United States v. American Society of Composers, Authors & Publishers

831 F. Supp. 137 (1993)

UNITED STATES of America, Plaintiff,
v.
AMERICAN SOCIETY OF COMPOSERS, AUTHORS AND PUBLISHERS, et al., Defendants.
In the Matter of the Applications of CAPITAL CITIES/ABC, INC. and CBS, Inc., Applicants,
For the Determination of Reasonable License Fees for Their Television Networks.

No. Civ. 13-95 (WCC).

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

August 11, 1993.

Paul, Weiss, Rifkind, Wharton & Garrison, New York City (Jay Topkis, Allan Blumstein, Robert N. Kravitz, John F. O'Sullivan, Nicole L. Felton, of counsel), and Bernard Korman, New York City (Richard Reimer, of counsel), for ASCAP.

Weil, Gotshal & Manges by R. Bruce Rich, New York City (Evie C. Goldstein, Beth K. Neelman, Martha Applebaum, and Charles Stanford, Capital Cities/ABC, Susanna M. Lowy, CBS, Inc., of counsel), for applicants.

OPINION AND ORDER

WILLIAM C. CONNER, District Judge.

These applications, progeny of the consummation of an historic shotgun union, are made to this Court in its enduring capacity as the so-called "rate-court" under Section IX of the Amended Consent Judgement ("Consent Decree") entered in United States v. ASCAP, 1950-51 Trade Cases (CCH) ¶ 62,595 (S.D.N.Y.1950). The Consent Decree, originally entered in 1941 and subsequently amended in 1950, settled the United States' antitrust suit against the American Society of Composers, Authors and Publishers ("ASCAP"). Today, the terms of the Consent Decree continue to regulate the manner in which ASCAP licenses its music inventory. This Court's jurisdiction, an artifact of the Consent Decree, Section XVII, is retained to oversee the ongoing implementation of these provisions.

As amended, the Consent Decree requires ASCAP to offer to users of music a "blanket license," permitting the non-exclusive right to perform, in unlimited fashion, any music contained in the ASCAP repertory. Section IX(A) of the Consent Decree provides that ASCAP and the users of its music are to attempt, in the first instance, to negotiate a license fee; failing to reach agreement after 60 days, the prospective licensee may then apply to this Court "for the determination of a reasonable fee."

Applicants in the instant proceeding are two television networks owned, respectively, by Capital Cities/ABC, Inc. ("ABC"), and CBS Inc. ("CBS"). Both seek a judicial determination of the reasonable fee to be paid to ASCAP for blanket licenses that authorize the performance of ASCAP music in the programming transmitted by each network.^[1] ABC seeks determination of a fee for the period January 1, 1986 to December 31, 1993; CBS seeks determination of a fee for the period January 1, 1991 to December 31, 1993.

BACKGROUND

ASCAP is an unincorporated membership association that licenses public performing rights to the copyrighted musical compositions of its members. ASCAP's members include over 50,000 music composers, lyric writers, and publishers who own the copyrights to a vast number of musical compositions, and who have granted ASCAP a non-exclusive right to license the performing rights to these compositions. SF ¶ 1.^[2] The society serves both as the licensing agent and as the collector and distributor of royalties for licensed performances. ASCAP v. Showtime/The Movie Channel, Inc., 912 F.2d 563, 573 (2d Cir.1990). ASCAP also endeavors to monitor the public performances of its members' music to assure that such performances are licensed. See Broadcast Music, Inc. v. CBS, 441 U.S. 1, 20-23, 99 S. Ct. 1551, 1562-64, 60 L. Ed. 2d 1 (1979); Tr. at 673-80. ASCAP's repertory contains over three million compositions. The performing rights for these compositions are licensed by ASCAP to a wide variety of users, including television and radio networks and stations, cable program services, restaurants, clubs, bars, and other establishments that publicly perform music. SF ¶ 2.

Because ASCAP represents a pooling by members of their copyrights which, among other advantages, enhances their commercial posture in negotiating with music users, the society became subject to an antitrust suit filed by the United States Department of Justice. The suit was settled in 1941 when the parties entered into a Consent Decree that imposed certain limitations on ASCAP. See, United States v. ASCAP, 1940-43 Trade Cases (CCH) ¶ 56,104 (S.D.N.Y.1941). In broad terms, the Consent Decree, as amended in 1950, permits ASCAP to obtain from its members only a non-exclusive agency to issue performance licenses. The members retain the right to negotiate directly the performance licenses for their own compositions, or to assign that role to another entity, and ASCAP is prohibited from interfering with a member's prerogative to pursue these alternatives. The Decree addresses the manner in which ASCAP is to issue performance licenses and requires, inter alia, that ASCAP "use its best efforts to avoid any discrimination among the respective fees fixed for the various types of licenses which would deprive the licensees or prospective licensees of a genuine choice from among such various types of licenses." Consent Decree, Sec. VIII.^[3]

As previously noted, the Consent Decree, Section VI, requires ASCAP to offer a blanket license covering all of the compositions in its repertory.^[4] A substantially similar blanket license is offered by Broadcast Music, Inc. ("BMI"), the other major music performing rights licensing organization. Showtime, 912 F.2d at 565. BMI's repertory also contains a vast number of musical compositions, though fewer than the number in the ASCAP repertory. SF ¶ 7. The background and nature of the blanket license has been discussed in several opinions considering antitrust challenges to its validity. See Broadcast Music, Inc. v. CBS, 441 U.S. 1, 99 S. Ct. 1551, 60 L. Ed. 2d 1 (1979); Buffalo Broadcasting Co. v. ASCAP, 744 F.2d 917 (2d Cir.1984), cert. denied, 469 U.S. 1211, 105 S. Ct. 1181, 84 L. Ed. 2d 329 (1985); Columbia Broadcasting System, Inc. v. ASCAP, 620 F.2d 930 (2d Cir.1980), cert. denied, 450 U.S. 970, 101 S. Ct. 1491, 67 L. Ed. 2d 621 (1981).

Applicant ABC operates the ABC Television Network, which transmits programs, commercial, promotional and public service announcements (collectively "broadcast material") broadcast by more than 200 affiliated local stations, including eight stations that are owned and operated by ABC. SF ¶ 3. Applicant CBS operates the CBS Television Network, which transmits broadcast material aired by more than 200 affiliated local stations, including ten that are owned and operated by CBS. SF ¶ 4. Similar broadcast material is aired by the NBC Television Network operated by the National Broadcasting Company ("NBC"), with over 200 affiliated stations, including seven that are owned and operated by NBC. SF ¶ 6.

ABC and CBS acquire, produce and distribute broadcast material to affiliated stations who then broadcast such programs to the viewers within their area. SF ¶ 8. Typically, a majority of the programs transmitted by networks to affiliates are supplied to the networks by independent producers; the balance are produced by the networks themselves. SF ¶ 10. While programs may be sourced from a multitude of producers or packagers, each network customarily deals with only a limited number of such suppliers. SF ¶ 11. Under the network-affiliate relationship, the networks generate revenues primarily through the sale of commercial announcements that are aired by affiliates in conjunction with the programs furnished by the networks.^[5] SF ¶ 9.

While the uses of music in broadcast material aired by the networks vary, they may generally be classified as either feature, theme, or background music. Feature music is the principal focus of audience attention, such as a song sung on a variety show. Theme music is played at the start or conclusion of a program and serves to enhance the identification of the program. Music in a program that is neither feature nor theme is generally considered background music. SF ¶ 13.

