’s headline was ‘UK music down-
loads in decline for first time as Spotify eats into iTunes sales’), though Ek (2014) con-
tends that to do so is to confuse correlation with causation. Instead, he points to Canada,
where download sales were falling before streaming services emerged. It should also be
noted that growth in download sales had been flattening out considerably since 2008
(which Mulligan (2014c) attributes to increased spending on apps).
In the long run, it seems reasonable to assume that the changing nature of the musical
commodity means that streaming will replace downloading but it is difficult to make pre-
dictions about the speed and the nature of the transition as current data is ambiguous.
Nonetheless, ‘while industry executives initially refused to attribute the early signs of
weakness in digital sales to the rise in streaming, in the second half of 2013 many con-
ceded that ad-supported and paid subscription services did seem to be cannibalizing digi-
tal sales’ (Christman 2014).
184 Lee Marshall

Spotify and the mainstream industry

Despite concerns that streaming music could be undermining digital sales, the major
record labels (Universal Music Group, Sony Music Entertainment and Warner Music
Group) have been very supportive of both Spotify and on-demand streaming more
broadly. Many of the bigger independent labels, such as Beggars Group and XL, have
also been supportive of the service. The majors’ support of Spotify is in stark contrast to
the early years of the twenty-first century, during which they were hostile to the emer-
gence of digital music and ultimately had to be bullied by Apple into committing them-
selves to the digital marketplace (Knopper 2009, 170–177). What has changed in the
meantime to effect such differing responses?14
Understanding the change in attitude requires understanding the changing contours of
piracy in the last 15 years or so. In the early twenty-first century, the record industry’s
overwhelming concern was that digital files generated massive amounts of piracy which
undermined legitimate sales. Initial responses to these developments involved searching
for ways in which music could be ‘locked’ through Digital Rights Management systems
and aggressive litigation campaigns, against both companies involved in piracy (such as
Napster) and individual downloaders (Marshall 2012). Around 2005, however, the record
industry changed its approach to online music consumption. While still publicly criticis-
ing piracy and campaigning for improved copyright enforcement, the record industry
apparently accepted that it would have to adapt to consumer behaviour. The challenge for
the labels was finding a way to incorporate the behaviour of a generation which had
grown up with music being free and accessible. This has been Spotify’s key selling-point:
it has consistently presented itself as offering a platform which ‘monetises’ existing con-
sumer behaviour. In its response to the withdrawal of the Century and Mute catalogues,
Spotify stated that the service:

was launched out of a desire to develop a better, more convenient and legal alternative to
music piracy. Spotify now monetises an audience the large majority of whom were down-
loading illegally (and therefore not making any money for the industry) before Spotify was
available. (in Bruno 2011)

The connection between Spotify and piracy is less than coincidental: Spotify founder
Daniel Ek was formerly CEO of uTorrent, the most popular client for bittorrent sharing
and the Spotify software itself depends on the peer-to-peer technologies that facilitate
widespread file-sharing (when a user requests a song, encrypted data is transferred from
other users’ computers) (Greeley 2011). Despite this heritage, however, a major selling
point of on-demand streaming services is that they are generating income for rights hold-
ers out of consumer behaviour that had previously generated none.15
However, surely the main reason for the major labels’ support of Spotify and similar
services is that the ‘consumption-based’ logic of streaming services actually fits the logic
of the pre-digital record industry quite well. The success of the major labels has always
depended upon ‘consumption-based’ rather than ‘unit-based’ profits; their strategy was to
release lots of records in the knowledge that only a small number of them would be suc-
cessful but that the rewards from a small number of hits would outweigh the losses of the
remaining releases. Thus, from the major label’s perspective, so long as some records
made a profit, it did not matter which records sold a lot and which did not. It mattered to
the careers of specific individuals (the artist, say, or the A&R executive that signed
them), but not to the label overall, whose success depended upon overall levels of music
consumption.
 Creative Industries Journal 185

