26 January, 2001

Impacts of Digital Distribution on the Music Industry

 

Introduction

The music industry has been through a tremendous cycle of discovery and change over the past year as both industry players and consumers of music have become increasingly aware of a precipitous trend that will revolutionise the business &endash; the transition of music distribution from physical to digital channels.

Music has been digitally recorded and mastered for quite a while now, and today's most popular format, the CD, is of course a high quality digital rendering. However a natural corollary of the digitisation of music coupled with the growth of the internet is that distribution of music will also transition from physical to digital, resulting in a transformation of the industry as we know it.

While digital distribution (involving transmission of files via network rather than in a physical form) is widely accepted as the end game, neither the industry nor expert observers are clear as to when this will happen, how quickly it will happen, and how it will affect the industry. Clearly digital distribution has started having an impact already, but the longer term effects of this phenomenon are unclear. In this paper, Durlacher outlines its views on digital distribution, examines implications for consumers, and projects the impacts that this change will have on the industry value chain.

 

Summary Hypothesis

Over the next few years we will see steady growth in online purchases of music &endash; the US online music market is projected to grow from $836 million in 2000 to $4.219 billion in 2004, and the European online music market from $140 million in 2000 to $1.932 billion in 2004. However, the majority of these online purchases will comprise online ordering with physical delivery (e.g. of CD's) rather than online ordering with digital delivery (e.g. via internet).

Digital distribution will continue to remain a very small part of the music market for the next few years. By 2004 we expect less than 2% of total European music spend to be for digital delivery. Adoption of digital distribution will be held back by a number of key inhibiting factors. However, these inhibitors will gradually be eroded as the market evolves, and digital distribution will see its most dramatic growth beyond 2004 (detailed market forecasts are provided later in this report).

In the longer term, we believe that digital distribution will become the mainstream (dominant) form of distribution for music as distribution is driven by the desire to reduce costs (while maintaining the integrity of the supply chain). Music is inherently an information product rather than a physical one, and, as human behaviour, industry structure and technological capabilities evolve, digital distribution becomes the most appropriate distribution channel for music.

A number of early digital distribution models are now being tried, including owners of intellectual property (i.e. artists and record labels) communicating directly with customer through websites, and new online music sites attempting to discover and develop bands bypassing the record labels. However, we believe that the emerging digital music value chain will see the emergence of new wholesale intermediaries that interface between the IPR owners and consumer-facing sites. These intermediaries will manage rights and royalties and will provide, in some cases, white label (unbranded) infrastructure and hosting services for the consumer-facing sites. In general, the vast majority of mainstream music will continue to be owned by the major record labels. While a few of the most popular artists may succeed in becoming independent of record labels and interfacing directly with consumers through their website, broadly, IPR owners will not succeed in distributing directly to consumers themselves. We do not believe new music sites will succeed in bypassing major record labels except in a few niche markets.

We believe that over time the industry will see a shift in the relationship between artists and record labels. Currently, record labels provide a bundled array of services including marketing, PR, distribution etc. We believe that distribution will become increasingly commoditised and will transition towards the independent wholesale intermediaries, leaving the record labels primarily as marketing and PR houses which may be contractually engaged by the artists (rather then the other way round).

While consumer uptake of digital distribution will be slow over the next few years, there are immediate opportunities for digital distribution to business end-users, including stores, broadcasters, entertainment venues and production studios. In addition we believe there are significant business opportunities in the provision of infrastructure for digital music, including conversion, compression, watermarking, encryption, digital rights management, payment, music discovery, value-added information and services, content distribution and ASP type services including archiving and library management.

A key consumer market opportunity for digital music will be provided by the growth in functionality and storage capacity on mobile devices. While music may not take advantage of mobile phone network capacity for some time, there are immediate opportunities to combine the functions of music player and phone/PDA into a single device and to utilise bluetooth or removable storage media for transfer of files to these devices.

Capabilities that are available to digital music consumers today, including the ability to 'rip' files from a CD, store in MP3 format, and distribute freely via email, websites and peer to peer networks, will always be available to them. It will not be possible to create and impose new standards that make these capabilities impossible, and we believe that the secure digital music initiative (SDMI) will fail in this respect. In addition, Durlacher's recent UK residential internet user survey demonstrates that consumers have few qualms about infringing copyrights through electronic distribution (see survey question below). This creates a particular challenge for the industry. By definition, any new standards or services created to protect copyright owners will be restrictive compared to existing MP3 file exchange. There is, therefore, only one way in which to encourage users to adopt newer protective standards and services, and that is by providing sufficient quality, convenience and value to drive users to the new systems voluntarily.

In Durlacher's recent survey of 3,800 residential households we found that 17% of all UK internet users have downloaded MP3 files downloaded from the internet. We also asked the following question to UK internet users:

Q: Would you have any qualms or concerns about downloading music without paying for it?

