In Re Application of MobiTV, Inc. ( 2010 )

712 F. Supp. 2d 206 (2010)

In re APPLICATION OF MOBITV, INC.
Related to United States of America, Plaintiff,
v.
American Society of Composers, Authors, and Publishers, Defendant.

Nos. 09 Civ. 7071 (DLC), 41 Civ. 1395 (DLC).

United States District Court, S.D. New York.

May 11, 2010.

Kenneth L. Steinthal, Joseph R. Wetzel, Harris Cohen, Matthew L. Reagan, Greenberg Traurig, LLP, San Francisco, CA, for applicant MobiTV, Inc.

David Leichtman, Hillel I. Parness, Bryan J. Vogel, Oren D. Langer, Robins, Kaplan, Miller & Ciresi, LLP, New York, NY, and Richard H. Reimer, Christine A. Pepe, New York, NY, for ASCAP.

REDACTED OPINION AND ORDER

DENISE COTE, District Judge.

Table of Contents
INTRODUCTION ........................................................................ 209
FINDINGS OF FACT .................................................................... 211
I. ASCAP and Its Two Competitors ................................................. 211
II. The Cable Television Industry ................................................. 212
III. The Creation of a Television Program .......................................... 214
IV. The Music Video Business and NDMAs ............................................ 215
V. Mobi .......................................................................... 216
A. Early Years and Growth ..................................................... 216
B. Mobi's Product Lineup ...................................................... 217
C. The Economics of Mobi's Business ........................................... 219
D. Mobi's SESAC License: Its First PRO License ................................ 221
VI. ASCAP Licenses ................................................................ 222
A. ASCAP's Post-Turner Licenses ...................................... 222
B. The Terms of the Post-Turner Licenses ............................. 223
C. Application of the Post-Turner Licenses to Wireless Distribution
Systems .................................................................. 224
D. ASCAP Licenses with Cable System Operators ................................. 226
E. Audio-Only Programming ..................................................... 226
CONCLUSIONS OF LAW .................................................................. 227
I. The Non-Dramatic Performance Right ............................................ 227
II. The Consent Decree ............................................................ 228
A. Origins of the Consent Decree .............................................. 228
B. The 1990s Turner Litigation ....................................... 228

C. The Negotiation of AFJ2's TTTA License Provision ........................... 229
D. Other Provisions of AFJ2 ................................................... 231
III. Fair Market Value .............................................................. 232
A. Music Choice: Retail or Wholesale Price as a Superior Indicator of
Fair Market Value? ......................................................... 233
B. AOL Rate Court Ruling ....................................................... 235
IV.  A Reasonable Rate .............................................................. 236
A. ASCAP Rate Proposal ......................................................... 236
B. Mobi's Fee Proposal ......................................................... 244
C. Adoption of a Reasonable Rate ............................................... 246
CONCLUSION .......................................................................... 255

INTRODUCTION

It is hard to overstate the importance of music in most of our lives. Every concert we attend, every song we listen to, virtually every entertainment we enjoy reinforces that lesson. The task at hand is to determine the fair market value of a blanket license for the public performance of music. The challenges of that task include discerning a rate that will give composers an economic incentive to keep enriching our lives with music, that avoids compensating composers for contributions made by others either to the creative work or to the delivery of that work to the public, and that does not create distorting incentives in the marketplace that will improperly affect the choices made by composers, inventors, investors, consumers and other economic players.

This Opinion addresses these decades-old issues in the context of a relatively new phenomenon: the delivery of television programming to mobile telephones ("handsets"), an innovation made possible by the digital revolution. MobiTV ("Mobi") is a middleman between the cable television networks that create programming and the wireless carriers to which consumers subscribe to obtain wireless service on their handsets. Mobi does principally three things: it provides the back-end technology that permits television program content to be delivered seamlessly to handsets, it licenses cable television program content from the networks and provides it to the wireless carriers, and it programs music video channels that it provides to wireless carriers.

Mobi has asked for a through-to-the-audience ("TTTA") license from the American Society of Composers, Authors and Publishers ("ASCAP"), an organization representing almost half of American composers and music publishers in their negotiations of public performance rights, to cover the programming content that Mobi provides to wireless carriers. The parties have been unable to reach agreement on the terms of a TTTA license, and, pursuant to an antitrust consent judgment, they now request that this Court set a rate for that license. The parties dispute, among other things, whether the revenue base for the calculated fee should be the retail revenue received by the wireless carriers or the amounts Mobi pays to content providers, and the rates that should be applied to that revenue base. Their fee proposals differ by tens of millions of dollars.

ASCAP applied to the Court to set a reasonable rate on May 5, 2008.^[1] A bench trial was held from April 12 to 26, 2010, to determine a reasonable rate pursuant to Mobi's license application to ASCAP and ASCAP's application to this Court. This Opinion constitutes the Court's findings of fact and conclusions of law following that trial. The factual findings are principally set forth in the first section of this Opinion, but appear as well in the final section.

With the parties' consent, the trial was conducted in accordance with the Court's Individual Practices. On March 18, the direct testimony of the witnesses was presented through affidavit and submitted with the joint pretrial order, along with the parties' trial exhibits and proposed findings of fact and conclusions of law.

ASCAP presented affidavits constituting the direct testimony from four of its employees, two experts and two composers. Its employee-witnesses were Christopher Amenita, Senior Vice President for Broadcast Operations and New Media Licensing ("Amenita"); Matthew DeFilippis, Vice President of the New Media and Technology Department ("DeFilippis"); Dr. Peter M. Boyle, Chief Economist ("Boyle"); and Ray Schwind, Vice President and Director of Broadcast Licensing ("Schwind"). ASCAP's two experts were Dr. Jennifer Vanderhart, an economist ("Vanderhart"), and Cliff Petrovsky, who reported on his observations about Mobi's programming ("Petrovsky").^[2] The two composers were David Vanacore ("Vanacore") and David Wolfert ("Wolfert"), each of whom is an accomplished composer of music for television programming.

Mobi presented affidavits constituting the direct testimony from five of its current or former employees and three experts. The Mobi employees were Raymond DeRenzo, Chief Marketing Officer ("DeRenzo"); Terri Falcone, Vice President of Finance and Controller ("Falcone"); Carl Ghoreichi, Principal Research Analyst ("Ghoreichi"); Andrew Missan, currently Vice President and General Counsel of Bytemobile, Inc. and Vice President and General Counsel for Mobi from May 2005 until April 2008 ("Missan"); and Paul Scanlan, Co-Founder and President of Mobi ("Scanlan"). Mobi's experts were Dr. Roger Noll, an economist ("Noll"); Larry Gerbrandt, an expert on the cable and satellite industries ("Gerbrandt"); and Paul Vidich, an expert on the history, production, and licensing of music videos ("Vidich").^[3] All of the parties' witnesses appeared at trial and were available for cross-examination with the exception of Wolfert.

In addition, the parties designated deposition testimony from many of the trial witnesses and nineteen additional witnesses. The additional witnesses for whom ASCAP offered deposition testimony were Jeff Bartee, Mobi's Vice President of Content; Kevin Grant, Mobi's Vice President of Sales; Ellen McDonald, Mobi's General Counsel; Sarah Walter, Mobi's Senior Director of Financial Planning and Analysis; William Colitre, an attorney and Director of Business and Legal Affairs of Music Reports, Inc. ("MRI"); Mark Nagel, Manager of Music Products for AT & T Mobility LLC; and Suzanne Hellwig, a Director of Marketing Management for AT & T. The additional witnesses for whom Mobi offered deposition testimony were Vincent Candilora, ASCAP's Senior Vice President of Licensing; Kevin Gage, ASCAP's Senior Vice President for Strategic Planning and Digital Development; John LoFrumento, ASCAP's chief executive officer; Julie Peng, ASCAP's Manager of New Media Licensing; Richard Conlon, the Vice President of New Media and Strategic Development at Broadcast Music, Inc. ("BMI"); Mark Eisenberg, Executive Vice President, Global Digital Business Group, and head of Business and Legal Affairs at Sony Music Entertainment; and Elliott Peters, head of Digital Legal Affairs at Warner Music Group. The parties each offered designated deposition testimony of John Siu, Vice President of Accounting for U.S. Networks for Discovery Communications, Inc.; Jeff Klaumann, TV Product Manager at Sprint Nextel Corporation; Christopher Lamb, Product Manager for Music Services and Video Services at Verizon Wireless; Patricia Bowes, Marketing Director for AT & T Mobility; and Brandon Shaw, Senior Product Manager of Music Products at AT & T Mobility.

FINDINGS OF FACT

I. ASCAP and Its Two Competitors

ASCAP is almost a century old; it was formed in 1914. It is an unincorporated membership association and performing rights organization created and controlled by music composers, writers, and publishers. It has about 380,000 members and its repertory contains over 8.5 million copyrighted musical works. The members grant ASCAP the non-exclusive right to license non-dramatic public performances of their music. ASCAP's board of directors is composed of writers and music publishers.

ASCAP operates under a consent decree first issued in 1941. The current consent decree, issued in 2001, will be described in some detail below. Among other things, the current consent decree regulates the terms on which ASCAP may offer music users a blanket license to publicly perform music from the ASCAP repertory.

A blanket license is a license that gives the music user the right to perform all of the works in the repertory of a performing rights organization ("PRO") such as ASCAP, the fee for which does not vary depending on how much of the music from the repertory the user actually uses. ASCAP negotiates with and collects license fees from entities that perform music publicly. ASCAP then distributes the collected royalties to its members based on a system of performance surveys and credits. Among other things, ASCAP aims to pay the money collected from one medium to the members whose works are performed in that medium.^[4]

ASCAP competes with two other United States PROs: Broadcast Music, Inc. ("BMI") and SESAC, Inc. ("SESAC"), each of whom also offers blanket licenses. PROs and their ability to grant blanket licenses covering a large number of compositions create significant economies of scale in negotiating a license and in policing the marketplace to prevent infringement of copyright rights. BMI, which is slightly smaller than ASCAP, operates under a consent decree that is similar to the one that governs ASCAP's licenses. See United States v. BMI (In re Application of Music Choice), 316 F.3d 189, 190 (2d Cir. 2003) ("Music Choice II").

SESAC does not publicly report its revenue or compositions, but is understood to hold a share of the total number of musical compositions in the single digit range. SESAC is an invitation-only membership organization, aggressively courts composers it identifies as creating "high value," and is not subject to a consent decree.

II. The Cable Television Industry

Because the content at issue here is largely cable television programming, and because the rates at which ASCAP licenses the public performance of music over cable television networks are integral to the proposals made by both ASCAP and Mobi in this trial, it is essential to understand the cable television industry. Commercial television was launched in the United States in the late 1940s. Cable television's roots can be traced back almost that far — to the early 1950s — although it took another twenty years for cable television to emerge in the form we recognize today.

In 1972, Home Box Office or HBO, a national cable network, was launched. With the launch of a new generation of geostationary satellites in 1976, television programming could be delivered nationally without leasing telephone lines and there was an explosion in the number of cable programming networks, including the founding of ESPN and CNN.

There are two major categories of cable programming^[5]: (1) channels made available to cable subscribers on a bundled basis or "basic networks;" and (2) networks sold à la carte for a separate subscription fee, or "pay" and "premium" channels. Basic networks usually carry advertising, while most premium channels do not. Advertising revenue for basic cable networks comes from minutes that are reserved or available for commercials or "avails" each hour, a few minutes of which are reserved for local advertising. The portion not reserved for local avails can be sold to national advertisers or used for internal program promotions or public service announcements.

The content assembled by networks is distributed by cable, satellite, telephone lines, and more recently and of special importance in this case, by wireless distributors and operators.^[6] Cable operators receive programming from cable networks via satellite systems and then transmit the signals through the cable network that passes to the households in its franchised region.^[7] Similarly, satellite systems allow individual subscribers to view hundreds of cable program channels using receiving dishes and are particularly important in sparsely populated areas. More recently, telephone companies, such as AT & T and Verizon, have entered the television delivery business and deliver programming in a number of ways, one example being IPTV, which is Internet Protocol Television. Very recently, in October 2009, United States broadcasters adopted a standard for mobile television service. As of now, about seventy television stations have announced that they will be offering service that conforms to that standard.

Along with the evolution of technologies that distribute television programming, there have been major improvements in the quality of the television signal as well. Color television was introduced in the 1950s. More recently, digital television and high-definition television ("HDTV") have been adopted. Similarly, the quality of the audio portion of the transmission has improved. Stereo television, the addition of a second audio program (which is often a Spanish-language feed), closed captioning, and Dolby Digital 5.1 are among the principal improvements.

Despite the changes in the modes of delivering television programming, the method by which cable television networks obtain revenue has remained the same. The networks license the rights to distribute their programming through affiliation agreements with distributors under financial terms generally premised on the number of subscribers served. These affiliate fees, along with revenue from advertising, are the primary sources of revenue that cable networks use to cover their expenses of acquiring, creating and delivering programming.

Affiliate fees reflect both a cable television network's cost structure and the perceived value of its programming. For example, ESPN's popularity permits it to demand a comparatively high affiliate fee. It is noteworthy for the discussion that follows that affiliate fees are not tied to the retail prices that the cable system operator charges its subscribers. Instead, networks simply license as many distributors of their programming as they can to reach as many viewers as possible in order to maximize their return. In contrast, the retail price that a distributor charges its subscribers is based primarily on the competitive factors in that business, such as its cost of technology, service enhancements associated with the delivery platform and the composition of its channel lineups.

The invention and adoption of digital technology has had at least two significant impacts on the cable system transmission architecture: cable operators have expanded their channel capacity dramatically and subscribers have increased control over the viewing experience. Video-on-demand or VOD is an example of the latter phenomenon. In contrast to linear programming, which requires a viewer to watch a program at its scheduled time, VOD permits subscribers to select a program from a menu and pause, rewind or fast-forward.^[8] Some cable operators also allow subscribers to watch a limited number of program episodes during a window of time after the initial airing on the cable network.

As Gerbrandt testified, despite the evolution of content delivery over a variety of distribution systems, "the viewing experience for the consumer — watching television on a screen — has remained fundamentally the same." And notwithstanding the evolution in delivery systems, networks typically still create and acquire content which they schedule into a linear lineup, and license third-party distributors through affiliation agreements premised on the payment of monthly per-subscriber fees to distribute that content via linear or VOD delivery methods to consumers. And despite the revolutions in the quality of the video and audio presentation of content delivered over cable television networks, the networks still derive their revenue principally from advertising revenues and from the number of subscriptions. To the extent that improvements in technology have increased the number of cable television subscribers, those improvements have been captured in the increased revenues earned by the cable networks from affiliate fees and advertising.

This revolution in digital transmission has blurred the distinctions among communications technologies and services. For example, while telephone networks were once devoted exclusively to point-to-point voice communications, and over-the-air and cable transmissions were dedicated to delivering television programs, digital technology has enabled telephone networks to support not just the traditional two-person telephone conversation, but has allowed it to deliver as well access to the internet, television programming, computing and a host of other services. Thus, wireless carriers and television stations increasingly compete in the delivery of mobile video entertainment to consumers.

With the advancements in mobile telephone technology, there has been an explosion in the number of wireless telephone subscribers. As of June 2009, it is estimated that there were over 277 million wireless telephone subscribers in the United States.^[9] In contrast, the number of land-line telephones peaked in 2000, and has been falling ever since. Using another yardstick, the monthly minutes of use of handsets has risen from about ten billion in 1999 to almost 200 billion in 2009.

III. The Creation of a Television Program

Another component of the evidence at trial was testimony that described the creation of television programming and the role of music in that programming. Of course, each television program may be comprised of many pieces of intellectual property, from scripts to scores, and of work by many individuals, from writers, directors, actors and musicians to composers.

Generally, program content contributors are compensated with up-front fees and with a percentage of revenue received by the owner of the audiovisual work from the exploitation of the program, including through returns in syndication or re-run markets. These creative contributors do not customarily receive compensation tied to the advertising or affiliate license fee revenues that cable television networks generate or to the down-the-line revenues that cable operators or satellite or telephone companies generate from subscriptions to their television distribution services. Moreover, the compensation to the contributors generally does not vary based on the technical means by which the program content reaches the consumer.^[10]

Vanacore, one of the two main composers for the television show Survivor, and a contributor to many other successful television shows, testified about his experience in composing music for a television program or series. Generally, he is hired on a composer-for-hire basis, which means that he does not own the musical work that he creates. To compensate him for his work, he is paid something "up front," which may be a relatively small payment, and is granted back what he characterizes as "the `writer's share' of the publishing rights, including, most significantly, the right to collect public performance royalties directly" from the PRO of his choice.^[11] This gives him the right to a 50% share of public performance royalties for compositions for which he is the sole writer; the remainder goes to the music publisher. In television, the publishing company is usually owned by the program producer itself.

