What options do music distributors have to maintain or improve both their revenue and profit models in view of streaming technology and emerging new players in the music industry?

Master Thesis Executive Program for Management Studies (EPMS)

University of Amsterdam

What options do music distributors have to maintain or improve both their revenue

and profit models in view of streaming technology and emerging new players in the

music industry?

Maarten Los

Student Number 10901809

MSc EPMS

Statement of Originality

This document is written by Student Maarten F.J. Los who declares to take full responsibility for the contents of this document.

I declare that the text and the work presented in this document is original and that no sources other than those mentioned in the text and its references have been used in creating it.

The Faculty of Economics and Business is responsible solely for the supervision of completion of the work, not for the contents.

Table of Content

Introduction …...….... Page 4 Literature Research …... Page 6

Strategy in 2-sided markets ... Page 6 Strategy in the music industry until the year 2000 …... Page 7 Strategy in the music industry after the digital revolution …... Page 8 Strategy in 3-sided markets... Page 9 Financial performance of music streaming services ... Page 11 Strength and weaknesses of streaming media …... Page 12 Financial performance of the music distributors ... Page 14 Meaning of the financials for the music distributors …... Page 15 The research question and the propositions ...Page 16 Method and work plan ... Page 21 Results …... Page 23 Conclusion and Discussion …... Page 36 Literature: scientific papers …... Page 38 Literature: news bulletins, websites and interviews …... Page 40 Appendix A: interview questions …... Page 42 Appendix B: conceptual model …... Page 43

Introduction

The music industry, which comprises the entire revenue from record sales and live performance fees, is a huge market, with an estimated 37 billion US dollars in 2000 worldwide (Throsby 2002). Most of these revenues, around 80%, were earned by a small group of music recording and publishing companies (also called music distributors): EMI, BMG, The Warner Music Group, Sony Music Entertainment and Universal Music Group. EMI dissolved in 2012 and BMG was dissolved and re-established later, leaving a group of three big music distributors, often called 'the big three'. Other record companies are often referred to as 'independent labels'. The music industry has been around for centuries, but became big through technological improvements, such as radio and vinyl records, increasing the availability and reach of music (Blanning, 2008). Though the industry has been changing every decade since the 1950's, from 45 Rpm discs to LPs to cassettes to CDs, the music distributors were doing very well for themselves. They paid royalties to artists and were price makers: a record cost a certain amount of money and people who really wanted to 'own' a song would have to pay the set price for it. Artists needed the record companies for art work, distribution, marketing and sales etcetera, and only very successful artists could negotiate lucrative contracts for themselves. A major challenge to the music distributors came in the wake of the digital revolution: Napster. Napster was a peer-to-peer file sharing software program on which people could share their songs with other users, thereby circumventing any royalties associated with the songs. Napster, and several imitators, were shut down after lawsuits, reducing the threat of piracy on the revenues of the music distributors (Alexander 2002). However, the size of the music industry had been drastically reduced at that point, with an estimate of 15 billion US Dollars in total in 2014 (IFPI report 2015). The launch of Apple's iTunes in 2003, where consumers would pay a small amount per song instead of buying entire albums, greatly removed the threat of piracy: consumers were willing to pay for their music and favored accesibility over the low costs. iTunes reduced revenues from album sales, instead selling more singles. In a sense, iTunes took

over the role of the retailers: physical sales reduced and digital sales increased, but the structure of the music industry stayed relatively intact. Several years later, the rise of streaming services such as Pandora and Spotify not only re-challenged the business models of the music industry, they also did it in a way that challenges the foundations of the current ideas on corporate strategy.

Literature Research

In strategy, it was always assumed that a firm can choose from 3 different generic strategies in any market: cost-leadership, differentiation and the niche strategy (Porter, 1979), though certain combinations are possible (Porter and Millar, 1985). And with one of these strategies, firms would enter or create a market where customers typically buy a service or product from the firm for money ,i.e. a separation of customers and firms. The firm typically buys its input from suppliers, or they have (parts of) their supply integrated in the firm. They may also have differentiated and sell different kinds of goods or services to customers. But in any case, the customers pays the firm for their product (see figure 1). There are no differences here between the market based view of firms and the competence/resource based views of the firm: they differ in their view on strategies, from a simple fit to a more stretched form with no simple strategy for any problem (Barney, 1991). But they all maintain one business strategy for a market with a separation of production and consumer, with the firm adding the most value to the product or service.

Supplier

F

Figure 1: simple schematic of the supply-production-output relation: product going from supplier to customer, money going from customer to supplier

It is important to make a distinction between the revenue model and the profit model. The revenue model refers to how the firm generates revenues, e.g. by flat fees, subscriptions or retailing their service or product. So it helps a company to determine the market or type of consumers that will buy each Supplier

Supplier

product or service that the company can create (Afuah 2004). Profit model simply refers to the gross profit a company makes, or revenues minus costs (but includes cost structure and capital intensity). So while revenues and profit are often related, there is no causal relation between the two.

Strategy in the music industry until the year 2000

In the music industry, as described in the previous chapter, this meant that music distributors paid artists, music recorders and studios royalty fees for their music, which they in turn marketed and distributed around the world. Retailers bought records (LPs, CDs) from the distributors and sold them to customers from their physical stores. In the beginning, sales were focused on hit singles, with often no more than 2-4 songs total on both sides of the 78 rpm vinyl records. These records could often not play more than 5 minutes of music per side. The introduction of the LP increased play time per side to more than 20 minutes. This lead music distributors to change their revenue model, with a focus on entire albums instead of the shorter and cheaper singles. In practice, customers were forced to buy entire albums when they liked a certain artist, paying for many songs instead of the few songs they would have wanted. This greatly increased revenues and profits for the music distributors. The songs on albums that people like less, and for which they would not have paid if the songs were on different albums or singles, can be seen as an economic inefficiency, increasing revenues for the music distributors but reducing the price-quality ratio for consumers. The cassettes may have inproved the ease of use (they can be listened to in a car) and CDs have improved the music quality when they appeared (LAWeekly, 2015), but they have not changed the revenue model of the music distributors. In the 90'ies of the 20th _{century, the rise of internet and the digital revolution it caused did change the} revenue and profit models of the music distributors (Kusek et al. 2005).