The correlation between the ‘consumption-based logic’ of the major labels and
streaming services can be seen in the way that royalty payments are calculated, which
is inherently beneficial to those with large catalogues. Spotify’s payments are based on
a ‘market share’ system. Simplifying slightly, the basic logic of the system is that an
artist is paid a percentage of Spotify’s overall revenue determined by the percentage
of overall streams that the artist accounted for in a particular month (Spotify 2013).16
There is thus no connection between an individual listener’s payment and their listen-
ing preferences. This is beneficial to artists and labels who receive a high number of
streams because it means that they receive a high proportion of income regardless of
how many paying users actually streamed their music. For example, if a listener pays
£10 but listens to only one obscure Finnish rapper for the entire month, the rapper will
still only receive a very small proportion of the listener’s money as the majority of it
will be distributed to those artists at the top of the Spotify charts. Independent and less
popular artists would benefit from an alternative system of payment that directly dis-
tributed an individual’s subscription to only those artists to which the individual had
actually listened.

Conclusion
The specific complaints about royalty rates raised by independent musicians and labels
reflect, after a period of uncertainty, some significant structural recalibrating of the
recording industry. Rather than internet technologies providing liberation from old indus-
try dynamics, what we may be seeing is a consolidation of long-established power struc-
tures (as predicted by Bukart and McCourt (2004) though not via the precise mechanisms
they suggested; see also Rogers 2013). Furthermore, the introduction of streaming may
be making the ground outside of the mainstream even more uncertain given that indepen-
dent artists and labels do not have the luxury of extensive back catalogues and are much
less insulated against declining sales. If streaming is ‘cannibalising’ sales, therefore, it
will be disproportionately impacting upon smaller players. Artists signed to bigger labels
can be insulated against the changing rhythms of payment for recorded music if, say, it
takes 5 years of streaming income for an album to recoup its costs rather than 2 years of
sales income, those with larger catalogues/financial reserves will be able to weather the
storm. However, even if streaming services can scale sufficiently to offset the global
decline in incomes from record sales, Spotify’s current payment system seems skewed
towards the major labels.
Thus, despite the seemingly radical shift from ownership to access-based models, the
business model of on-demand streaming is one with which the major labels are familiar
and which suits their existing strategies. It is unsurprising that they have been supportive
of the new services. Indeed, this support has gone beyond the realm of public statement
as the major labels are actually shareholders in Spotify: financial filings reported in 2009
state that the major labels have a 16.5% in Spotify, with Merlin, an aggregator for the big-
gest independent labels, owning a further 1% (Arrington 2009; Jerrang 2009).17 Further-
more, in late 2012, Access Industries, owners of Warner Music Group, invested
$130 million dollars in Deezer for an undisclosed stake. This, again, is a repeat of earlier
industry structures, in which each of the majors owned their own CD distribution compa-
nies (Negus 1999, 55–60). Needless to say, the fact that the labels have a stake in Spotify
has merely added to the suspicion and distrust towards the company and its royalty sys-
tem. The lack of transparency in the entire system has been something that Spotify has
tried to address (through its Spotify Artists portal), but they remain limited in information
186 Lee Marshall

they can release and streaming royalty statements remain incomprehensible to even the
most seasoned analysts.
From this perspective, the concerns raised by artists and independent labels about
royalty payments from Spotify seem very similar to arguments made about the record
industry in the past and, rather than being a ‘disruptive technology’ (Barr 2013), Spo-
tify and other streaming music services may actually (or also) be quite conservative
technologies. There are historical precedents here. In Pop Song Piracy, Barry Kernfeld
argues that the history of copyright infringement in the music industry follows a repet-
itive pattern: new technologies or social practices emerge which undermine music
rights holders’ established forms of profit-making, who respond by yelling ‘piracy!’.
The public recognises genuinely transformative new practices from blatant copying,
however, and, in such cases, rights-holders end up having to adapt to the new practices
yet, in doing so, find ways of incorporating them into their existing profit-making prac-
tices (2011, 2 4).
Though the story is not complete, it is relatively easy to fit the story of digital music
into such a narrative, with on-demand streaming services being the chapter in which the
powerful corporations begin to incorporate the disruptive practices and reassert domi-
nance. The recorded music landscape in the streaming era is beginning to bear many simi-
larities to that of the CD era: financial success depends upon scale and catalogue, the
major labels have a stake in music distribution networks, and the vast majority of artists
do not make any money.