The music contained in programs supplied to the networks by independent producers is typically selected for those programs by the producers.^[6] SF ¶ 14. The producers obtain from the copyright owners of the music selected the right to record the music on the soundtrack of the program's film or videotape in synchronization with the action. Acquisition of this so-called "synch" right does not carry with it the separate right to perform the music. SF ¶ 14. By longstanding practice, the right to perform music is licensed to the networks by ASCAP and BMI. SF ¶ 14.

During trial, applicants presented considerable testimony concerning the current licensing regime under which, in order to broadcast the programs acquired from producers, the networks must acquire music performing rights for music which is already fully integrated into the programs. Tr. at 373-75, 660-70. The frequently mentioned alternatives to acquiring a music performing license from ASCAP and BMI are "source" or "direct" licensing. Source licensing entails the program producer procuring performing rights to the music used in the program from the copyright proprietors, and conveying those rights to the network in conjunction with the performing rights to all other components of the program. Direct licensing entails the network obtaining music performing rights directly from individual copyright proprietors. As noted, however, the networks typically have not secured source or direct licenses, but rather customarily obtained from ASCAP and BMI blanket licenses permitting television performance of all the music in the repertories of these societies. SF ¶ 14; see, Buffalo Broadcasting, 744 F.2d at 920-22.

Since 1949, the networks have entered into various blanket license agreements with ASCAP. SF ¶ 20. Of particular relevance to this proceeding are the four most recent agreements reached between ASCAP and the networks. In June, 1981, CBS entered an agreement ("1981 CBS Agreement"), which retroactively finalized license fees for the 1970-1980 period,^[7] and prospectively established fees for the 1981-1985 period. JX 21. The agreement called for total payments to ASCAP of $51.0 million, $6.2 million of which was allocated to the 1970-1980 period,^[8] and the balance, $44.8 million, used to establish final fees for the 1981-85 period. The 1981-1985 fees were not apportioned evenly, but rather began at $8.0 million in 1981 and escalated to $9.8 million in 1985.^[9]

In November, 1985, ABC concluded an agreement with ASCAP, ("1985 ABC Agreement") retroactively finalizing fees for the 1977-1985 period. JX 12. For the period 1977-1980, ABC fees were finalized at $18.84 million, allocated over the entire period rather than on a per-year basis.^[10] For the period 1981-1985, ABC's final fees totaled $44.8 million, the same amount as paid by CBS for that period. But where the CBS payment schedule called for escalating fees, ABC allocated a fee of $8.96 million to each year. JX 12.

In December, 1985, CBS negotiated an agreement ("1985 CBS Agreement"), which prospectively established fees for the 1986-1990 period. JX 22. The fees began at $9.8 million in 1986, and escalated to $11.3 million in 1990.^[11]

In April, 1992, NBC consummated an agreement with ASCAP ("1992 NBC Agreement"), retroactively finalizing fees for the period 1976-1991, and prospectively finalizing fees for the period 1992-1993.^[12] JX 32. The agreement called for an additional payment by NBC of $17.5 million to finalize fees for the period 1976-1990. For 1991, NBC's fees to ASCAP were finalized at $11.3 million. For 1992, NBC agreed to pay an amount equal to 0.44 percent of NBC's average gross network revenue^[13] for the years 1991 and 1993.^[14] For 1993, NBC agreed to pay an amount equal to 0.44 percent of the gross network revenues earned in 1993. JX 32.

Applicant CBS remains "open" with regard to final fees for the period 1991-1993. During this period, CBS has paid interim fees to ASCAP in the amount of $9.8 million per year for the years 1991 and 1992. SF ¶ 21. Applicant ABC remains open with regard to final fees for the period 1986-1993. During this period, ABC has also paid interim fees to ASCAP in the amount of $9.8 million per year for the years 1986-1992. SF ¶ 21. Both applicants seek to finalize fees for the years they respectively remain open.

DISCUSSION

As noted, the Consent Decree establishes a procedure by which the fees payable for an ASCAP license are to be determined. Upon receipt of a written application for a license, Section IX requires ASCAP to "advise the applicant in writing of the fee which it deems reasonable for the license requested." If the parties are unable to agree upon a reasonable fee within 60 days, the applicant may apply to the Court for the determination of a reasonable fee. In such a proceeding, the burden of proof falls upon ASCAP to establish the reasonableness of its fee proposal. Consent Decree, Section IX.

The Consent Decree does not provide further direction as to how the Court should arrive at a license fee, nor does it specify any criteria by which to gauge what constitutes a reasonable royalty. In ASCAP v. Showtime, 912 F.2d at 569, the Second Circuit indicated that the task of assessing such a fee necessitates an appraisal of "fair market value" — an appraisal based essentially on an estimation of "the price that a willing buyer and a willing seller would agree to in an arms-length transaction." And in the lower court opinion in Showtime, affirmed on appeal, Magistrate Judge Dolinger noted that a proper analysis should seek to "define a rate or range of rates that approximates the rates that would be set in a competitive market." See ASCAP v. Showtime, 912 F.2d at 576 (opinion of trial court).^[15]

Yet the very need for such an approximation reveals the problematic nature of the analysis envisioned and the enigmatic task thrust upon a court that must undertake this evaluation. For underlying such an analysis is the recognition that the market for blanket licenses does not typify the model of a competitive market, in the sense that it is not characterized by multiple sellers, each possessing negligible market power. Rather, ASCAP and BMI are in the position of being the sellers of a unique product to their licensees, and the providers of a unique service to their constituent members. See BMI v. CBS, 441 U.S. at 20, 99 S.Ct. at 1562. Inasmuch as applicants in this proceeding, and others similarly situated in like proceedings, may prefer to depict the product offered by ASCAP as the composite of individual musical compositions, the blanket license is "truly greater than the sum of its parts." BMI v. CBS, 441 U.S. at 21, 99 S.Ct. at 1563. Beyond the performing rights to the music actually used, the license offers the flexibility of immediate and unlimited access to a vast repertory of compositions, without the cost and delay of consummating individual agreements, and without the concern of exposure to liability for copyright infringement. With regard to ASCAP members, the blanket license achieves efficiencies in the monitoring and enforcement of individual copyrights; its absence would significantly raise the cost of licensing, and undoubtedly alter the supply and cost of using the desired compositions. The efficiencies obtained from an aggregate license explain the genesis of this atypical market: "a bulk license of some type is a necessary consequence of the integration necessary to achieve these efficiencies, and a necessary consequence of an aggregate license is that its price must be established." BMI v. CBS, 441 U.S. at 21, 99 S.Ct. at 1563. In short, the market for blanket licenses appears to be one whose natural consequence is the lack of broad-based competition. To postulate what prices would prevail were such a market "competitive" is perplexing in theory, impractical in practice, and dubious in outcome given the efficiencies obtained due to the aggregating nature of the service rendered. Little can be adduced as to the expected behavior of such a market were it populated by multiple sellers. It is questionable whether such a market could sustain many sellers; whether the market would function efficiently and the price levels at which its supply and demand would converge are even more uncertain.^[16] In addition, limited evidence is discernible as to the incremental costs facing ASCAP.^[17] Consequently, any rate-setting standard that calls, in the abstract, for a theoretic construct of a competitive market in blanket licensing must confront the reality that there exists minimal evidence as to what that market would look like, much less the prices it would yield.