59% 'No, definitely not'

'13% Yes, under certain circumstances'

'21% Yes, definitely'

Historically, new music formats have been defined, developed and introduced to the consumer by the industry. Now, for the first time, we are witnessing the evolution of a music industry that is consumer-driven rather than industry led. It is the consumer that is adopting the MP3 format with such alacricity that the industry is being forced to embrace it. This phenomenon takes some getting used to &endash; in fact the industry still appears to hold the belief that it will be able to create and introduce the next standard, which will supersede and ultimately eliminate MP3. We do not believe this to be possible &endash; if there is another standard to supersede MP3, it will do so only because it is embraced and driven by the consumer on account of the increased functionality, convenience or value it provides.

 

The current music value chain

Development In the music industry, record labels will actively seek to sign up bands and artists on long term exclusive contracts. The key to success in development is to spot talent and to sign it up early. Record labels spend considerable amounts in the development function in order to stay ahead of the game. Production Production is relatively cheap in the music industry, where the cost of digital recording equipment and production of CDs is falling rapidly. It is now possible for artists and bands to produce music (on CD) in their own homes. Major labels will finance production and provide advances to artists. Sales and Marketing Major record labels hold power in the sales and marketing process through their established relationships with music stores, radio stations, press and media. Major record labels will determine the amount of promotional spend on a given album / single based on its potential popularity and profitability. Distribution Majors typically have a global network of branch offices to handle the sales, marketing and distribution process, while independent labels would need to license to local distributors. Major record labels will typically do the marketing and licensing themselves and may outsource the physical distribution process. Wholesale While distribution companies will typically sell directly to the big retail chains and to some of the smaller chains, some small shops will opt to purchase from a wholesaler (so they are not tied to purchasing minimum amounts from the distributor). Retail Distribution to record stores. Retailers put in orders to the wholesalers as and when albums and singles are required. In the UK the retail chains are dominated by HMV, Virgin/Our Price/Smiths, Tower etc. These chains account for about 80% of the market.

• In the music industry, artists are typically signed on for long term contracts. This is because popularity in the music industry tends to be artist-driven (i.e. if a few tracks from a given artist are popular it is likely that subsequent tracks will be popular too). This does not happen in the film industry, where artists are commissioned on a script by script basis.
• The major labels typically extend their activities right the way down the value chain from development to distribution. Typically major labels do not have their own retail outlets.

 

Royalty Structure

Rights

There are a variety of copyrights that are created when individual events occur in the music value chain. A simple representation of some of these rights is shown below. Once a right has been created there are numerous ways in which this right may be exploited.

• When an artist/writer composes a song/piece of music, a publishing copyright is created. This right represents ownership of the concept of the music itself, and is called a publishing right because in early days the concept was represented as a musical score which was published. Any subsequent exploitation of this track (whether by recording, performance or broadcast) requires a payment to the owner of the owner of the publishing rights.
• Once a track is performed in public (whether by broadcast of the performance, recording, sale, or re-performance), subsequent performances require a payment to the owner of the performing right.
• When a track is recorded, payment must be made to the owner of the mechanical rights. Subsequent exploitation of this recording (whether by broadcast, sale or reproduction) requires a license from the owner of the performing rights.
• Typically professional artists will request royalties for exploitation of rights. Each country has collection agencies which manage the collection and distribution of royalties on behalf of the industry. In the UK this is PRS (the performing rights society)[www.prs.co.uk] and MCPS ( the mechanical copyright protection society)[www.mcps.co.uk].

Exploitation of rights

• A copyright owner has the option of carving up the rights in multiple ways in order to create and price different licenses for their exploitation. One representation of how the components of these rights may be separated is shown below:

So, for example, a local distributor in Germany may be assigned the right to distribute the CD version and cassette version of a given track for a five year period in the German market. Labels that wish to minimise the risk of producing a new track from a known artist may opt to pre-sell the rights before the track has been created. Alternatively, major labels will typically absorb higher risk by carving up the rights only after the track has been produced &endash; a good track will produce a much higher return in this way.

 

Distribution of Royalties

The following diagram shows how the cost of a typical CD is split among the various players in the value chain.

Most licensing and royalty arrangements are made on the basis of the published price to the dealer. The retailer will typically charge a margin of 30% on top of this (though this varies widely).

The effect of digital distribution on the size of the music market

While the music industry is vigorously discussing the threats and opportunities presented by digital distribution, one interesting question that is not asked too frequently is: "What will be the net effect of all these changes on the total size of the music industry?"

The global music business has maintained a relatively steady size of around $40 billion per year for the past few years. Historically, major new formats including the audio cassette and the CD have each increased the overall size of the business as they introduced better convenience or higher quality. In each case, these formats also drove a surge in demand as users replaced their existing collections in the new format. The music industry has historically been able to capture the full value of this increased demand as existing players were able to exert control over the value chain and over prices.

