Download PDF Mr. Chairman and Members of the Subcommittee, thank you for the opportunity to appear before you today to testify on issues relating to music licensing.
I currently serve as a Managing Director at Navigant Economics, a Visiting Scholar at the American Enterprise Institute and an adjunct professor at George Mason University Law School, where I teach the course on Regulated Industries. In all of these capacities, and for much of the past two decades, I have written about and taught on topics relevant to the subject of today's hearing. While some of the research upon which my testimony today is based was supported in part by the musicFIRST coalition, I am appearing today solely on my own behalf, and the views I will express are exclusively my own.
My testimony today focuses on the sound recording performance right and, in particular, on what is commonly referred to as the digital performance right.1 As the Subcommittee knows well, until recently, owners of sound recording performance rights were granted reproduction and distribution rights, but – unlike the holders of musical work rights – were not granted a performance right. Thus, copyright holders of sound recordings could monetize the copying and distribution of their recordings, but could not charge for "performances," such as when radio stations (or webcasters) played copyrighted music. In the absence of such a property right, naturally, there was no market for sound recording performances.
Beginning with passage of the Digital Performance Right in Sound Recordings Act (DPRA) in 1995,2 Congress has moved gradually in the direction of both creating performance rights and putting in place the conditions to allow such rights to be traded at market (that is, economically efficient) rates. The DPRA established the first sound recording performance right in the form of the digital performance right. Then, in the 1998 Digital Millennium Copyright Act (DMCA), Congress established the principle that license terms and royalty rates would either be negotiated directly between the parties or, in the case of rights subject to a compulsory license, would "represent the rates and terms that would have been negotiated in the marketplace between a willing buyer and a willing seller."3 Twice in recent years, this subcommittee has passed legislation that would have further advanced market-based principles by extending the sound performance right to the over-the-air broadcasts of terrestrial broadcasters.4
The central point of my testimony today is that Congress should continue to move in the direction of using market-based mechanisms for setting the terms and rates by which sound recording performance rights are licensed among rights holders and users. Equally important, it should resist entreaties to backslide by passing legislation that would replace the current marketbased standard for royalty rates with one designed to tilt the playing field in such a way as to subsidize a particular class of copyright users.
I am referring, of course, to the proposed Internet Radio Fairness Act (IRFA, H.R. 6480/S. 3609). While the IRFA contains a number of provisions designed to tilt the rate-setting process in favor of copyright users and against copyright holders, at its core is its proposal to replace the market-oriented willing buyer/willing seller standard with the uneconomic, four-part standard under Section 801(b)(1) of the Copyright Act of 1976 (the "801(b) standard"). To do so would represent a significant step in the wrong direction, both because the rates likely to emerge from the rate setting process would be below those that would emerge from a competitive market, and thus reduce economic welfare, and because the "non-disruption" standard contained in Section 801(b)(1)(D) would create perverse incentives that are fundamentally at odds with the innovative, dynamic nature of the market for online music.
Specifically, replacing the willing buyer/willing seller standard with the 801(b) standard and making the other changes proposed by the IRFA would harm consumers for four primary reasons.
(1) Market-based rates result in the efficient – i.e., consumer-welfare-maximizing – allocation of society's resources, and the willing buyer/willing seller standard embodies the principle of market-based rates.
(2) The lower rates that would result from the IRFA are not necessary to preserve a vibrant, growing market for online music, and would harm the market for content creation.
(3) The non-disruption standard contained in Section 801(b)(1)(D) is fundamentally inconsistent market-based incentives for efficiency and innovation, especially in a dynamic market such as the market for digital music.
(4) Adoption of the IRFA would distort the rate setting process and likely result in the further politicization of rate setting for sound performance rights.
Let me expand briefly on each reason.
First, market-based rates maximize consumer welfare by ensuring that society's resources are directed to their highest valued uses. In a market-based economy like ours, prices serve as the key signaling mechanism telling economic actors how capital and labor should be directed to produce the products and services valued most highly by consumers at the lowest possible cost. Prices set above market-clearing levels result in too many resources being directed towards production, while at the same time too little of the resulting output is demanded by consumers. Prices set below market clearing levels have the opposite effect – too little is produced, and consumers are unable to procure the amount, or the quality, of products they desire.
As I detail in Attachment A, the willing buyer/willing seller standard has been implemented in such a way as to produce royalty rates consistent with those that would likely result from a freely functioning market. In particular, the arbitration bodies that have set rates in the major Webcaster proceedings have based their determinations on freely negotiated rates for analogous products, e.g., the rates for interactive services, which are not subject to a compulsory license. While no rate setting process is perfect, the procedures followed by the Copyright Arbitration Royalty Panel (in Webcaster I) and by the Copyright Royalty Board (in Webcaster II and Webcaster III) – which include opportunities for all sides to fully present their positions, supported by expert economic and industry testimony, as well as both administrative and judicial review – have likely yielded rates that reasonably approximate those that would have resulted from voluntary negotiations in a freely operating market, and thus are presumptively consumer-welfare- maximizing.