IV. The Music Video Business and NDMAs

Mobi distributes not only cable television programming but also procures rights to music videos and creates music video channels that it then distributes to wireless carriers. Therefore, before describing Mobi, it is useful to describe the music video business, in particular its financial structure.

Music videos are essentially short-form films. They are copyrighted works that involve video production, direction, writing, choreography and editing, as well as the acting and visual performance of the recording artist. They are designed to create a visual connection between recording artists, their songs and their fans.

The music video art form was born in the early 1980s when MTV and later other cable television channels created outlets for the broadcast of the format. One of the ground-breaking music videos was Michael Jackson's "Thriller," released in 1983. The video lasted fourteen minutes (more than twice as long as the underlying sound recording), was co-written and directed by a successful movie director, and included intricate costumes and vivid special effects. Soon thereafter, young directors were starting their careers by making music videos before moving on to direct films and television shows.

While music videos promote sales of sound recordings, the record label companies ("record labels") that produce the music videos seek an independent stream of revenue by licensing the music videos themselves. A record label's license for distribution of a music video conveys a bundle of rights, including the copyright in the integrated audiovisual work.^[12] As part of the production process, the record label also secures and pays for any choreography and script copyright rights that may be involved. Thus, a license of a music video generally conveys not just the right to reproduce, distribute and transmit the audiovisual work, but also warrants that other rights associated with the video have been "cleared" or secured by the record company. The license also typically authorizes the reproduction and transmission of the sound recording as part of the audiovisual work.

The license for the distribution of a music video will also generally convey rights in the composition embodied in the music video. These rights include the right to reproduce the composition embodied in the music video, known as the synchronization or "sync" right. (On occasion, the license may provide that the record label has secured the sync right.) It is the custom for the record labels to secure the sync rights either directly from recording artist-songwriters incident to those artists' recording agreement with the companies or to secure them separately from music publishers.

If the record label obtains the sync right from music publishers, the record labels get either a sync license for a single composition or a catalog-wide license such as the New Digital Media Agreement ("NDMA"). NDMAs have been entered between [REDACTED]. NDMAs allow record labels to pass through certain rights in the compositions to the companies that transmit or sell the licensed products to consumers.^[13]

There is one right, though, that licenses for music videos do not convey. They do not typically convey the public performance right in the composition embodied in the music video. Frequently, the licensee must procure such rights separately. Thus, in this litigation, ASCAP seeks a fee for the public performance of music in music videos distributed by Mobi.

V. Mobi

Mobi is a privately held company located in California. It was founded in 1999 with six employees and has grown since then to employ nearly 200 individuals.^[14]

A. Early Years and Growth

Mobi concentrated in its early years on creating and then licensing technology that would allow for the efficient transfer of data across wireless networks. By 2002, Mobi shifted its business to the recreation of the cable television experience over wireless networks.

Mobi describes itself today as a leading provider and platform for content delivery over wireless networks. Mobi's technology is essentially directed to the transfer of large amounts of data in a fluid manner over wireless networks, whether Mobi has a role in assembling the content or not.^[15] The transfer is enabled through Mobi software which was either pre-loaded in the handset or downloaded into the handset by the consumer.

Mobi also aggregates television and radio channel content from third parties, programs its own music video channels and other specialty channels, and assembles this content into a product for delivery to wireless carriers through Mobi's technology platform. As a content provider, Mobi tries to assemble a broad range of programming with a mix of brand-name content and specialty content and offer a mix of live,^[16] clip-linear,^[17] and VOD^[18] content. The range of programming includes a mix of channels in categories such as news, finance, sports, music, comedy, cartoon/kids, entertainment, women and weather. News has been particularly important to Mobi, with news channels among its most frequently viewed channels.

Mobi distributed the world's first live mobile television product when it distributed programming on the Sprint network in November 2003 under the brand name MobiTV. This service delivered continuously streamed, linear television channels to handsets, including television channels such as CNBC, Discovery, College Sports Television, and MSNBC. While Mobi began by offering fewer than twenty channels in a single package, it may offer as many sixty different channels of programming in a single package today. To replicate the cable or satellite television channel guides with which consumers are generally familiar, Mobi signed a patent license agreement with Gemstar-TV Guide in 2007. By 2009, Mobi was touting that it had passed the seven million subscriber mark.

B. Mobi's Product Lineup

Mobi refers to the subscription television and radio services that it distributes through wireless carriers as "products." A product is a package of television and/or radio channels that it or the wireless carriers license from cable television networks or other content providers. The products that are branded with Mobi's name include MobiTV, MobiRadio, and MobiVJ. Other products are branded with the wireless carriers' names, for instance, Sprint TV and AT & T XM Radio.

What follows is a description of some of Mobi's products, presented roughly in the chronological order in which they appeared. Throughout its brief history, though, the majority of Mobi's revenue has been earned from products appearing on the Sprint network.

1. MobiTV

MobiTV is known as an à la carte product because subscriptions to it are sold separately by the wireless carriers and not as part of a bundle that includes other products. As of today, MobiTV includes roughly 60 television channels distributed over half a dozen or so wireless carrier networks. The channels distributed over any particular wireless carrier's network vary slightly. Some of the most-watched channels that are distributed to every wireless carrier are Animal Planet, CNBC, Discovery, The Mic HipHop, The Weather Channel, Toon World TV Classics, and V40 Top Hits.^[19]

After beginning with the distribution of MobiTV over the Sprint network in November 2003, Mobi began in January 2005 to distribute MobiTV through Cingular and the Midwest Wireless Networks in January 2005. Later in 2005, MobiTV began to be distributed over the Alltel, now Verizon, network.

2. MobiRadio

Mobi's radio product, MobiRadio, was first distributed on the Cingular wireless network in November 2005. It consisted of 40 channels of commercial-free digital music supplied by Music Choice. In July 2006, Mobi added other stations such as ESPN Radio, The NPR Program Stream, and The Weather Channel Radio Network. In late 2007, the radio channels offered by DMX, Inc. replaced the Music Choice channels.

3. Music Videos

In January 2006, Mobi began to program its own music video channels by licensing music videos from two or three of the four major record labels. In the years since then, the number of record labels supplying music videos to Mobi has decreased. Over time, Mobi has created several different music video channels which it has inserted into many of its product offerings.

4. Sprint TV

Less than a year after Sprint began to offer MobiTV to its customers, it changed the name of the product to Sprint TV. Sprint TV was launched in August 2004, and is delivered to Sprint customers as a component of a Sprint "bundle," that is, a subscription service that provides the data necessary to use the internet, text messaging, and email services, as well as the Sprint TV programming, or, for customers who have already paid for data plans, as a stand-alone product for an additional monthly subscription fee.

In May 2006, Mobi released a Spanish-language live television service called Sprint TV En Vivo. It consists of fifteen channels. As of mid-2009, this service is Mobi's fifth largest à la carte television product, measured in terms of revenues.

While Sprint relies on Mobi's back-end technological services to provide content over its wireless network, it has become less dependent on Mobi to acquire the rights to that content. After beginning to acquire a small amount of its own content in late 2005, by the Spring of 2007, Sprint had independently acquired a large proportion of its television programming content directly from cable television networks.^[20] Mobi and Sprint cooperated on this transition, which significantly reduced Mobi's role as a supplier of television programming content to Sprint. Mobi, however, still provides the programming content for the Sprint TV Extra and Sprint TV en Vivo products on the Sprint network.

Mobi also provided Sprint with Mobi-programmed music video channels for Sprint to include in a bundled offering called Music Pack. Music Pack also includes web browsing, navigation and television and radio content that Sprint independently licenses from others, including from cable television networks and other content providers. The agreement under which Mobi supplied four music video channels to the Music Pack expired in March 2009.

5. AT & T (formerly Cingular)

In May 2006, Mobi announced the launch of a live television package available to laptop users over AT & T's WiFi network. The product included fifteen channels on either a per month or 24-hour session rate. Later that same year, Mobi released AT & T Broadband TV, a twenty channel subscription live television service available on PCs for a monthly fee. These laptop/PC products were discontinued in 2009.

In the Fall of 2006, Mobi began providing back-end technology services for Cingular's XM Radio product, which consisted of up to 25 channels of programming. Cingular was responsible for securing all necessary rights directly from XM Radio.

In October 2007, Mobi launched its MobiVJ product on AT & T handsets. It features Mobi-programmed music video channels, but also includes radio and television content such as Access Hollywood and Fashion TV, to give two examples.

6. Mobi4BIZ

In September 2008, Mobi announced the creation of Mobi4BIZ, which as its name suggests, includes programming of business or financial news such as Bloomberg Television, CNBC, and TheStreet.com. A subscriber can customize Mobi4BIZ to follow certain companies or stocks in a ticker at the bottom of the screen.^[21]

7. Apple

In March 2009, Mobi released its first application ("app") for the Apple iPhone, the NCAA March Madness on Demand product. This product permits purchasers to watch live video of all sixty-three games of the NCAA men's college basketball championship tournament. In 2009, in partnership with NBC Sports, it added all of Notre Dame's home football games as the Notre Dame Central app.

C. The Economics of Mobi's Business

Most of Mobi's à la carte television products are offered by wireless carriers at retail for roughly $10 per subscriber per month. The wireless carriers pay [REDACTED] to Mobi for the content rights and the technical services that underlie the wireless delivery of the Mobi à la carte offerings.^[22] Mobi tries to spend about [REDACTED] of the [REDACTED] on the purchase of content, but on occasion spends more than that.

When Mobi's aggregated programming is bundled with other products offered by wireless carriers, Mobi receives a flat dollar figure per subscriber per month based on the relative value of the Mobi service to the bundle. This flat fee can be significantly less than [REDACTED] per subscriber per month. If the product is sold as part of a bundled package, the wireless carrier does not supply Mobi with an accounting of how much the carrier received from the customer for that sale. The payments to Mobi from the carriers are not affected by whether a subscriber actually watches any Mobi-supplied content.^[23]

Mobi gives wireless carriers a report that ranks the channels in a product in terms of minutes viewed, minutes per subscriber, megabytes consumed,^[24] usage per type of handset, etc. The carriers use this data to monitor usage trends and the load on their networks. The audio-only products that Mobi delivers consume only about one-quarter of the data per minute that audiovisual content consumes.

Mobi markets itself to the wireless carriers as a way to increase the number of consumers that subscribe to unlimited data plans and to increase purchases of more expensive handsets to play Mobi content. It portrays itself as a synergistic partner to the wireless carriers, helping drive demand for data packages that are lucrative to the carriers.

To obtain content, Mobi negotiates with executives at cable television networks, broadcast networks and others, such as executives at major sports leagues. Mobi argues in these negotiations that its service provides another stream of revenue (in the form of licensing fees from Mobi) and a growing subscriber base for the programming. The rates negotiated by Mobi depend on factors such as the number of subscribers or the popularity of the channel as compared to other Mobi channels.

Mobi pays almost all of its content providers a monthly per-subscriber fee for the right to distribute their content.^[25] Mobi gives content providers a monthly report showing total usage in minutes per day aggregated across all products in which the channel is supplied. It does not show the content provider a breakdown by wireless carrier.

This structure for Mobi's affiliation agreements with the content providers is similar to the affiliation agreements that cable television networks typically enter when distributing their content. These per-subscriber affiliate fees, combined with whatever advertising the television network is able to sell, provide the television network with the compensation that it will receive for distribution of its product.

All of the licenses between Mobi and content providers provide that the content providers will secure and pay for every copyright right and other payment right associated with every person or entity that contributes to the creation of the programming with a single exception. Every content license separately and explicitly addresses the obligation to secure the public performance license for musical compositions embedded in the programming. Many content providers have assumed the obligation to secure such licenses; others have not.

Another aspect of Mobi's business is the music video channels that it programs and sells to carriers. Mobi pays the major record labels at [REDACTED]. Mobi has sold these programmed channels to carriers either as a stand-alone product for a per-subscriber fee or in an undifferentiated bundle that includes the music video programming, other channels, and fees for back-end technology services.

Finally, Mobi earns revenue by supplying the technology services that permit wireless carriers to deliver content that the carriers themselves have licensed directly from content providers. Providing such back-end technology services can earn Mobi less than [REDACTED] per subscriber per month. The contracts in such situations require the wireless carrier to secure all of the rights associated with the content.

While Mobi's revenue has grown every year, [REDACTED]. Mobi projects that it will [REDACTED]. Mobi's income statement for the 2008 fiscal year illustrates the flow of its revenue and its impact on Mobi's financial condition.

In fiscal year 2008, Mobi earned just over [REDACTED] in revenue, with over [REDACTED] of this figure being attributed to subscription revenue, whether from the supply of back-end technology services to wireless carriers or to the provision of content as well as that technology.^[26] Sixty-two percent of Mobi's subscription revenue is attributable to products in which Mobi secured the content that was distributed over the handsets. In fiscal year 2008, net advertising revenue totaled about [REDACTED].^[27] Mobi's cost of revenue, which amounted to just over [REDACTED], includes content-provider fees of over [REDACTED].^[28]

D. Mobi's SESAC License: Its First PRO License

Initially, Mobi used MRI to help it obtain licenses for the public performance of music. Pursuant to MRI's advice, Mobi requested blanket licenses from both ASCAP and BMI in 2003. On February 15, 2005, SESAC accused Mobi of being a willful copyright infringer; Mobi had not yet requested or obtained a SESAC license.

During the ensuing negotiations, SESAC represented that it controlled approximately 10% of the market for public performance rights in compositions used in U.S. television content. Expecting to have to pay roughly 1% of revenue to the three United States PROs, and understanding that SESAC should receive about 10% of that figure, Mobi offered a fee roughly equivalent to 0.1% of Mobi's revenue. Ultimately, the parties agreed in a June 25, 2007 agreement that Mobi would pay more than that. The settlement agreement, which included a release of any copyright infringement claims related to Mobi's prior public performance of SESAC music, required Mobi to pay SESAC a lump sum for prior years "in full satisfaction and settlement of any and all SESAC claims... for such period," a lump sum for 2007,^[29] and on a going-forward basis for 2008, the greater of (i) a set figure or (ii) [REDACTED] of Mobi's gross revenues. This gross revenue calculation excluded revenues that Mobi received for its [REDACTED]. The license contains the following limitation:

[REDACTED]

Even if SESAC could be shown to control 10% of the public performance rights in the programming at issue here, Mobi paid SESAC a fee that is proportionately higher than the fee it contends that it should pay ASCAP (or BMI). For the reasons that will be explained below, Mobi's payment to SESAC is a poor yardstick to measure the fairness of any fee to ASCAP and has limited relevance to this proceeding.

VI. ASCAP Licenses

A. ASCAP's Post-Turner Licenses

Shortly after the turn of the new century, and after lengthy litigation known as the Turner litigation,^[30] ASCAP and the cable television networks settled their disputes over the valuation of a public performance TTTA license and agreed on a three-tiered rate structure to be calculated as a percentage of cable television network revenue. The tiers depended on the music intensity of the programming offered by the network, specifically whether the programmed network was "music intensive," "general entertainment," or "news and sports," with licensing rates, respectively, of 0.9%, 0.375%, and 0.1375% ("Post-Turner Licenses").^[31] The first set of Post-Turner Licenses, executed in 2002, included both backward-looking settlements, usually a lump-sum payment covering a span of years, and forward-looking fee provisions adopting one of the three rates. By 2004, ASCAP had adopted three form license agreements, one for each of the three rates. More than 100 Post-Turner Licenses have been issued by ASCAP, with the most recent having been executed in 2009. A dozen or so of the Post-Turner Licenses will expire during 2012.^[32]

ASCAP's "music intensive" license generally covers channels that feature music videos, music-focused feature films, and music variety shows. Examples of music intensive networks licensed at the rate of 0.9% under the form Post-Turner License are the Gospel Music Channel Network, Fuse Network, and Artists & Fans Network. An example of a general entertainment cable television network is Turner Broadcasting's TNT Network, whose [REDACTED] license had the rate of 0.375%. Examples of networks that have been licensed on the ASCAP sports and news form license, at a rate of 0.1375%, are the Fox Sports and the Fox News networks.