Notes
1. Pandora was available only in the United States until December 2012, when it launched in
Australia and New Zealand and where it now has 2 million listeners.
2. Clear Channel has subsequently changed its name to iHeartMedia.
3. WiMP has subsequently been purchased and rebranded as Tidal.
4. Although often not recognised as such, YouTube is actually the largest music streaming ser-
vice, and is the de facto source of music listening for significant numbers of young internet
users. It has approximately 140 million weekly music video users (Mulligan 2014a).
5. For the original Swedish news story, see Andersson (2009).
6. The amount related to songwriter’s publishing royalty only (not the income from the recording
itself, which conventionally is about nine times higher), after collecting society deductions, for
a song that was co-written, covered only a short period of time just after Spotify had launched
and related to streams in Sweden only (see Barnett 2010).
7. Unless you’re a fan of Californian metal or UK dance, in which case it really sucks.
8. Perhaps most active is David Lowery, member of alternative rock bands Camper Van
Beethoven and Cracker who has also worked as a financial analyst and derivatives trader.
Lowery runs the blog ‘The Trichordist’ (subtitled ‘artists for an ethical and sustainable inter-
net’), where he provides analytical and provocative posts on a range of music industry issues.
9. Recently, Spotify has begun to subtly deflect pressure from itself onto the record labels, both
by making streaming data directly available to artists (via its ‘Spotify Artists’ service) and
through implicit public comments about the distribution of revenue.
10. It should also be noted that Spotify’s contracts with the major labels are subject to non-disclo-
sure agreements which make it impossible to ascertain how much, and in what ways, specific
labels are remunerated. It is commonly accepted that the major labels received lump sum pay-
ments for making their catalogues available on the service and that this income is not shared
with artists.
11. In Ek’s public response to Taylor Swift, he announced that Spotify has paid $2bn to rights
holders (Ek 2014).
12. The bulk of that 65% 70% is paid to the rights holder of the sound recording, with 10.5% (of
total revenue) paid to owner of publishing rights for mechanical royalties and performance
rights.
 Creative Industries Journal 187

13. In ‘Spotify explained’ (2013), the company used 40 million subscribers as their basis for
explaining the potential of the service for musicians. Two important provisos need to be made
about this argument, however. The first is that it assumes subscription prices remain constant,
but there are many commentators who think that, for streaming to really break though into the
mainstream, lower price points will have to be introduced (for example, Mulligan 2014d).
Second, it assumes that new subscribers are converted from existing free tier users. Spotify
has thus far remained consistent in converting 25% of users into subscribers. If, however, the
number of free users increases at a faster rate than the number of subscribers, then revenue
compared to number of streams could conceivably go down rather than up.
14. It should not be assumed that the major labels’ reaction to Spotify was uniformly positive. In
particular, there was resistance to the idea that access to music should be free and, before
launching in the United States, Spotify implemented some limits on the amount of music that
could be listened to by non-subscribers. These limits have since been removed, however.
15. There is not space in this paper to evaluate the claim that streaming monetises ‘pirate’ behav-
iour. There are certainly examples of countries, most notably Spain, in which rates of piracy
have reduced as adoption of streaming has increased. There is also some consumer survey
data which also suggests a connection between increasing streaming services and declining
rates of piracy (for example, NPD Group 2012). Ek’s aforementioned caution about confusing
correlation with causation should be noted, however, given that other factors, such as the dra-
matic rise in web takedown notices and the growth of YouTube, will have played a part in
declining piracy rates.
16. The actual equation offered by Spotify is ‘Spotify’s monthly revenue x (artist’s Spotify
streams/total Spotify streams) x approximately 70% to rights-holders’.
17. According to Arrington, Sony owns 5.8%, Universal 4.8%, Warners, 3.8%, EMI 1.9% and
Merlin 1%. Presumably, Universal assumed EMI’s share when it completed its takeover in
2012. There is dispute over how much the labels actually paid for their stakes: Jerrang con-
tends that it was just €8,800 (implying they received shares in exchange for licensing their
catalogues) while Arrington disputes this, claiming inside sources state that the labels paid the
same rate as other external investors and thus Jerrang’s figure is missing three zeros. It is
impossible to verify these figures, but it is commonly accepted that the major labels and
Merlin do have a stake in Spotify.