Lacking the instruments with which to construct from start a model to price fairly the rights at issue, it becomes necessary to establish a tangible foundation from which to commence this inquiry. Thus, as in prior proceedings, the Court finds it appropriate to consider previous agreements voluntarily entered between the parties, or those similarly situated, as the starting point of its analysis. See Showtime, 912 F.2d at 577 (opinion of trial court) ("we must look to very imperfect surrogates, particularly agreements reached either by these parties or by others for the purchase of comparable rights"). Such an approach offers a workable means by which to advance the analysis in a manner consonant with the stated aims of the Consent Decree. See Showtime, 912 F.2d at 576-77 (opinion of trial court) (deliberating the countervailing considerations relevant to a rate-setting inquiry, beyond policy of encouraging price restraint upon ASCAP). The Decree demands the determination of a "reasonable" fee; as ASCAP remarks, it does not require the Court to create the "platonic ideal" of a competitive market. ASCAP Br. at 4. Rather than speak in terms of competitive market pricing, when such a term carries vague significance within the context of the blanket license market, we believe it more instructive to view the Court's role as a moderating influence on ASCAP that serves "to minimize the likelihood that ASCAP's evident market leverage may be exerted to obtain unacceptably inflated price levels for its licensees." Showtime, 912 F.2d at 576 (opinion of trial court) (citing cases). Performing this function does not require the Court to ignore the history of the parties' preferences as expressed in their prior agreements; on the contrary, prices negotiated voluntarily in an arms-length transaction offer the only palpable point from which to proceed towards an estimation of fair value for later periods. See Showtime, 912 F.2d at 569; see also, United States v. ASCAP; Application of Home Box Office, Slip Op. at 11 (S.D.N.Y. July 11, 1986) (interim fee opinion) ("invocation of a previously negotiated agreement as a guide to interim fees is based on the premise that both parties' uncoerced acquiescence to the terms of the contract reflects their conclusion that those terms are reasonable under then-current conditions").

Of course, we do not merely endorse as appropriate for today the terms of compromises concluded yesterday. Rather, this inquiry must faithfully address both challenges to the validity of negotiated agreements as reliable benchmarks of reasonable rates at the time entered, as well as changed circumstances that may make prior benchmarks outdated measures of fair value. The very existence of a rate court necessitates the initial inquiry:

Though the rate court's existence does not mean that ASCAP has violated the antitrust law, the court need not conduct itself without regard to the context in which it was created. 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.

Showtime, 912 F.2d at 570. Where prior agreements form the starting point of a rate-setting inquiry, the Court will consider the distinctive conditions impacting those agreements, and evaluate claims that the agreements were the product of a disparity in bargaining leverage. With regard to changed circumstances, the Court must account for alterations in the economic conditions confronting the parties, as well as appraise variations in the nature and value of the rights at issue.

With these general principles in mind, we examine the divergent proposals submitted by the parties as measures of reasonable blanket license fees. Since ASCAP bears the burden of establishing the reasonableness of its approach, its proposal will be evaluated first.

A. ASCAP's Fee Proposal

ASCAP seeks to discharge its burden by relying primarily on the terms of an agreement recently entered into between ASCAP and NBC. The 1992 NBC agreement retroactively finalized fees for the period 1976-1991, and prospectively finalized fees for the period 1992-1993. ASCAP's fee proposal suggests for ABC and CBS terms identical to as those agreed to by NBC for the particular period for which each applicant seeks to finalize its fees. Thus ASCAP proposes for ABC the same terms agreed to by NBC for the years 1986 to 1993, and for CBS the terms agreed to by NBC for the years 1991 to 1993. The fees assented to by NBC for these years are as follows:

    1986     [redacted]
    1987     [redacted]
    1988     [redacted]
    1989     [redacted]
    1990     [redacted]
    1991     $11,300,000
    1992     0.44% of 1991/1993
             average gross revenue[18]
    1993     0.44% of gross revenue[19]

SF ¶ 21. According to applicants' projections, ASCAP's percentage-of-revenue proposal for 1992 and 1993 would yield the following fees:

            ABC          CBS
   1992     [redacted]   [redacted]
   1993     [redacted]   [redacted]

AX 127, 152. ASCAP's proposal also seeks from the applicants interest on the amounts by which the recommended fees exceed previous interim fee payments; interest is to be computed from April 30, 1992, the date of the NBC agreement. Tr. at 37; ASCAP Br. at 6-7.

The principal argument advanced by ASCAP in support of its proposal rests on the theory that there is no better measure of a reasonable fee than the price recently agreed to voluntarily, in an arms-length transaction, by a network situated similarly to the two applicants. Indeed, the Consent Decree itself adjures ASCAP to treat users evenhandedly; Section IV(C) enjoins discrimination in license fees between licensees similarly situated. In broad terms, ABC and CBS do appear to be situated similarly to NBC. All three networks are approximately the same size,^[20] operate in the same manner,^[21] earn roughly equivalent revenues,^[22] and use ASCAP music in a similar fashion and in approximately the same amounts. Tr. at 79-80; DX Q.

Additionally, ASCAP's proposal professes to respect the experience of applicants' own history; the fees resulting from the present proposal are "in line," ASCAP maintains, with fees generated by prior agreements entered into by ABC and CBS. ASCAP Br. at 7-9; Tr. at 50-51 (testimony of ASCAP chief economist Dr. Peter Boyle that ASCAP's proposal is "status-quo type deal"). With regard to CBS, the proposed 1991 fee of $11.3 million is the same as the fee paid by CBS in 1990; adjusted for inflation, the fee would be lower than the prices CBS had previously agreed to for each of the last ten years. Tr. at 55-56; DX N.^[23] Using applicants' projections, ASCAP's percentage-of-revenue proposal for 1992 and 1993 yields, in absolute dollar terms, fees equal to or lower than the amounts paid by CBS in 1989 and 1990. AX 127, 152; SF ¶ 21. As a percentage of gross revenue, the 0.44% rate sought for 1992 and 1993 is lower than the average rate reflected in the fixed fees paid by CBS during the years 1981 to 1990 (an average of [redacted] of gross network revenue).^[24] DX D. With regard to ABC, while the proposed fees are higher in absolute dollar terms than the final fees ABC has paid in prior years, adjusted for inflation, the average of the fees proposed for the years 1986 through 1991 is lower than the average of the fees paid by ABC during the years 1981 through 1985. See DX M.^[25] As a percentage of gross revenue, the 0.44% rate for 1992 and 1993 is close to the average rate yielded by the fixed fees paid by ABC during 1981 to 1985 (an average of [redacted] of network revenue).^[26] DX OOO; see Tr. at 51-60.