Beginning in late 2007, ASCAP began to issue interim fee agreements with cable television networks, rather than renewing or executing new Post-Turner Licenses. ASCAP apparently intends to enter into a new round of negotiations with the cable television network industry regarding the licensing of ASCAP rights.

B. The Terms of the Post-Turner Licenses

Because of their importance to the analysis that follows, it is helpful to describe in detail some of the terms of the Post-Turner Licenses. In brief, while the threetiered Post-Turner Licenses distinguish among networks based on the content of their programming, they do not distinguish among them based on the manner in which that content is delivered. For example, they do not distinguish among the various kinds of distribution mechanisms through which a network may arrange to transmit its programming to the public and they do not distinguish between content that is delivered in a linear versus a VOD format.

ASCAP's form Post-Turner License, which is entitled the ASCAP Cable Television Programming Service Blanket License Agreement, classifies music videos and motion pictures and films in the musical genre as non-dramatic performances covered by the license. Its all-encompassing definition of a Distribution System includes:

a cable television system, MMDS, SMATV, TVRO, satellite direct to home system, direct broadcast satellite system, Internet, broad band or any other means or method which is or hereafter may be used to transmit or receive a "Programming Service" (including transmission or reception on a video-on-demand, or near-video-on-demand, subscription video-on-demand pay-per-view or pay-per-period basis) except that Distribution System shall exclude free over-the-air broadcast television, and any transmission via the Internet or other online service that is not at least one of: (i) broadcasts, transmissions or retransmissions of substantially the same program content as that of a "Programming Service" transmitted or received via a cable television system, MMDS, SMATV, TVRO, satellite direct to home system or direct broadcast satellite system (or other similar system as may hereafter be used for similar purposes); or (ii) an Internet site, or musical compositions broadcast, transmitted or retransmitted on an Internet site, used primarily to promote any "Programming Service" and/or the exhibition of any "Program."

(Emphasis supplied.)

The definition of a Distribution System, as just shown, depends in part on the definition of a Programming Service. The "Programming Service" refers to the network to which the license applies, and a Program is defined as

any show, motion picture or film (including motion pictures or films in the "musical" genre), music video, television program, sports and sports related program, commercial advertisement (of any duration), promotional announcement, interstitial program, public service announcement or other type of audiovisual material transmitted as part of any Programming Service.

(Emphasis supplied.)

The all-encompassing nature of the license is reflected as well in the description of the TTTA license itself. The license defines a "Through-To-The-Viewer License" as "a license which authorizes the transmission or retransmission of the Programming Service to LICENSEE's Distribution Systems and the simultaneous retransmission of the Programming Service by any of LICENSEE's Distribution Systems by whatever means to subscribers or viewers."^[33] (Emphasis supplied.)

The fee from these licenses is based upon the network's gross revenues, and the license makes no mention of the subscription revenue generated by any affiliate that is responsible for the direct distribution of the content to the consumer. Gross revenues are thus defined in pertinent part to comprise

(i) monies or other consideration received by LICENSEE from Distribution Systems and directly from subscribers to LICENSEE's Programming Service;

(ii) advertising revenues or other monies received by LICENSEE from sponsors if any ....

(Emphasis supplied.) Gross revenues exclude "payments to Distribution Systems for LICENSEE's Programming Service." (Emphasis supplied). Thus, and to underscore this point again, the Post-Turner License does not require that the network pay ASCAP any portion of the revenues collected at the retail level by the network affiliates. In terms of the scope of the license, the license forbids the licensee from granting others the right to publicly perform ASCAP music "except that nothing in this paragraph shall limit or curtail the effect of the Through-to-the-Viewer License granted by this Agreement."^[34]

C. Application of the Post-Turner Licenses to Wireless Distribution Systems

The parties hotly contest whether ASCAP, after its adoption of the form Post-Turner Licenses, ever acted to include the distribution of programming content over a wireless distribution system as a method of delivery encompassed within the license. Despite contrary testimony from ASCAP's witnesses,^[35] the documentary record establishes that it was ASCAP's practice until recently to try to collect licensing fees from cable television networks based on revenues from wireless distribution services. That this is true is unsurprising given the sweeping definitions in the Post-Turner Licenses of a Distribution System and other terms.

Thus, as early as 2006, ASCAP confirmed in correspondence with licensee [REDACTED] that its Post-Turner License would encompass wireless distribution activities. In its October 2, 2006 letter to [REDACTED] during their licensing negotiations, ASCAP represented that their "mutual understanding" of their Post-Turner License was that "The Definition of `Distribution System'" in the accompanying signed Agreement includes "wireless networks and any other forms of video programming distribution (now or hereafter devised) and any Internet site operated by LICENSEE." (Emphasis supplied.) And because [REDACTED] license was a TTTA license, ASCAP acknowledged that it "would not be entitled to license separately performances contained in Programs transmitted as part of any Programming Service licensed under the Agreement by any `Distribution System.'"

Then, in a series of audits of its Post-Turner licensees that ASCAP initiated as recently as 2008 and 2009, ASCAP asked its licensees for revenue information for wireless distribution systems.^[36] In at least five such instances, ASCAP's audit request referred specifically to Mobi. In audits in 2008 and 2009 of Comedy Central, the Bravo Network, USA Networks and Sci-Fi Channel, and Discovery, ASCAP demanded to see revenues that these cable television networks received from Mobi. In these demands, ASCAP consistently characterized Mobi as a "Distribution System," as that term is defined in its Post-Turner Licenses.

In July 2008, for example, the audit notice that ASCAP's outside auditor sent to the USA Network and Sci-Fi Channel stated: "In conformity with USA's and Sci-Fi's license agreements with ASCAP authorizing the examination, we will require the following information .... 9) Access to all contracts between USA and Sci-Fi and the following Distribution Systems[:] Comcast Cable[,] Time Warner Cable[,]... Dish Network[,] ... Mobi TV[,] Verizon VCast, ATT[,] Sprint ...." ASCAP had given its auditor this list of Distribution Systems.

In a January 2009 "Draft Report" of the audit results for both the Bravo Network and the Sci-Fi and USA Networks, the auditor stated: "The examination was performed within the scope of the License Agreements, the relevant provisions of which were the same in both License Agreements." The draft then copies the definitions of "Distribution Systems," "Programming Service," and "Gross Revenues" from Bravo's License Agreement. It continues: "Accordingly, our examination endeavored to determine the royalties to be paid during the Examination Period by the three Networks using such definitions and focused on performances of the Programming Services via three different Distribution System areas: (1) MSOs (i.e. cable/satellite systems); Internet web sites; and (3) wireless services." (Emphasis supplied.)

ASCAP's Schwind wrote to Comedy Central in September 2008, demanding the right to audit revenue from "non-linear platforms," and Mobi specifically. He stated:

I assume that by `non-linear platforms' you mean such businesses as Apple iTunes, MobiTV or Sprint, as well as various Internet distributors. I refer you to the provisions of the License Agreement, a copy of which is enclosed for your convenience, that define both `Distribution System' (Section XVIII(4)) and `Gross Revenues' (Section XVIII(9)). Stated simply, these provisions entitle ASCAP to verify that the gross revenues reported to ASCAP by Comedy Central include revenues, or the value of other consideration, Comedy Central has received from, among others, all of such `non-linear' distribution systems ....

(Emphasis supplied.)

Similarly, a March 17, 2009 letter from an ASCAP auditor to Discovery stated: "Pursuant to the audit notification letter dated January 22, 2009, ... we will require the following information in order to complete our examination .... Access to all statements from and contracts between Discovery and it's [sic] distribution systems, including but not limited to" a list that included "Mobi TV."

At some point after the commencement of this litigation, probably by mid-2008, and certainly by May of 2009, ASCAP made a decision that it no longer wanted to construe the Post-Turner Licenses as including the requirement that licensees pay ASCAP based on wireless distribution revenue. In late 2008, ASCAP made a change to the language in its Post-Turner License with [REDACTED] to exclude content delivered through a wireless service from the TTTA license. Thus, ASCAP's November 18, 2008 blanket license agreement with [REDACTED] added to the list of exclusions from the definition of Distribution System: "any wireless service, system or carriers." Then, in the Spring of 2009, ASCAP even went so far as to return a small payment made by Discovery due to Discovery's determination that it had underpaid ASCAP for the revenues received from wireless distribution activities.^[37]

D. ASCAP Licenses with Cable System Operators

Because the Post-Turner Licenses do not cover all uses of music in television programming, ASCAP also directly licenses cable system operators. Cable operators are licensed under a form agreement that resulted from negotiations with an industry trade association and that currently requires payments at an annual per-subscriber rate of $0.099.^[38] These agreements cover content that is not covered by any other ASCAP license, such as locally originated programming and the advertising that runs with it, programming on leased access channels, advertising material that is inserted at a local level into other programming, VOD programming that is not otherwise licensed, and some local or regional news and sports programming.

E. Audio-Only Programming

Mobi not only provides audio-visual programming over handsets, but also provides audio-only products. MobiRadio is Mobi's audio-only product, for which Mobi licenses music from DMX (and in prior years, from Music Choice). Mobi also provides back-end technology services for XM Satellite Radio, Inc. ("XM Radio") and Sirius Satellite Radio, Inc. ("Sirius") products.

ASCAP has licensed audio-only channels programmed by Music Choice, [REDACTED]. Music Choice provides nearly fifty music channels to televisions via cable and satellite and to computers over the internet. The ASCAP licensing agreement with Music Choice [REDACTED] requires Music Choice to pay ASCAP a licensing fee of 2.5% of its gross revenue.

The 2.5% rate of gross revenue applies as well to the licenses currently in effect between ASCAP and [REDACTED] for their programmed all-music channels that are provided over a [REDACTED] service to subscribers' homes. ASCAP also has a different license with [REDACTED] for its subscription-based [REDACTED]. This service is a mix of roughly half-music and half-talk radio channels, and is licensed at a rate of 1.44% of gross revenue.

CONCLUSIONS OF LAW

On November 11, 2003, Mobi applied for an ASCAP TTTA license "only to the extent required" for its "service that allows mobile handset users to access television and other content by aggregating television and other audio/visual content for transmission over telecommunications networks."^[39] After the parties failed to reach agreement on the terms of such a license, ASCAP filed this lawsuit on May 5, 2008 in order for the rate court to determine "a reasonable fee retroactive to the date of the written request for a license." AFJ2 § IX. ASCAP contends that Mobi owes $41.3 million for the period November 11, 2003 through the end of 2011. Mobi contends that it owes only a fraction of that amount, specifically $301,257.99 for the period from December 2003 through July 2009.

These conclusions of law will begin with a description of the Copyright Act section at issue in this litigation, the origin of the consent decree, the ASCAP rate court litigation in the 1990s that addressed the rate that would apply to licenses in the cable television industry (the Turner litigation), and the amendment of the consent decree in the antitrust litigation that created AFJ2 in 2001, before proceeding to describe the jurisprudence addressed to the aspects of a TTTA license in dispute here. After these principles are described, the governing legal standard will be applied to ASCAP's proposal for a reasonable rate, and then to Mobi's proposal. Throughout this determination, the burden rests on ASCAP to demonstrate "the reasonableness of the fee it seeks." AFJ § IX(B). Scattered throughout this discussion will be additional findings of fact.

I. The Non-Dramatic Performance Right

This dispute arises from one of the rights created by the Copyright Act that may be owned and conveyed by composers of music. See United States v. ASCAP (In re Application of Cellco P'ship d/b/a/ Verizon Wireless), 663 F. Supp. 2d 363, 368-69 (S.D.N.Y.2009); 17 U.S.C. §§ 106, 201. That right is the right to publicly perform the musical work. 17 U.S.C. § 106(4).

Most of the music that is used in the type of television programming at issue in this dispute is created under composer-for-hire agreements with the producers of the television programs. Under the Copyright Act, "[i]n the case of a work made for hire, the employer or other person for whom the work was prepared is considered the author ..., and, unless the parties have expressly agreed otherwise..., owns all of the rights comprised in the copyright." 17 U.S.C. § 201(b). By historical practice, however, the producer retains all of the rights comprised in the copyright except the public performance right, which is granted back to the composer and/or a music publishing company with whom the composer has a contract. See Buffalo Broad. Co., Inc. v. ASCAP, 744 F.2d 917, 921 (2d Cir.1984) ("Buffalo Broadcasting"). See also Alden-Rochelle, Inc. v. ASCAP, 80 F. Supp. 888, 892 (S.D.N.Y.1948) (explaining the origin of this assignment practice in the film industry). The result is that a broadcaster who transmits programming that contains copyrighted musical works without a license from the copyright holder faces liability for copyright infringement. ASCAP exists to efficiently license these public performance rights of its composer and publisher members.

II. The Consent Decree

A. Origins of the Consent Decree

Because ASCAP collectively licenses its members' songs, which enhances their market power in negotiating with music users, it attracted the attention of the United States Department of Justice's antitrust division ("DOJ"). DOJ sued ASCAP in 1941 for alleged violations of the Sherman Antitrust Act related to ASCAP's licensing of the public performance right to copyrighted music within ASCAP's repertory. The lawsuit resulted in a 1941 consent decree.

In 1948, a court decision of importance to the development of the TTTA license was issued. The court entered an injunction restraining ASCAP's members from refusing to grant to motion picture producers the right to publicly perform music through the exhibition of motion pictures. Alden-Rochelle, 80 F.Supp. at 900. The court rejected the contention that this requirement that ASCAP's members offer a license to the upstream distributor of the works would not result in any payment by the exhibitors for the downstream public performance:

But that in some way the value of the performing rights would be claimed by the copyright owner and eventually would be passed on to the exhibitor, I have no doubt at all. The ultimate result would be that the exhibitor would not be separately charged for the performance rights, as he is now through Ascap, but he would be charged for those rights in the total rental he would pay for the film.

Id. at 896.

In response to this injunction and the increased popularity of television, the 1941 consent decree was amended substantially. The Amended Final Judgment was entered in 1950 ("the AFJ"). The AFJ prohibited ASCAP from acquiring exclusive music performing rights and from limiting, restricting, or interfering with the right of any member to issue to any user a non-exclusive license. Additionally, the AFJ required ASCAP to grant a license to any applicant who requested one and created the rate court to resolve disputes in the event the parties were unable to agree on reasonable fees. Buffalo Broadcasting, 744 F.2d at 922-23.

The AFJ was amended in 2001 to create the Second Amended Final Judgment ("the AFJ2" or "Consent Decree"). The Consent Decree currently regulates how ASCAP may participate in the music industry and gives this Court exclusive jurisdiction to oversee the implementation of AFJ2. This context explains why "rate-setting courts must take seriously the fact that they exist as a result of monopolists exercising disproportionate power over the market for music rights." United States v. BMI (In re Application of Music Choice), 426 F.3d 91, 96 (2d Cir.2005) ("Music Choice IV").

B. The 1990s Turner Litigation

Litigation in the 1990s that came to be known as the Turner litigation established the principle that cable television networks were as entitled to a TTTA license as were over-the-air television broadcasters. By the terms of the AFJ, ASCAP was required to issue TTTA licenses to broadcasting networks upon request. It read in pertinent part:

Defendant ASCAP is hereby ordered and directed to issue, upon request, licenses for rights of public performance of compositions in the ASCAP repertory... [t]o a radio broadcasting network, telecasting network or wired music service ..., on terms which authorize the simultaneous and so-called "delayed" performance by broadcasting or telecasting, or simultaneous performance by wired music service, as the case may be, of the ASCAP repertory by any, some or all of the stations in the United States affiliated with such radio network or television network or by all subscriber outlets in the United States affiliated with any wired music service and [t]o not require a separate license for each station or subscriber for such performances[.]

AFJ § V(A) (quoted in Turner, 782 F.Supp. at 784-85 (emphasis supplied)). This right was referred to as either "licensing at the source" or licensing "through to the viewer," Turner, 782 F.Supp. at 781, and is referred to herein as the TTTA license.

Confronted with a new industry, ASCAP took the position that the TTTA licensing requirement in § V(A) of the AFJ applied to traditional television broadcasting networks, like NBC, but did not apply to cable television. In 1989, Turner Broadcasting System, Inc., a cable program supplier, sought an order from the rate court that it was also entitled to a TTTA license. ASCAP contended that the cable industry required two separate licenses, or "split" licenses: one from the cable television network to transmit the programs to the cable system operators, and another from those system operators to transmit the programming to viewers. Id. at 781-82. ASCAP protested that the cable television industry was so different from over-the-air television "in its technology and financial structure" that the AFJ drafters could not have intended the AFJ to require it to offer a TTTA license to the cable program suppliers. Id. at 799.