Strategy in the music industry after the digital revolution

Improvements in computer software made it possible to reduce the size of digitalized songs, improvements in hardware increased storage capacity of songs and the rise of internet enabled people to electronically share songs with everyone (Gopal et al. 2004). The sharing, or peer-to-peer, of songs enabled people to acquire music without paying royalties for the music: something called piracy. Logically, music piracy greatly reduced revenue and profits for music distributors (Marshall, 2004). Companies making money off piracy, such as Napster and Kazaa, were quickly sued by the music distributors for dodging royalties and they soon disappeared. These programs did allow people to acquire music on a song-basis, and not on an album-basis, as was common in that period. Apple inc. noticed and created iTunes, an online store were consumers could buy songs per piece (for .99$) or per album, whichever they preferred. Gift-cards were available at any supermarket or convenience store. iTunes was also fully legal and decreased online piracy of music (Koh et al. 2014). Music distributors still received their royalties and had the benefit of not having to physically distribute their CDs, which reduced transport costs, wastage, theft etc. So far however, these effects have not turned the tables, and music distributors still generate less revenues than they did in the year 2000, as can be seen under 'Financial performance of music streaming services'.

Another phenomenon also appeared and grew after the year 2000: streaming services. These services allow consumers to listen or view content online without the need to download it. The most famous example of this type of media is YouTube. And even though many of these streaming media were offering illegal content (such as MegaUpload, which was closed and banned in 2012), others offered legal content. The largest music streaming media as of January 2016 are Spotify and Pandora. These companies use an interesting revenue model, with a mix of advertisers and paying customers, called the freemium model (a combination of 'free' and 'premium'). And the more customers they have, the more

interesting the company becomes for advertisers, which in turn enables these companies to further increase their customer base, yet again increasing either monthly fees or advertisement prices they can ask the advertisers (Evans, 2003). They do not have a typical 2-sided market business, but a 3-sided market business model.

Strategy in 3-sided markets

Today, there are many new firms that basically circumvent the old ideas about strategy and are able to shake up an entire industry: big bang disruptions (Downes and Nunes, 2013) (Bower and Christensen, 1995). What happens is that many new firms are able to beat the competition by improving on all three of the generic strategies: they provide better products (differentiation) at lower cost (cost-leadership) and for a broader customer base (niche). The distance between customer and firm can also be reduced, up to a point where the customer is actually involved in the creation of the product, as can happen in video games (Arakji and Lang, 2007). This process has occurred in the entertainment industry in the last decade. In the movie industry, Netflix started as a cheaper, more convenient alternative than common video rentals such as Blockbuster, with a broader selection on top of that. Netflix welcomed it's 75 millionth customer in January 2016. Blockbuster filed for bankruptcy in 2011. And as stated before, in the music industry, companies such as Spotify and Pandora transformed their revenue model from a 2-sided market into a 3-sided market (see figure 2) (Klein 2015) (Bockstedt et al. 2005) (Bourreau et al. 2008)

Figure 2: the 3-sided market: advertisers pay streaming programs, streaming programs pay music distributors through royalties, consumers can either pay or get the product for free, be it in lower quality

This 3-sided market draws its money mostly from either advertisers, who pay to get noticed by the consumers, or by the consumers who pay subsciption fees. Consumers are drawn by the quality and amount of music the services can provide. And record companies are drawn by the money making potential, though there is still much discussion about the value appropriation in these markets. The most interesting part of this type of market market is the option for consumers to use the service for free. At first, this looks much like radio, but with most streaming services, consumers can choose the content. The streaming media also offer more music than a physical store could ever hope to offer, with easier access as well: no physical distance and no opening hours. So the streaming services fall well into the big bang disruption category, with their different type of strategy and beating traditional retail stores on all three generic strategies. The reduction of physical stores was predicted even before the streaming media appeared (Zentner, 2008) (Vaccaro and Cohn, 2004), and recent bankruptcies in the Netherlands has proven this (retaildetail.nl, 2014).

Record companies are no longer having to take care of physical distribution and retailing of music (Lorenzen and Frederiksen, 2009), but are still collecting royalties for every song bought or played (Clemons et al, 2002). In the case of Spotify, artists and record companies have their music placed on Spotify and sometimes even pay for having their songs heard by a large audience. Consumers can listen

Advertisers Pandora/Spotify Music distributors/artists

for free if they accept advertisements, or pay a monthly subscription fee to not hear the ads. Spotify has focused more on the premium users, whereas Pandora derives most of its income from advertisers, based on the amount of free users. iTunes functions more as a traditional record store, where people just buy the right to permanently listen to a certain song. Spotify and Pandora have created a model in which consumers have the option of getting a service, music in this case, for free. Artists/distributors can either earn money from music sales, or pay one of the streaming programs for the potential of reaching lots of music listeners and increase their reputation. Advertisers pay the streaming programs to reach the music listeners, the more the better. And the more consumers use the streaming programs, the more artists are willing to have their music available on the site, and the more the advertisers will pay the streaming media. So clearly, these sites not only benefit from having a large number of subscriptions: their entire existence depends on it.

Financial performance of music streaming services

The main expenditures of streaming sites are the royalties they have to pay to record companies and artists. Spotify claims to pay royalty fees of 0.006$ to 0.008$ per streamed song in total, with a fraction of that going to the artist and composer. Pandora claims to pay between 0.0014$ and 0.0025$ per song. Both pay higher royalty fees for subscribers than they do for non-paying users, although they do not give clear insight in how they calculate the paid amounts. Financial results in 2014 were not good for Spotify, despite its 15 million paying users at the time: it reported a loss of 197 million US Dollars, increasing its loss from the previous years (The Guardian 2015). Pandora reported a 49 million US Dollars loss in March 2015 over 2014 despite its 80 million active users (Pandora Financial Statement, 2015). Both companies did spend much on growth and expansion, which is the main reason for the losses they reported. Other streaming services have much lower revenues than these two have, but show the same trends: losses due to growth and expansion which may be temporary.

Strength and weaknesses of streaming media

The first thing that comes to one's notice are the financial results of the last few years: Pandora and Spotify are losing money. Most emerging companies lose money at first, but at a certain point they either turn profitable, or they disappear. But both companies have been losing money for quite some years, with the royalties as their highest costs at more than 80% or revenue. They claim they are still investing in increasing their customer base, hence the losses. This does raise the question that if 75 to 80 million subscribers still lead to negative performances, how many subscribers do they want and how many do they need to make a profit? They can try to move more people using the programs for free into paying monthly fees, though this may be difficult (Datta et al. 2015). The amount and diversity of songs the music streaming services offer is huge: both Spotify and Apple Music claim a library of more than 30 million songs. But people cannot listen to 30 million songs and they are probably the most interested in a small number of artists, the best-selling ones in most cases. This makes the streaming media quite dependent on several musicians and their music labels. For example, Taylor Swift, one of the best selling artists in the world recently, refuses to have any of her songs streamed on Spotify. Adele, another bestselling artist, decided to publish her latest album '25' physically and digitally, but not on any streaming sites. If in any case a large number of famous artists (or their music distributors) decide to collude and all boycott one or more streaming services, the streaming services are in deep trouble, potentially losing the majority of their customers. This does raise the question what would happen in this case, since the streaming services pay for the royalties. Perhaps a reduction in accesibility could increase piracy.