There is every reason to assume that the transition to digital distribution will once again increase demand and consumption of music overall. There are several reasons why consumption of music will increase:

• People will be able to find the music they want more easily (by lyrics, moods, genres, preferences etc).
• People will be able to acquire the music they want more easily (without requiring physical delivery).
• People will be able to acquire music immediately &endash; there is less of a time delay between the point at which user emotion generates desire for a track and the time at which the track is acquired and listened to. This will mean that a higher proportion of desires will be converted into acquisitions.
• There will be a wider choice of form factors. People will be able to acquire music more flexibly, whether as an individual track, as an album, as a personalised compilation or as a service packaged with value added extras.
• There will be a wider choice of value propositions &endash; payment models will allow pay per play, eat as much as you like, annual subscriptions to bands, 'radio stations' or genres, limited use licenses etc. This makes it more likely that there will be a suitable value proposition for your usage pattern.
• People will have access to more of the music they really enjoy, and hence will listen to it more frequently.
• People will be able to access their music from more places, including work, home, the car, and abroad (even if they are not carrying their collection with them).
• As music becomes more readily accessible, people's mode of consumption may change to one that includes higher usage of music as a background to other tasks (such as listening to Spinner.com while surfing).

Within the industry there has been an implicit assumption that the transition to a new distribution mechanism, coupled with increased consumption patterns, mean that the total size of the industry must inevitably rise. In fact at many conferences there is talk of the music industry being a $100 billion industry in a $40 billion body, with the hope that digital distribution will be the key that unlocks the industry's hidden potential. Digital distribution will enable the industry to more readily experiment with elements of the supply-demand curve, such as testing elasticity of demand at different price points etc. Outside of digital distribution there are other reasons to be hopeful including growing prosperity and awareness of consumers, particularly in under-saturated markets such as Asia, Eastern Europe, Latin America and Africa.

However the key question with regard to digital distribution is: "Will the music industry successfully capture the full value of increased music consumption?". The answers to this question are not so easy. In the emerging environment established players will no longer exert the same level of control over prices. Any new propositions presented by the music industry will need to be compelling in terms of functionality and convenience, but will also have to represent such good value that users will pay for it in preference to using existing file swapping mechanisms. It is likely that while consumption will increase, the average revenue generated per track consumed will fall dramatically (though not uniformly). Pricing pressure is brought about by several factors:

• Prices will need to be highly attractive in order to deter users from resorting to piracy (which is free and is likely to remain relatively consequence-free).
• The industry is likely to converge upon eat-as-much-as-you-like payment models to attract and retain users. Such price plans typically generate lower revenues per track than pay-per-use models.
• The economics of physical distribution mean that users are often forced to pay for complete albums or singles with superfluous filler tracks when purchasing a desired track. Digital distribution is likely to result in an unbundling of this package (particularly for certain genres). This means that people will increasingly pay for just the track they want rather than for the complete package.
• Mainstream labels are likely to face increased competition from a wider variety of artists and independent labels, as well as from user generated content. It is already known that the music market is fragmenting and that user tastes are becoming more eclectic. The internet will enable a wider variety of content creators to reach consumers. Many of these content creators will pitch their offerings at lower prices than mainstream content. This means that major labels may capture a smaller part of the music pie.
• Users will expect lower prices because digital distribution will initially have some perceived disadvantages (user will consider the 'switching cost' of adopting a new standard which will not be fully supported at first). In Durlacher's recent residential survey we asked UK internet users to "Consider the most recent music album or single that you purchased. What percentage cheaper would it have had to be for you to consider buying it as a digital download?". The majority of respondents said it would have to be more than 25% cheaper, with the mean response being 39.9% cheaper (this would mean that an £11.99 CD would need to be supplied online for £8.99 or less to attract the majority of users).

In the early days of digital distribution it has become clear that the music industry has not been able to capture the value of increased consumption. For example, studies show that as usage of Napster spread like wildfire across university dormitories in the US, students listened to more music but provided less revenue to the industry than before. While this is to be expected in the early days, the industry will need to come up with mature platforms and compelling business models very quickly if they are not to miss the boat on the revenue opportunities of digital distribution. A number of things must come into place for this to happen:

• A digital rights management and security architecture must emerge (and be widely adopted) that generates sufficient confidence within the industry to spur the availability of mainstream content in digitally distributed form. Content owners will need to converge upon a standard in this area.
• The industry must develop music services of sufficient quality, flexibility, convenience, and added value to drive consumers to use them in preference to piracy.
• The industry must develop a range of flexible payment models at a sufficiently compelling price point.