Second, while IRFA would almost certainly produce the lower royalty rates its supporters seek, there is no valid economic or public policy basis for forcing content providers to subsidize webcasters by charging them below-market rates. The market for online music is intensely vibrant and growing rapidly. Tens of thousands of new listeners are signing up to services like Pandora and Spotify every week, and existing listeners are using the services more and more intensely every year. Online advertising revenues are growing 30 percent per year, new firms are entering the market at a rapid pace, and existing firms are garnering billion dollar market valuations.
IRFA's leading supporter, Pandora, makes much of the fact that content acquisition accounts for a large proportion of its revenues, but in fact its content costs as a proportion of revenues are comparable to other, similar firms. For example, as I detail in the attachment, the proportion of revenues accounted for by content costs for Netflix and Pandora have been nearly identical over the last three years (2009-2011) for which data is available from both firms; indeed, for each of the last two years, Netflix has paid a higher proportion of its revenues for content acquisition than has Pandora.
Moreover, and crucially, the ratio of Pandora's content costs to its revenues is well within Pandora's control: To raise its revenues, it need only choose to sell additional advertising. As The New York Times reported recently, "Throughout the music industry there is a wide belief that Pandora could solve its financial problems … by simply selling more ads."5
Third, the Section 801(b)(1)(D) non-disruption standard would fundamentally distort the rate setting process by granting users a de facto right to perpetual profitability based on their current business models. Indeed, as I detail in the attachment, experts testifying on behalf of copyright users in the current SDARS II proceeding have argued that the non-disruption standard not only requires rates to be set so as to guarantee copyright users profits on their initial investments, apparently in perpetuity, but even to ensure that they can "recover the financial cost of capital for forward-looking investments," since rates that fail to give users incentives to continue investing in their businesses would be "disruptive."
In the dynamic world of online content delivery – in which new and improved business models are constantly replacing old, obsolete ones – the creation of what licensees argue is a de facto right to perpetual profitability is a recipe for technological and marketplace stagnation.6
Fourth, and finally, both the act of passing the IRFA and a number of its specific provisions would distort the rate setting process and likely result in the further politicization of rate setting for sound recording performance rights. As I detail in the attachment, a number of the IRFA's specific provisions, including the changes it would make to the appointment process and qualifications of the copyright royalty judges, would threaten to reduce the objectivity and independence of the CRB.
More broadly, it is a truism that all market participants would prefer to pay lower prices for their inputs – car manufacturers would like to pay less for steel, gas stations less for gas and soft drinks, aluminum plants less for electricity. In the absence of market failures, however, market forces ensure that the prices paid for such inputs are, to paraphrase Goldilocks, neither too high nor too low, but "just right." The politicization of pricing decisions, on the other hand, results in prices which favor the actors with the greatest capacities for political influence. In this case, Congress should not allow the fact that webcasters have the demonstrated capacity to generate a large volume of emails from their listeners to lead to a result that would, in the end, harm those very same consumers by retarding innovation and destroying incentives for content creation.
 I have recently authored a study on the sound performance recording right which addresses many of the issues
discussed herein. It is included in this written statement as Attachment A.
 Digital Performance Right in Sound Recordings Act of 1995, Pub. L. No. 104–39.
 17 U.S.C. § 114(f)(2)(B).
 See The Performance Rights Act of 2007 (H.R. 4789/S. 2500) and its successor, The Performance Rights Act of 2009 (H.R. 848/S. 379).
 See Ben Sisario, “Proposed Bill Could Change Royalty Rates for Internet Radio,” The New York Times (September 23, 2012) (available at http://www.nytimes.com/2012/09/24/business/media/proposed-bill-could-change-royalty-rates-for-internetradio. html). See also Richard Greenfield, “Congress Should be Working to Raise Royalty Rates on Pandora, Not Lower Them,” BTIG Research (September 24, 2012) (available at http://www.btigresearch.com/2012/09/24/congress-should-be-working-toraise-royalty-rates-on-pandora-not-lower-them/) (“[T]he reason why companies such as Pandora pay such high royalty rates as a percentage of revenues is because they severely limit audio advertising to protect the user experience and keep people on the platform. If Pandora ran several minutes of audio ads per hour (the way terrestrial radio does) vs. just a few 15 sec. spots, the % of revenues paid out as royalties would be dramatically lower and would be more in line with satellite radio or cable TV. Interestingly, Spotify’s radio product runs substantially more advertising per hour than Pandora.”).
 By contrast, as the D.C. Circuit explained in reviewing the Webcaster II decision, the willing buyer/willing seller standard does not require rates to be set so as to preserve inefficient business models. See Intercollegiate Broadcast System v. Copyright Royalty Board 574 F. 3d 748, 761 (D.C. Cir. 2009) (“[I]t was not error for the Judges to reject the small commercial webcasters’ pleas that paying per performance would wreck their inefficient business models. The Judges made clear they could not ‘guarantee a profitable business to every market entrant.’ The Judges are not required to preserve the business of every participant in a market.”) (emphasis added).