There is also the possibility that the large music distributors will copy the model of Spotify or Pandora. They are the ones that own the rights to a huge amount of songs, so they will not incur the high costs that the streaming media do: they could cut out the middle man so to say. And the algorithms that music

streaming services use, in order to present consumers with playlists they will fancy, are mostly based on the sheer amount of users and the playlists that they create. These may be relatively easy to imitate (QZ.com, 2015). Or in the market based view of a firm: the streaming media do not possess a resource that is rare, inimitable or non-substitutable (Barney, 1991).

From the upside, the streaming media do have quite a strong brand name, many subscribers (up to 100 million) and many paying advertisers. And though their algorithms are imitable or substitutable, they are functioning well at the moment, as does all of their software. Their ability to offer a wide variety of music at a very low cost, combined with easy access to many across the globe, has been quite a big bang disruption.

Another interesting aspect to look at the differences with other streaming media, such as YouTube and Netflix, who both offer video material. YouTube is owned by the Aphabet conglomeration (formerly known as Google), and has countless numbers of videos on its website, ranging in the millions. YouTube has all sorts of video material in its library including music videos, with all of the 50 most watched videos being music videos (youtube.com). Their income derives from advertisers, who pay a certain amount every time their advertisement is shown to a group of viewers. A part of the advertisement money is paid to the uploader of the video. Unfortunately, Alphabet does not provide information on the financials of YouTube, so it is unknown whether YouTube is profitable or not. Alphabet is a large conglomerate with huge revenues and could easily maintain a non-profitable YouTube for a long period of time for whatever reasons.

Netflix on the other hand is publicly listed and does give information on its financials. Netflix was founded as a DVD-per-mail-rental firm, but transformed in a streaming service starting in 2007. Netflix

is a subscription-only service, though the first trial month is free. They expanded to many countries and offer TV shows, movies and documentaries. Most of these are produced by other companies, but Netflix also makes it own content (such as 'House of cards' and 'Orange is the new black'). So while they do not own any royalties on their own productions, they do incur high production costs for their own productions (Variety.com, 2016). Their subscriptions-only business model leaves no room for advertisers or freemium models, making theirs a more traditional 2-sided business model. Their revenues in 2015 were 6.8 billion US Dollars, with profits of 122 million US Dollars. So despite their huge royalty costs of 70-80% of their entire revenue, Netflix has managed to be profitable for many years in a row now, despite investing heavily in growth (Netflix financials 2015).

Finally, there is of course traditional radio, both analog and digital. Many radio stations are privately-owned, instead of being government-run, meaning they need to turn a profit, which they have been doing for decades. Radio channels are of course free for consumers and run on advertizing money. Consumers can not choose which music to listen to, although radio stations adjust their choice of played music to their audiences. This does give them full control over the royalties that will need to be paid to the music distributors.

Financial performance of the music distributors

With streaming programs recording huge losses, it raises questions about the financial performance of the music distributors. They get royalty fees from the streaming programs, which may be less than they would have gotten from direct sales, but they can cut costs on physical distribution, such as manufacturing, inventory losses and costs associated with retailer contracts. Of the 'big three' music distributors, Sony Music Entertainment increased its revenue from 31 March 2014 – 30 March 2015 with more than 8% over the previous year (Sony Financial Statement, 2015) to 4.5 billion US Dollars,

for all elements of its business, such as video clips, live performance fees, direct sales and royalties. March 2013 – March 2014 also saw an increase as compared to the previous year, from 3.6 to 4.14 billion US Dollars. Revenues in 2010 were approximately 3.96 billion US Dollars, only slightly lower than they were in 2015.

Universal Music Group saw a drop of almost 7% in revenue in 2014 as compared to 2013, to 4.5 billion US Dollars, caused by a decline in sales of physical music. Digital music revenues did not change significantly. And though downloads were down, received royalties increased (Universal financial statement, 2015). 2013 did see an increase from 2012: from 4.5 billion US DOllars to almost 4.9 billion US Dollars. The revenues in 2010 were approximately 4.45 billion US Dollars, so revenues have not changed much over the past years.

Warner Music Group has seen a decline from September 2014 – September 2015 of 2% as compared to the previous year, though digital revenues have gone up almost 4% (Warner Financial Statement). 2014 saw an increase of 5.4% from 2013. Revenues in 2010 were around 2.3 billion US Dollars, 28% lower than the 2.97 billion US Dollars revenue in 2015.

Meaning of the financials for the music distributors

So while the streaming services lose money, the music distributors have not increased their revenues the last few years, with the exception of Warner Music Group (be-it a small increase). Not decreasing in revenue can be considered good in view of recent events and changes. But the world population has grown enormously the last decade, as has the amount of the people with enough money to spend money on entertainment such as music, especially in China. So this may indicate that the potential market for music has grown, but that music distributors have not yet benefited from this growth. Also, the music

industry was almost twice the size in the year 2000, as compared to the year 2015. With the big streaming media losing money and the large music distributors hardly growing at all, it is interesting to know what is happening to the market currently. Has the market shrunk, despite a growing world population? Not likely, with the improvements in living standards. Are the smaller music distributors increasing their revenues, and thus increasing their market share? That may be possible, but there does not seem to be a specific reason to clarify this. Are other parties increasing their revenues? Not likely. Retailers have gone online or they have disappeared. And artists are still dependent on the music distributors, with no reason for them to have changed the contracts with their artists. The main reasons are a shift from album to single sales and an increase in amount of streams. So consumers are listening to perhaps more music than ever, but are not paying more for it. On the contrary, free use of streaming services has lowered the costs for most consumers. In a way, this can ben seen as a loss of dead weight in the market, or an increase in public wealth. This raises the question whether the music distributors can maintain their current revenue and profit model: the streaming media may grow so large they can dictate terms to the music distributors, or artists may decide they do not need the music distributors any more.

The research question and the propositions

The development of these 3-sided markets raises some interesting questions: will artists be able to benefit more from their music through downloads and streaming than they would have made before internet became so important? Or will their music become a marketing tool that they will use to increase revenues from live performances? Will record companies cease to exist or will they be in a strong position to improve their revenue and profit models? Are the streaming media here to stay or are they merely a temporary phenomenon? (Wagner et al 2013). The music distributors currently have deep pockets and a large infrastructure which could enable them to turn the tables and return to a situation

comparable to before the digital revolution (Collins and Young, 2014). Or there could be another big bang disruption which will fundamentally reshape the industry again. In any case, it is very interesting to theorize what the music distributors will do in the current situation. This leads to the following research question:

What options do music distributors have to maintain or improve both their revenue and profit models in view of streaming technology and emerging new players in the music industry?