1. Prior Benchmarks

Applicants vigorously contest the validity of reliance on the prior agreements between the networks and ASCAP. With respect to replicating the 1992 NBC agreement on the assumption that it is probative of reasonable fees for ABC and CBS, applicants maintain that they are not situated in a position similar to that of NBC at the time that network negotiated its contract. Testimony from the NBC executive who directed the negotiations culminating in the 1992 agreement profiles the circumstances unique to NBC that formed the basis for this agreement. Tr. at 249-87 (testimony of Richard Cotton, NBC Executive Vice-President and General Counsel). Prior to 1992, NBC remained open with respect to finalizing its ASCAP fees for a prolonged period dating from the last quarter of 1976. Tr. at 250. NBC's interim payments to ASCAP for the 1976-1990 period totalled approximately $15 million less than the amounts paid as final fees by CBS. Tr. at 126, 257; App.Br. at 23. In its fee discussions with NBC, ASCAP appears to have asserted that the network should make-up the $15 million shortfall as well as pay interest for the substantial period it had use of this money. Tr. 126-28, 255-57, 266-68; DX A. While NBC repeatedly resisted the notion of paying interest attributable to the 1976-1990 period, the network apparently viewed ASCAP's interest claim seriously. Tr. 255-60, 266-68. The trial record reveals assessments of NBC's interest exposure, ranging from NBC's appraisal of $25 to $45 million,^[27] to ASCAP's estimate of $15 to $17 million.^[28] Tr. at 257, 128. The parties appear to have agreed, in their fee discussions, to settle the 1976-1990 open period in conjunction with the determination of fees for the period 1991 to 1993. Tr. at 260-62, 265-68, AX 4. The 1992 NBC agreement ostensibly represents the compromise struck to settle the "whole ball of wax;" the means by which to set fees on a forward basis through 1993, as well as to arrive at a final disposition for the previous period, including a resolution of the interest issue. Tr. at 266-278. NBC asserts that it viewed the deal as an explicit exchange: ASCAP bartering its claim to a payment designated as interest in return for both premium fees during the 1991 to 1993 period and ASCAP's preferred percentage of revenue format for 1992 and 1993.^[29] Tr. 271-79, 282-83. Based on this testimony, applicants urge that the agreement entered into by NBC was induced by concerns particular to NBC and cannot be considered as the benchmark of reasonable fees for ABC and CBS.App.Br. at 22-27.

ASCAP disputes NBC's characterization of the 1992 agreement. Initially, ASCAP notes that it was far from a foregone conclusion that NBC would match the fees CBS had paid for the 1976 to 1990 period, much less pay a sizable sum of interest on the shortfall.^[30] ASCAP's interest claim rested largely on the award of interest in Showtime, Slip Op. at 3, (October 13, 1989); generally, prior agreements between the parties did not provide for interest payments. ASCAP Br. at 15. ASCAP also argues that the fees prescribed by the NBC agreement "are consistent with" the fees that CBS accepted in its most recent agreement, undermining the contention that the NBC deal was distinctive due to NBC's peculiar position and thus inappropriate for ABC and CBS. Finally, ASCAP suggests that the claim that NBC would agree to pay a premium in fees to ASCAP during any period between 1986 to 1993 imputes a significant degree of imprudence on the network's part with regard to its fee obligations to BMI; NBC's contracts with BMI required the network to pay final fees equalling 85% of NBC's final fees to ASCAP for the period 1986 to 1990, and 100% of the final fees NBC paid to ASCAP for the years 1991 and 1992. Tr. 287-301; ASCAP Br. at 16-18; JX 94-97. Consequently, it would be reckless for NBC to agree to pay substantial excesses to ASCAP for the 1986 to 1992 period, when the premiums reflected in those fees would have to be duplicated in large part in the fees paid to BMI.

Despite the explications offered by ASCAP, we think applicants have raised sufficient questions as to the similarity in circumstance between themselves and NBC to cast doubts on the validity of replicating the terms of the NBC agreement for ABC and CBS. Even if it remained uncertain that NBC would ultimately have to match CBS's payments to ASCAP for the 1976 — 1990 period, as well as pay interest on the shortfall, it is readily apparent that ASCAP pressed its interest claims and that NBC took these claims seriously. The magnitude of the interest calculations arrived at by NBC, the network's awareness of the Showtime^[31] decision and the interest levels assessed in that case, the repeated references in fee discussions to ASCAP's interest claims and NBC's view that an unfavorable resolution of the question was a "dealbreaker," all support the conclusion that NBC's interest exposure influenced the terms to which it acquiesced. The fact that the fees dictated by the NBC agreement are roughly "consistent with" the amounts under the 1985 CBS agreement^[32] does not address squarely the claim that in 1992, NBC deemed a lower fee to be reasonable and may have struck a more favorable bargain if unencumbered by its peculiar circumstances. With regard to the reference to NBC's BMI obligations, we agree that the magnitude of the premium that applicants, in their post-trial briefs, claim is reflected in the NBC fees does not indicate an entirely sound or balanced approach by NBC towards its total cost for purchasing music performance licenses: it seems imprudent to pay premiums of $3 to $4 million a year to ASCAP to offset an interest exposure that ASCAP itself estimated to be in the range of $15 to $17 million, and for which ASCAP had indicated it would accept a $9 million payment in settlement, if those premiums were to be duplicated, dollar for dollar, in the fees paid to BMI for 1991 and 1992.^[33]See App.Br. at 24-27, ASCAP Rep.Br. at 5-6; Tr. at 128, 255-56, 267. However, the fact that music users are apt to make exaggerated estimates of the amounts paid in excess to ASCAP does not invalidate entirely the claim that there was some degree of premium reflected in the fees agreed to by NBC to compensate ASCAP for interest due over the considerable period that NBC had failed to finalize its fees. If the premiums are appraised at a lower order of magnitude, the NBC agreement appears more sound with respect to the additional royalties due BMI. Indeed, if we take Mr. Cotton's testimony at face value, it would appear that NBC was "so focused on avoiding the payment of any interest fee in the ASCAP arrangement," that the network did not consider the consequences as to BMI.^[34] Tr. at 292. Even if we were to discount Mr. Cotton's testimony as ASCAP would prefer, it is simply inconceivable that the 1992 NBC agreement would not reflect the duration of the open period and the interest claims accumulated therefrom. To accept this position would compel us to ignore the reality of NBC's open period dating to 1976, ASCAP's repeated demands for interest and NBC's steady resistance, the network's insistence that a resolution of these questions be part of any agreement reached between the parties, ASCAP's ability to obtain its preferred percentage-of-revenue format, which was a substantial departure from prior practice,^[35] and NBC's success in avoiding a lump sum payment to ASCAP designated as interest—all facts with which ASCAP does not take issue, and all facts that strongly suggest a compromise. Even if the agreement may not have been as explicit a "horse trade" as applicants claim, the record reveals that it was at least implicitly influenced, if not entirely induced by the unique conditions confronting NBC. Consequently, we do not find ourselves obliged to prescribe fees for applicants on the basis of what ASCAP was able to obtain from another user when the facts indicate a high order of probability that the fees obtained were through circumstances particular to that party. See Showtime, 912 F.2d at 570. We thus agree with applicants that ASCAP has failed to meet its burden of demonstrating that the 1992 NBC agreement is probative of reasonable fees for ABC and CBS.

Applicants also take issue with the validity of relying on any of the other three prior agreements between ASCAP and the networks as benchmarks of reasonable royalties. With regard to the 1981 CBS agreement, applicants contend that the loss of CBS's antitrust suit in 1981 eliminated a meaningful constraint on ASCAP's market power; the subsequent agreement reached between CBS and ASCAP reflects the performance society's unbridled ability to price the blanket license at levels reflecting the limited alternatives available to CBS. App. Br. at 32-34, Tr. at 706-07, 710-13 (testimony of applicants' expert economist, Professor George Benston). Nothing else, applicants contend, can account for the precipitous increase in fees between 1980 and 1981.

