Rockonomics

by Alan B Krueger

  Amazon’s Steve Boom thinks streaming is not a winner-take-all technology, and that multiple streaming companies can co-exist without a single dominant player (that is, unlike Facebook in social media and Google in search engines).36 Such an assessment is correct only if the companies can distinguish themselves, perhaps by exploiting their monopoly on users’ listening histories to provide personalized service and other unique benefits to their customers.

  Marc Geiger conjectures that in the future streaming services will bundle movies, music, and other content.37 Currently, Spotify and Apple provide music streaming for monthly subscription payments, and Netflix provides movies for subscription payments; it would be more convenient for consumers if these separate streams merged. Geiger predicts that Netflix will buy Spotify and provide one-stop shopping for all entertainment needs. Amazon Prime is already in both the movie and music businesses and could become, well, the Amazon of bundled entertainment streaming. Others, such as Facebook and Google, could also compete in Geiger’s “bundled services revolution.” A key question for Geiger and his clients is how the revenue will be divided if we do see bundling of movies, music, and other entertainment services.

  Another development to watch is that streaming companies can attempt to compete with traditional record labels and publishers, much as Netflix has done with movie studios. Spotify has encouraged artists to post music on Spotify directly, without a label, and has built a recording studio in its New York City office. Amazon is also recording some music that is not copyright protected. With their informational advantages, streaming companies have unique insights into the most profitable types of music to mimic and supplant (e.g., “Happy Birthday” and other children’s music). And Apple has launched its own music publishing company.

  Yet another area that is sure to change involves audio technology. The innovative company Sonos, for example, has created a highly successful line of smart speakers. A new generation of smart speakers could displace Google Home, Amazon Echo Dot, and other devices that are currently being used to stream and listen to music.

  Amazon’s founder and CEO, Jeff Bezos, once told me that he is often asked, “What is going to change in the future?” But he is rarely asked, “What is not going to change?” “That second question is actually more important than the first,” he said, “because you can build a strategy around it.”38 That seems like the type of perspective that enabled Bezos to become the wealthiest businessperson on the planet. When it comes to music streaming, and media more generally, it seems to me that customers are always going to want convenience, affordability, wide variety, and good listening recommendations. Streaming services that can deliver on those priorities are likely to be among the handful of platforms that survive and thrive in the twenty-first century.

  Outro

  We are likely in the early stages of the streaming revolution. The multitude of streaming services will surely face a shakeout in the future, and it is unclear what form successful business models will take. Streaming companies may displace record labels, or they may be absorbed by labels. A handful of streaming platforms may survive as stand-alone companies and dominate the market. Or streaming platforms may all become—or, in the case of Amazon Music, Apple Music, and Google Play, remain—subsidiaries of dominant Internet companies that benefit from complementarities. New players, such as Netflix and Facebook, may become competitors or strategic partners for streaming companies. And new devices, such as future generations of smart speakers, may significantly alter the contours and possibilities of streaming.

  All of these possible developments—and more that cannot currently be predicted—can significantly change the incentives that affect music production and music discovery. Twists and turns in the streaming landscape will continue to affect the size and division of the music pie. The only thing that is certain is that the future of streaming will pose many intriguing issues for musicians, managers, labels, publishers, performing rights organizations, policy makers, and others that can be studied with the tools of economics. Rockonomics will be an active field for decades to come.

  *1 In inflation-adjusted dollars, revenue from recorded music fell by 68 percent from the peak in 1999 to the trough in 2015.

  *2 One exception to this prediction is Japan, where CDs are still king. See Chapter 10.

  *3 The day she returned her music to Spotify also happened to coincide with the release of rival pop star Katy Perry’s album Witness.

  *4 In the company’s defense, Apple’s Eddy Cue noted that the company “had originally negotiated these deals based on paying them a higher royalty rate on an ongoing basis to compensate for this brief [trial] time.” Apple agreed, however, to compensate artists during the trial period and keep the royalty rate at the previously negotiated level.

  CHAPTER 9

  Blurred Lines: Intellectual Property in a Digital World

  Music fans cannot expect their favorite musicians to continue to produce quality albums if they are not willing to pay. People, including musicians, expect to be rewarded for a job well done. It’s all about supply and demand. If there is not demand, there will eventually be no supply.

  —Sheryl Crow

  Sitting for a deposition early in the afternoon on April 21, 2014, Pharrell Williams looked uncomfortable answering questions. In fact, he said so on several occasions. The deposition was part of a landmark trial to determine whether Pharrell Williams and Robin Thicke’s hit 2013 song “Blurred Lines” crossed a legal line and infringed the copyright of Marvin Gaye’s 1977 song “Got to Give It Up.”