In order to further understand this, I would like to propose several propositions. The first proposition assumes that in the current situation, conditions will not improve for the music distributors, but will remain stable as they have been the past few years.

P1: In the current situation, it is unlikely that the music distributors will return to the profitability they enjoyed before the digital revolution took off.

But the situation may very well change, leading to the following propositions:

P2a: The music streaming programs have thus far incurred such high losses due to the royalties that they may never turn profitable with their current business models.

P2b: Musicians who enjoy high sales and popularity may prefer to distribute their music without use of the music streaming media, believing they can increase their earnings. Musicians with lower sales and popularity will likely opt to keep their music on the music streaming media

effectively ending their existence. This will increase use of the more 'traditional' programs/sites such as iTunes, with fixed rates per song. Sites such as YouTube might introduce a pay-per-view system for their music videos and Vevo (a YouTube based video hosting service, owned by Universal Music Group and Sony Music Entertainment) may decide to leave YouTube and start their own channel. This scenario would again increase the revenues for the music distributors and would decrease free-riding for consumers. To further complicate matters for the streaming programs, more artists will follow Taylor Swift in prohibiting streaming programs from playing their music, greatly reducing the quality of the music streaming programs can offer their consumers.

The gap that the disappearance of the streaming media leaves will most likely be filled by other participants. The music distributors, who own most of the music rights, seem the most likely candidates:

P3a: The large music distributors may attempt to acquire some of the streaming media programs and turn them profitable, using their existing infrastructure and reputation while avoiding the high royalty costs.

P3b: The large music distributors may found their own streaming programs in order to compete with the existing music streaming programs, and to avoid any possible acquisition costs.

P3c: The large music distributors could enter the music streaming market together, creating an oligopoly in the music streaming business

This move would transfer all the revenue of streaming sites to the large record companies. But they will not incur the high royalty costs, since they own most of the rights of the music. They will have to pay the artists, like they have always done, but they will be without high distribution costs. This move

would transfer huge amounts of power to the record companies, who control production and distribution, making it easier for the companies to put pressure on the artists for lower royalty fees from their side. And this would greatly improve financial performance of the record companies. This could be done either in the 3-sided market model, with advertisers providing revenue, or through a more traditional model with premium users. Or through a combination of the two.

Another option would be that artists unify themselves and start negotiating with streaming media and iTunes themselves, cutting out the record companies. In this case, artists could receive increased fees, while the costs for the streaming media would be reduced, perhaps even turning them profitable. Record companies would lose their function of 'middle man', greatly reducing their financial performance. One company that tries to sidetrack the music distributors is Global Music Rights (GMR). GMR negotiates directly with streaming media, promising higher returns to the artists they represent. But while GMR has the connections and is gaining the market power to negotiate good deals for artists, they are currently not at the organizational level of music distributors in terms of marketing, recording etc. If they would ever reach that organizational level, few things would separate them from the music distributors they tried to supplant in the first place.

If artists decide to remain under contract with the music distributors, they may force them into accepting so called 360 deals. These deals are the opposite of the traditional contracts where artists receive royalties from the distributors: in these types of contracts, distributors receive certain percentages of all the revenues an artist collects with record sales, streaming and live performances. Though these deals can be very lucrative, they are not as beneficial as the traditional deals where the distributors got the majority of all revenues:

P4a: Even though some artists will find ways to circumvent the use of music distributors when producing or distributing their music, the majority will remain under contract with the current music distributors

P4b: The more famous artists will force the music distributors in accepting 360-deals, reducing revenues for the music distributors

And finally, there could be new players in the field. Emerging new players are notoriously difficult to predict: many companies try, few of them actually succeed. For example, Pandora and Spotify have 75-80 million subscribers. Many other companies active in streaming have less than 10% of their amount of users, or have been acquired by bigger companies. Apple has recently started Apple Music, another streaming program with monthly subscription fees (9.99$ a month in this case), with around 10 million users in February 2016. Of interesting note it that Apple Music offers their users a 3-month trial instead of a 1-month trial that Spotify offers (Billboard.com). At this point, Apple Music looks to be the biggest threat to both the other streaming media as to (the revenue and profit models of) the music distributors. An interesting phenomenon which has occurred several times is the 'pay-what-you-want' idea, where people can download songs or an album for free and then decide whether they will pay and if so, how much they will give. Even though this model is risky, results so far have been quite good (Regner and Barria, 2009) (Pitchfork.com), though it has not been tried often so far. But perhaps there are other new players or technologies that will shake up the entire market, as a big bang disruption:

P5: A new emerging player or technology will fundamentally change the music industry, which could afflict the music distributors negatively

Method and work plan

In order to answer the research question and its propositions, several companies (cases), both music distributors and streaming media companies, will be studied in the context of the changing music industry. The recent changes in the industry can be studies in descriptive form, future options and possibilities for the music distributors and streaming media companies, which rely heavily on opinions and future actions, will be studied in an exploratory form. The unit of analysis will be the decisions already taken by the companies, as well as any plans or ideas they have. These are not easily quantifiable, but will be able to answer the propositions and the research question in a qualitative manner.

Two types of sources will be used for each case, to strengthen construct validity through data triangulation. The first, most important, one is a combination of research papers, music magazine articles, financial statements and news reports. Data summaries will be used for answering the research question and for strengthening of the interview questions. The second source is data collected from interviews with executives of music distributors who are well known with the music market and take operational and strategic decisions for their companies. Hence, they are expected to understand the challenges their companies face and may have already have formulated or executed solutions to these challenges. Executives from streaming media will also be interviewed since they provide information on the other side of the spectrum. Interviewees will at first be found through the network and through acquisition. After that, snowball selection will be used to increase the number of interviewees, since the people already interviewed are likely to know which of their colleagues or acquaintances can further help answering the research question. The amount of people to be interviewed depends entirely on the given answers: until saturation is reached. The transcripts of the recorded interviews will be used to

answer the research question, using axial coding. The analyzing technique will be one of explanation building, since the research is partly exploratory.

The major limitation of this study is the willingness of the executives to share their (future) decisions and specifics. However, expected profits and revenues will not be asked, only the general strategic decisions. The major strength is the possibility to look into both sides of the music market and to indulge in their ideas and decisions.

As for the work plan, 5 months are the time that is available for writing a full introduction, setting up appointments with interviewees and executing these interviews, write up the interviews and results, write the conclusion and write the methods section. I plan on setting up the first appointments as soon as the thesis proposal has been approved, since it may take a while before the interview can take place. Interviews can take place in March and April, with possible postponing into May. I plan on using February for writing the methods section and writing down the interview questions. Results from the interviews will be written down in May, or possibly earlier if possible. June will be used for writing the whole and finishing the thesis, well on time before the first deadline of June 30th_.