Consider the following equation:

So far, we have examined how volume of consumption is likely to increase, but unit prices are likely to fall dramatically. The third factor to take into account is the cost of distribution. Digital distribution will significantly reduce costs for the industry because:

• Digital distribution eliminates a number of intermediaries from the value chain, or replaces existing intermediaries (e.g. physical retailers) with intermediaries that have a lower cost base (e.g. online retailers).
• Digital distribution replaces the costs of packaging, production, logistics and physical distribution with lower cost electronic equivalents.
• Digital distribution coupled with increasing usage of the internet for rich media content consumption could reduce the cost of marketing to potential buyers.

Clearly the net effect of all these changes will depend on the relative speed and magnitude with which each effect occurs. It is conceivable that over several years the music industry could see increased consumption, lower prices resulting in reduced revenues, and reduced costs, which could mean that profitability is maintained. These effects are likely to occur only when digital distribution begins to have a significant impact (largely from 2005 onwards). For the following section on market forecasts, we have assumed that between now and 2005 the total size of the music industry will increase very slightly.

 

Medium term music market forecasts

A summary of our findings for the US and European markets are provided below. Full projections are tabulated in Appendix A

Music Market

• We estimate that the US online music market will reach $4.2 billion by 2004, of which $920 million will be digitally distributed.
• We believe that the European online music market will reach $1.9 billion by 2004, of which $284 million will be digitally distributed.
• Durlacher's projections are derived from internal analysis of evolving trends combined with data from Jupiter, IFPI, MBI World Report and others. We have assumed that the European market will continue to capture around 32% of the global music market (as it currently does, per IFPI) and that growth within Europe will be matched by growth in countries outside of Europe and the US.
• We have assumed that European uptake of online music purchases will lag two years behind the US, and conversion from online to digitally distributed music will follow about a year behind the US.
• Europe is taken to include the thirty-one European countries quantified in the IFPI World Sales factbook.
• The online music market is taken to comprise two parts: online purchases with physical distribution and online purchases with digital distribution.

Source: Durlacher analysis combined with data from Jupiter, Forrester and MBI World Report

Source: Durlacher analysis combined with data from Jupiter, Forrester and MBI World Report

Streaming vs. downloads

• Jupiter in its latest projections of July 2000 estimates that by 2005, around 65% of digitally distributed music revenues will be for subscription based, whereas 35% will originate from downloaded music. The remainder of the online music market consists of music ordered online and physically delivered (also known as a 'hybrid' purchases).
• Durlacher does not believe that there is a clear correlation between subscriptions and streaming, or between pay-per-play and downloads. i.e. different payment methods could be applied independent of whether the music is streamed or downloaded. In addition, the distinction between streaming and downloads will no longer remain clear because:

  1. With content distribution networks using local caching, streaming music will be streamed from closer and closer to the user.

  2. With the growth in home networks and local storage devices, music may be streamed within your house from a downloaded file on your own home server.

  3. With software such as Real Jukebox, streamed music is cached locally when played for future reference &endash; effectively meaning that music is both streamed and downloaded at the same time.

• We believe that a valid distinction will be to differentiate between pricing models. e.g., subscription, pay per play, limited use licenses, etc. However, it is too early to make a call on how revenues will divide as a variety of payment plans are yet to emerge.

 

Factors affecting adoption of digital distribution

We believe that take-up of digital distribution of music on the consumer side will remain very low for the next few years, but will accelerate rapidly thereafter. Key contributing factors include the following:

• User behaviour changes slowly
  People are slow to adjust their habits when a radical shift first begins to take hold. However, as a critical mass of people adopting the new trend emerges, network effects begin to take hold and people begin switching at an increasingly rapid rate. Witness this phenomenon in the adoption of mobile phones in the UK, which have been around since the early 1980's.
  
  
• Rollout of broadband connections will be slow
  Most consumers access the internet via modem today and, outside of the USA, most are subject to metered telephony rates. In many European countries, the telecommunications markets continue to be regulated by necessity (since the vast majority of infrastructure is still owned by dominant PTTs). These telcos do not have a strong incentive to roll out affordable broadband connection to consumers and, in the absence of regulatory changes, are likely to drag their heels in rolling out DSL. However, de-regulation is under way, and in some markets regulatory bodies have announced 'unbundling' of the local loop (see broadband section later). Additionally cable companies are beginning to roll out cable modems, which will increase the pace of competition in broadband access. Consumer lag and high price differentials with current ISP offerings will slow consumer adoption. Finally cost and the limitations of the core network mean that data rates are likely to be no more than about 500 kbit/s in the short term.
  
  
• Adequate copyright protection is not in place
  In the current market, it is extremely difficult for the music industry to control copying and unauthorised distribution of digital music files. Although secure formats (such as Liquid Audio and Windows Media) are available, songs currently released on CD can be insecure. Thus, any new track can be 'ripped' (uploaded and converted into MP3 format) from CD, and can be distributed in an uncontrolled fashion. While SDMI (the secure digital music initiative) has had relatively little impact on the market to date, we believe that the completed implementation of SDMI phase 2 (currently scheduled for the end of this year) will increase confidence. SDMI will not solve the problem of illegal copying. It may, however, create a security framework (including a method of identifying the origin of tracks) that provides the music industry with sufficient confidence to make their entire catalogues digitally available through selected partners. While copyright protection will not cease to be an issue, we believe it will no longer remain a key inhibiting factor.
  