The rate court disagreed, and held that the TTTA provision in the AFJ was both an "effort at inclusiveness" of different types of communications technology and an effort to equalize bargaining power. Id. at 791. As the court explained it, the right to request a license at the source of the programming placed the negotiating parties on more equal footing. It observed,

The problem that Article V(A) [of the AFJ] addresses is potentially found whenever programming is packaged by an entity for transmission to the public by another entity, and the solution adopted by the Decree rests on the fact that the packager has greater ability to negotiate on equal terms with ASCAP than does the affiliated telecaster. These considerations are not at all affected by the technology of transmission or by the financial arrangements between the two entities.

Id. (emphasis supplied). The court construed the term "telecasting network" in AFJ § V(A) in its functional sense, i.e., to cover not only the traditional networks but also the supplying of programming by a packager to another entity for transmission to viewers. Id. at 795. See also United States v. ASCAP (In re Application of the Nat'l Cable Television Ass'n), No. 41 Civ. 1395(WCC)(MHD), 1999 WL 335376, at *8-9 (S.D.N.Y. May 26, 1999) (construing Turner).

The upshot of the 1991 Turner decision and ensuing litigation was a lengthy period of negotiation and ultimately the issuance of a host of TTTA licenses to cable television networks. Those licenses were entered pursuant to the successor to AFJ, that is, in the wake of AFJ2.

C. The Negotiation of AFJ2's TTTA License Provision

In the late 1990s, DOJ and ASCAP began to negotiate a proposed second amended final judgment to replace AFJ. Throughout these negotiations, the requirement that ASCAP issue a TTTA license at the source of programming, in the event one was requested, was not in dispute. What was the subject of discussion, however, was how to capture the concept that the value of the downstream performance should be part of that licensing calculation. A letter that DOJ wrote on March 11, 1999 to ASCAP describes its approach as of that date:

In form but not in substance, thru-to-the audience has been changed substantially. The idea is that a thru-to-the-audience license would cover any firm in the downstream economic chain of distribution but would not cover anyone who did not in some way contract for the performance or the program containing the performance (e.g., the bar that turns on the radio).... We have also added an express provision permitting fees to reflect downstream usage.

In the draft circulating between the parties at the time that letter was written, the final sentence of § V read: "The fee for any such license may reasonably reflect its expanded scope."

Ultimately the current form of AFJ2 § V emerged. That provision reads:

ASCAP is hereby ordered and directed to issue, upon request, a through-to-the-audience license to a broadcaster, an on-line user, a background/foreground music service, and an operator of any yet-to-be-developed technology that transmits content to other music users with whom it has an economic relationship relating to that content .... The fee for a through-to-the-audience license shall take into account the value of all performances made pursuant to the license.

AFJ2 § V (emphasis supplied). A TTTA license is defined in AFJ2 as a "license that authorizes the simultaneous or so-called `delayed' performances of ASCAP music that are contained in content transmitted or delivered by a music user to another music user with whom the licensee has an economic relationship relating to that content." AFJ2 § II(S).

When this language was published during the public comment period, organizations and companies representing the "vast majority" of those providing license income to ASCAP, and representing both "new and traditional media," requested, inter alia, that the final sentence of § V be deleted. In their December 4, 2000 submission, they complained that, despite the existence of the Consent Decree and a rate court, there was still no meaningful competition in the marketplace for music performing rights. They complained that ASCAP's intransigence during negotiations had resulted in lengthy and costly rate court proceedings. They pointed to the Turner litigation and noted ASCAP's attempt to stymie cable television requests for a TTTA license. As for the final sentence in § V, they argued that it was unnecessary and

lends itself to misuse by ASCAP, as it conceivably could be cited by ASCAP as alleged support for its position in the pending cable rate proceeding that it is `entitled' to fees based upon the revenues not only of the cable networks but also of their distributors.

It proposed that "[a]t the very least, the word `value' in this sentence should be changed to the phrase `economic significance, if any.' The word `value,' as used in this provision, is completely ambiguous" and "is a meaningless economic concept."

DOJ responded to the comments made during the public notice period in a lengthy memorandum, noting that it was adopting some of the changes proposed in the comments to enhance the pro-competitive features of AFJ2, but was not incorporating all of the proposed amendments. As for § V, it noted that, when combined with the new definition of broadcaster in AFJ2, the section expanded the class of music users entitled to a TTTA license. As for that section's final sentence, it observed:

One comment objected to the requirement that fees for a through-to-the-audience license `take into account the value of all performances made' on the grounds that ASCAP could seek fees based on the revenues of both the licensee and its distributors. The provision does not support such an interpretation, but rather is simply an acknowledgement that the fee for a through-to-the-audience license shall properly reflect its expanded scope. In the case of cable television networks ... ASCAP would bear the burden of proving why the economic relationship between the network and the distributor does not already reflect the value of the music in the network programming licensed to the distributor (e.g., why a percentage of the network's revenue is not sufficient.)

In its March 28, 2001 rejoinder to this response, ASCAP argued for a different interpretation of the final sentence of § V:

If [DOJ] is merely suggesting ... that ASCAP bears the burden of proof on most issues in rate proceedings, it is correct. If on the other hand, it is suggesting that the fee for a [TTTA license] should not recognize the benefits realized by all persons who perform music pursuant to the license, it is incorrect. Unless an "upstream" user obtains a [TTTA license], "downstream" users would require separate licenses to perform music contained in any content they receive from their suppliers ..... The appropriate license fee depends on the value of the license to all the music users whose performances its [sic] authorizes. There may be numerous situations in which an upstream user's revenue might not reflect much, if any, of the value of downstream performances. In those situations, the license fee should reflect the value of downstream users' performances by taking their revenue into account ....

AFJ2 was entered by the court on June 11, 2001, and went into effect on September 11, 2001. It governs this proceeding.

D. Other Provisions of AFJ2

There are other provisions of AFJ2, in addition to § V's TTTA licensing requirement, that are pertinent to this litigation. Section IX of AFJ2 governs the determination of reasonable fees. When a party requests a license from ASCAP to publicly perform works from the ASCAP repertory, ASCAP must advise the applicant of a fee it deems reasonable or ask for more information in order to determine a reasonable fee for the license requested. If the parties are unable to reach agreement on a reasonable fee, either ASCAP or the applicant may apply to the rate court for the determination of a reasonable fee. AFJ2 § IX(A).

In any fee-setting proceeding, ASCAP bears the burden of proof that the fee it seeks is reasonable, absent exceptions not relevant here.^[40] AFJ2 § IX(B). If ASCAP does not establish the reasonableness of its proposed fee, "the Court shall determine a reasonable fee based upon all the evidence." AFJ2 § IX(D). Finally, "[w]hen a reasonable fee has been determined by the Court, ASCAP shall be required to offer a license at a comparable fee to all other similarly situated music users who shall thereafter request a license of ASCAP." AFJ2 § IX(G).

Similarly, AFJ2 prohibits ASCAP from "[e]ntering into, recognizing, enforcing or claiming any rights under any license for rights of public performance which discriminates in license fees or other terms and conditions between licensees similarly situated." AFJ2 § IV(C). Music users or licensees are "similarly situated" if they are

in the same industry that perform ASCAP music and that operate similar businesses and use music in similar ways and with similar frequency; factors relevant to determining whether music users or licensees are similarly situated include, but are not limited to, the nature and frequency of musical performances, ASCAP's costs of administering licenses, whether the music users or licensees compete with one another, and the amount and source of the music users' revenue[.]

AFJ2 § II(R).

AFJ2 also describes the types of licenses that ASCAP must grant. An ASCAP license can take one of three forms, which differ in the comprehensiveness of the programming that they cover. At issue in this case is the "blanket license," which is a "non-exclusive license that authorizes a music user to perform ASCAP music, the fee for which does not vary depending on the extent to which the music user in fact performs ASCAP music." AFJ2 § II(E). It grants a music user the right to use any song in the ASCAP repertory in any program any number of times, for one fee. Buffalo Broadcasting, 744 F.2d at 922. If a music user requests a "per-program" or "per-segment" license instead of a blanket license, ASCAP must grant that type of license and quote a fee for it.

III. Fair Market Value

While AFJ2 imposes on the rate court the task of setting a "reasonable" fee for the requested license, that term is not defined in AFJ2. Governing precedent dictates, however, that in determining the reasonableness of a licensing fee, a court "attempts to make a determination of the fair market value — the price that a willing buyer and a willing seller would agree to in an arm's length transaction." United States v. BMI (In re Application of Music Choice), 316 F.3d 189, 194 (2d Cir.2003) ("Music Choice II") (quoting ASCAP v. Showtime/The Movie Channel, 912 F.2d 563, 569 (2d Cir.1990)). Fair market value is, of course, a "hypothetical" matter. Showtime, 912 F.2d at 569. "[S]ince there is no competitive market in music rights, the parties and the Court lack any economic data that may be readily translated into a measure of competitive pricing for the rights in question". Id. at 577 (district court opinion). Furthermore, in conducting the analysis of a reasonable fee, a court may consider "the context in which [the rate court] was created." Id. at 570. "The opportunity of users of music rights to resort to the rate court whenever they apprehend that ASCAP's market power may subject them to unreasonably high fees would have little meaning if that court were obliged to set a `reasonable' fee solely or even primarily on the basis of the fees ASCAP had successfully obtained from other users." Id.

A determination of the fair market value "is often facilitated by the use of a benchmark — that is, reasoning by analogy to an agreement reached after arms' length negotiation between similarly situated parties." Music Choice II, 316 F.3d at 194. When choosing a benchmark and deciding how that benchmark should be adjusted for the application it is considering,

a rate court must determine the degree of comparability of the negotiating parties to the parties contending in the rate proceeding, the comparability of the rights in question, and the similarity of the economic circumstances affecting the earlier negotiators and the current litigants, as well as the degree to which the assertedly analogous market under examination reflects an adequate degree of competition to justify reliance on agreements that it has spawned.

Music Choice IV, 426 F.3d at 95 (citation omitted).

A. Music Choice: Retail or Wholesale Price as a Superior Indicator of Fair Market Value?

The Court of Appeals for the Second Circuit recently discussed the role that the price that consumers pay to receive a product — the retail price — may have in setting the fair market value of a TTTA license requested by a broadcaster. In Music Choice II, the applicant Music Choice was a partnership composed of some of the largest entertainment and communications companies in America. At that time, Music Choice provided genre-based channels of music to consumers' homes via satellite or cable television.^[41] It originally paid BMI an amount equal to 2% of the cable/satellite systems operators' retail sales revenue from sales of Music Choice. Music Choice first offered its programmed channels through cable providers on an à la carte basis. By the time of the trial before the rate court, however, almost all of the service was offered as part of a bundle of cable or satellite television programming for which customers paid a blanket fee. With this transition, it became difficult to identify the portion of the retail revenues attributable to the customers' purchase of the Music Choice programming. Music Choice II, 316 F.3d at 192.

Both Music Choice and BMI agreed during the rate court proceeding that the fee should be expressed as a percentage of the wholesale revenue only, that is, the price paid to Music Choice by the cable/satellite operators, but they disagreed as to the appropriate benchmark for that rate and as to the rate itself. Id. at 193. BMI argued for a percentage that would approximate the amount BMI received under the old, retail revenue-based formula. Assuming that wholesale revenues were about half of the retail revenues, it argued for a 3.75% rate, or a bit less than double what it considered the appropriate rate for retail revenues.

The rate court set the fee at 1.75% of wholesale revenue, reasoning that the roughly 2% of wholesale revenues that were meant to approximate the extra amount due from retail revenues should not be included. Id. at 193-94. The rate court explained that, in its view, BMI's proposed percentage, designed to approximate retail revenue, overstated the fair market value of the music:

The other components of the package for which the subscriber pays (the former tuner, the cable, the connections, the labor of installation, etc.) are not contributed by the author of the music, and there is no reason why the author should be compensated for their cost. Quite to the contrary, the true value of the music is expressed at the earlier stage where it is incorporated into Music Choice's programs. The blanket license authorizes the use of the music, and should have no regard to whether the mechanics of delivery are cheaper or costlier. Thus, the idea that to recover the full value of the music, the blanket license rate should include a component based on the cable or satellite operators' revenues, is misconceived.

United States v. BMI (In re Application of Music Choice), No. 64 Civ. 3787(LLS), 2001 WL 829874, at *6 (S.D.N.Y. July 23, 2001) ("Music Choice I") (emphasis supplied).

The Court of Appeals vacated the rate court decision. It noted that the rate court's decision had led it to set BMI's rate "at approximately half of the rate that had prevailed under its previous contract with Music Choice, as well as less than one half of the rate recently negotiated between BMI" and a principal competitor of Music Choice. Music Choice II, 316 F.3d at 195. It further observed, that, "in the absence of factors suggesting a different measure," id.,

what retail customers pay to receive the product or service in question (in this case, the recorded music) seems to us to be an excellent indicator of its fair market value. While in some instances there may be reason to approximate fair market value on the basis of something other than the prices paid by consumers, in the absence of factors suggesting a different measure the price willing buyers and sellers agree upon in arm's-length transactions appears to be the best measure.

Id. (emphasis supplied).

Nonetheless, in Music Choice II and its successor Music Choice IV, the Court of Appeals emphasized that there may well be good reasons not to use retail revenue as a measure of the music's fair market value. For example, "where customers pay a single fee for a package of audio and visual programming, which includes the music, it will be difficult to determine what part of the fees paid was for the music, as opposed to other programming," Music Choice II, 316 F.3d at 195 n. 2, or "[i]f it were demonstrated that retail purchasers were motivated to pay more because of advantages that resulted from a particular mode of delivery, such as better quality, better accessability [sic] or whatever," id. at 196 n. 3, then it may be preferable to base a rate on wholesale revenue.

Thus, the Court of Appeals cautioned that it did not want its rejection of the rate court's decision to be read as a rejection of estimating fair market value "by first fixing on wholesale revenue and then making appropriate adjustments to that figure to approximate fair market value." Id. at 197 n. 5. Indeed, basing a rate on wholesale revenue would be "[e]specially" appropriate where retail revenue attributable to the music is difficult to ascertain "because of the bundled packages offered at retail, while wholesale revenues attributable to the music are easily determined." Id.^[42] Despite this endorsement of the use of a wholesale revenue base for calculating fair market value in the appropriate case, however, the Court of Appeals continued to caution that such a calculation "is very different from proposing that wholesale revenue, in itself, represents the fair market value of what is being sold."^[43]Id.

When the litigation returned to the Court of Appeals for a second time in 2005, a different panel concluded that the rate court had again erred, but that it did so this time by adopting as a benchmark an agreement that it had earlier rejected "without explaining why its earlier rejection of that rate ... was wrong." Music Choice IV, 426 F.3d at 95 (citation omitted). The district court had set the rate for Music Choice at the same level as that in that benchmark agreement, 3.75% of gross revenues, which was designed to reflect a 2% increment attributable to retail revenues. The Second Circuit again reversed, noting: "[I]n spite of our endorsement of retail price as generally a good marker for fair market value, we did not require it to be used in all circumstances, but only `absent some valid reason for using a different measure.'" Music Choice IV, 426 F.3d at 97 (citing Music Choice II, 316 F.3d at 195). It exhorted the rate court that "it was free to find fair market value on any basis adequately supported by the record." Id. The Court of Appeals observed that "as long as the District Court includes retail value of the music in its valuation of the music rights (either as enshrined in a previous Agreement or as an approximation of the wholesale price or any other way it finds reasonable)," then it is free to choose a basis other than retail revenue as the measure of fair market value. Id. at 98 n. 8 (emphasis supplied). Ultimately, the Music Choice litigation resulted in the parties' agreement to a fee set on the basis of wholesale prices and at a rate of 1.75% of Music Choice's gross revenue for the period 1990-2005 and 2.5% of gross revenue for the period 2006-2010. See United States v. ASCAP (In re Application of America Online, et al.), 559 F. Supp. 2d 332, 382 ("AOL").

B. AOL Rate Court Ruling

Before proceeding to the issues raised by this litigation, there is only one other decision which should be described in any detail. ASCAP places great reliance on the April 2008 rate setting decision rendered in AOL, 559 F. Supp. 2d 332, which is now pending on appeal.^[44] The 2.5% rate applied in the Music Choice agreement with BMI to wholesale revenues for pure music channels delivered by satellite and cable transmission to televisions became, along with other benchmarks, the foundation of the rate set in AOL.