Results

It is unlikely that the music distributors will return to the profitability they enjoyed before the digital revolution took off

The first proposition that needs answering is whether the music distributors can ever return to the profitability they once enjoyed. In order to do this, we must first determine the strength of the music distributors in their old situation, in the years before 2000. There are many ways to theorize on their position, but two of the most important and famous ways to look at it are through the market-based view of the firm and through the resource-based view of the firm.

In Porter's market-based view of the firm (Porter, 1979), any company is subject to 5 different variables that determine the ability the company has to make profits in that specific market. In the old situation, all 5 forces working in the music distribution market were low, meaning the music distributors were in a very profitable position: the first force, the threat of new entrants in the market, was very low due to the oligopoly the largest four distributors had created, owning almost 90% of the market. Independent distributors acted in niche markets almost, and the sheer size, branding and market power the 'big four' had, made it near impossible for rivals to enter the market. The second force refers to the bargaining power of suppliers, artists and recording studios in this case. With countless artists and recording studios operating in the industry, and only several big distributors, artists had little bargaining power; they could only try negotiating with so many distributors. This severely reduced their bargaining power. The bargaining power of buyers, the third force in the model, was also limited. Millions of people were buying music worldwide from many different retailers. Wishes or demands from one group of buyers or one retailer could easily be ignored, especially since a few companies owned almost the entire market (where would a record store be if they could not sell 30% of all music produced that year?). A threat of substitutes, the fourth force, did not exist at that time. Music distributors had

long-term contracts with artists and retailers and there was hardly a way for any of them to circumvent the distributors, except for the independent labels perhaps. Technology for copying was limited at that time, especially since quality of the recording drops when copying a cassette or vinyl record. The CD burner increased this threat, as CDs do not lose any quality of sound when copied, but could not be scaled easily, partly due to it being illegal (and having a high chance of being caught). The final force is rivalry within the industry itself. This was also limited as almost 90% of the industry was controlled by only 4 companies (later three), with the remaining 10% claimed my numerous independent labels.

So what changed in the past decade and a half? Industry rivalry is the same as it was 15 years ago and the threat of new music distributors entering the market is still low. As for the power of buyers, electronic stores such as iTunes have largely replaced the 'brick' retailers, but they are still selling the distributors' music. Of note is that electronic stores sell per song, whereas retailers most often sold albums, fetching higher prices per purchase. The threat of substitutes, however, has increased immensely. Not only did piracy give consumers an alternative to buying music in retail or electronic stores, it did it for free. And competing with free products is often tough, even when the alternative is not legal. The other substitute, streaming services, give royalties every time a song has been listened too. In practice, this gives low revenues for poor-performing songs, but high revenues for good-performing songs (more on this later). But while the streaming services are fully legal and are paying royalties to the distributors, they also interfere with the sale of albums. Imagine that distributors sold every album, with 10-12 songs, to retailers for 10$, consumers on Spotify, assuming 1 cent per stream, would have to listen to songs of that album approximately 1000 times. For a 1$ iTunes song, a consumer would have to listen to that song 100 times. Something the streaming services have also done in some way is to increase the possibility of artists to distribute their own music without the large distributors, often for free. Revenues from live performances can make up for this. For more famous

artists, this effect is more ambiguous: does a larger consumer base increase the amount and price of live performance tickets? Or did artists use live performances as a way of promoting their albums, hence selling tickets underpriced, and have streaming services reduced the need for low ticket prices? (Krueger, 2005). In both cases, the bargaining power of suppliers has increased somewhat. In conclusion for the market-based view of the firm, the position of the large music distributors has reduced significantly. This has reduced their profitability and, with the substitutes still in place, there seems to be no return to the high profitability of before.

In the resource-based view of the firm, as theorized by Barney (1991), a company will perform well if it has one or more resources that are valuable, rare, inimitable and non-substitutable. The definition of a resource is very abstract and broad. In this case, the resource would be defined as the ability to acquire and distribute music across the globe. This ability is clearly valuable, as seen by the profits the distributors made in the past. It was rare in a sense that the market was largely controlled by a few players, barring entry to other large distributors who could amass the same network and capital as the existing players. Their ability to distribute music was also not imitable, since all of their products were copyrighted. Only small scale copying of cassettes, vinyl records and later CDs would count as an imitation, which would not have been enough to severely impact the financial performance of the distributors. The distribution, or the music itself, could also not be substituted. Physical sales were the only sales there were and there are no alternatives to 'listening to music', other than perhaps live performances or making music at home. So in the resource-based view of the firm, the large companies had a resource which would guarantee them high incomes.

In the current situation, the digital revolution has made distribution of music imitable: piracy caused large-scale distribution over the internet, for which the distributors would receive nothing. Streaming

services reduced the substitutability, paying per stream instead of paying for entire albums and in the process killing retail. And these factors have reduced the value of having a good distribution network, leaving the music distributors in a less competitive state than they were years ago.

So in summary, the music distributors are not likely to return to the profitability they once enjoyed. The financials described earlier seem to support this claim. Their revenues and profits have been stable the last few years though, so a further reduction is also unlikely.

The music streaming programs have thus far incurred such high losses due to the royalties, that they may never turn profitable with their current business models

Unfortunately, the streaming services were unable to provide comments on this proposition. The latest financial statement of Spotify over the whole of 2015 gives an 80% increase of total revenue, from 1.22 billion in 2014 to 1.95 billion US Dollars a year (Financial Times, 2016). Losses have increased from 162 million in 2014 to 173 million US Dollars in 2015. Losses have increased, although they are relatively lower. Spotify paid 1.63 billion US Dollars in royalties to music distributors and spent 196 million on advertizing. That leaves 297 million open for technical updates and expansions, maintenance, salaries and overhead. The majority of the costs, the royalties, are fixed costs, unless Spotify can negotiate more lucrative deals with the distributors. Incomes were 1.74 billion from the paying subscribers, but only 196 million dollars from advertisers. Spotify does not give costs or revenues per customer, which cannot be calculated with the freemium model they use. This makes it hard to accurately predict the business model when costs for growth and expansion are neglected. But it can be estimated, assuming an equal amount of free and paying users, a growth in subscribers for 2015 distributed equally over the year (meaning the average user listened to music for 6 months), equal advertizing revenues per free user, equal operating costs and equal streaming costs per user, neglecting

the slightly changed ratio of paying/free users in favor of the paying subscribers and neglecting corporate taxes. As for revenues, paying subscribers amount to 28 million people paying 10$ per month on average for 12 months, totaling 3.35 billion dollars in 2016. Royalties will obviously increase: they increased from 882.5 million in 2014 to 1.63 billion in 2015, with 29 million extra users listening for 6 months on average. For a full year, total royalties would increase with 1.96 billion, giving a total of 2.84 billion. Advertizing revenues would increase to 293 million dollars per year. With operating costs stable at 297 million and without any growth costs, Spotify would make an estimated 740 million dollars a year. This if of course a wild estimate, but the business model in its current form may be profitable, at least for the biggest streaming service. It is interesting to see that the profitability for paying users seems to be much higher than for free users. The conversion rate of free users to becoming paying users is approximately 25% (Reuters, 2016), which would indicate that a large part of costs associated with expanding yields little results. A business model with paying customers only, like Apple Music has, may be more profitable than services with the freemium model. Of interesting note is that the launch of Apple Music in June 2015 did not decrease the amount of subscribers for Spotify (Reuters, 2016), indicating that the music streaming market is still growing.

