THE DIGITAL REVOLUTION

The music industry (particularly the Big Five companies) faces tremendous challenges from the explosion of digital technology in the late-1990s, challenges that promise to transform every aspect of the industry. The outcome of battles over online music will create the template for industrial policies concerning online delivery of text, video, and other media, a convergence already indicated in the industrial trend to lump together writers, artists, musicians, actors, and others as content providers. The following section examines key issues in the controversy over online music and discusses some of their potential implications for the recording industry.

The Napster Imbroglio

The earliest form of digital music delivery was the WAV standard; 3-minute songs in this format, however, required hours to download. In 1987, the Motion Picture Experts Group (MPEG, a branch of the Geneva-based Inter- national Organization for Standardization) developed new digital com- pression software. The most powerful version, MPEG-1 Layer 3 (MP3), could compress a 40-megabyte file to one tenth of its original size. Modem

ROTHENBUHLER AND McCOURT

speeds also increased, and recordings could now be downloaded easily onto the hard drives of home computers. However, MP3 developed outside of the Big Five’s control and offered no intrinsic protections against copy- ing. MP3s therefore threatened the music industry by holding out the “pos- sibility of a business model that links artists directly to consumers, bypassing the record companies completely” (Garofalo, 1999, p. 349).

Although MP3 undoubtedly will be succeeded by systems that afford greater possibilities for copy protection, it currently has a momentum that diminishes chances for the immediate adoption of a different format. The domestic recording industry claims to lose $300 million per year to pirate recordings; a report prepared for the recording industry predicted that by 2002, an estimated 16% of all U.S. music sales, or $985 million, would be lost to online piracy (Foege, 2000). The Big Five focused their mounting con- cerns about piracy in all formats on Napster, which was released on the Internet in August 1999. Napster functioned as a music search engine that linked participants to a huge and constantly updated library of user-provided MP3s. Its key feature was an online database of song titles and performers, searchable by keyword. Napster ’s brokered architecture effectively coordinated users and increased search effectiveness, and its interface was highly user-friendly. Between February and August 2000, the number of Napster users rose from 1.1 million to 6.7 million, making it the fastest growing software application ever recorded.

No sooner had Napster become a “killer app” than legal woes beset the company. The RIAA filed suit against Napster on December 7, 1999, claim- ing that the free service cut into sales of CDs. Their case against Napster turned on the fact that although it did not generate revenue, the service supplied users with software and provided a brokering service that man- aged a real-time index of available music files. This combination of prod- ucts and services, the RIAA argued, effectively turned Napster into a music piracy service. Napster’s defenders claimed that its users enjoyed First Amendment protection, so the state could not enforce a prior restraint on the speech of Napster’s user and publishers. Its attorneys also argued that the service’s “substantial, non-infringing uses” included allowing users to sample new music and “space-shift” their collections between delivery systems like CD’s and hard drives (Gomes, 2000; M. Lewis, 2000).

In late-July 2000, Federal judge Marilyn Patel ordered an injunction against Napster, finding that the service was used primarily to download copyrighted music and rejecting Napster’s arguments. In February 2001, a three-judge panel unanimously upheld the injunction, and Napster soon began filtering its system to block copyrighted material before shutting down altogether. Despite the RIAA’s claims that Napster-driven piracy was eating into profits, recorded music sales in the United States reached an all-time high of 785.1 million units in 2000, up 4% from 1999. The RIAA

11. ECONOMICS OF THE RECORDING INDUSTRY

claimed that sales of CD singles dropped 39% in 2000 and inferred that Napster was to blame, yet fewer CD singles were released as the industry cut production. Some market research suggests that users used Napster not primarily to “steal” music, but to sample it before purchase. Users also were drawn to the huge array of music it presented, the obscure as well as the popular—a vast catalog (including out-of-print material) that was other- wise inaccessible.

Despite Napster’s defeat, Internet trading of music files continued un- abated through a number of other services, which, lacking Napster’s cen- trality, were virtually impossible to police. The recording industry has attempted to reassert its control through a variety of means, including legislation, buyouts, and digital rights management technologies. Each of these are discussed in turn.