In AOL, the rate court set a fee for the blanket license to America Online, Real-Networks, and Yahoo!, Inc. for the streaming of content over the internet, including radio programs, music videos, movie trailers, sports, news, and games. The rate court's formula for its fee began with each company's total domestic revenue from licensed services, less some customary deductions. This revenue amount is adjusted using a "music use adjustment factor" that is a "fraction whose numerator is the number of hours music is streamed from the site and whose denominator is the total hours that visitors are using the site." Id. at 412. The resulting number is the relevant revenue base. To that base, the rate court applied a 2.5% rate to determine the fee due to ASCAP, citing Music Choice's agreement to pay 2.5% of its gross revenue as one relevant benchmark among several. Id. at 413. The AOL applicants agreed that the 2.5% figure was an appropriate rate for the ASCAP fee, but only insofar as it was applied to revenue directly generated from on-demand audio music performances.^[45]Id. at 400. They strongly disagreed with the other portions of the calculations used by the rate court.

IV. A Reasonable Rate

The parties do not dispute that Mobi requires a license for its role in distributing music to users of handsets. Their agreement essentially ends there. While their disagreements are many, the principal thrust of their dispute can be reduced to three major points of contention.

First, Mobi asserts that ASCAP should not seek to collect a licensing fee from Mobi for the public performance of music unless Mobi has secured the content (or in the parties' words, "sourced" the content) that is delivered to the handsets. Mobi asserts that ASCAP should look elsewhere for a licensing fee when another entity has procured the rights from a content producer to use content that contains music and when Mobi's only role in the distribution is the provision of the back-end technology services.^[46] ASCAP's fee proposal is premised on both Mobi-sourced content and content secured by others where Mobi provides back-end technology services. Next, ASCAP asserts that the revenue base for calculating the TTTA licensing fee that Mobi will pay ASCAP should be the retail revenues received by the wireless carriers; Mobi contends that that revenue base should largely be the monies it spends to purchase the content that it distributes. And, finally, the parties dispute whether the Post-Turner Licenses issued to the cable television networks should be used as a benchmark for the rate that should be applied to the revenue base in arriving at the licensing fee.

Because ASCAP bears the burden of proving that its proposed fee is reasonable, the discussion that follows begins with an analysis of ASCAP's fee proposal. Since ASCAP has failed to show that its proposal results in a reasonable fee for Mobi's TTTA license, the discussion will then address Mobi's proposal. This Opinion concludes, inter alia, that Mobi's fee should not include any payment when Mobi's role in the distribution of musical content is restricted to providing back-end technology services; that the revenue base should in principal part be based on Mobi's costs of procuring content; and that the Post-Turner Licenses are an appropriate benchmark for selecting a rate to be applied to that revenue base.

A. ASCAP Rate Proposal

ASCAP, principally through the testimony of Vanderhart, advocates choosing wireless carriers' retail revenues as the revenue base for the license fee to be paid for the public performance of music by Mobi. It contends that the revenue base should capture not just the amount that consumers pay to the wireless carriers for products that Mobi "sources" and for ones for which Mobi provides only back-end technology services, but should capture as well revenue that the wireless carriers receive from the sale of the data plans that consumers must buy to gain access to such content. Vanderhart asserts that using an upstream revenue base, for instance one based on the revenue received by the cable television networks from Mobi, would not account for all the value of the public performance of the music that is derived by participants downstream. ASCAP argues that it is particularly appropriate to include wireless carriers' data plan revenues in the revenue base since Mobi has urged wireless carriers to include Mobi products on their handsets by predicting that the availability of Mobi's data-intensive products will motivate consumers to select unlimited data plans, which are particularly lucrative for the carriers.

Because ASCAP's proposal begins with an extraordinarily large revenue base, Vanderhart employed a series of adjustments to arrive at an adjusted revenue base to which the rate is applied. Briefly described, her calculations start with revenue which she labels "beginning" revenue of over $54 billion, moves to MobiTV-Use-Adjusted ("MUA") revenue of almost $4 billion, continues to MobiTV-Use and Music-Intensity-Use-Adjusted ("MIUA") revenue of over $1.6 billion, and then arrives at a licensing fee of over $41 million for the years 2003 through 2011 after the application of a flat 2.5% rate.

Vanderhart calculated the beginning revenue in different ways for different wireless carriers.^[47] In the case of only two carriers — Sprint and AT & T — is the beginning revenue figure different than the MUA revenue figure.

As noted above, after collecting this pool of retail revenue, Vanderhart applies two different "use adjustment" factors. First, she applies a MUA factor based on minutes of usage in an effort to reduce the revenue base of the wireless carriers to one that she contends is attributable in some way to Mobi products. At this stage of her calculation, the adjusted revenue base includes: (1) the amounts associated with the sale of Mobi products à la carte; (2) an allocation of data plan revenue that is associated with the use of Mobi à la carte products; and (3) when the Mobi service is bundled with other services, an allocation of the revenue associated with the Mobi products and the data used to access Mobi products.

Second, Vanderhart further adjusts the revenue base through the application of a music-intensity-use adjustment factor to account for the intensity of music within the various products offered by Mobi or supported by Mobi technology. This results in an "adjusted revenue base." In this stage of the calculation — the adjustment for music use intensity — ASCAP adopts the three-tiered division of content based on music intensity from its Post-Turner Licenses, that is, the categories of general entertainment, news and sports, and music intensive programming, and applies them to the audio-visual content offered by Mobi. Vanderhart recognizes that "the ratios of the rates that were negotiated at that time [that the cable television networks entered into their licenses with ASCAP] are useful for establishing relative music-intensity-use-adjustments for the different products in this proceeding." But, Vanderhart does not use the rates themselves. Instead she converts them into a ratio, using the rates from the cable licenses in the numerator and the rate used for music intensive cable programming in the denominator. Specifically, the MIUA calculation uses 0.9, 0.375, or 0.1375 in the numerator, and 0.9 in the denominator, so that the intensity adjustment factor is 1 for music intensive programming; 0.4167 for general entertainment programming; and 0.1528 for sports and news programming. For purely audio programming (as opposed to audio visual programming) and music videos, Vanderhart does not apply the MIUA, which implicitly equates these products, as far as music-intensity goes, with other music intensive programming that receives an MIUA of 1.

In her fourth and final stage of the calculation, Vanderhart relies extensively on the AOL decision and determines that the appropriate license rate to be applied to the adjusted revenue base is 2.5%. She calculates the license fee by multiplying the adjusted revenue base and this rate. As a result of these steps, Vanderhart arrives at a figure of $41,275,429 owed by Mobi to obtain a TTTA license for the years 2003 through 2011.

Vanderhart applies a modified Georgia-Pacific test, weighing eight factors (as opposed to the fifteen factors that appear in patent case law), to test whether her proposal results in a reasonable royalty. To further support the reasonableness of the final fee, she compares her result to the sums paid by traditional television channels such as ABC, NBC, and CBS and by the cable television networks. She also looks at the licenses that SESAC entered with Mobi and others. Finally, Vanderhart estimates an average per-subscriber fee of $0.11. ASCAP contends that all of these analyses confirm the reasonableness of its proposed license fee.

ASCAP has not carried its burden of showing that this formula will result in a reasonable fee for Mobi's TTTA license. Put another way, ASCAP has failed to show that over $41 million — or any dollar figure in that order of magnitude — represents the fair market value for such a license.

1. Vanderhart's Qualifications as an Expert in These Matters

Vanderhart's testimony was principally useful in laying out ASCAP's calculations and its legal and factual arguments in support of its fee request. These matters, of course, would have been better presented through an econometrician and briefing by counsel.^[48] While it is evident that Vanderhart has worked extremely hard to make and present her complex calculations and cannot be faulted for devoting herself to the specific task assigned to her by ASCAP, her testimony included surprisingly little economic analysis. Vanderhart also has comparatively limited expertise as an economist on the types of issues that lie at the heart of this case.

Vanderhart earned a Ph.D. in economics from Texas A & M University in 2000. While at Texas A & M, she was a graduate instructor in the Department of Economics and a lecturer in the Department of Management. She is a member of several professional organizations, including the Licensing Executives Society and the American Economic Association. In between her undergraduate and graduate programs, she worked for a period of time at Nathan Associates, a consulting firm whose clients have included ASCAP. She is currently employed as a Managing Director at Invotex Group, specializing in the areas of trademark and patent litigation, breach of contract, and commercial damages. She has little or no experience with copyright issues, or the music, cable or television industries.

Vanderhart does not rest her calculations on an explanation of guiding economic principles or any coherent theory. It is often difficult to discern the basis for the choices she makes other than her reliance on her understanding of legal authority, including in particular AOL. As a result, ASCAP's proposed formula must rest on its own inherent strength.

2. Revenue Base

The revenue base upon which ASCAP builds its fee calculation is over $54 billion; ASCAP labels this sum as the beginning or starting revenue. This figure is largely composed of revenues from Sprint and AT & T. In the case of all other carriers, ASCAP begins with what it labels the MUA revenue.

Sprint's beginning revenue figure includes all revenue from à la carte sales of products supported by Mobi technology and Sprint's "total wireless data revenue," including revenue from bundled data plans that support audio-visual programming and deliver text messaging and other services. In some process that was not fully explained at trial, Vanderhart made an effort to exclude from this revenue figure bundled data plans or certain portions of those plans that she determined had no connection to the delivery of audio-visual content. As for AT & T, Vanderhart similarly included the à la carte revenue received by AT & T for products supported by Mobi technology and a portion of AT & T's data revenue from two data plans.

Even though Vanderhart made an effort to exclude entirely unrelated revenue, the starting point for her calculations remained an enormous figure that was poorly understood. After all, the revenue data that ASCAP wanted to use to calculate its fee was not in Mobi's possession and the records from the wireless carriers could only be obtained through costly and burdensome discovery directed to non-parties. This process involved two motions to compel.^[49] To compound the problem, the wireless carriers do not maintain their documents in the ordinary course of their business in the ways that are useful to ASCAP's theory of its case. Accordingly, the foundation of ASCAP's calculations was a gargantuan, poorly understood figure that was vastly over-inclusive.

3. Mobi-Use-Adjusted Revenue Base

Other deficiencies associated with ASCAP's choice of methodology can best be illustrated by examining the string of calculations that it applies to a single carrier's retail revenue. Since Sprint is the largest contributor to the revenue base — accounting for over three-quarters of the revenue base as well as the final fee — it will serve as the example.^[50]

As noted, ASCAP begins with two components of the Sprint revenue base: the à la carte subscription revenue from the "Mobi products" and the Sprint wireless data revenue. Even the first category is overbroad, since the à la carte revenue includes not just the products like MobiTV where Mobi provides the content, but also Sprint television products where Mobi provides only the back-end technology services. Vanderhart leaves this à la carte revenue figure unadjusted at this point in her calculations; she does not apply a MUA to the à la carte services.

The second component of the revenue base starts with a great deal, but not the entirety, of Sprint's wireless data revenue. Because it includes bundled data plan revenue, however, it includes revenue from data plans that support not only the viewing of television programs but also emailing, text messaging, and browsing the web. To reduce this roughly $30 billion figure, Vanderhart constructed a calculation based on minutes of usage and succeeded in reducing the starting figure to a tenth of its original size.^[51]

The decision to construct a minutes-based calculation required Vanderhart to overcome a significant hurdle, though. Sprint only keeps usage information in kilobytes and, moreover, does not break out the kilobyte data among its various services. As a result, Sprint's records do not allow any easy way to isolate the usage of audio-visual content or any other relevant data, whether provided by Mobi or sourced by Sprint itself, from the rest of the data plan usage. To address this conundrum, Vanderhart relied on a single document to convert kilobyte data into minutes.^[52]

The single document on which ASCAP relied to convert kilobytes to minutes was a document that Mobi had created and provided to Sprint. It listed monthly megabytes and minutes of usage for the twenty-three months between April 2007 and February 2009, a period during which Sprint had taken over the task of securing most of the television content offered on Sprint handsets. The only testimony about what was being measured in the document was provided by a Sprint witness in his deposition, where he guessed that Mobi was measuring some component of the channels encompassed by the Sprint TV product. From this testimony, Vanderhart concluded that the conversion ratio applied to Mobi audio-visual products only. Since the Sprint witness who described the document could not identify which channels or "packs" of Mobi channels were being measured in the document, however, Vanderhart did not know precisely what was included in the conversion chart.

Having adopted a conversion ratio of kilobytes to minutes, Vanderhart used Mobi and Sprint documents to construct a ratio: minutes of Mobi-supported content usage were in the numerator and the total minutes of her selected universe of Sprint data plan usage were in the denominator. The documents spanned the years 2003 to 2009, and thus included service data from roughly three years before and several months after the data that produced the conversion ratio. The numerator was based on Mobi records for all minutes that users spent accessing content from any product for which Mobi provided back-end technology support.

The denominator of this MUA presents other problems. Audio-visual content is far more data intensive than other services, such as text messaging or emailing. For instance, using the relationship between minutes and megabytes from the document upon which Vanderhart relied in creating the conversion ratio, Mobi demonstrated that viewing one minute of audio-visual content consumes as many kilobytes as 6,200 text messages. And streaming audio-visual content is roughly four times more data intensive than streaming audio-only content. As a result, a kilobyte-to-minute conversion based solely on audio-visual content will make it appear that Mobi's percentage of the denominator is far larger than it actually is. Put another way, it will shrink the denominator.

This single assumption—that the conversion ratio from kilobytes to minutes for audio-visual products could be applied to all Sprint kilobytes because all types of Sprint customer data plan usage consumed data at the same rate—is an error of significant proportions. Vanderhart admitted at trial that the denominator for her calculation was understated, but she did not know by how much.

There is another significant problem embedded in this level of Vanderhart's calculation. It is the assumption that all minutes and/or all kilobytes are of equal value even though the products to which they relate are sold in a bundle and a consumer has not placed a separate dollar value on each of the components of the bundled package. It assumes, for example, that a consumer would pay the same amount of money to receive a kilobyte of text messaging as a kilobyte of television programming even though a kilobyte might give a consumer several back-and-forth exchanges of text messages but only the briefest glimpse of an image from a television program.^[53]

Finally, ASCAP produced no evidence as to how many, or even whether there were any, subscribers to Mobi-sourced content among the subscribers to each of the data plans whose data usage figures were included in the denominator. In sum, the creation of the numerator, the creation of the denominator, and the relationship between them raise a host of issues regarding the extent to which the services that they measure are tied in any meaningful way to Mobi-sourced content and how those services should be valued.

4. Adjusted Revenue Base

Vanderhart's next step in calculating a fee for Mobi's TTTA license was the construction of a set of ratios to account for the intensity of music use of the products that were either provided by or technologically supported by Mobi. To reach what she termed an adjusted revenue base, she made use of the three-tiered rates from the Post-Turner Licenses, but used them to craft a new set of multipliers. Treating the 0.9% fee for music-intensive services as 100% (to arrive at a MIUA of 1), the news/sports rate of 0.1375% becomes 0.1528, and the general entertainment ratio of 0.375% becomes 0.417. Using Sprint again as an example of how this works, Sprint's Mobi-Use-Adjusted revenue base of roughly $3.3 billion becomes an adjusted revenue base of $1.3 billion when these multipliers are applied to segments of the programming content.

While this newly-minted set of multipliers functions to decrease the revenue base, the choice of these precise multipliers is puzzling. Vanderhart's logic appears to have been that the revenue base was still very large and did not yet distinguish among channels based on the amount of music that was used in them. And by adopting the Post-Turner License rates to create new multipliers, ASCAP would seem to admit that cable television programming is the most apt comparison for the majority of the content distributed by Mobi. But, Vanderhart has failed to explain why she adopted the Post-Turner rates for the purpose of creating a multiplier, but did not just apply the Post-Turner License rates themselves to reach a fee. And, having determined that it was necessary to create a new set of multipliers, she did not adequately explain why the denominator for the ratio should be 0.9 rather than any other number, such as the 2.5% rate that Mobi would prefer. In short, Vanderhart does not explain in any persuasive way why her particular multipliers are appropriate.