  
• Conventional music equipment does not accept digital formats
  Although a plethora of portable MP3 players have emerged on the market, and PCs are being used to play music, these devices are primarily used by early adopters. Take-up of portable digital players (themselves limited by high flash ram costs), as well as CD burners, has been relatively slow. Mainstream music devices including hi-fi and midi systems, car audio and professional music equipment do not have the storage capabilities, or the requisite digital inputs to accept MP3 and other digital music formats. We believe this will change as mainstream labels begin to adopt digital music and as mainstream manufacturers embrace device convergence. We are already seeing hi-fi 'music server' components beginning to appear, and other audio systems will begin to accept digital music (a number of MP3 car audio players are already on the market). Sony is in the vanguard of convergence between consumer electronics and computers, and a future of networked devices incorporating digital storage, internet connections, and bluetooth is beginning to emerge.
  
  
• The quality of digital music is not high
  It is important to note that CD sound is itself an uncompressed digital format, and is widely regarded as a high quality benchmark for sound. However, most of today's compression technologies are "lossy", meaning that in compressing the file, some data is lost and the quality can be compromised. The MP3 format encompasses a range of quality levels that can approach CD quality. However, for file swapping and downloads, users currently use lower quality levels to optimise file size. We believe that as new formats are released in an environment of increasing bandwidth, quality levels will meet or exceed CD quality, which is more than sufficient for the human ear.
  
  
• Users are not willing to pay for digital music
  In the current environment users face a stark choice when accessing digital music:

  (1) Use MP3: Find a huge selection of tracks from every available artist including mainstream artists from a variety of locations on the internet, and download for free (arguably an illegal practice). Copy and distribute freely.

  (2) Use officially sanctioned formats: Find a very limited selection of a few artists from selected record label sites which charge a large one time fee for downloads using officially approved formats. Consume on limited devices in limited locations.

  We believe that the success of Napster has demonstrated huge latent demand for ready access to mainstream music. As record labels grow confident of copyright protection, and make all of their mainstream tracks readily available we believe that a range of alternative pricing plans will emerge, including pay per play, limited use licenses, subscription based models, eat as much as you like models etc. From this initial complexity a range of acceptable pricing models will emerge and we believe that users will be willing to pay for the freedom and convenience that digital formats will provide.

• Mainstream artists are not available because record labels face channel conflict
  Currently, record labels are making a few tracks digitally available on selected sites, but they are reluctant to make a wide range of mainstream artists available at competitive prices. This is because they are afraid of creating conflict with the expansive physical distribution networks they have nurtured over the years. Physical distribution is likely to remain the largest part of the market well into 2005, and hence, maintaining these physical distribution networks will remain important. Therefore, we believe that channel conflict will cause record labels to move slowly and unaggressively into digital distribution for the foreseeable future.
  
  
• It is hard to locate and download tracks online
  In the current environment it is actually quite difficult to locate and download the MP3 tracks you are looking for. MP3 search engines often reveal links to disused ftp/http servers/sites or limited availability ftp sites. Many MP3 ftp servers are also "ratio" servers that only permit downloads once a certain number of uploads have been made. The process has been made simpler by the introduction of file exchanging mechanisms (peer to peer networks) such as:

  (1) Client-server-client file exchanges (e.g. Napster) where the server co-ordinates the clients' access to other clients.

  (2) Client-multiple client file exchanges (e.g. Gnutella) where clients link directly to other clients and get access to the clients that they, in turn, are linked to.

  We believe that the emergence of legitimate support to the digital download industry from the mainstream labels will result in more comprehensive professionally managed sites as well as peer-to-peer networks that enable better location and download of music. Bertelsmann's agreement with Napster to develop a subscription based service is a step in this direction.

 

The long term game

Because of the rapidly changing nature of the technology environment, most industry outlooks and forecasts do not look beyond the four to five year horizon. Within this time horizon, we have outlined various reasons above why we believe that digital distribution of music will grow relatively slowly and will make up a small part of the music market.

However, we have also outlined why the principle market inhibitors will be steadily eroded over time. We strongly believe that digital distribution of music will become a high growth phenomenon of tremendous business significance beyond the four year horizon. There are a number of reasons why digital distribution of music is attractive:

• Music is an information product
  Music is inherently an information product rather than a physical product, and thus digital distribution is inherently more efficient. We believe it is inevitable that the music industry will be dominated by electronic distribution in the long term.
  
  
• Digital enables new products
  New kinds of products will emerge as a result of the changing economics of the distribution chain. Issuing a single track is uneconomic in the physical world (even 'singles' contain superfluous filler tracks to justify their value). However the single track is the most natural unit of transaction in the digital world - the consumer is usually paying for a given track.
  