  After a weeklong trial and two days of deliberation, the jury ruled in favor of Marvin Gaye’s family. Williams and Thicke appealed to the Ninth Circuit, which covers the western region of the United States, but lost a split 2–1 decision in March 2018. They and their publisher were ordered to pay the Gayes $5.3 million in damages and share 50 percent of future songwriting and publishing royalties. In rejecting Williams’s argument, the court held that “musical compositions are not confined to a narrow range of expression.”1 In her dissenting opinion, Judge Jacqueline H. Nguyen wrote, “ ‘Blurred Lines’ and ‘Got to Give It Up’ are not objectively similar. They differ in melody, harmony, and rhythm. Yet by refusing to compare the two works, the majority establishes a dangerous precedent that strikes a devastating blow to future musicians and composers everywhere.”

  The theatrics of the trial aside—and there were many—the case of Williams v. Gaye illustrates the difficulty of determining copyright to a song, and the stakes involved.*1 The Copyright Act of 1831 extended copyright protection to original musical compositions. When I listen to “Blurred Lines” and “Got to Give It Up” with my untrained ears, I don’t hear enough similarities to conclude that Williams plagiarized Marvin Gaye. And since the legal case was confined to the composition, not the recorded performance, the similarities are even less apparent if you strip out the rhythmic cowbell tones and falsetto singing that both songs have in common. Williams lost not because he copied Gaye’s expression, but because he was inspired by the feel and groove of Gaye’s music. Listen to the songs when you have a chance, and see if you agree with the ruling.

  It is unclear whether the case will set a precedent for future cases and other jurisdictions. But it is likely that more claims will be brought and settled out of court, rather than risk an embarrassing trial and uncertain verdict. Even before the verdict in Williams v. Gaye was reached, Sam Smith chose to share credit for his Grammy-winning “Stay with Me” with Tom Petty and Jeff Lynne, after Tom Petty said it sounded a lot like his 1989 hit “I Won’t Back Down.”2 And Mark Ronson settled two copyright challenges to his 2014 hit with Bruno Mars, “Uptown Funk”; a third challenge is ongoing.3 Another high-profile case, in which it was initially ruled that Led Zeppelin’s “Stairway to Heaven” did not plagiarize its iconic opening from Spirit’s instrumental “Taurus,” will be retried as a result of an appellate co
urt ruling, and could set a precedent that expands or narrows copyright protection, depending on the outcome.4

  It is an understatement to say that the lines between legitimate inspiration and illegal copyright infringement are blurred. Quincy Jones told me that the amazing thing about music is that it all involves “the same 12 notes being played over and over again,” so some overlap is inevitable.5 There is no Shazam-like app, however, that can tell a composer or a judge whether a song infringes another song’s copyright (although maybe there should be). Mark Twain was on to something when he quipped, “Only one thing is impossible for God: To find any sense in any copyright law on the planet.”

  Copyright assigns a legal right to allow authors to exclude others from using their original work for a period of time. Copyright law as applied to music is even more complicated and confusing than it is for books and other written works. There are two copyrights to music: a right to the sound recording and a right to the underlying composition and lyrics.6 The rights that must be secured to use a song depend on the particular use, such as a movie soundtrack, streaming platform catalog, videogame sound effect, phone ringtone, or terrestrial radio broadcast. The fees for obtaining the relevant rights vary as well, depending on the use of the work. Access to the rights for some uses is compulsory (i.e., no permission required) at a preset price, while the rights and fees for other uses must be negotiated on a case-by-case basis.

  The goal of copyright law is to grant a temporary monopoly to creators so they have a financial incentive to create new music, books, or other works. Copyright protection makes particularly good economic sense when the cost of producing a new work is high but the cost of producing subsequent copies is low. “Drawing the line of copyright infringement too short will fail to give the original author his due,” writes Stanford Law School’s Paul Goldstein, “but extending it too far will make it hard for other writers to earn theirs.”7 As the dispute between Pharrell Williams and Marvin Gaye’s heirs indicates, the line must be drawn between different creators. Copyright must also strike a balance between copyright owners and copyright users—that is, between those who create music and those who broadcast it or listen to it. Similar issues arise with other forms of intellectual property rights, from iconic photographs to scientific patents.

  The music copyright system was originally developed for rights involving sheet music in the days of player pianos. An elaborate legal and administrative system has evolved over more than a century to (1) determine the rights that must be secured to use music, (2) set royalty rates, and (3) collect and distribute payments to rights holders. Not surprisingly, such a system is antiquated in the digital era. Any normal human being’s head spins faster than a turntable when confronted with the varying rights, collection agencies, and processes that determine fees for the myriad different uses of music. In this chapter I sample some of the key economic issues that arise in copyright protection in a digital era.

  Digital technology reduces the cost of sharing identical copies of already-produced copyrighted material to essentially zero, and accelerates the speed and cuts the cost of creating and distributing new music. Neither economics nor any other field can provide complete or unambiguous answers to fundamental questions about where copyright lines should be drawn, but economics research can help frame the questions and provide insights into the various trade-offs involved.