Disclosure statement
No potential conflict of interest was reported by the authors.

Notes on contributors
Lee Marshall is a Reader in sociology at the University of Bristol. His research interests centre on
issues concerning authorship, stardom and intellectual property, with a particular focus on the music
industry. Previous books include Bootlegging: Romanticism and Copyright in the Music Industry
(2005), Bob Dylan: The Never Ending Star (2007), Music and Copyright Second Edition (with
Simon Frith, 2004), The International Recording Industries (2012) and Popular Music Matters:
Essays in Honour of Simon Frith (with Dave Laing).

References
Anderson, C. 2007. The Long Tail: How Endless Choice is Creating Unlimited Demand. London:
Random House.

Andersson, C. 2009. “Lady Gaga tj€anar 1150 kronor pa Spotify.” Expressen.se 20 November.
Accessed 10 January 2015. http://www.expressen.se/noje/lady-gaga-tjanar-1150-kronor-pa-spotify/
Arrington, M. 2009. “This is Quite Possibly the Spotify Cap Table.” Techncrunch.com, August 7.
Accessed 10 January 2015. http://techcrunch.com/2009/08/07/this-is-quite-possibly-the-spo
tify-cap-table/
Barnett, E. 2010. “Spotify Rejects Claims That it ‘Rips off Artists’.” The Telegraph, April 14.
Accessed 10 January 2015. http://www.telegraph.co.uk/technology/7590782/Spotify-rejects-
claims-that-it-rips-off-artists.html
188 Lee Marshall

Barr, K. 2013. “Theorizing Music Streaming: Preliminary Investigations.” Scottish Music Review 3.
http://www.scottishmusicreview.org/index.php/SMR/article/view/40
Baym, N. 2011. “The Swedish Model: Balancing Markets and Gifts in the Music Industry.” Popular
Communication 9 (1): 22 38.
Bruno, A. 2011. “Spotify Responds to Indie Criticism.” Billboard, August 10. Accessed 10 June
2013. http://www.billboard.com/biz/articles/news/1176562/spotify-responds-to-indie-criticism
Bukart, P., and T. McCourt. 2004. “Infrastructure for the Celestial Jukebox.” Popular Music 23 (3):
349 362.
Century Media. 2011. “Pull Their Repertoire from Spotify!” News Relase, August 8. Accessed 10 Jan-
uary 2015. http://www.centurymedia.com/newsdetailed.aspx?IdNewsD10180&IdCompanyD3
Christman, E. 2014. “What’s Behind the Digital Download’s Decline and Can Streaming Save the
Day?” Billboard, January 7. Accessed 9 January 2015. http://www.billboard.com/biz/articles/
news/record-labels/5869521/whats-behind-the-digital-downloads-decline-and-can-streaming
Dickey, J. 2014. “Taylor Swift on 1989, Spotify, Her Next Tour and Female Role Models.” Time
Magazine, November 13. Accessed 10 January 2015. http://time.com/3578249/taylor-swift-
interview/
Dredge, S. 2012. “Spotify’s Daniel Ek: ‘We Want Artists to be Able to Afford to Create the Music
They Want to Create’.” The Guardian, December 6. Accessed 10 January 2015. http://www.the
guardian.com/technology/2012/dec/06/spotify-daniel-ek-interview
Dredge, S. 2013. “Thom Yorke Calls Spotify ‘The Last Desperate Fart of a Dying Corpse’.” The
Guardian, October 7. Accessed 10 January 2015. http://www.theguardian.com/technology/
2013/oct/07/spotify-thom-yorke-dying-corpse
Ek, D. 2014. “$2 Billion and Counting.” Spotify Blog, November 11. Accessed 10 January 2015.
https://news.spotify.com/se/2014/11/11/2-billion-and-counting/
Frith, S. 1986. “Art versus Technology: The Strange Case of Popular Music.” Media, Culture and
Society 8 (3): 263 280.
Greeley, B. 2011. “Spotify’s Ek Wins over Music Pirates with Labels’ Approval.” Bloomberg.com,
July 14. Accessed 6 August 2012. http://www.bloomberg.com/news/2011-07-14/spotify-wins-
over-music-pirates-with-labels-approval-correct-.html
Houghton, B. 2011. “Spotify Responds to Artist Payments Controversy.” Hypebot, September 19.
Accessed 13 January 2013. http://www.hypebot.com/hypebot/2011/09/spotify-responds-to-art
ist-payments-controversy.html
IFPI. 2013. The Recording Industry in Numbers 2012. London: IFPI.
IFPI Norge. 2014. “Music Sales in Norway 2013.” Accessed 13 January 2015. http://ifpi.no/flere-
nyheter/item/57-musikkmarkedet-2013
IFPI Sverige. 2014. “Musikf€ors€aljningen €okade Med 5,1% Under 2013.” Accessed 13 January 2015.
http://www.ifpi.se/nyheter/musikforsaljningen-okade-med-51-under-2013