  
• Digital distribution opens up opportunities for a wider range of artists
  Digital distribution is a very low cost distribution channel which opens up opportunities for a wider array of artists and independent labels (particularly those who cannot afford to sell into radio or make videos for new tracks/artists). The result is a proliferation of new artists (some of whom will use digital to build presence before being picked up by a label). While we believe that music will still be dominated by major labels and mainstream artists, we believe there will be a larger independent music scene and greater scope for individual consumers to be contributors of music content.
  
  
• Digital enables greater selection / availability
  Digital distribution of music not only enables national or international economies of scale for retailers and aggregators, but eliminates the need for holding of physical stock and the limitations that brings. Mainstream consumers will be looking for the widest range and high availability.
  
  
• Digital enables immediate gratification
  Instant delivery of requested music is clearly an advantage for the consumer. My.mp3.com and other music locker services are in the process of signing agreements with record labels (following lawsuits against my.mp3.com) to enable digital consumption of music as soon as a physical purchase has been made. While on the face of it this adds value to the physical purchase, in reality such customer exposure to digital delivery sets the foundation for a migration from physical to purely digital delivery.
  
  
• Digital reduces distributions costs
  Digital distribution lowers the cost of sale (CD, CD-replication, packaging, distribution are no longer costs). Cost advantages accrue to both distributor and consumer. Price plans that reduce the financial barrier to consumption of music could potentially open up latent demand and create opportunities to monetise back catalogue that is uneconomical in the physical world. Price will be a key driver for digital music. Many consumers are already unhappy with the price of CDs (hence the lawsuit by 28 US states against leading labels and retailers alleging CD price fixing).
  

  "Mi2N conducted a survey on retail pricing of music CDs in May 2000. The majority of participants indicated that today's pricing levels were not only unreasonable (74%), but excessive as compared to other forms of entertainment (70%). To reinforce the point, 40% stated that they would significantly increase their CD purchasing if prices were to drop to $9.99, while 45% indicated that they would marginally increase their purchasing. But an even greater measure of consumer frustration with CD prices is the fact that when some major labels announced that they would probably price digital downloads at levels comparable to CDs, 96% of respondents indicated that they would be unwilling to buy digital downloads of major recording artists' work at those levels."
  Sample size and methodology unknown.

• Digital enables flexible consumption and usage
  Music in digital format will allow people to more easily manage play lists (most people don't make mix tapes, but there is demand for customised play lists). The meta data that will be built around digital music databases will enable music discovery (through proactive recommendations based on music preferences) and a host of other new applications. Provided copyright concerns can be overcome, digital music can be e-mailed to friends, distributed or transferred to portable player, cars, or accessed from anywhere through centralised storage. It can also be processed, mixed, spliced, added to video content etc. more easily. Technologies such as X10's MP3 Anywhere and also Bluetooth will move MP3 files from PC to the hi-fi / Car Audio system helping to spur the market.

 

The Evolving Music Value Chain

Digital distribution to consumer facing businesses

We believe that the digital music distribution chain will evolve as follows:

(where the size of the ellipses is related to the number of players in the given space).

In the current, early stages of this market, some artists and labels are conducting trials through which they provide digital music directly to the consumer. However, the long term model will involve the emergence of a few B2B aggregators who will license distribution rights from a wide array of labels and artists. There will be some direct sales, but this will not constitute a large part of the market.

 

We believe this model will develop because:

• It is not feasible on a large scale for individual retailers and portals to negotiate digital distribution rights directly with artists and labels. Artists will wish to negotiate with a few large aggregators rather than 1000s of individual consumer facing sites.
• Consumers will want the widest selection of artists and labels on any given distribution site.
• E-tailers and portals will find it easier and more cost effective to outsource the digital distribution component to a B2B aggregator which will offer a plug and play solution for their websites.

There will only be a few aggregators because:

• Economies of scale will be important in this business
• Portals and retailers will want to do business with the aggregator that has distribution rights from the largest range of artists and labels
• The need for scale will drive industry consolidation

Winners will be those that have existing portal/retailer and artist/label relationships. It will be a strong advantage to have existing physical distribution infrastructure as we believe the migration to digital distribution will take place over a long period of time (4-10 years) and will occur slowly at first. It will also be beneficial to possess a large existing digital music database.

Aggregators can complement their services by providing a range of value added services around the database, including music search and discovery, aggregated usage data and perhaps even merchandise and other value added products around the basic purchase.

Today, record companies play a consolidated role combining several different functions including artist discovery, assumption of risk through advance payments to signed artists, recording, distribution, marketing etc. While in the physical world it made sense for all these activities to be bundled, in the digital environment this role may be disaggregated.