5. 2.5% Rate

In the final stage of her calculation of Mobi's TTTA license fee, Vanderhart selected 2.5% as the royalty rate to be applied to the adjusted revenue base. Application of this rate results in Mobi owing a license fee of $41.3 million for the years 2003 to 2011. Since this fee will have to be essentially doubled to pay both ASCAP and BMI, the upshot is that Mobi will owe these two PROs about $80 million. To put this figure in context, in the year 2008, Mobi spent less than [REDACTED] to procure all of its rights to distribute content—with the exception of its duty to pay ASCAP and BMI for the right to publicly perform any of their music contained in that content. In the last five years, Mobi paid all of its content providers less than [REDACTED].

While the final fee is an extraordinarily high figure, the rate applied to arrive at that figure is also extraordinarily high. Vanderhart explains that the "starting point" for her analysis of the correct royalty rate was the 2.5% rate applied in AOL.^[54] The 2.5% rate, however, has its genesis in the Music Choice litigation and is the current rate applied to wholesale revenues in order to obtain a TTTA license for the distribution of pure music through television. From that perspective alone, a 2.5% rate is an artificially high rate to apply to retail revenue, or to apply to audio-visual content.

6. Modified Georgia-Pacific Test

In selecting her rate, Vanderhart did not begin with any traditional benchmark analysis.^[55] Instead, Vanderhart analyzes the reasonableness of her proposed licensing fee and the 2.5% rate by applying a modified Georgia-Pacific test. See Georgia-Pacific Corp. v. U.S. Plywood Corp., 318 F. Supp. 1116 (S.D.N.Y.1970), mod. and aff'd, 446 F.2d 295 (2d Cir.1971). The Georgia-Pacific test is applied in patent cases to determine a reasonable royalty rate for a patent infringer to pay the patentee as money damages. Foster v. Am. Machine & Foundry Co., 492 F.2d 1317, 1320 (2d Cir.1974); see also Rosco, Inc. v. Mirror Lite Co., 626 F. Supp. 2d 319, 328 n. 9 (E.D.N.Y.2009).

ASCAP has been unable to identify any copyright case that has applied the Georgia-Pacific factors, and this Court declines the opportunity to be the first to do so. Applying the Georgia-Pacific test in the context of a rate court proceeding would be inappropriate. Patent law exists to protect an inventor's right to exploit the monopoly he is granted in his invention in order to induce more creative activity. By contrast, the rate court's role is to protect the market for licenses to play ASCAP music. Rate courts must "take seriously the fact that they exist as a result of monopolists exercising disproportionate power over the market for music rights." Music Choice IV, 426 F.3d at 96.^[56]

7. ASCAP's Alternative Proposal: A Split License

In apparent recognition of the fact that its request of a license fee of over $41 million from Mobi is tantamount to the refusal to grant any license to Mobi, ASCAP asks in the alternative that Mobi pay a portion of the $41 million proposed fee and that the wireless carriers pay the remainder. Vanderhart suggests that Mobi pay roughly $3.7 million of the $41 million. This figure is calculated from revenues that Mobi received from the wireless carriers. Vanderhart admits, however, that breaking up the license is not "ideal" from the standpoint of reducing transaction costs.

ASCAP's proposal essentially denies Mobi the right to obtain a TTTA license, and is reminiscent of ASCAP's efforts in the 1980s to require separate licenses from cable television networks and the cable operating systems that distributed their programming. Of course, ASCAP is entitled to a reasonable fee whatever the financial strength of the applicant may be, and an applicant's financial condition or other factors may make its application for a TTTA unrealistic, but ASCAP has not shown that such a situation prevails here. In this vein, ASCAP argues that it is not required "to subsidize" Mobi's business model. Quoting Intercollegiate Broadcast System, Inc. v. Copyright Royalty Board, 574 F.3d 748 (D.C.Cir.2009), ASCAP argues, "[i]f small commercial webcasters cannot pay the same rate as other willing buyers and still earn a profit, then the Judges are not required to accommodate them." Id. at 761. Intercollegiate Broadcast is inapposite. It arises in a different context—the setting of copyright royalties to be paid to the owners of sound recordings for the performance delivered by digital audio transmission. Moreover, ASCAP has not shown that there have been any other "willing buyers," id., of an ASCAP license premised on a payment of 2.5% of wireless carrier retail revenue. Thus, the issue remains whether ASCAP has shown that its fee proposal is reasonable under the legal standards described at the beginning of this discussion.

8. The Overall Reasonableness

At the conclusion of trial, ASCAP essentially abandoned its fee proposal. It explained that Vanderhart and ASCAP had relied in good faith on the guidance given by AOL, but that the result had been "messy." ASCAP declined to defend its calculation of a final fee and essentially confined itself in summation to raising questions about the reasonableness of Mobi's fee proposal.^[57]

ASCAP did argue, however, in summation that the many difficulties with Vanderhart's calculations and formula could be overcome by simply adopting her per subscriber rate of $0.11. This rate, however, is the outcome of the flawed calculations and formula. It is, therefore, no more reliable than the analysis which produced it. Thus, it must be rejected as well.

In sum, ASCAP has not carried its burden of showing that its proposed fee for a TTTA license for Mobi is reasonable. It has not shown that it located a revenue base with a sufficient nexus to content "sourced" by Mobi within the wireless carriers' revenue base or that it is entitled to a fee built upon any broader revenue base. Because ASCAP chose a vastly inflated revenue base it faced the Herculean task of contracting that base through a series of calculations. Each of those layers of calculations was laden with unsupported and faulty assumptions. The final fee request is a demonstrably unreasonable number. The fair market value for a TTTA license for the public performance of music cannot cost more money than the combined price demanded by every other contributor to that content. And, of course, as the discussion of the complexity of the Vanderhart calculations illustrates, the adoption of this vast revenue base, along with the layers of calculations required to reduce it to a fee proposal, imposes enormous transaction costs on the parties that are entirely out of line with the commercial realities faced by both ASCAP and all but perhaps the very largest communications companies in America.

B. Mobi's Fee Proposal

Mobi proposes that the appropriate revenue base and rates for calculating the fee that Mobi owes ASCAP for a TTTA license are, with two exceptions, those found in the Post-Turner Licenses, to wit, those that have governed the distribution of cable television programming during the past decade. The two exceptions relate to Mobi's distribution of audio programming and its programming of music video channels. For the pure music content that Mobi distributes, it advocates applying the 2.5% rate from Music Choice to its payments to content providers. For the music video channels that Mobi programs, Mobi asserts that it is functioning like a television network and that the 0.9% rate for music intensive services should be applied to its advertising revenues from that programming (should there be any) and the revenues it receives from the wireless carriers. All told, Mobi asserts that a TTTA license fee of $301,257.99 for the period from Mobi's launch in 2003 through July 2009 is a reasonable fee.

In support of its proposal, Mobi offered expert testimony regarding economics, the cable television industry, and music videos. In particular, Noll provided a detailed exposition of relevant economic principles and endorsed using wholesale revenue as the revenue base for the calculation of Mobi's licensing fee.

Noll's discussion of economic principles includes a description of the principles of monopolistic competition, competitive pricing,^[58] and derived demand. Noll argues for using the cable television networks' revenue as the revenue base for several reasons, including the fact that "wholesale prices for the television and radio programming in which musical works are embedded are more closely linked to its musical content than are the retail prices charged by wireless carriers for the services they sell to consumers." An additional advantage of using the networks' revenue as the revenue base is avoidance of discrimination among the several technologies that deliver the same content to the consumer.

Noll identified essentially four reasons why the most appropriate revenue base for assessing ASCAP performance rights fees is the revenue of the cable network that supplies the programming containing the musical works. One, it includes all of the relevant revenue, including advertising revenues that accrue to the channel supplier. Two, it does not include revenue that should be excluded, such as the compensation that the program distributor receives for aggregating content, providing technical services that enable distribution, or for bundled services such as high-speed data and telephone services. Three, if the proper benchmarks for a reasonable fee are the ASCAP cable television and radio licenses, then choosing this revenue base permits a proper comparison to be made, that is, one that is not affected by irrelevant costs such as distribution costs. Four, this choice of a revenue base also has the advantage of reducing the transaction costs associated with implementing a rate system and avoids the complex and inherently inaccurate task of attributing revenue between music content and other products.

Conversely, in Noll's view, using retail revenue has several undesirable characteristics, including the fact that retail prices embody the value contributed by many inputs, including investments in wireless technology, wireless networks, and the handsets themselves. In addition, the products sold to consumers are often sold in bundles and any effort to separate the revenues attributable to music will inevitably be ridden with error and arbitrary. As a result, a fee-setting system that depends on retail revenue will necessarily be a highly inefficient and costly system to implement and manage.

Noll has provided a theoretical basis for addressing this rate determination. This framework arises from his unquestioned expertise as an economist and his deep engagement with the industries at issue here. Both his qualifications and his detailed exposition of his analysis entitle his opinion to careful consideration.

Noll earned a Ph.D. in economics from Harvard University and is a Professor Emeritus of Economics at Stanford University, a Senior Fellow in the Stanford Institute for Economic Policy Research, and Co-Director of the Institute's program in Regulatory Policy. He describes his research interests as including antitrust and regulation and technology policy. He has served on many boards and committees for local, state and federal governments; he has consulted for class plaintiffs in several antitrust actions; he has authored or co-authored fourteen books and over 300 articles. His research interests have included the entertainment industry and antitrust and intellectual property issues associated with the digital revolution. As a consultant for the Federal Communications Commission, he helped develop the 1980 regulations regarding the relationships between cable television and producers of content.

C. Adoption of a Reasonable Rate

1. Revenue Base

Mobi has made a compelling showing that wholesale revenue provides the appropriate revenue base from which to measure the value of the public performance of the music at issue here. The economic concept of derived demand explains why the "value," to use the word of art from AFJ2, of a downstream performance is reflected in the wholesale prices of the musical content. See AFJ2 § V. After all, essentially all other rights of those who contribute to the content are priced and compensated at the time the content is created, and there does not appear to be any inherent impediment to placing a value on the public performance right at that time as well. Indeed, there is an advantage in doing so. Pricing the public performance right at the time the content is first sold gives direct and immediate feedback to content producers about the value of a component of their product.

Mobi has shown that the value of the public performance of the music at the retail level is indeed captured at the wholesale level, not just theoretically by the concept of derived demand, but also functionally from the fact that the cable television networks principally generate their revenues by measuring the number of subscribers for their programming. To the extent that a channel's content becomes popular among consumers, the seller of content demands a higher rate of compensation from advertisers and from purchasers of the content. And, Mobi's payments to the cable television networks for the programming it distributes are driven by the subscriber data that Mobi tracks and conveys to the networks.

Moreover, the administrative advantages of basing a rate on wholesale revenue are amply illustrated in this case by the challenges that ASCAP's expert sought to surmount as she endeavored to construct calculations that might result in a reasonable fee and to justify those calculations. Fully appreciating that the retail revenue base vastly overstates the value of the public performance of the musical composition, since it reflects so many inputs that bear little or no relation to that content, she designed layers of formulae to reduce the retail revenue base. At each step of the process, those formulae raised a multitude of questions about their intellectual rigor, fairness, and the cost and burden associated with their implementation.

Nor, despite ASCAP's repeated argument to the contrary, do Music Choice II and IV prevent the choice of wholesale revenue as the appropriate revenue base here. The very factors identified by the Court of Appeals in those two cases as potentially counseling against using retail revenues as a revenue base are emphatically present here. This is a case in which many of the retail customers pay a single fee for a bundle of programming, making it difficult to determine what part of the fees is paid for music. Music Choice II, 316 F.3d at 195 n. 2. And the digital revolution, which has turned handsets into computers and permitted television programming to be included among the many innovative products to which a consumer has immediate and constant access, makes it an extremely complex task to tease out one component of the retail price and identify the extent to which retail price is driven by the musical content of television programming. Id. at 196 n. 3.

Thus, in this case, the value of the public performance at the retail level is fairly captured through a calculation that begins with the wholesale price for the musical content. In other words, a wholesale revenue base reflects the "value" of downstream users' performances by taking their value "into account." See AFJ2 § V.

The revenue base for the calculation of Mobi's TTTA licensing fee will be the following. For the programming that Mobi licenses from content providers, aggregates, and conveys to wireless carriers, the revenue base shall be the amounts that Mobi pays to the cable television networks or other providers to license the content, plus any revenue from advertising Mobi inserts into that programming, including revenue that is shared with wireless carriers or potentially content suppliers.^[59] For the music videos that Mobi obtains from record labels, programs into music video channels, and then provides to the carriers, the revenue base will be the revenue that Mobi receives from the wireless carriers for this programming, plus any revenue from advertising Mobi inserts into that programming, including revenue that is shared with wireless carriers or potentially content suppliers.^[60] This difference in the revenue base is necessary because Mobi is acting like a network when it programs the music video channels.

2. Rates

In setting a reasonable rate to be applied to the revenue base, it is appropriate to look at an agreement reached after arms' length negotiation between similarly situated parties. Music Choice II, 316 F.3d at 194. In the case of each of the categories of services that Mobi offers, appropriate and reliable benchmarks are near at hand.

The three-tiered Post-Turner Licenses negotiated between the cable television networks and ASCAP after years of litigation provide a ready-made benchmark for the television programming that Mobi supplies to the wireless carriers. Those benchmarks have been in existence for almost a decade, are reflected in at least a hundred separate licenses, and will govern the fees paid by some licensees through 2012. Although ASCAP has recently been issuing interim licenses based on the same formula, it has not yet revised the formula. Indeed, ASCAP's own expert in this litigation, Vanderhart, has incorporated the three-tiered rates from the Post-Turner Licenses as key components of a multiplier that she uses in her calculations.

The Post-Turner Licenses are also appropriate benchmarks when measured against the comparability of the parties, the similarity of the rights in question, and the similarity of economic circumstances. See Music Choice IV, 426 F.3d at 95. They concern the identical content and rights: a fee for a license to distribute television programming downstream to the public. Indeed, their definitions are so broadly written that they require the networks to pay licensing fees to ASCAP for the very content at issue here. Mobi delivers that content unedited to the wireless carriers. The negotiating parties are also comparable. ASCAP is the PRO in both cases. While Mobi is not a cable network, the rates are ones to which ASCAP has already agreed when negotiating with Mobi's upstream content-providers, and thus ASCAP cannot be said to be disadvantaged by applying those rates to essentially an identical revenue base.

It is true, however, that Mobi has significantly less negotiating power in dealing with ASCAP than a cable network. It is a start-up company that has [REDACTED]. But again, this fact cannot be said to disadvantage ASCAP unfairly. In fact, ASCAP has boasted that its cable television network licenses are a prime example of ASCAP's success in structuring deals "that allow [ASCAP's] revenues to grow as the new media revenues grow." Mobi is quintessentially another new media venture.

Finally, the cable network license settlement agreements were reached after extensive litigation and involvement by the rate court. They reflect, therefore, to the extent possible in this industry, the degree of competition or leveling of the playing field that justifies reliance on the three-tiered Post-Turner Licenses as a benchmark.

Similarly, an excellent benchmark exists for the rate that should be applied to the all-audio channels that Mobi supplies to the wireless carriers. The 2.5% rate that the parties negotiated following the Music Choice litigation is still in effect, and is a rate upon which ASCAP relies heavily in the formulae that it has constructed in this case. Again, this rate is being applied to essentially the same revenue base: Mobi's cost of obtaining the content, which is a component of the revenue base of the content provider.

Working through the Music Choice factors for evaluating the reliability of a benchmark, see Music Choice IV, 426 F.3d at 95, the 2.5% rate passes this examination as well. The rights at stake are comparable: they are the right to publicly perform music contained in pure audio channels delivered over a distribution system that also conveys audio-visual television content. The revenue base to which the rate will be applied is comparable as well, that is, the wholesale revenues received by the content provider. The parties at issue are also comparable. The PRO in Music Choice was BMI, which represents a similar number of musicians and publishers to those represented by ASCAP. While the applicants in Music Choice were among the largest entertainment and communications companies in America, and Mobi is new entrant into the communications arena, that single difference does not detract from the suitability of the Music Choice benchmark. Finally, the economic circumstances affecting the earlier negotiators and the market circumstances under which agreement on this rate was reached also strongly favor application of this rate. The rate was set after extensive litigation, including multiple proceedings before the rate court and the appellate court, and hard-fought negotiations between major components of the entertainment industry. Thus, the adoption of the 2.5% rate is appropriate.