Jerrang, M. 2009. “Sa fick Spotify skivbolagen Med Sig.” ComputerSweden, August 7. Accessed 10
January 2015. http://www.idg.se/2.1085/1.239983/sa-fick-spotify-skivbolagen-med-sig
Kernfeld, B. 2011. Pop Song Piracy: Disobedient Music Distribution Since 1929. Chicago, MI: U
of Chicago Press.
King, M. 2012. “Spotify’s D.A. Wallach Explains How Spotify Pays Artists.” Hypebot.com,
September 6. Accessed 10 January 2015. http://www.hypebot.com/hypebot/2012/09/spotifys-
da-wallach-explains-how-spotify-pays-artists.html
Knopper, S. 2009. Appetite for Self-Destruction: The Spectacular Crash of the Record Industry in
the Digital Age, 53 74. London: Simon and Schuster.
Marsden, R. 2012. “A Tight Spotify: Is There a Better Way to Make Music Streaming Sites Pay?”
The Independent, November 8. Accessed 10 January 2015. http://www.independent.co.uk/arts-
entertainment/music/features/a-tight-spotify-is-there-a-better-way-to-make-music-streaming-
sites-pay-8294714.html
Marshall, L. 2012. “The Recording Industry in the Twenty-First Century.” In The International
Recording Industries, edited by L. Marshall. London: Routledge.
Metal Underground. 2011. “Napalm Records Pulls Catalog from Spotify.” September 14. Accessed
10 January 2015. http://www.metalunderground.com/news/details.cfm?newsidD71709
Morris, J.W. 2011. “Sounds in the Cloud: Cloud Computing and the Digital Music Commodity.”
First Monday 16 (5 2). Accessed 29 July 2012. http://firstmonday.org/htbin/cgiwrap/bin/ojs/
index.php/fm/article/viewArticle/3391
Mulligan, M. 2014a. “10 Thoughts on YouTube Music Key.” Music Industry Blog, November 12.
Accessed 10 January 2015. https://musicindustryblog.wordpress.com/2014/11/12/10-thoughts-
on-youtube-music-key/
 Creative Industries Journal 189