Applicants' theory is untenable. First, the CBS 1981 agreement finalized fees retroactively for the 1970 to 1980 period, as well as prospectively for the 1981 to 1985 period. That the same agreement established the fees for both periods quickly undermines the hypothesis that the pendency of the antitrust litigation was the reason for the difference in fee levels between 1980 and 1981.^[36] Second, CBS cannot adequately explain why it did not seek rate-court adjudication of its 1981 to 1985 fees, if it truly believed the "precipitous increase" reflected, in large part, a premium over the reasonable sums due. Third, the fact that the 1985 ABC agreement established the same total amount as that paid by CBS for the 1981 to 1985 period, while simultaneously setting lower fees than CBS's payments for 1977 to 1980, supports the contention that the 1981 to 1985 fees were deemed reasonable. Finally, ASCAP offers an assortment of factors—the growth enjoyed by the networks in the 1970's, the surge in inflation during the period at issue, and the dynamics of ASCAP-network relations in the 1970's — that offer more viable explanations for the increase in ASCAP fees during the early 1980's than the one posited by applicants.

With respect to the 1985 ABC agreement that finalized fees for the 1977 to 1985 period, applicants appear to advance the same argument as that raised against the 1981 CBS agreement. App.Br. at 43-45. Applicants seem to suggest that the failure of the CBS antitrust suit, which resulted in CBS agreeing to inflated fees for the 1980 to 1985 period, also saw ABC inevitably faced in 1985 with no viable alternative than to agree to the rates acquiesced in by CBS in 1981. As noted above, we find little merit in this theory. In fact, ABC's position is even less tenable than that of CBS: ABC's fees were established retroactively—thus the network had the perspective of CBS's experience as well as the benefit of hindsight.

In their post-trial brief, applicants also allude to their difficulties in obtaining viable per-program licenses from ASCAP, despite a persistent five-year effort beginning in 1981. App.Br. at 44-45. The networks appear to imply that the lack of viable per-program alternatives may have influenced ABC to accept reluctantly the terms ASCAP offered for blanket licenses in 1985. The record provides sparse support for this proposition. Moreover, the rate-court alternative remained ever-present; applicants do not explain why ABC failed to pursue its objective further through adjudication, if the network had in fact favored per-program licenses. See Consent Decree, Sec. VIII (enjoining ASCAP from discriminating among fees for various types of licenses so as to deprive users of genuine choice); see also, United States v. ASCAP, 586 F. Supp. 727, 729 (S.D.N.Y.1984) ("mandatory per-program option remains an integral part of the injunctive relief provided for by the decree, necessary to provide users with a viable alternative to the blanket license"). In any event, there is meager evidence in the record that ABC's frustration in accessing the per-program route induced it to enter the 1985 agreement. Because applicants did not vigorously press this claim, we do not discuss it at greater length.

Applicants make two claims with respect to the unsuitability of the 1985 CBS agreement as a benchmark of reasonable royalties. First, applicants claim that the agreement was the outcome of an overly optimistic outlook on the network's part as to its future prospects. Based on "rosy" projections contained in a series of reports entitled The Road to 1990, prepared under the direction of David Poltrack, CBS Senior Vice President of Planning and Research, the network submits that it anticipated robust economic health into the 1990's; the fees CBS agreed to pay ASCAP reflect that mistaken optimism. App.Br. at 40-43.

The record contains few if any facts that support this claim. There is scant evidence that CBS relied on the Poltrack forecasts in arriving at a fee for ASCAP. While the testimony of CBS's able negotiator George Vradenburg suggests that the forecasts were the source of "general corporate optimism," there is no indication that these reports influenced, much less formed the basis for the fees negotiated with ASCAP.^[37] Tr. 425-26, 435-37. Absent from the record is any credible connection between the Poltrack predictions and the claim that CBS was amenable to increased outlays in its programming costs. Nor do The Road to 1990 reports appear to be in the nature of a financial planning guide to which CBS may have referred in assessing the expenses it was willing to incur. The forecasts are silent with regard to the changing cost structure confronting the networks and thus, by themselves, are an inappropriate basis for financial decision-making of any sort. Tr. at 348-50. Moreover, the reports do not aspire to a careful appraisal of CBS's individual economic prospects, but rather speak broadly in terms of cumulative network growth, three-network share, and collective network performance. It appears that in the face of gloomy forecasts for network decline^[38], The Road to 1990 reports were offered as assuring estimates of the promising performance of network television in an environment sporting new competitive challenges — testaments to the enduring viability of the network "dinosaurs" in the evolving climate envisioned ahead. See Tr. at 308, 350-53, 436. The Poltrack predictions thus seem to be more in the nature of reports prepared for public relations or for designing marketing and sales strategies, than as reliable guides for financial planning, or as the basis of decisions respecting the costs of production or operation. Tr. at 308. In any event, the record is barren of facts that can persuade the Court that a sophisticated and seasoned market participant such as CBS was moved to generosity in the music licensing fees it was willing to pay by pie-in-the-sky forecasts of the nature contained in The Road to 1990.

Second, applicants also claim that the 1985 CBS agreement was the reluctant result of the network's failure to achieve alternative source and direct licensing for music performing rights. George Vradenburg, CBS's former general counsel, testified to a four-year effort by the network to obtain agreements with composers, publishers and program packagers to convey their music performing rights to CBS, in the event CBS chose at some future date not to operate under a blanket license. Tr. 403-06. The testimony suggests that despite fairly "serious and intense" efforts, CBS's venture proved fruitless: while the network was able to obtain cooperation from the composers it hired for programs produced in-house, it was able to obtain agreement from only a handful of program packagers. Tr. 416-24. Even if CBS had been successful in overcoming the reluctance of many to grant the options it sought, the network estimated the total cost would have exceeded the cost of a blanket license due to two factors: first, the incentive for individual suppliers to hold-out and charge premium prices for music rights knowing the network was obliged to obtain all licenses at the source;^[39] second, the "inflated royalties" ASCAP furnishes to members whose works have been performed on networks.^[40] The result of these failures, argue applicants, was that CBS had no alternative but to agree to the terms ASCAP demanded for a blanket license.

The evidence submitted provides an inadequate basis to find that the 1985 agreement saw a frustrated CBS resigned to paying an unreasonable price for a blanket license after the failure of exhaustive efforts to secure alternative forms of licensing. Evidence regarding the source and direct licensing initiatives pursued by CBS remains insufficient for the Court to assess the intensity or purposefulness of the efforts undertaken.^[41] Nor are we able to determine the posture under which these initiatives arose; whether they were exploratory—in the nature of attempts to obtain leverage against ASCAP or to procure a competitive yardstick for the blanket license—or whether they were serious efforts to establish a system of alternative licensing capable of prompt and practical implementation. In any event, the difficulties CBS faced in opening the source and direct licensing routes—including any hurdles erected by hold-outs and by ASCAP's distribution methods —are relevant here only insofar as they affect the prices CBS agreed to pay in the 1985 agreement. On this score, the record is conspicuously silent. There was no testimony indicating either the prices CBS suggested as reasonable, or the premiums it believed it paid due to its failure to secure source and direct licenses. While CBS declares that it saw no alternative but to come to terms with ASCAP on blanket license fees, there is nothing in the record to indicate the amount of excess fees, if any, that ASCAP was able to extract as a result.^[42] On the contrary, the record suggests that fees agreed upon were acceptable to CBS:

[T]he absence of a competitive measure of the value of the ASCAP/BMI licensed rights is a real, business issue; as a practical matter, however, if a license is available to CBS at current rates in years after 1984, there probably won't be an "issue" which will require people to focus on trying to find alternatives to current practices.