Finally, there is strong evidence to support adoption of the 0.9% rate for musicintensive programming from the Post-Turner Licenses as the benchmark for the rate to be applied to Mobi's programmed music video channels. This rate was applied to equivalent channels that were covered by the Post-Turner Licenses, such as the Artists and Fans Network. As explained above, the revenue base to which this rate will be applied is an equivalent revenue base, to wit, Mobi's revenue from the distribution of the music video channels that it programs.^[61] For all the reasons already described, the Post-Turner Licenses are appropriate benchmarks.^[62]

It is worth noting as well, that the effective rates for the ASCAP licensing of broadcast television channels fall comfortably within this same range of rates, and serve as additional benchmarks supporting these determinations. Vanderhart calculated that the three broadcast networks effectively pay ASCAP at a rate that falls between the sports and news and general entertainment rates contained in the Post-Turner Licenses, even though their own Post-Turner licensing agreements with ASCAP are flat-fee agreements.

3. ASCAP's Objections

ASCAP makes ten principal arguments against the use of Mobi's costs to procure content as the revenue base, the selection of rates outlined above, and the overall reasonableness of a fee that in large part adopts Mobi's proposal.^[63] ASCAP's four arguments concerning the adopted revenue base are that 1) it is unprecedented to use Mobi's costs to procure content as opposed to its revenues as the revenue base; 2) the advertising revenue earned by content providers will not be captured; 3) Mobi pays no licensing fees to acquire some content, which would preclude ASCAP from receiving any payments from this content; and 4) Mobi's innovative revenue sharing arrangements with content providers will deprive ASCAP of its fair share of revenue. ASCAP's four principal attacks on the selected rates are that the Post-Turner Licenses are 1) outdated benchmarks; 2) inapplicable to the audiovisual content on handsets; 3) inapplicable to an aggregator of content like Mobi; and that 4) a 2.5% rate should be applied to music video content rather than the 0.9% rate. Finally, ASCAP complains about the overall magnitude of the fee award because it is too small in comparison to 1) the fee Mobi agreed to pay SESAC; and 2) Mobi's payments to record labels for music videos. Each of these arguments will be considered in turn. None of them persuades the Court that its adopted formula should be altered.

(a) ASCAP's Revenue Base Objections

ASCAP protests that the revenue base for calculating Mobi's licensing fee for audio-visual content procured from the cable television networks should not be Mobi's payments to those networks for that content. ASCAP argues that it is unprecedented for the revenue base to be the applicant's own costs instead of its revenues. One party's costs, however, are another party's revenue, and categorizing a revenue base as one party's "costs" sheds little light on the wisdom of selecting any particular revenue base for the calculation of a fee. The choice of a revenue base should be context-specific, reflecting an understanding of both the industry at issue and an applicant's position within that industry.

Here, the Post-Turner Licenses reflect an established, recent pattern in which ASCAP and producers of television content chose, in scores of arms-length negotiations, the networks' revenue as the revenue base for calculating ASCAP's fee for a TTTA license. Mobi has shown that this revenue base is similarly appropriate for their fee to ASCAP. The Post-Turner Licenses encompass the delivery of program content through "any . . . means or method," and thereby include the very programming that Mobi distributes. At least one network, Discovery, has regularly made payments to ASCAP under its Post-Turner License based on revenue received from Mobi. Thus, Mobi has demonstrated that its payments to the content providers are the appropriate revenue base for the channels it does not program, that is, those channels other than the music video channels.

ASCAP next argues that there is a risk that it will not receive any licensing fee income from an important revenue stream: the advertising revenue received by the cable television networks. As of now, there is no mechanism for Mobi to learn of any advertising revenue generated by content-providers for themselves. While a theoretically important loophole, this gap need not detain us long. ASCAP did not show at trial that there was any network providing television programming content to Mobi with whom ASCAP did not already have a license covering the years at issue. Under the terms of such licenses, ASCAP has already received a licensing fee for all of that advertising revenue.^[64] There is no suggestion that ASCAP should be paid twice from that same revenue stream.^[65]

There are also about ten or so channel providers who have provided content to Mobi for free in the hope that their channels will develop a subscriber base that will permit them to attract advertisers and begin charging Mobi and other distributors for the content. As ASCAP correctly points out, any licensing fee that uses Mobi's payments to content providers as the revenue base will result in no payment to ASCAP for the distribution and public performance of this "free" content.

As Mobi showed at trial, it is not uncommon for new entrants into the programming market to offer Mobi the opportunity to distribute their content for free. In choosing among these requests, Mobi tries to select those channels it believes are the most likely to succeed,^[66] and typically allows the channel about a year to see if it can build an audience.^[67] This phenomenon came to loom larger at trial than it had in the parties' pretrial submissions. As a result, the factual record regarding the phenomenon was not well-developed.^[68] Nonetheless, there is a strong argument that this phenomenon requires no adjustment to the formula for defining a revenue base: The market for television content has spoken, and this content would receive no airing at all through a wireless distribution platform unless the content were provided for free and given the opportunity to build a base. To the extent the programming succeeds in building an audience, composers and publishers, and indeed all those who contribute to the creation of the content, may eventually reap reward.^[69]

Nor is it unusual for content contributors to reduce their initial fees to support new content. Vanacore explained at trial that he will "often take on new assignments for unreleased shows . . . for free," in the hope that he will receive meaningful compensation if the show is successful. While it would be possible to impose a small minimum fee for ASCAP on all programming content that Mobi acquires, the record at this trial shows that the fair market value for an ASCAP license in this experimental market is $0. The Post-Turner Licenses contain no minimum fee requirement, and none will be imposed here.

ASCAP's final objection to the proposed revenue base is that one of Mobi's contractual arrangements with a content provider allows the cost for the content to be paid directly by the wireless carrier to the content provider. ASCAP is rightly concerned that such an arrangement might deprive ASCAP of a fee to the extent that the fee is calculated solely on the basis of Mobi's payments to content providers. In this single instance, which arose in a 2010 contract, if the total revenue paid by the carrier exceeds a certain threshold, the network shares the revenue with Mobi; if the revenue falls short of that threshold, Mobi must pay the shortfall. Because this formula triggers Mobi's receipt of revenue and payment obligations, its books track the totality of the payments to the content provider. In this and any similar instance, the revenue base for Mobi's TTTA license fee must be the amount the content provider receives for the content Mobi has contractually procured, whether that content provider receives the revenue from Mobi or from a third party or from both of them.

(b) ASCAP's Objections to the Rates

ASCAP contends that the Post-Turner Licenses should not be used as benchmarks for the rates to be applied to the audiovisual programming that Mobi procures. It asserts that those licenses are "outdated," that the programming that Mobi procures and distributes is far different from that encompassed by those licenses, and that Mobi is more akin to a cable system operator than the signatories of the Post-Turner Licenses, which were cable television networks. None of these attacks on the use of the Post-Turner Licenses as benchmarks is persuasive.

First, the Post-Turner licensing program is extant and provides a strong benchmark for a TTTA license to Mobi that will cover the years 2003 through 2011. While in late 2007 ASCAP largely decided to stop issuing such licenses, offering interim licensing arrangements in their stead, there are Post-Turner Licenses that will extend through the year 2012. Therefore, these licenses will have been in effect for the entire period encompassed by the Mobi license. Their obvious relevance to the Mobi application is underscored by the fact that ASCAP's expert, Vanderhart, used the rates from the Post-Turner Licenses in her formula.

ASCAP next argues that Mobi's wireless distribution system and its programming are just too different from ordinary television viewing for the Post-Turner Licenses to have any relevance. ASCAP points out that the handsets on which consumers receive Mobi's programming are not confined to "brick and mortar" locations, and contends that Mobi's programming, in contrast to ordinary television programming, is dominated by "on demand" programming.

Nothing in the Post-Turner Licenses suggests that either of these distinctions matters. In their definition of a Distribution System, the licenses list every system of distribution that had ever been devised and then add as a catchall for good measure: "or any other means or method which is or hereafter may be used to transmit or receive a `Programming Service.'" And, ASCAP's correspondence with licensees and audit history—at least until 2009 or so—reflect a custom and practice of capturing wireless distribution revenue through these licenses, including revenue received from Mobi itself. Moreover, there is nothing in the licenses to suggest that any term within them hinges on the location of the public performance.

Against the plain meaning of the Post-Turner Licenses, which ASCAP admits are unambiguous, ASCAP sought to offer evidence that in drafting the form Post-Turner Licenses it never intended them to apply to wireless distribution of cable television programming. Mobi's objection to the offer of such parol evidence was upheld. Absent an ambiguity, contract interpretation is a legal question. JA Apparel Corp. v. Abboud, 568 F.3d 390, 397 (2d Cir.2009). Moreover, the license agreements in question contain merger clauses. Where a contract is unambiguous and complete, no extrinsic evidence as to the parties' intentions will be considered. Id.; R/S Assocs. v. N.Y. Job Dev. Auth., 98 N.Y.2d 29, 744 N.Y.S.2d 358, 771 N.E.2d 240, 242 (2002) (citation omitted).^[70]

As for ASCAP's contention that the television content that Mobi delivers is largely "on demand" programming, the record at trial did not bear that out. Fewer than 25% of Mobi's programs are available on an on-demand basis. In any event, the Post-Turner Licenses explicitly include VOD within their definition of Programming Service, and nothing in those licenses suggests that the extent to which viewers make use of that option is of any moment in the calculation of ASCAP's fee.

ASCAP next asserts that the Post-Turner Licenses are not a good benchmark since Mobi's role in the distribution of content is more akin to that of a cable system operator rather than a cable channel programmer. Mobi performs several functions in the distribution of television content and operates in a middle space between the cable television network and the wireless carrier. Its position in this middle space, however, does not rob the Post-Turner Licenses of their utility in this rate determination. The rates that those licenses establish are being applied to essentially the same revenue base to value the public performances of analogous, and often identical, content, and thus serve the same function as the rates in the Post-Turner licenses. The cable television networks have created the programming that will be governed by these rates and Mobi delivers it unedited to the wireless carriers. Valuing the ASCAP music differently because of the precise place that Mobi occupies in the delivery chain would elevate form over substance. The fee Mobi will pay reflects the contribution of ASCAP music to the value of the content and the value of the TTTA license for the public performance of that music.

Finally, ASCAP contends that the benchmark for the rate that should be applied to Mobi's revenue in connection with its programming of channels of music videos should be the 2.5% rate adopted in AOL and the YouTube interim fee decision, United States v. ASCAP (In re Application of YouTube, LLC), 616 F. Supp. 2d 447 (S.D.N.Y.2009), rather than the Post-Turner License 0.9% rate for music intensive channels. Mobi has shown that the music video channels that Mobi programs are akin to the channels of music intensive services that historically have been covered by the Post-Turner License rate of 0.9%, and that in programming those channels it is functioning like the networks whose licensing fees were computed under those licenses. In contrast, ASCAP did not perform a benchmark analysis to show that either the AOL decision or the YouTube interim fee decision (which in any event has limited precedential value) should apply here.

(c) ASCAP's Objections Regarding the Overall Reasonableness of the Mobi Fee

ASCAP makes two arguments about the overall reasonableness of the fee. First, ASCAP argues that Mobi's fee to ASCAP is unreasonably small as compared to the fee that it paid to SESAC to publicly perform SESAC music for the period through 2008. But, Mobi's settlement agreement with SESAC does not suggest that a licensing fee to ASCAP premised on the revenue base and rates adopted in the Opinion will be unreasonable.

ASCAP did not use the SESAC settlement agreement as a benchmark in this litigation. If it had, of course, that agreement would have suggested that ASCAP's own proposed fee was wildly off the mark. In any event, SESAC agreements have never been used as benchmarks in ASCAP rate court proceedings and a departure from that practice is not warranted here.

SESAC threatened to sue Mobi and ended up settling with Mobi for a fee that was insubstantial in terms of the risks that Mobi faced in litigation. Because Mobi requested an ASCAP license, it is not a copyright infringer and this proceeding will determine the fair market value of its TTTA license. ASCAP has failed to offer any analysis of the size of the discount that must be applied to the SESAC license to account for this distinction. This alone is sufficient to undermine the use of the SESAC settlement agreement as a reliable measure of the reasonableness of the fee adopted here.

Second, ASCAP argues that Mobi's payments to the record labels for the rights to the music videos that make up Mobi's music video channels should be used as a "reasonableness check" on the overall reasonableness of the fee that Mobi pays to ASCAP, and that those payments demonstrate that Mobi's proposed fee to ASCAP is unreasonably small. For the period from 2003 to 2009, Mobi has paid more than [REDACTED] to three record labels for the music video content. Mobi proposes to pay ASCAP $144,000 for the public performance of music in its music video channels for the same period. ASCAP calculates that this results in a [REDACTED] ratio of payments.

ASCAP's argument fails. Its own expert, Vanderhart, admitted that Mobi's fee to ASCAP should not hinge on its payments to the record labels. ASCAP has pointed to no licensing agreement in which its payments were affected by the size of the licensees' payments to the record labels or by any changes in the amounts of such payments. For example, ASCAP's fee under its license with [REDACTED], whose programming is dominated by music videos, is completely untethered to any payments made to record labels.

The absence of any tie between a payment to a record label and to ASCAP is understandable. After all, the market for rights to music videos is different from the market for the rights to publicly perform music. The record in this case has the benefit of largely undisputed evidence from Mobi's music video expert, Vidich. Vidich explained that record labels make very large financial investments in the creation and marketing of music videos and that the musical composition is only one of many components necessary to create a music video.

Additionally, the payments Mobi makes to record labels cover a host of rights. Therefore, it is understandable that record labels command a significant payment for this broad set of rights. As Vidich opines, the amounts paid by Mobi to record companies to transmit music videos "cannot fairly be equated" to the value of public performance of compositions in those videos.^[71] For all of these reasons, ASCAP's objections to the determinations reached in this Opinion are rejected.

CONCLUSION

Following an April 2010 bench trial between Mobi and ASCAP, this rate court has determined that a reasonable fee for a TTTA license for Mobi for the years 2003 through 2011 is as follows. The revenue base upon which the licensing fee will be calculated is (1) for the content that Mobi licenses from content providers, aggregates, and conveys to wireless carriers, the amounts that Mobi pays to the cable television networks or other providers to license the content, plus any revenue from advertising Mobi inserts into that programming; and (2) for the music video channels and any other channel programmed by Mobi, the payments Mobi receives from the wireless carriers for those channels, plus any revenue from advertising that Mobi inserts into that programming. The rate to be applied to that revenue base is 0.1375% for new and sports content; 0.375% for general entertainment; 0.9% for music intensive programming, to include Mobi's music video channels; and 2.5% for all-audio offerings.

SO ORDERED:

NOTES

[1] By the terms of the consent decree, the matter was to be ready for trial within one year of that date, subject to an extension of no more than one year, absent good cause for a longer extension. AFJ2 § IX(E). Mobi and ASCAP reached agreement in February 2009 on interim fees to be paid pending the outcome of the rate court litigation. The case was reassigned to this Court on July 22, 2009. Pursuant to scheduling orders of August 28, 2009 and January 8, 2010, the parties completed fact and then expert discovery, and a trial was scheduled for the spring of 2010.

[2] Mobi moved in limine to strike the testimony provided by both Vanderhart and Petrovsky. The deficiencies in Vanderhart's testimony will be discussed at length below. Petrovsky is an investigator. He reported on his use of handsets to access Mobi programming and demonstrated in court the features of that programming that he chose to highlight. Although the motion to strike his testimony was largely denied, the Court explained at the final pretrial conference that Petrovsky's testimony would not be received as expert testimony.

[3] ASCAP moved in limine to strike portions of the Noll declaration regarding an agreement between ASCAP and DirecTV. That motion was denied. ASCAP also moved in limine to strike certain portions of Vidich's testimony. That motion was granted in part.

[4] ASCAP formed a New Media and Technology Department in 1995. This department has had a major role in shaping ASCAP's current view of how it should negotiate licenses encompassing the public performance of musical compositions over the internet and by wireless technology.

[5] Within the television industry, cable programming refers to television programming carried by cable, which can include the same channels being distributed via satellite, over telephone lines, and by other means.

[6] Other technologies that have or still do deliver cable subscription programming to consumers include Community Antenna Television, or CATV; a now-defunct line-of-sight transmission system using microwave frequencies known as the multi-point distribution system and its variant the multiple multi-point distribution system; and satellite master antenna television services which use a combination of CATV and a satellite dish to distribute both broadcast and cable programming to a community or to a large institution such as a hospital or condominium complex.