Apple does not give details about Apple Music, but 13 million people have signed up to Apple Music as of late April 2016. Many of them still in the 3-month free trial, many of them in the 15$ family plan and the rest in the individual 10$ per month plan (Techtimes, 2016). Current revenues are unknown, but with the amount of users, it can be estimated to be around one billion US dollars and growing. One major strength that Apple Music has is its distribution system: Apple Music is preinstalled on all iPhones, making is easily accessible. Apple also has huge revenues so they are independent from outside investors. It would seem unlikely that Apple Music pays less royalties than Spotify does for their paying subscribers. However, in the case of Spotify, subscriptions are more profitable than

advertizing revenues. So Apple Music may very well be quite profitable, when we do not consider costs related to expansion.

Pandora made a 170 million US dollars loss in 2015 (Pandora financial statement 2015), with revenues of 1.164 billion. Advertizing brought in the majority of revenue with 933 million dollars. Royalties accounted for 610 million, sales and marketing for 398 million and general and administrative costs for 154 million. 80% of costs, as compared to total revenue, were accounted for by royalties, sales and operating costs. Core investments made up 26% of expenditures (such as marketing and technical adjustments), which all lead to a 5% loss overall. Without any costs related to expansion, Pandora could have probably had a 200 million profit. So while Pandora is still losing money (forecast for 2016: a 60 million operating loss), the revenues are higher than the underlying operating costs.

So all 3 companies seem to be able to operate profitably using their respective business model. The big question is how long these companies will keep expanding and more importantly, how big the total music streaming market is. It seems unlikely that consumers in Latin America, Africa and most of Asia can afford 10$ a month for music streams when they can get it for free. And with a combined total of 200-250 million people already using one of the music streaming services, the market may be saturated quite soon. This may explain why the companies expand so aggressively. A saturated market will in general lead to decreased profitability, putting the worst performers out of business. But until that time, the music streaming industry can be profitable with any of the current business models, at least for the large streaming services.

But what for the smaller services? Deezer operates mostly on a subscription basis (10$ per month) and gets almost 50% of its revenues from France, 141 million in 2014 and 93 million for the first 6 months

of 2015. Deezer, with 3.8 million subscribers, made operating losses of 29 million in 2012, 22.5 million in 2013, 27 million in 2014 and 12.2 million in the first 6 months of 2015. 80% of its costs go to royalty payments. Deezer has minimum license agreements with several music distributors, which means that the company pays a minimum amount to the distributors for certain content, regardless of how much the subscribers listen to the content. The company recognizes these contracts as the main reason to invest in growth, and acknowledges its lack of profits at the current scale. It is uncertain about the profitability in the future (Deezer financial statement, 2015). 8tracks, a free-to-use streaming service that requires its new subscribers to create an 8-track long playlist from the company's library, claims it has been profitable since 2012, with a total investment of 3 million US dollars. They now have 8 million users, but will likely announce a crowdfunding action to be able to expand and deal with the ever expanding rivals (Thenextweb.com, 2016).

The list of streaming that services that went bankrupt the last 2 years is long: Grooveshark, Rara, Last.fm, MixRadio, Simfy Germany, Ubuntu One Music and Rdio all shut down, and several others were acquired by other services such as Beats Music and iTunes Radio, Music Unlimited, WiMP and Songza. Most of these services had less than 10 million active users at their zenith. This could make a valid point for the remaining streaming services to expand and benefit from a large scale, in order to negotiate better deals with distributors and decrease marketing costs.

In summary, the music streaming market seems to be big enough for several large services to be profitable in the coming years. This means that despite the current losses the streaming services incur, they are here to stay.

Musicians who enjoy high sales and popularity may prefer to distribute their music without use of the music streaming media, believing they can increase their earnings. Musicians with lower sales and popularity will likely opt to keep their music on the music streaming media

So far, there have been only a few cases in which this has happened. Taylor Swift retracted her music from Spotify, criticizing the company's payments of royalties. She did, however, let Apple Music stream her latest album exclusively (Macworld.com, 2015). Adele refused to have her latest album '25' streamed, using art/music-related arguments. The hundreds of millions she made with the physical and digital sale of the album would have taken forever to make with streaming royalties. However, streaming does not necessarily influence album sales (digitalspy.com, 2015). Some other artists only have one streaming service stream their music. For example, Prince's songs are available on Tidal, as are Neil Young's. Dr Dre had his music put on Apple Music exclusively (consequenceofsound.com, 2016).

In the absence of new cases in the first half of 2016 and ambiguity over the effect of streaming on physical album sales for individual artists, it does not seem that artists will refrain from having their music streamed. Experts from the field agree on this: they believe that not only the revenues for almost all of the artists are too low to pull their music from streaming services (many lesser known musicians live on extermely low incomes), but that most artists simply love having their music played/streamed to audiences, regardless of their income (Interview Menno Timmerman, 2016)(Interview Paul Zijlstra, 2016). It is more likely that artists will sign exclusive deals with one service, creating more competition between the streaming services (Engadget.com, 2016). An interesting aspect is that many artists do not control all the rights to their music, but share them with the recorders and composers. This leaves them unable to control what streaming services can stream their music (Blacc et al. 2015)

The large music distributors may attempt to acquire some of the streaming media programs and turn them profitable, using their existing infrastructure and reputation while avoiding the high royalty costs.