AX 27 (Vradenburg memorandum describing discussion with ASCAP general counsel Bernard Korman in December 1984 regarding CBS's source licensing efforts). The prices agreed upon were in fact consistent with then-current rates: the 1986 fee was identical to that of 1985, while subsequent years showed moderate increases in line with those of the prior period ($1.5 million over five years).

Related to the claims concerning CBS's experiences at seeking source and direct licensing, applicants also attempt to characterize each of the prior agreements between the parties as products of the substantial market power that the blanket license affords ASCAP. Given the nature of music use in network television, a licensing system combining direct and source licenses could function efficiently, argue applicants, were it not for the restraints imposed by the blanket license regime. Most network programming is pre-recorded, and most of the music contained therein is composed specifically for the program pursuant to composer-for-hire agreements. SF ¶ 10. Tr. 373-74, 584-84. In these transactions, the producers select composers and negotiate the prices for the services performed, as well as the fees for the licensing of synchronization rights; music performing rights, however, are reserved for licensing by the composer's chosen music performing rights society. Applicants maintain that absent the blanket licensing system, there would be no reason why the producer could not also acquire from the composer music performing rights on the networks' behalf. The "all-or-nothing" aspect of the blanket license, however, makes it prohibitively costly for parties to pursue this course;^[43] in order to benefit from accessing licenses at the source, the network would have to be prepared to forgo the blanket license entirely. Thus, before crossing such a rubicon, the network would have to succeed in obtaining performing rights to all the music in its programs, a daunting task both in theory and practice.

The issues raised by these claims sound strikingly similar to issues deliberated at great length in the antitrust cases brought against the performing rights societies. See, e.g., CBS v. ASCAP, 400 F. Supp. 737 (S.D.N.Y.1975), rev'd, 562 F.2d 130 (2d Cir. 1977), rev'd, BMI v. CBS, 441 U.S. 1, 99 S. Ct. 1551, 60 L. Ed. 2d 1 (1979), on remand, 620 F.2d 930 (2d Cir.1980), cert. denied, 450 U.S. 970, 101 S. Ct. 1491, 67 L. Ed. 2d 621 (1981); Buffalo Broadcasting Co. v. ASCAP, 546 F. Supp. 274 (S.D.N.Y.1982), rev'd, 744 F.2d 917 (2d Cir.1984), cert. denied, 469 U.S. 1211, 105 S. Ct. 1181, 84 L. Ed. 2d 329 (1985). The final result of those cases was a rejection of the claim that the blanket license prevents music users from licensing performing rights through alternative means so as to function as a restraint of trade. Applicants urge that the context in which their theories are raised in this proceeding distinguishes the conclusions reached in those cases. The focus of the antitrust cases was on whether the blanket license operates an unlawful restraint of trade; the inquiry centered on plaintiffs sustaining their burden of establishing that they lacked realistic alternatives to the blanket license. The focus in the instant case is on whether ASCAP has discharged its burden of establishing that its fees are reasonable; the inquiry raised by applicants' claims concerns whether the blanket license empowers ASCAP to extract excessive fees from the networks, so as to undermine reliance on the prior agreements entered between the parties. See Showtime, 912 F.2d at 570 (failure of antitrust plaintiffs to prove an antitrust violation does not mean that Magistrate lacked evidence sufficient to support a finding that ASCAP enjoys more market power than it would in a freely competitive market for music rights). Inasmuch as the Court was not eager to retry, in this limited rate proceeding, issues raised and resolved in prior antitrust cases, we permitted applicants to proceed with their inquiry, confined to the specific purpose of demonstrating the consequences of blanket licensing on the prices the networks ultimately consented to in their agreements with ASCAP.^[44]

The cornerstone of applicants' claim is that the "all-or-nothing" aspect of the blanket license makes other licensing alternatives virtually impossible. Yet as the Second Circuit found, and as applicants themselves seem to recognize,^[45] the availability of per-program licenses substantially undermines this claim. See Buffalo Broadcasting v. ASCAP, 744 F.2d at 926-29 ("the availability of the program license enables [the stations] to forgo the blanket license and still obtain music rights for any program for which direct licensing proves infeasible"). While applicants assert that ASCAP has consistently resisted offering per-program licenses, they are unable to explain why the networks did not resort to the Court if ASCAP's fee demands deprived them of a "genuine choice" among licenses. Indeed the path remained ever-open; in 1984, this Court denied ASCAP's motion to amend the Consent Decree to permit ASCAP the prerogative to deny perprogram licenses to broadcasters that held blanket licenses with BMI. See United States v. ASCAP, 586 F. Supp. 727 (S.D.N.Y. 1984). While ABC did pursue per-program license negotiations with ASCAP that ultimately faltered, none of the networks availed themselves of the rate-court alternative. They should not be heard now to complain that high fee levels precluded resort to per-program licenses:

The availability of a judicially enforceable requirement of a `reasonable' fee precludes any claim that the program license rate is too high, especially in the context of television stations regularly represented by a vigorous committee with the demonstrated resources, skill, and willingness to invoke the rate-adjustment process.

Buffalo Broadcasting v. ASCAP, 744 F.2d at 927. Consequently, it is difficult for the Court to accept that the nature of the blanket license forecloses access to other licensing options when the vital path bridging these alternatives does not appear to have been vigorously pursued by any of the interested parties. Cf. id. at 926-33 (characterizing per-program license as bridge to source and direct licensing).

ASCAP, of course, submits that the blanket license has always been the license of choice due to the efficiencies it provides. Indeed, this claim finds much support in the language of the antitrust cases ASCAP has survived. See, e.g., Buffalo Broadcasting, 744 F.2d at 934 (Winter, J., concurring) ("the lack of use of alternatives does not signal a restraint on competition but merely reflects the competitive superiority of the blanket license"). Applicants, however, maintain that irrespective of the efficiencies the blanket license may generally afford, within the context of network music use, source licensing can offer an equally efficient alternative. Tr. 668-85 (testimony of Professor George Benston). Accepting applicants' theory at face value, the prospect that networks may be able to access performing licenses at the source without incurring prohibitive transaction costs could appear as a viable alternative from the networks' perspective; however, any theory extolling the feasibility of such alternatives must address the efficiencies lost from the perspective of composers and publishers. In the absence of licensing through ASCAP, alternative mechanisms for monitoring music use and enforcing copyrights must be engaged. Applicants' theory does not appear to account fully for the time and capacity required to fulfill these functions in situations where a license conferring blanket immunity upon a user is no longer available. Indeed, it is not surprising that producers would be reluctant to fly to the undiscovered country of source and direct licensing, when the blanket license has been known over the years to be an efficient manner by which to license music as well as facilitate the protection of copyrights. In short, we remain highly skeptical of the claim that source or direct licensing alternatives would be able to match the overall efficiencies afforded by the blanket license. More pertinent from the perspective of this proceeding, we are unpersuaded that the blanket license has obstructed the networks' assiduous efforts to develop a system of source or direct licensing and that, as a result, ASCAP has been persistently able to obtain inflated royalties for its license from the networks.