[7] Occasionally, a cable operator also creates programming. The largest operator, Comcast, produces E! and Golf Channel, among others. Time Warner, the second largest operator, owns entities that produce Cartoon Network, CNN, Court TV, HBO and TNT.

[8] In contrast, pay-per-view or PPV technology allows a subscriber to purchase programs but requires the subscriber to watch the programs at a specified time.

[9] Noll provided this statistic from the CTIA— The Wireless Association, the organization on which the Federal Communications Commission relies for data on mobile telephones.

[10] At trial, Gerbrandt admitted that the Writers Guild of America recently secured a different compensation structure for works appearing in "new media outlets."

[11] Vanacore typically receives as well the writer's share associated with the mechanical rights of the musical works that he has composed. But, since few television shows ever produce or distribute the sound recordings of his compositions for the television programs, that is not a meaningful avenue of compensation. There is no mechanical royalty associated with the audiovisual work.

[12] The copyright in the integrated audiovisual work is usually controlled by the record label.

[13] The digital distribution of music videos was one of the factors motivating the development and negotiation of the terms of NDMAs.

[14] Mobi was founded under the name Idetic, Inc., and changed its name to MobiTV in 2005.

[15] Mobi holds ten patents and has forty-seven more pending for inventions related to video distribution and related services. The technological challenges inherent in Mobi's business include minimizing bandwidth consumption, configuring the content of hundreds of different mobile telephone combinations, and delivering high-quality content across the nation. Mobi's technological achievements have won it a number of prestigious awards, including the 2005 Emmy Engineering Award for Outstanding Achievement in Engineering Development, which it shared with Sprint Nextel Corporation.

[16] Live programming refers to a 24-hour, seven-days-a-week feed of programming that Mobi relays from cable television networks over wireless networks to mobile telephone subscribers. There is a delay of only a few seconds in the transmission. This is Mobi's most popular and expensive content.

[17] A clip-linear channel does not deliver a live feed from a cable television network. Instead, it delivers episodes of a program in a sequence defined by the network, for instance, an eight hour loop of programming that is refreshed periodically. The content provider is responsible for the programming decisions.

[18] Mobi's VOD channels give subscribers the ability to select a particular television episode or clip (segments of an episode) and watch it when they want to watch it. For instance, the viewer could select full episodes or clips of the television program The Office. VOD programming is largely confined to audiovisual content; it is not typically available for music videos and never available for radio. In the last few months, however, Mobi has entered contracts pursuant to which a small amount of on-demand music video or music performance content is available via Mobi products. For example, a music video by the star of the Nickelodeon show iCarly that was shown at the end of a television episode could be a stand-alone clip for that iCarly episode. Less than 25% of the programs offered by Mobi are offered on an on-demand basis.

[19] The Mic HipHop and V40 Top Hits are MobiTV-programmed music video channels.

[20] Apparently, Mobi still procures all of the television programming content for certain Sprint handsets that carry the MobiTV product.

[21] Mobi4Biz has been rebranded as Market-NowTV.

[22] Many Mobi contracts with wireless carriers provide that the carrier will pay [REDACTED]. Mobi4BIZ contracts provide for payment to Mobi of [REDACTED].

[23] On average, in any given month, only [REDACTED] of customers who subscribe to any Mobi service actually use it. Customers tend to use Mobi services more if they purchased the services à la carte than if they purchased the services as part of a bundled data package.

[24] A megabyte is a unit of digital information equal to 1024 kilobytes or 1,048,576 bytes.

[25] In at least one instance, a content provider shared advertising revenue with Mobi.

[26] Subscription revenue from the delivery of audio-visual programming far exceeds Mobi's revenue from the delivery of audio-only content.

[27] The net advertising revenue excludes advertising revenue received by Mobi that was shared with wireless carriers pursuant to license agreements. Mobi's advertising revenue has remained small since it is still developing a subscriber base and there is no nationally recognized tracking service that is measuring the extent of its audience.

[28] In the last five plus years, Mobi has paid over [REDACTED] to content providers, with about [REDACTED] going to television channel programmers, just under [REDACTED] going to record labels for the right to program and distribute music video channels, and about [REDACTED] being paid to obtain radio services such as Music Choice and DMX.

[29] These two lump sums represented [REDACTED] of Mobi's gross revenues for the relevant periods, less a discount for the time period through 2006.

[30] See United States v. ASCAP (In re Application of Turner Broadcasting System, Inc.), 782 F. Supp. 778 (S.D.N.Y.1991), aff'd, 956 F.2d 21 (2d Cir.1992) ("Turner").

[31] The Post-Turner Licenses were entered after the issuance of the antitrust consent decree known as AFJ2, which is discussed below. Not all cable networks are licensed under Post-Turner Licenses. In fact, HBO has a license that is based on per-subscriber charges. Some other cable networks and the three broadcast networks — NBC, ABC, and CBS — pay fixed fees.

[32] Among the most prominent of the dozen or so Post-Turner Licenses that will extend into 2012 is the ASCAP agreement with [REDACTED]. In 2007, ASCAP offered a final agreement to [REDACTED], a cable television network with [REDACTED], with an expiration date in December 2012. ASCAP executed that agreement on May 1, 2008. Additionally, ASCAP entered into Post-Turner Licenses in 2007 and 2008 with networks such as [REDACTED].

[33] In a license that ASCAP entered in 2008 with the cable television network [REDACTED], the word "simultaneous" was omitted from the definition of the TTTA license. The same omission occurs as well in other recently-issued licenses.

[34] As for internet sites, the form license grants the licensee the right to publicly perform ASCAP musical compositions on its own internet sites, that is, internet sites operated or maintained by LICENSEE or any other party ... to the extent that any such Internet site is one of LICENSEE's Distribution Systems, and then only to the extent that (i) it is engaged in the distribution of LICENSEE'S Programming Service; or (ii) used to promote the Programming Service and/or the exhibition of any Program containing musical compositions licensed under this Agreement.

[35] The efforts of several ASCAP witnesses to deny the clear import of this record were not credible.

[36] During this litigation, ASCAP apparently only conducted discovery of one Post-Turner licensee, Discovery Communications, Inc. ("Discovery"). That discovery practice produced evidence that Discovery's stream of licensing payments to ASCAP had indeed included revenues generated from distribution of Discovery programming over wireless distribution platforms.

[37] At trial, evidence was admitted that showed that Discovery had been paying ASCAP based on revenues it received from distribution of its programming through Mobi as far back as 2006. On the other hand, at least three cable television networks have taken the position that their ASCAP licenses do not, or do not always, cover performances delivered to wireless products and over the internet. These include [REDACTED].

[38] ASCAP negotiates licenses with cable television operators by negotiating with a committee of the National Cable & Telecommunications Association ("NCTA"). After reaching an industry-wide agreement with NCTA, ASCAP offered that agreement to the roughly 100 Multiple Systems Operators ("MSOs"), each of which is a cable operator that runs cable systems in more than one community.

[39] Mobi adapted a form letter offered to it by MRI for its application to ASCAP.

[40] Citing several copyright infringement cases, in its closing argument ASCAP suggested for the first time that the burden of proof regarding a reasonable fee should be shifted to Mobi because the wireless carriers bundle so many of the Mobi-supplied products with other services and it was difficult to untangle the bundled products and their revenue streams. This invitation, which is at odds with the clear language of AFJ2 and is without precedence in rate court jurisprudence, was declined by the Court. Mobi is before this rate court not as an infringer, but as an applicant for a TTTA ASCAP license.

[41] A portion of the Music Choice service was available as well to internet subscribers, of which there were only 1,500 at the time of the rate proceeding, as compared to 21 million cable or satellite customers.

[42] The Court of Appeals observed that the challenges encountered with bundling are not unique to retail product packaging. After all, "the wholesale price, like the retail price, also cover[s] costs for services and materials other than the music." Music Choice II, 316 F.3d at 196.

[43] In his concurrence, the Honorable Robert Katzmann explained that he would leave open for the moment whether fair market value "is best measured by wholesale revenues, retail revenues, or some other measure." Music Choice II, 316 F.3d at 198 (Katzmann, J., concurring).

[44] In its summation, ASCAP explained that its formulation of a fee proposal for Mobi was largely driven by ASCAP's reliance on the analysis in AOL.

[45] An on-demand audio music performance is one in which the consumer can select which song she would like to play at any time.

[46] Mobi does not dispute that it would have liability under the Copyright Act for infringement if no one sought a license for the distribution of that music. Its argument is rather that responsibility for paying licensing fees should rest in the first instance, for several reasons explained below, with the entity that secured the rights to distribute that content.

[47] Vanderhart had not identified her beginning revenue calculations in her direct testimony, but did provide those calculations in response to the Court's request at trial.

[48] Mobi's motion in limine to strike the entirety or at least significant portions of Vanderhart's testimony was denied before trial. While there were strong grounds in support of the motion to strike, the parties were advised that the issues raised by Mobi would be taken into consideration in evaluating the weight that would be given to Vanderhart's testimony. See Olin Corp. v. Certain Underwriters at Lloyd's London, 468 F.3d 120, 133-34 (2d Cir.2006); McCullock v. H.B. Fuller Co., 61 F.3d 1038, 1043 (2d Cir.1995).

[49] ASCAP does not explain how the retail revenue figures and analysis upon which its expert relies could ever be obtained outside of a rate court proceeding. Conscious of this limitation, it suggests that it will never again have to initiate a rate court proceeding against Mobi or a Mobi-like entity since it now understands the industry better and has converted its formulae into a per-subscriber fee. The per-subscriber fee, however, is no more reliable than the calculations from which it is derived, and in any event, ASCAP is not the only party entitled to seek a reasonable fee from the rate court. Under AFJ2, an applicant for an ASCAP license is also entitled to seek an adjudication from the rate court.

[50] The ASCAP methodologies for calculating a fee premised on the other wireless carriers' retail revenues had their own separate deficiencies and were no more reliable than those it employed in connection with Sprint's retail revenue.

[51] When deciding to create a minutes-based use adjustment, Vanderhart was apparently relying on the AOL decision, and its measurement of minutes of access to musical content to reduce its own vast revenue base. See AOL, 559 F.Supp.2d at 399.

[52] Two other Mobi documents generally confirmed the conversion ratio.

[53] ASCAP itself does not value musical compositions on the basis simply of their length. It uses a variety of criteria in deciding how to allocate compensation among its composer and publisher members.

[54] The issues in AOL are sufficiently different from those confronted here that AOL can provide at best only limited guidance.

[55] With minimal discussion, Vanderhart rejects agreements between AT & T and BMI, and between SESAC and Mobi as appropriate benchmarks, describing them as "neutral" during her discussion of her modified Georgia-Pacific factors. Vanderhart does, however, focus at length on the profitability of the wireless carriers, which is all but irrelevant to the task at hand.

[56] Because of these different contexts, it is unsurprising that Vanderhart was unable to apply the Georgia-Pacific factors directly to her rate and fee determinations, but had to modify these factors. In doing so, Vanderhart made several questionable adjustments to the factors. To give but one example, factor 13 of the Georgia-Pacific test examines the portion of the "realizable profit" that should be credited to the invention as distinguished from other elements, while Vanderhart purported to examine the "allocation of the selling price that should be made to the public performance of music" as distinguished from other elements.

[57] ASCAP also tried, unsuccessfully, to blame Mobi for the fact that ASCAP's fee proposal included revenue for products for which Mobi only provided back-end technology services. It is far preferable from every party's perspective for ASCAP to issue a TTTA license to the entity which provides or procures the content in which the music is embedded. That entity will have the relevant financial records and the transaction costs associated with obtaining those records and negotiating a reasonable fee will be minimized. That entity is also in a position to have that negotiated price appropriately reflected in commercial transactions related to the content.

[58] Noll explained that competitive pricing reflects three attributes. One, competitive prices are cost-based and a change in cost causes the price to change in the same direction. Two, competitive prices are based on usage, such that fees for performance rights would be higher for content that contains more music. Indeed, competition among types of content is stifled if performance fees are not based on the intensity of music use. Three, competitive prices are non-discriminatory; comparable services are charged comparable rates. Any other system would run the risk that the choices of consumers among available technologies would be distorted.

[59] There was no evidence at trial that the wireless carrier inserted any advertising into content received from Mobi. To the extent that a wireless carrier does earn advertising revenue from advertising it inserts into the programming and that is not covered by Mobi's license from ASCAP, then the carrier is separately responsible for payments to ASCAP for that advertising revenue. United States v. ASCAP (In re Application of the Nat'l Cable Television Ass'n et al), No. 41 Civ. 13-95(WCC)(MHD), 1999 WL 335376, at *8 (S.D.N.Y. May 26, 1999).

[60] There was no evidence at trial that the wireless carriers inserted any advertising into the music video channels.

[61] To the extent that Mobi has programmed any other channel, such as a cartoon channel, then the revenue base shall be Mobi's revenue from distribution of that channel and the revenue from any advertising it inserts.

[62] The rate to be paid by Mobi to ASCAP for any advertising revenue will be the same rate that applies to the channel in which the advertising appears, as is the case under the Post-Turner licenses.

[63] While ASCAP raised additional issues during the course of the trial, these ten arguments were the ones it chose to emphasize. Its remaining arguments have been considered and do not alter the findings or results set forth in this Opinion.

[64] Mobi pointed out in summation that it is proposing to pay ASCAP based on all of its payments to the cable television networks and similar content providers even if ASCAP has already received a stream of licensing fees from the networks themselves based on revenue received from Mobi. Mobi suggested that its proposal would necessarily result in a double payment to ASCAP in every instance in which the content provider had made such payments to ASCAP, as is true in the case of Discovery, and would more than compensate ASCAP for any advertising revenue that had somehow escaped the reach of the ASCAP licensing program.

[65] Mobi offered during summation to devise some contractual arrangement with content providers on a going-forward basis to allow ASCAP to confirm that it has received a licensing fee based on advertising revenues collected by content providers. ASCAP and Mobi are invited to discuss such arrangements to ensure that they meet with the approval of both parties.

[66] These arrangements are not without cost to Mobi, because it must still bear the cost of distributing the programming.

[67] An example of a channel that did succeed in building an audience is the Ataku Channel, which shows anime cartoons, and now commands a payment from Mobi for the distribution of its content.

[68] The factual record regarding placement of promotional audiovisual content within Mobi's products was even less developed at trial and was not addressed in the parties' submissions. To the extent the parties need a ruling from the Court on this issue, they may make a separate application.

[69] ASCAP points out that the individual composer who contributes to the "free" content may not be the composer who contributes to the program and earns a licensing fee after the program is sufficiently popular to demand a distribution fee. That is certainly true. On the other hand, if an individual composer's contribution is sufficiently tied to the success of the programming, then it is likely that the initial composer will at least have an opportunity to contribute to the programming and benefit when the audience for the program grows.

[70] ASCAP's arguments to the contrary fail. ASCAP has cited no case that has admitted evidence to "corroborate" the meaning of an unambiguous, fully-integrated contract. Further, ASCAP's argument that the parol evidence rule applies only to cases that are "contractual in nature," see People v. Dean, 56 A.D.2d 242, 392 N.Y.S.2d 134, 142 (N.Y.App. Div.1977), also fails to persuade. The rate court sits in a unique context, creating a reasonable licensing fee where the parties have failed to agree on contract terms. Because ASCAP is obligated to offer licenses at comparable fees to all similarly situated music users after the Court has set a rate for one such user (AFJ § IX(G)), the terms of the contracts between ASCAP and music users alleged to be similarly situated have special import. Finally, ASCAP argues that the "stranger to the contract" exception to the parol evidence rule applies. While it was at one point fairly settled that the parol evidence rule only applied to actions between the parties to the contract and their privies, that rule is no longer absolute under New York law. Instead, "in the case of a fully integrated agreement, where parol evidence is offered to vary its terms, the rule operates to protect all whose rights depend upon the instrument even though they were not parties to it." Oxford Commercial Corp. v. Landau, 12 N.Y.2d 362, 239 N.Y.S.2d 865, 190 N.E.2d 230, 231 (1963) (citation omitted). Because the Court must interpret ASCAP's other license agreements, including the Post-Turner Licenses, in order to do a benchmark analysis in the course of determining a reasonable fee, the interpretation of those contracts is essential to both ASCAP's and Mobi's right to a reasonable fee.

[71] The parties also debate the relevance of the NDMAs entered into between record labels and music publishers to the overall reasonableness of the fee payable to ASCAP, but it is unnecessary to discuss that debate here.