In the digital world, independent labels will be better able to reach customers, and some artists may decide to build independent and direct relationships with the market. However, the key role the record industry plays in financing bands up front and taking risk of failure will ensure that they have a continued if slightly reduced position of dominance in the industry. The instant customer feedback available in the digital environment will enable them to increase ROI on new albums.

Digital distribution to business end-users

Digital distribution to business customers (i.e. where the end user is a business) is compelling because:

• Digital provides functional advantages such as access to a vast library of music without local physical inventory.
• Digital allows intelligent searches using a variety of metadata
• Audio is relatively low bandwidth among rich media content types, and businesses are well served with high bandwidth connections, making the distribution of audio to businesses an immediate market opportunity.
• We are seeing increasing digitisation of the media production environment.
• Businesses are likely to be willing to pay significantly for the increased functionality and access that digital enables.
• Digital enables businesses to outsource more of the audio library management process that are non-core to their business such as indexing, archiving and retrieval of content.

We believe that the digital music distribution chain to business end users will evolve as follows:

In the business value chain there are far fewer customers than in the consumer value chain. The importance of having a few centralised aggregators is therefore proportionately less. The key factors that will drive business users to deal through an aggregator, rather than dealing directly with rights owners, will be the additional functionality and convenience (value added services) that the aggregator can offer. We believe that while digital distribution to consumer facing sites will grow very slowly, there are immediate opportunities for digital distribution of music to business customers including broadcasters and film producers (existing business), web sites, music stores, and radio stations (new business models).

 

Music on mobile devices

There are two distinct reasons for using devices such as mobile phones and PDAs for listening to music:

1. Convenience &endash; having multiple functionality in one device is simpler, saves weight and can save on overall cost if some electronics can be shared,

2. Connectivity &endash; the ability to access music or related information via mobile networks in a spontaneous and location independent manner.

Current GSM mobile phones have:

• a very limited data rate of rate of 9.6 kbit/s
• payment systems linked to time of usage
• limited functionality (processing power, memory)

These factors have, so far, limited the appeal of mobile phones as a platform for music consumption.

This is beginning to change as both (i) the network and (ii) the reception devices develop.

 

Changing Mobile Network Bandwidths:

GSM mobile (1999) 9.6 kbit/s

GPRS mobile (mid 2000 &endash; 2001) 100 kbit/s

GPRS - EDGE mobile (2002) 384 kbit/s

UMTS / 3G mobile (2003 &endash; 2004) 384 kbit/s - 2Mbit/s

 

The usage of music on mobiles

There will be a variety of transfer music to a mobile device:

File transfer through local connectivity:

Given the lower cost of internet over fixed links (compared to wireless) and the rapidly increasing storage capacity of portable devices (at a given price), local file transfer is likely to have significant attractions. File transfer can occur via:

• Wire connectivity
• Bluetooth
• Wireless LAN network
• Removable storage media (e.g. flash memory cards)
• File transfer over the air
• Transfer of music files over the air may occur via
• Mobile networks, including GSM, GPRS, EDGE and UMTS
• Digital audio broadcast (DAB), or 'digital radio'

Use of the mobile network provides the advantage of spontaneity and convenience but the disadvantage of relatively high cost and, in the short term, limited bandwidths. Mobile networks are a 'narrowcast' medium in that each individual receives their own private dedicated stream of data. This enables music to be delivered to individual requirements 'on demand', but has the disadvantage of being an inefficient means of utilising the network, particularly for mainstream content that most people will want to listen to. In addition we do not believe that connectivity (even with 3G / UMTS networks) will be of sufficient quality to enable uninterrupted live audio streaming in real world conditions. As we already know, the quality of mobile coverage varies as we move around, and may be coverage may be obscured by buildings, metal or tunnels. Voice calls on the move are already frequently interrupted, and users are unlikely to pay for unreliable mobile music services that depend on live streaming.

However, we are likely to see a significant expansion in the local storage and processing capacity of mobile devices as storage costs decrease and processing power continues to rise for a given cost. We see the emergence of a thick client model rather than a server-centric thin client model for mobile applications &endash; mobile devices will store data locally and will conduct the bulk of processing on the device, using the network sparingly when required. This enables a new form of mobile network usage where music may be trickled down to the device at low speed, stored in memory and consumed at full speed, even when disconnected. For example, a user may subscribe to the Top 40 chart tracks, which are trickled down to the device overnight using spare network capacity for local consumption on demand.

Digital Audio Broadcast is a particularly interesting way of getting rich media content to mobile devices. This channel (which was designed for digital radio and is optimised to function in moving vehicles) already has sufficient bandwidth and quality for live music, and is available almost nationwide (in most European countries). Being a broadcast medium, it is a particularly efficient way of getting mainstream content to the majority of users (for example, Top 40 tracks could be broadcast only once and stored on every mobile device for consumption on demand. Compare this to streaming these tracks individually to every mobile user via the mobile network). Content could also be constantly updated and would be then be available for access when desired. Some services could be focused on specific groupings for example information and entertainment targeted at the London commuter. A content management system would constantly update the stored material according to the preference of the individual user and the availability of network connections. Clearly, broadcast does not work efficiently for niche content that may only be required by a few users.