This is exactly what Apple Music did when they bought Beats Music (and Beat Electronics) in 2014, for 3 billion worth in cash and stock (Apple press info, 2014). They eventually turned all of the Beats Music subscriptions in Apple Music subscriptions, when they launched in June 2015. With it, they also acquired some licenses, such as the ones on Dr. Dre's songs. But besides these few licenses, Apple Music still has to pay the distributors for playing their songs. So while Apple has used the existing infrastructure and reputation of Beats Music, they most likely still have to pay the full price for most of the royalties. If a music distributor would acquire a streaming service, they would receive the full price for their royalties. A merger would cut out the middle man, meaning that the whole 10$ a month per subscriber would go to the license holder instead of having to share with the likes of Spotify and Apple Music. The risk of cannibalization, where distributors would see decreased sales per album or song due to streaming, seems extremely low, especially since streaming has become a large market with several large players in it. However, the large distributors would only have their own music and would still have to pay other distributors for additional songs. Also, the music streaming market has grown much the last few years, with some players at 80 million subscribers. These factors makes it hard for the music distributors to acquire an existing music streaming service, one that would either be very expensive or struggling, with only their own music as an advantage over other sites. Blocking the use of their own music on other streaming services, in order to lure subscribers to the own service, may backfire and lead to a loss in sales and increased piracy. So this option may have been very profitable several years ago, but seems very unlikely at the moment.

The large music distributors may launch their own streaming programs in order to compete with the existing music streaming programs, and to avoid any possible acquisition costs.

The disadvantages of acquisition also apply to a de-novo scenario: the distributor can only receive full amounts for their own music and will struggle to acquire subscribers, in this scenario even more than in acquisitions. And though billion dollar acquisition costs can be avoided, setting up a music streaming service will cost time and needs a lot of expertise. Expertise may be lacking at music distributors, but can be bought/hired. Time is not an ally however, as Spotify, Pandora, Apple Music, and others, keep growing and will saturate the market at a certain point. Saturated markets often lead to increased competition, lower prices and lower profits when one assumes that the players do not collude in any way. These possible reduced profits in the future make that this option is a very unlikely one.

The large music distributors could enter the music streaming market together, creating an oligopoly in the music streaming business

Theoretically, this is the best option that the music distributors have if they wish to enter the music streaming market. Acquisition or startup costs can be shared, just like knowledge. But the best thing about this option is the possibility to have a huge library of music and their rights especially when the 'big three' join hands. They own almost 90% of all music rights worldwide, so they can acquire almost the full amount for their royalties, without having to share with any other streaming services. This also gives them the option to block other streaming services from using their music in the future, pushing them out of business. The major problem of this option, however, is that authorities may consider it to be collusion, as a joined streaming service can monopolize the music streaming market. So this option will likely only work if (some of) the competition is left untouched, or no more than 2 of the big companies would cooperate in such a project, like what happens for the VEVO channel on YouTube. So while this option makes perfect sense business wise, it is very unlikely that it will ever occur due to

antitrust laws.

Of interesting note is that Warner, Sony and Universal own a minority of Spotify (not of other streaming services), with several percent of Spotify owned by each of the companies. With Spotify still in the investment and growing stage, the goals of the distributors are not to make easy profits. Most likely they are interested in gaining a foothold in the music streaming industry and perhaps in investigating whether music hits can be predicted using big data. With big data, people theorize that it can be predicted what consumers would like to hear, based on the amount of streams songs get. For example: if songs lasting 3-4 minutes get double the amount of streams on average than songs lasting more than 4 minutes or less than 2 minutes do, a successful song needs to be 3-4 minutes long. More data can be used, such as male vs. female voices, pianos vs. synthesizers and so on and so forth. All these data should predict whether a song will be streamed a lot or not. The only problem so far are the results: songs from certain artists that should have been there biggest hits were in fact one of their least successful ones (Interview Menno Timmerman). So while the algorithms are still in development, the results so far have been less than promising.

Even though some artists will find ways to circumvent the use of music distributors when producing or distributing their music, the majority will remain under contract with the current music distributors

The music streaming services do not have the limits on 'shelf space' that physical stores have, meaning that the amount of music they can offer is theoretically limitless. Artists can have their music put on the streaming services or even upload their music themselves on YouTube, without the help of a music distributor. However, with libraries of millions of songs, it is very hard for artists to stand out with their music. So unless they are either extremely lucky, operate is a niche market, or know a way to do their

own successful marketing, musicians will still need music distributors in order to have consumers listen to their music when they are not well known. For famous artists, this may be different. Musicians or groups that have sold millions of albums do not lack attention and can gather enough publicity on their name alone. For them the question is whether the distributors can get them so much extra sales and plays or live performances that the benefits of that offset the costs of using a distributor. So far, no famous artists have stopped working with distributors. This option is therefore not likely to happen in the (near) future. (Interview Menno Timmerman)(Interview Paul Zijlstra)

The more famous artists will force the music distributors in accepting 360-deals, reducing revenues for the music distributors

360-deals are currently signed with almost all musicians, though many of these contracts are not as well known as the Robbie Williams deal. But in almost every signed contract, there is a certain percentage of live performance revenues that goes to the distributor. The idea behind this is that the distributor has invested in the artist and his/her music, which the artist uses in the live performances. In the 80'ies and 90'ies of the 20th century, the huge revenues of the music distributors made these fees quite irrelevant. But the reduced revenues of the last 15-20 years changed that, shifting the focus towards other revenue streams. The percentages and the (amount of) rights and licenses the musicians have to give or sell to the distributors differ: some musicians have to 'pay back' their investment, while others can get huge fees upfront for future performances. So in fact, these deals have increased revenues for the distributors rather than reduce them, compensating for lower revenues generated from album and single sales. But these deals are not fundamentally changing the game for the music distributors. (Interview Menno Timmerman)(Interview Paul Zijlstra)

P5: A new emerging player or technology will fundamentally change the music industry, which could afflict the music distributors negatively

As said earlier, big bang disruptions are amost impossible to predict beforehand. And also when they are just getting started. So this proposition is difficult to answer. But currently, there are no signs that point to something happening with a huge impact on the industry. There are smaller changes happening: TV series have become more popular the last years, something Netflix and HBO are very well aware off. This makes for an interesting revenue stream and marketing opportunity for musicians, who would like to have their songs put in these series. TV series usually take longer than movies, so there is more potential for musicians in these than in movies. Reduced payments from streaming services, as opposed to album sales in the past, also makes musicians more interested in having their music put in commercials. But while these streams of revenues are very interesting for some, they do not change the music industry fundamentally. There have also not been any news bulletins, mergers or other situations that indicate a fundamental change for the industry or the distributor. (Interview Menno Timmerman)(Interview Paul Zijlstra)