  The Useful Arts

  Intellectual property protection is so important that it is enshrined in Article I of the United States Constitution. Specifically, Section 8, Clause 8, gives Congress the power “To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.” Popular music surely falls into the “useful Arts.”

  Songwriters are creators. Without lyricists and composers, there would be no music. Once a song is written or recorded, however, it is what economists call a non-rivalrous good, meaning that one person’s consumption does not diminish another person’s opportunity to consume the good. If you listen to or perform a song, I can listen to or perform the same song too, without diminishing your experience. By contrast, if you eat a ham sandwich, I cannot consume that same sandwich. Other examples of non-rivalrous goods are television and radio broadcasts, patented ideas, fashion designs, and architectural plans. There is a strong incentive for free riding when it comes to non-rivalrous goods—that is, for people to consume the good but avoid paying for it. James Madison and the other framers of the Constitution recognized that it is necessary to establish a property right to intellectual property, such as writings and inventions, in order to prevent free riding and encourage investment in the development of new writings and discoveries. This is the foundation of copyright and patent law.

  By establishing a limited property right—the legal authority to exclude others from using a creative work for a period of time—copyright law seeks to strike a balance between the interests of creators and the interests of consumers.8 And while in pre-Internet days one could plausibly argue that bringing a new book or song to market first provided protection against copiers and imitators—as a young Stephen Breyer once argued before he was named to the Supreme Court—that argument clearly does not hold water in the digital era, when perfect copies can be made and distributed throughout the world almost instantaneously.

  Property rights are essential in an economy. How can you trade something if you do not know who owns it? Why would you invest in something if you were unsure of whether you could share in the returns? One of the basic insights of economics is that well-defined property rights are necessary for a market to function well. This applies both to intellectual property, such as songs, and to physical property. The lack of copyright protection for music until recently in China is one of the reasons the most populous country on earth has produced such little indigenous music. But where should the lines be drawn? How long should copyright protection last? How should access to copyrighted songs be restricted? How should royalties be collected when there are more than two billion streams a day? And, in practice, how does copyright protection actually affect the total quantity and quality of music produced?

  Napster: A Case Study in Destructive Creation

  The advent of the file-sharing service Napster and other websites that made unauthorized, pirated digital music available on a mass scale laid bare the centrality of copyright protection for the music industry, and the fragility of the system. In principle, the availability of illegal file-sharing websites could have several effects on authorized sales of music. First, and most important, potential buyers could substitute unlawful downloading for purchases of legitimate recordings that they otherwise would have made. Second, file-sharing services could increase customers’ exposure to new music, and increase their purchases of such music as a result or increase their demand for live performances (i.e., file sharing could act like advertising). Third, those who listen to music on unlawful file-sharing sites could inform others in their network about the music, and those others might purchase it. The evidence overwhelmingly points to the dominance of the substitution effect: why buy music if you could obtain it for free?

  Music has always been subject to illegal copying. Before he was chairman of the Federal Reserve Board, Alan Greenspan conducted research for the Recording Industry Association of American that concluded, “From the available data, it appears that roughly half of the taping from borrowed records or tapes would have generated record or tape purchases had home audio taping not been possible….This represents lost sales of approximately 32 percent of the total volume of record sales in 1982.”9 Even discounting the interest of the sponsor, this suggests a lot of illegal copying. The record business was growing at a healthy clip in the 1980s, so piracy was not considered an existential threat. Some bands, like the Grateful Dead, even encouraged their fans to record their concerts, which likely cannibalized album sales (but increased deman
d for their live shows and fostered fan loyalty).

  Napster and its ilk, however, did represent an existential threat to the incomes of recording artists and record companies. As I noted earlier, record company revenue and musicians’ royalties plummeted because of digital piracy after Napster was launched in 1999 (see Figure 2.2). Napster provided much greater ease and wider access for users to obtain music without paying for it than homemade cassette tapes ever afforded. After the band Metallica successfully sued Napster for copyright infringement, the file-sharing site filed for bankruptcy and closed in 2001.10 (The streaming service Rhapsody was rebranded as Napster in June 2016.)11 A host of other file-sharing and torrent services subsequently moved into the space, including Gnutella, Freenet, Kazaa, LimeWire, Scour, Grokster, eDonkey2000, and Pirate Bay—some of which were also challenged for copyright infringement.12 Only in the last few years have legitimate streaming services offered an effective alternative to piracy.

  Economist Joel Waldfogel used the unprecedented growth of pirated music in the 2000s that was ushered in by Napster as a natural experiment to examine the effects of weakening copyright protection on the quantity and quality of music produced. After looking at a variety of measures of the quality of music produced each year (e.g., critics’ ratings, comparison of one vintage of music against another in terms of airplay or sales), he reached the surprising conclusion that “the quality of new recorded music has not fallen since the introduction of Napster.”13 And the total volume of music produced actually increased, rather than decreased.