The networks' theory fails to overcome one inescapable fact: the availability of the rate-court in instances where the fees demanded by ASCAP appeared unreasonable. Given the tenacity with which applicants present their case in this proceeding, one would expect that the networks, faced with intransigent demand by ASCAP, would fly to the Court for relief. Applicants blame their previous reluctance on the fact that invocation of the Court's jurisdiction is risky, costly, and time consuming; the availability of the rate-court mechanism is therefore a feeble constraint, they argue, on ASCAP's ability to exert its market power and extract excessive fees from the networks. The risks and costs of litigation, however, must be borne by both parties. If anything, these burdens would appear to weigh more onerously on ASCAP: the networks are likely to have the greater wherewithal with which to wage litigation; moreover ASCAP, in each proceeding, must bear the burden of proving that its proposals are reasonable. Consequently, it is difficult to accept that parties in the position of applicants, represented by skillful negotiators and able counsel, would be incapable of wielding the leverage afforded by the rate-court alternative. Nor can we give credence to the claim that the television networks were repeatedly willing to succumb to ASCAP's demand for substantially inflated fees rather than taking the matter to the Court to determine a reasonable fee.

The theories advanced by applicants are similar to those successfully advanced against ASCAP at the lower Court level in the Showtime proceeding. See, 912 F.2d 563, 573-98. Indeed, applicants repeatedly remind the Court of the standards enunciated and the findings made in Showtime. In that proceeding, Magistrate Judge Dolinger declined to use ASCAP agreements with two competing cable services, HBO and Disney, as benchmarks of reasonable fees for Showtime. The decision in Showtime, however, cannot support applicants disclaimers' in the present proceeding against the prior agreements which ABC and CBS themselves entered into with ASCAP. Several facts distinguish this case from Showtime. First, in Showtime, ASCAP sought to set fees for one user based on fees agreed to by another; the Court here has already rejected the NBC agreement as a reliable benchmark of fees for ABC and CBS, and now seeks only to rely on prices to which the two applicants themselves agreed.^[46] Second, in Showtime, Judge Dolinger specifically credited the testimony of cable company executives who stated they viewed the rate-court as "ASCAP-friendly," thus accepting the cable companies' "aversion" to invoke this alternative. 912 F.2d at 585. No similar testimony is available to applicants here to justify the absence of attempts to invoke the rate-court's jurisdiction in the face of unreasonable fee demands.^[47] Third, Judge Dolinger noted that per-program licenses were essentially unavailable to cable suppliers due to ASCAP's position that the Consent Decree did not require these to be offered to cable stations;^[48] applicants cannot claim here that the per-program alternative is similarly denied the networks. 912 F.2d at 585. Fourth, Showtime placed substantial weight on the fact that the ASCAP fee rates for HBO and Disney were far higher than the rates BMI was able to obtain from these users. 912 F.2d at 587-88. No such disparity exists between the fees ASCAP and BMI are able to generate from the networks.^[49] Finally, the Showtime opinion itself contrasts the cable suppliers with the television networks and stations in terms of the latter entities' substantially enhanced capacities to negotiate fairly with ASCAP. 912 F.2d at 584. For all these reasons, the claims successfully presented in Showtime will not sustain applicants' theories in this proceeding.

Thus, our evaluation of the prior agreements between the networks and ASCAP leads to the following conclusions. We agree with applicants insofar as rejecting the 1992 NBC agreement as probative of reasonable fees for ABC and CBS. With regard to each of the other three contracts, we do not believe there exist valid grounds to doubt their reliability as benchmarks of reasonable royalties at the time they were entered into.

2. Changed Circumstances

ASCAP contends that the fee levels it proposes are modest in comparison to the general upward trend in network programming costs. The rise in total broadcast expenses, however, does not appear to be due to across-the-board increases in all aspects of network programming cost. Rather, it appears that certain costs—notably license fees for sports and for popular prime-time programming, and salaries for sought-after scriptwriters, producers and actors—have risen, on some occasions dramatically, and have largely accounted for the increase in overall broadcast expense, despite rigorous cost-cutting measures implemented elsewhere by the networks. Tr. at 540-49 (testimony of Jay Gold, CBS Vice President of Finance); 623-33 (testimony of John Wolters, ABC Senior Vice President of Finance). Senior executives at CBS and ABC testified credibly as to the concerted cost containment efforts embraced by both networks; efforts entailing a range of administrative cuts, including reductions in staff, salaries, station compensation and the closing of news bureaus around the world, as well as changes in programming mix to introduce lower cost shows such as news and reality-based broadcasting. Tr. at 540-50, 623-33. Thus the argument that ASCAP's fees should increase, or at least should not be singled out for reduction, when all other network costs are generally rising, does not find support in the record. Rather, if reference must be made to other programming costs, it seems more precise, though admittedly more difficult, to determine whether music performing rights represent the sort of programming factor that should be deemed to fall within the category of factors subject to the networks' cost-cutting measures, or whether they represent an element of programming whose value would likely command fees consistent with or above prior levels.

In this regard, both parties sought at trial to illustrate the relative value of music to network programming. Applicants presented testimony to the effect that music does not play a substantial role in attracting or retaining viewers, and therefore does not fall within the category of the "make-or-break" elements of programming whose costs have historically escalated. Tr. at 584-90, 599-600 (testimony of producer Robert Berger). Applicants also suggested that the recent rise in news and reality-based programming has seen a marked diminution in the use and significance of music to network broadcasting. ASCAP, for its part, entertained the Court with two plays of one episode from the popular television drama series "Dallas"— once with music as would it normally be aired to a television audience, and once without music, in order to highlight the essential contribution music makes to a dramatic presentation.

Despite such inventive efforts, the Court must admit that it remains incapable of quantifying the value of music to any particular television program. Nor do we believe that the rate-setting function requires us to venture any such assessment. Surveying the fluctuations in the amount of music used by a network over time provides an adequate proxy by which to gauge whether the significance of music to network programming has changed relative to prior years; assuming all other factors remain constant, the direction in which a network's music use has headed should chart the course for the music licensing fees owed to ASCAP. See, United States v. ASCAP; Application of Turner Broadcasting, Inc., Slip Op. at 18-21, 1989 WL 222654 * 20-25 (S.D.N.Y. October 12, 1989). With reference to the above-mentioned efforts undertaken to trim costs at ABC and CBS, if the networks' consumption of music has remained fairly constant, or has increased over the years, the networks should not be heard to complain that their music licensing fees should be reduced in accordance with cuts implemented elsewhere. If, on the other hand, music use has steadily decreased, this would provide support for the position that music performing licenses have been relatively less valuable to network programming in recent years than in the past. In the absence of any other yardstick by which to measure the varying value of music to network programming over time, the necessity of incorporating changes in music use into the calculation of music licensing fees becomes apparent. Because ASCAP's percentage-of-revenue formula takes no account of changes in music use, it is not an acceptable method of arriving at a reasonable fee.

In defense of its proposal, ASCAP strenuously contests the claim that the networks' use of ASCAP music has declined markedly. Given that the most comprehensive source of data measuring music use on network programming comes from ASCAP, the debate over the direction in which music use has changed is ultimately a debate over the interpretation of ASCAP data. See Tr. at 137-155; 823-42. The dispute centers around the reliability of ASCAP data on music use in the so-called "other" category^[50]—a question that will be deliberated in due course. At this stage, it is not so relevant whether music use actually increased, decreased or remained constant for purposes of evaluating the reasonableness of a rate-setting formula from a methodological point of view. What is relevant is the fact that no formula which fails to account for the changes in the level of network music use can be a fair method of assessing a reasonable fee.^[51]

In addition to emphasizing that the networks' gross revenues bear little relationship to applicants' use of ASCAP music,^[52] applicants also challenge the other justifications offered for ASCAP's percentage-of-revenue proposal. With respect to the rationale that revenue-based licenses are common in the field of intellectual property, applicants point out that in these cases, the licensing agreements have generally been "product and user-specific."^[53] Tr. at 218-19. Consequently, the comparison to other intellectual property situations is inapt.^[54] With respect to fees the networks pay for other creative elements of television programming, the percentage-of-revenue approach represents an anomaly; ASCAP can point to no other element of network programming for which the price is determined solely as a percentage of gross revenues. Tr. at 105, 629. ASCAP's reliance on the 1992 NBC agreement as vindicating the suitability of a revenue-based formula does not support its application to ABC and CBS; as discussed already, it appears likely that the NBC deal was driven by considerations specific to NBC. As to the contention that flat fee arrangements engender spiraling disputes between ASCAP and the networks and occasion long "open" periods, ASCAP fails to present any convincing rationale as to why a formula based on revenues promises a panacea.^[55] We remain convinced that an approach based solely on calculating some percentage of a networks' gross revenue cannot serve as a reasonable measure of the value of a music performance license.