Conclusion and Discussion

The main question of this thesis was what options music distributors have to maintain or improve both their revenue and profit models in view of streaming technology and emerging new players in the music industry. The financials of the large music distributors show that the companies have been able to maintain their revenues and profits in the last years. The emerging new players in the market, currently the music streaming services with their respective business models, have not decreased revenues and profits, but have compensated for the losses suffered through piracy. However, the higher revenues gained from streaming have decreased physical sales for the distributors, meaning they now earn much more money with separate songs as opposed to full albums in the years before the digital revolution. And the albums where one of the most important reasons for the high revenues. The music streaming services, despite the heavy losses they incurred the last few years, seem to have business models than can turn in profits in the near future (and still attract enough money from positive investors), meaning they are here to stay. So a return to sales based mostly on albums instead of singles will probably not happen. It is unlikely that the music distributors will enter the music streaming market on a large scale in the (near) future, despite the largest three distributors owning part of Spotify. They are aware of the large presence and influence the music streaming services can have, so they are involved and cooperative at this point. Whether it will be possible to predict good selling songs beforehand using big data generated by streaming services is questionable. If it will ever be possible, this might be a big improvement for the distributors that have this information, although this would only work well if it increases overall music consumption. But is does not seem likely that this will be a new big bang development for the industry. And it does not look like there will be another development or game-changing technology in the industry soon, although those are notoriously difficult to predict. So while it will be very difficult, if not impossible, for the music distributors to increase their current

revenues and profits, they are able to maintain their current profits and revenues. Artists will still need them for marketing and promotion even if they are independent in the production process. So while the distributors do not have options to improve their current profit and revenue models, except for mergers perhaps, they still create value for consumers and artists, meaning they are in a good place to maintain their current operations. And they can keep exploring new developments and technologies so as to not miss out on them. For most people, the consumers, little will change the coming years. Consumers have benefited greatly from cheap music since the digital revolution, with cheap or even free streaming services they can use. And there is always the threat of piracy if music distributors or streaming services decide to increase the costs for consumers to much for their liking. So in conclusion, I believe the music industry to be an industry where consumers wil keep enjoying their cheap and diverse music, some streaming services will stay in business and be profitable despite heavy competition, artists will still need the music distributors if they want to make a good living and where the music distributors will stay in business, not returning to the golden age in the 20th _{century but operating stable and still} profitable.

Literature: scientific papers

- Afuah, A. 2004. Business Models: A Strategic Management Approach. New York: McGraw-Hill/Irwin, pp. 67-69

- Alexander PJ, Peer-to-Peer File Sharing: The Case of the Music Recording Industry, Review of Industrial

Organization, March 2002, Volume 20, Issue 2, pp 151-161

- Arakji RY, Lang KR, Digital consumer networks and producer-consumer collaboration: innovation and product development in the video game industry, J of Management Information Systems, Volume 24 Issue 2 2007 - Barney J, Firm resources and sustained competitive advantage, Journal of Management, March 1991, vol 17 no.1 pp

99-120

- Blacc A, Manta ID, Olson DS,, 2015, A Sustainable Music Industry for the 21st Century, Cornell Law Review Online, Forthcoming, Boston College Law School Legal Studies Research Paper No. 383, Hofstra Univ. Legal Studies Research Paper

- Blanning TCW, The triumph of music: the rise of composers, musicians and their art. Cambridge, Mass.: Harvard University Press, 2008

- Bockstedt J, Kauffman RJ, Riggins FJ, The move to artist-led online music distribution: Explaining structural changes in the digital music market, System Sciences 2005, HICCS' 05, Proceedings of the 38th _{annual Hawaii} International Conference

- Bourreau M, Moreau F, Gensollen M, The Digitization of the Recorded Music Industry: Impact on Business Models and Scenarios of Evolution, Telecom ParisTech Working Paper No. ESS-08-01

- Bower JL, Christensen CM, Disruptive technologies: catching the waves, Harvard Business Review, Feb 1995 Vol. 73 Issue 1 p43-53

- Clemons EK, Gu B, Lang KR, Newly vulnerable markets in an age of pure information products: an analysis of online music and online news, Journal of Management Information Systems, Volume 19, Issue 3, 2002

- Collins C, Young S, Beyond 2.0: the future of music. 2014, Equinox Publishing

- Datta H, Foubert B, Van Heerde HJ, The challenge of retaining customers acquired with free trials, Journal of Marketing Research (JMR). Apr2015, Vol. 52 Issue 2, p217-234.

- Downes L, Nunes PF, Big-bang disruption, Harvard Business Review, Mar 2013 Vol. 91 Issue 3, p44-56

- Gopal RD, Sanders GL, Bhattacharjee S, Agrawal M, Wagner SC, A Behavioral Model of Digital Music Piracy, Journal of Organizational Computing and Electronic Commerce, Volume 14, Issue 2, 2004, pages 89-105

- Lorenzen M, Frederiksen L, The management of projects and product experimentation: examples from the music industry, European Management Review Volume 2 Issue 3, p198-211, 2005

- Klein CC, The music industry as a vehicle for economic analysis, The Journal of Economic Education, Volume 46 Issue 4, 2015

- B Koh, BPS Murthi, S Raghunathan, Shifting demand: online music piracy, physical music sales, and digital music sales, Journal of Organizational Computing and Electronic Commerce, Volume 24, Issue 4, 2014, p366-387 - Krueger AB, The economics of real superstars: the market for rock concerts in the material world, The Journal of

Labor Economics, Vol. 23, No. 1 (January 2005) pp. 1-30

- Kusek D, Leonhard G, Lindsay SG, The future of music: manifesto for the digital music revolution, Ed. Susan Gedutis Lindsay. Boston: Berklee Press, 2005.

- Marshall L, The effects of piracy upon the music industry: A case study of bootlegging, Media Culture Society March 2004 vol. 26 no. 2 163-181

- Porter ME, How competitive forces shape strategy, Harvard Business Review 1979

- Porter ME and Millar VE, How information gives you competitive advantage, Harvard Business Review, Jul-Aug 1985,

p149-152

- Regner T and Barria JA, Do consumers pay voluntarily? The case of online music. Journal of Economic Behavior & Organization, 71(2), 395-406. doi:10.1016/j.jebo.2009.04.001

- Throsby D, The music industry in the new millennium: Global and local perspectives, Paper prepared for The Global Alliance for Cultural Diversity Division of Arts and Cultural Enterprise UNESCO, Paris. October 2002

- Vaccaro VL, Cohn DY, The evolution of business models and marketing strategies in the music industry, 2004, International journal on media management, 6(1-2), 46-58, doi:10.1080/14241277.2004.9669381,

- Wagner TM, Benlian A, Hess T, The Advertising Effect of Free--Do Free Basic Versions Promote Premium Versions within the Freemium Business Model of Music Services? System Sciences HICCS '13, Proceedings of the 46th annual Hawaii International Conference

- Zentner A, Online sales, internet use, file sharing, and the decline of retail music specialty stores. 2008, Information economics and policy, 20(3), 288-300. doi:10.1016/j.infoecopol.2008.06.006