Mulligan, M. 2014b. “What Acquiring Beats Could Do for Apple (And Everyone Else).” Music
Industry Blog, May 9. Accessed 13 January 2015. https://musicindustryblog.wordpress.com/
2014/05/09/what-acquiring-beats-could-do-for-apple-and-everyone-else/
Mulligan, M. 2014c. “What is Really Cannibalising Download Sales.” Music Industry Blog.
Accessed 9 January 2015. https://musicindustryblog.wordpress.com/2014/02/07/what-is-really-
cannibalising-download-sales/
Mulligan, M. 2014d. “Why it’s Time for a Streaming Pricing Reset.” Music Industry Blog, October
31. Accessed 13 January 2015. https://musicindustryblog.wordpress.com/2014/10/31/why-its-
time-for-a-streaming-pricing-reset/
Negus, K. 1999. Music Genres and Corporate Cultures. London: Routledge.
NPD Group. 2012. “Increased Use of Free Music Streaming Services Takes a Bite Out of Illegal
Peer-to-Peer Music File Sharing Activity.” February 26. Accessed 13 January 2015. https://
www.npd.com/wps/portal/npd/us/news/press-releases/the-npd-group-music-file-sharing-
declined-significantly-in-2012/
Peoples, G. 2012a. “Accounting Explains How Spotify’s Business Model Can Succeed.” Billboard,
October 5. Accessed 10 June 2013. http://www.billboard.com/biz/articles/news/1083499/busi
ness-matters-accounting-explains-how-spotifys-business-model-can
Peoples, G. 2012b. “Mumford & Sons’ ‘Babel’ Smashes Spotify Streaming Record.” Billboard,
October 2. Accessed 10 June 2013. http://www.billboard.com/biz/articles/news/1083573/busi
ness-matters-mumford-sons-babel-smashes-spotify-streaming-record
Peoples, G. 2013. “Pandora Surpasses 200 Million Registered Users.” Billboard.biz, April 9.
Accessed 10 January 2015. http://www.billboard.com/biz/articles/news/digital-and-mobile/
1556714/pandora-surpasses-200-million-registered-users
Peoples, G. 2014. “Exclusive Q&A: Deezer’s New Chief Marketing Officer Aims for U.S. Market.”
Billboard.biz, December 15. Accessed 10 January 2015. http://www.billboard.com/biz/articles/
news/digital-and-mobile/6406292/exclusive-qa-deezers-new-chief-marketing-officer-aims
Pham, A. “Streaming Made Up One-Fifth of U.S. Recorded Music Revenue in 2013.” Billboard.biz,
March 18. Accessed 8 January 2015. http://www.billboard.com/biz/articles/news/digital-and-
mobile/5937634/streaming-made-up-one-fifth-of-us-recorded-music
Resnikoff, P. 2011. “Distributor STHoldings Pulls 234 Labels from Spotify, Rdio, Others.” Digital
Music News, November 16. Accessed 10 January 2015. http://www.digitalmusicnews.com/per
malink/2011/11/16/stholdings
Resnikoff, P 2013. “Spotify: Not One Artist Can Demonstrate Lost Sales from Streaming….” Digi-
tal Music News, May 10. Accessed 16 January 2015. http://www.digitalmusicnews.com/perma
link/2013/05/10/spotify-artist
Rogers, J. 2013. The Death and Life of the Music Industry in the Digital Age. London: Bloomsbury.
Rosenthal, S. 2011. “Projekt is No Longer Available at Spotify, Grooveshark, Rdio, etc.” Projekt
Records Blog, September 23. Accessed 10 June 2013. http://www.projekt.com/newsarticles/pro
jekt-spotify.asp
Smith, Patrick,. 2009. “‘Fair Play? A Million Spotify Streams Earned Gaga $167’.” Gigaom.com,
23 November 23. Accessed 8 January 2015. https://gigaom.com/2009/11/23/419-fair-dos-a-mil
lion-spotify-streams-earned-gaga-167/
Solsman, J. 2014. “Pandora Rival iHeartRadio Hits 50 Million User Milestone.” Cnet.com, June 17.
Accessed 10 January 2015. http://www.cnet.com/news/pandora-rival-iheartradio-hits-50-mil
lion-user-milestone/
Spotify. 2013. “Spotify Explained.” Accessed 13 January 2015. http://www.spotifyartists.com/spo
tify-explained/
Spotify. 2014. “Information.” Accessed 10 January 2015. https://press.spotify.com/uk/information/
Spotify. 2015. “15 for ’15.” January 12. Accessed 13 January 2015. https://news.spotify.com/us/
2015/01/12/15-million-subscribers/
Stahl, M. 2013. Unfree Masters: Recording Artists and the Politics of Work. Durham: Duke UP.
Theberge, P. 2004. “Technology, Creative Practice and Copyright.” In Music and Copyright:
Second Edition, edited by S. Frith and L. Marshall, 139 156. Edinburgh: Edinburgh UP.
Titcomb, J. 2015. “UK Music Downloads in Decline for First Time as Spotify Eats into iTunes
Sales.” The Telegraph, January 1. Accessed 13 January 2015. http://www.telegraph.co.uk/
finance/newsbysector/mediatechnologyandtelecoms/media/11319348/UK-music-downloads-
in-decline-for-first-time-as-Spotify-eats-into-iTunes-sales.html
Wikstrom, P. 2009. The Music Industry. Cambridge: Polity.