Payment models for mobile data will be based either on data volumes (rather than time connected) or on application/server specific pricing models. The initial killer application for mobile data will be email and text messaging, followed by corporate data access for business users. Entertainment (including games and music) is not seen to be a "driver category" in mobile usage short term, but is likely to expand considerably once a critical mass of users is reached on next generation networks.

Business models for music on mobiles

• The device will contain locally stored music files that may have been downloaded over the air or from devices in the home linked to fixed networks.
• The user will be able to access streaming audio files from radio stations, record companies, artists and other content sites. Continuous streaming via mobile networks is likely to remain unattractive due to cost issues and accessibility.
• The user will pay for data transferred rather than airtime, although these data charges may themselves be incorporated into the value-added price of the 'service' the user is using.
• The user may have a stored selection of favourites and a collection of licenses / subscriptions to access specific types of content. These preferences, favourites and licenses may be stored locally on the device or on one or more portals / aggregators on the network, making them accessible from anywhere.
• Peer to peer transfer of selected favourite tunes - " have you heard this?" - subject to adequate rights safeguards
• Certain developments will be governed by currently accepted norms. For example, since users do not currently pay for access to radio stations, they may be reluctant to pay for 'radio station' type services in the mobile world.

 

Jay Marathe
Head of Consulting
Durlacher

Please contact us for further information on Diurlacher's research, consulting, or corporate finance services
http://www.durlacher.com/
jmarathe@durlacher.com 020 7459 3641

 

APPENDIX A: MUSIC MARKET PROJECTIONS

US MUSIC MARKET

($ millions)	Total music market	Offline music market	Offline as % of music market	Online music market	Online as % of music market	Online with digital distribution	Digital as % of online music market	Digital as % of total music market	Online with physical distribution (hybrid)	Hybrid as % of online music market	Hybrid as % of total music market
1997	12,333	12,296	99.70%	37	0.30%	0	0.00%	0.00%	37	100.00%	0.30%
1998	13,818	13,666	98.90%	152	1.10%	0	0.00%	0.00%	152	100.00%	1.10%
1999	14,217	13,830	97.30%	387	2.70%	0	0.00%	0.00%	387	100.00%	2.70%
2000	15,421	14,585	94.60%	836	5.40%	9	1.10%	0.10%	826	98.80%	5.40%
2001	16,433	14,964	91.10%	1,469	8.90%	39	2.70%	0.20%	1,431	97.40%	8.70%
2002	17,402	15,143	87.00%	2,259	13.00%	151	6.70%	0.90%	2,109	93.40%	12.10%
2003	18,300	15,119	82.60%	3,181	17.40%	467	14.70%	2.60%	2,713	85.30%	14.80%
2004	19,170	14,951	78.00%	4,219	22.00%	920	21.80%	4.80%	3,299	78.20%	17.20%
2005	20,100	14,736	73.30%	5,364	26.70%	1,511	28.20%	7.50%	3,853	71.80%	19.20%

EUROPEAN MUSIC MARKET

($ millions)	Total music market	Offline music market	Offline as % of music market	Online music market	Online as % of music market	Online with digital distribution	Digital as % of online music market	Digital as % of total music market	Online with physical distribution (hybrid)	Hybrid as % of online music market	Hybrid as % of total music market
1997	12,224	12,224	100.00%	0	0.00%	0	0.00%	0.00%	0	100.00%	0.00%
1998	12,275	12,275	100.00%	0	0.00%	0	0.00%	0.00%	0	100.00%	0.00%
1999	12,359	12,297	99.50%	62	0.50%	0	0.00%	0.00%	62	100.00%	0.50%
2000	12,744	12,604	98.90%	140	1.10%	0	0.00%	0.00%	140	100.00%	1.10%
2001	13,386	13,024	97.30%	361	2.70%	4	1.10%	0.00%	357	98.90%	2.70%
2002	13,931	13,179	94.60%	752	5.40%	20	2.70%	0.10%	732	97.30%	5.30%
2003	14,413	13,130	91.10%	1,283	8.90%	86	6.70%	0.60%	1,197	93.30%	8.30%
2004	14,862	12,930	87.00%	1,932	13.00%	284	14.70%	1.90%	1,648	85.30%	11.10%
2005	15,408	12,727	82.60%	2,681	17.40%	584	21.80%	3.80%	2,097	78.20%	13.60%

Durlacher projections derived from internal analysis of evolving trends combined with market data supplied by Jupiter, Forrester, IFPI and MBI World Report. The online music market comprises two parts: "online with digital distribution" and "online with physical distribution" (or hybrid).