Legislation

The Big Five have been instrumental in recent legislation concerning intellectual property. In addition to the Audio Home Recording Act of 1992, the Digital Performance Rights in Sound Recordings Act of 1995 gave the own- ers of recordings (i.e., the record companies) exclusive control over their music in online web casts. In contrast, radio stations have freedom to use music as they wish after acquiring a license from songwriters’ organiza- tions . The two most important legislative acts affecting the recording industry, however, were passed in 1998: the Sonny Bono Term Extension Act and the Digital Millennium Copyright Act (DMCA). In the former, Con- gress, responding to industry pressure, extended copyright protection for an additional 20 years. Authored works are covered for the life of the author plus 70 years, whereas corporate-owned “works for hire,” such as recordings, are covered for 95 years. Section 1201(a) of the DMCAmade it illegal to circumvent copy-protection technologies; the purpose of bypass is immate- rial. The DMCA essentially eliminated “fair use” provisions of the 1976 Federal Copyright Act, dismissing the tenet that we buy the right to make unlimited copies for personal use after purchasing an original copy of a recording. The DMCA also treated Internet service providers and telecom- munications networks as publishers, rather than common carriers, with the intent of forcing these networks to bar their users from sharing copyrighted material (Gomes, 2001).

Buyouts

The Big Five have privately hedged their bets through mergers and acquisi- tions that would allow file-sharing under their exclusive control. Shortly after the Napster decision, Bertelsmann broke ranks with the other major

ROTHENBUHLER AND McCOURT

record companies on October 31, 2000, and announced that it would loan Napster $50 million to develop a secure file-sharing system that would “preserve the Napster experience” while compensating copyright holders. Bertelsmann was attracted by Napster’s corporate identity, tangible assets, and software (including the protocol and interface). In exchange, Bertelsmann retained the right to take a 58% interest in Napster when the new service is developed (Gomes, 2000).

Bertelsmann’s actions regarding Napster follow the example of the Musicbank storage locker service, which obtained licenses for content from Universal, Warner, and Bertelsmann only after granting these companies an equity stake. The case of MP3.com is also instructive in this regard. MP3.com’s stock was valued as high as $63.61 before the company was hit by a barrage of copyright infringement lawsuits from artists, publishers, and record labels against its MyMP3.com storage locker. In May, 2001, 7 months after winning a $54.3 million judgment from the company, Vivendi purchased MP3.com for $372 million. (MP3.com had previously settled copyright infringement claims with the other four major record companies for $20 million each). Vivendi offered $5 per share for MP3.com’s stock, which had traded for only $3.01 per share before the acquisition was announced (Sorkin, 2001). Despite its legal liabilities, MP3.com was attractive to Vivendi because it was one of the few firms with the technological infra- structure in place to operate a large-scale online distribution service.

The Big Five repeatedly used high-profile lawsuits to deter venture capi- talists from providing second- and third-round funding to Internet startup companies. They then offer funding and content licenses to these startups in exchange for equity shares. The Big Five can thus acquire these opera- tions below their fair market value and also save research and development costs. Most importantly, they can thwart the creation of independent distribution systems (McCourt & Burkart, 2003).

Digital Rights Management

Afinal strategy for dealing with unauthorized digital music distribution in- volves Digital Rights Management (DRM) technologies, which “lock up” content through “trusted systems” in which copy protection is built into every component sold—the operating system, the artifact, and the player. An early attempt at creating a universal DRM system was proposed by the Se- cure Digital Music Initiative (SDMI), which was formed in December, 1998 by a consortium of record companies and hardware and software manufac- turers. SDMI’s 200 members included AOL, AT&T, IBM, Microsoft, Matsushita, Sony, RealNetworks, Liquid Audio, ASCAP, Intel, and Napster (no consumer or civil rights groups were represented). Based on water- marking technology, SDMI’s system was intended to serve as a gate

11. ECONOMICS OF THE RECORDING INDUSTRY

through which content must pass. The system enabled time limits on use, restricted the potential number of copies that the purchaser can make from an authenticated original, and allowed protected content to be traced back to the original purchaser.

However, development of the SDMI standard lagged far behind schedule. Its members had highly divergent and often antagonistic interests, which were aggravated by a problem intrinsic to software development: Every protection scheme can be broken. SDMI dissolved in early-2002, yet DRM may become a fait accompli. Such technology could become required by law, as was the case with SCMS. Or, equally likely, DRM could be im- posed through agreements between hardware and software divisions of colluding companies. DRM would impose new costs on consumers by rendering existing formats and hardware obsolete. It also would defeat one of the principles of intellectual property most nettlesome to corporate interests: Although copyright is designed to cover works for a limited amount of time, the incorporation of DRM into distribution would copy-protect them forever.

Implications for the Future Artists and Record Companies.

Much has been made about the Internet’s possibilities for artists to circumvent the control of record companies. The record industry rations an artist’s material in marketable doses to optimize supply and demand, which is additionally problematized by the existing preselection system. Certainly, the Internet affords artists greater flexibility in presentation, as digital technology allows for infinitely varied packaging and releasing. A release could contain 1 or 100 songs, and any schedule for release is possible. Performers could release a song the day it is recorded and release collections in standard and advanced packages, akin to a “director’s cut,” with varied price structures. Several artists, such as Prince and Todd Rundgren, presently use a fan-based patronage model, with recordings, concert tickets, online chats and additional materials available to subscribers through a tiered pricing system. This direct avail- ability to consumers negates the need for marketing subsidized by record companies; an artist who nets 50%, for example, on direct sales of 70,000 CDs earns more revenue than an artist who earns 10% royalties on 300,000 sales—not even counting deductions for production costs.

Record companies are likely to experience a faster diffusion of informa- tion and greater volatility of demand, as new genres and performers emerge and fade rapidly. These companies will have to maintain more flexible links with independent labels, which tend to follow new market devel- opments more closely, and organize themselves in a more flexible fashion (Dolfsma, 2000). The Internet will also provide record companies with de-

ROTHENBUHLER AND McCOURT

tailed consumer profiles and the ability to track “hits,” or downloads, for individual artists. The Internet is likely to produce independence from the Big Five at the extremes, with a growing disparity between stars who culti- vate their fan bases and fledgling performers who must essentially give away their recordings to promote interest. Yet promotion, which is the Big Five’s metier, will remain particularly important as choices proliferate, and traditional marketing will still be needed to break new artists or recordings.

Subscriptions. The Big Five may find a subscription model to be most lucrative, whereby users pay a flat monthly fee to access record company catalogs via computers, fixed and portable stereos, cell phones, and “Internet appliances.” In April, 2001 two subscription systems were announced: Duet, a project of Universal, Sony, and Yahoo; and MusicNet (now PressPlay), comprised of BMG, Warner, EMI, AOL, and RealNet- works. Unlike onetime sales, subscriptions generate steady cash flow and provide a convenient benchmark by which to measure growth. Licensing, rather than sale, also provides a direct link between vendor and purchaser that makes it easier to enforce limitations on use. Because subscriptions usually are paid in advance, they avoid the volatility of retail sales or pay-per-play. Subscriptions can produce more revenue than would sales from infrequent users, while encouraging increased use among heavier users, and allow the provider to charge higher rates to advertisers (Meyers, 2001). These companies can also harness a growing collection of customer databases derived from Web activity to reduce marketing uncertainty and provide revenues through resale to other vendors (Gandy, 1993).

Subscriptions present new challenges, however. Prices would need to make up in volume what is lost in profitability, suggesting cost pressures and even price wars among music services. Subscriptions also would pe- nalize chain music stores and retail outlets, which now account for 80% of sales in the popular music market. Record company consortiums may try to offer their own subscription models, but ultimately will have to license their catalogs to each other to attract the largest number of users. Yet licensing content between the Big Five invites antitrust action, and subscription services offered by combinations of the Big Five to date have been largely unsuccessful.

Running a successful subscription service would require the record companies to operate more like catalogs, magazines, or radio stations, and very little in their history has prepared them for this model. Both their infra- structure and their established methods of doing business are built on mass producing, distributing, promoting, and selling mechanical copies of recordings. Almost everything in the business is built around the number of physical copies sold. The recording industry has become adept over the years at maintaining a catalog, but attracting and holding attention via cor-

11. ECONOMICS OF THE RECORDING INDUSTRY

porate branding and providing reliable service directly to consumers are all foreign ideas from another world of business.

THE DIGITAL REVOLUTION