On the other hand, considerable advantages may be gained by relying on the change in a networks' gross revenue over time as one of the key variables involved in the calculation of a music performance licensing fee. As ASCAP points out, a formula that allows for fluctuations in gross revenues takes into account a host of considerations. It adjusts automatically for changes in the economy—primarily inflation—as well as changes in the financial fortunes of the network, such as the consequences of competition from other sources of home entertainment.^[56] Tr. at 40, 82-83, 199. It accounts for the financial effect on the network of variations in viewership or audience share. Tr. 82-83. On a forward-looking basis, it allows for a complex of changing circumstances that cannot be anticipated and shares between the parties the risk therefrom. Tr. at 199. Finally it facilitates even-handed treatment among licensees to the extent that it self-adjusts for variations in the relative financial fortunes of each network. Tr. at 40. See Consent Decree, Sec. IV(C) (encouraging even-handed treatment between licensees similarly situated). In short, gauging the changes in gross revenue provides a practical proxy by which to incorporate a range of considerations. Thus it appears to the Court that a formula that factors into the calculation of a royalty for the right to perform music on network television, the changes in both the levels of gross income earned by a network and the degree to which music is used by a network, provides an approach that addresses many of the concerns raised by the parties.

Applicants would prefer the Court to focus on the networks' profitability, the "key economic indicator" from their perspective. App.Br. at 53; Tr. 631. During trial, applicants presented evidence of their deteriorating net income positions, and the net operating losses sustained in the recent past.^[57] AX 126, 129, 151; Tr. 549-51, 631. It seems applicants would invite the Court to use net profits rather than gross revenues as a factor in fee-setting. The Court declines to do so. The perception that profitability is an element inherently difficult to measure meaningfully and is easily susceptible to manipulation, was substantiated by the trial testimony of witnesses from both parties. Tr. 81-82, 761-62. Gauging profitability would potentially open the Court to a critical audit of the networks' financial statements and a probing inspection of the networks' operations, as well as absorb us in complex calculations involving accounting and allocation issues. Indeed, applicants' own expert, Professor George Benston, agreed that the concept of tying fees to profitability was "not a very good idea." Tr. at 761. Moreover, we do not believe that network profitability is a matter entirely pertinent to the task of assessing the fees for the rights to perform music in network programming. Profitability is a function of many variables that have nothing to do with the use of ASCAP music;^[58] in broad terms, profitability accounts for the cost of all the inputs involved in generating a network's income. We are not persuaded that the cost of all other factors of production, most of which are entirely unrelated to the input at issue, must be factored into the price assessed for this particular input.^[59] Applicants do not vigorously press this position, nor do we believe there is substantial support in theory or practice for its application here. In sum, we think a focus on profitability would only serve to obfuscate rather than illuminate our analysis.^[60]

B. Applicants' Counter-Proposal

Applicants present an alternative fee-setting formula that gives force to the various factors the networks believe are relevant in measuring the reasonable royalties for music performance licensing. In the absence of any market measure of the blanket license, applicants' fee proposal admits the necessity of relying on prior negotiated prices as starting points from which to calculate the fees currently reasonable. The approach presented involves two steps: first, a determination of the prior agreement on which to rely as most closely approximating fair value,^[61] and second, the adjustments required to "carry forward" the fees to the periods at issue. App.Br. at 59.

1. Choice of Base Fee

Applicants' preferred approach is to use the fees paid by ABC and CBS in 1980 as the base level from which to make the adjustments necessary to carry forward prior fee levels to the present period. The rationale offered for reaching this far back is that the 1980 rates reflect the pendency of the CBS antitrust suit against the music performing rights societies; the suit, applicants submit, operated as a meaningful constraint on ASCAP's market power and thus the 1980 fees, being so constrained, should be viewed as being "closer to a competitive market price" than the fees for the years following the loss of the antitrust case. App.Br. at 60. The Court has already addressed and rejected this rationale in discussing the exceptions applicants posed with regard to the CBS 1981 agreement. See note 36 supra and accompanying text. The fact that the 1980 fees were finalized retroactively in 1981, subsequent to the termination of CBS's antitrust suit, effectively undermines any claim that the fee levels were the consequence of the pendency of the antitrust litigation.^[62] There thus remains no viable rationale that warrants reaching back to 1980 to find reasonable rates to use as base levels in implementing this analysis.

Moreover, to the extent that our evaluation of the parties' prior agreements found them to be benchmarks of reasonable royalties at the time entered, we feel justified in relying on fees fixed by later contracts between the parties. Applicants themselves offer the fees paid in 1985 as an alternative to their preferred 1980 base-year approach. We think it appropriate to rely on the price paid by each of the networks in 1985 as the suitable base level upon which to predicate the present analysis. With regard to ABC, the 1985 rates are fitting in that they represent the latest fees finalized by the parties; further, these fees were set retroactively so that both parties had the benefit of hindsight in evaluating their reasonableness.

With regard to CBS, the 1985 prices also represent a suitable starting point, even though the network entered a later contract —the 1985 CBS agreement—finalizing fees prospectively through 1990. First, the fees CBS paid for 1985 were identical to the fees set for 1986 by the last agreement between the parties, reinforcing the reliability of the 1985 rate as an appropriate base. Second, given the claim that substantial changes in circumstance, particularly in the networks' use of music, have made the prices fixed by prior agreements outdated measures of present fair value, it seems improper to take 1990 as the base year from which to commence analysis when much of the changes to which applicants refer occurred over the 1986 to 1990 period. Though the prices set prospectively by the 1985 CBS agreement are deemed to have appeared reasonable to the parties at the time the agreement was reached, the substance of applicants' claim is that subsequent changes—in the economic conditions confronting the networks, and in the value to the networks of the rights represented in the blanket license —make the prices prospectively fixed over this period no longer valid bases upon which to establish fees for future periods. Starting with 1985 as a base permits the Court to evaluate the merits of this claim; to use the fee levels of later years as a base would be to gloss over any unanticipated changes that occurred, and assume, perhaps quite erroneously, that the fees paid for these later years constitute reasonable royalties under then-prevailing conditions.

There remains the question of what the actual base level figure for 1985 should be. ABC's 1985 fee was $8.96 million, while CBS's fee was $9.8 million. The total fees paid by ABC and CBS for the 1981-1985 period, however, were identical ($44.8 million). While CBS's fees were set prospectively by the 1981 agreement, ABC's fees were finalized retroactively in 1985, providing ABC the benefit of CBS's fee-setting experience. The difference in fee levels for 1985 reflects the independent allocation decisions reached by the networks in their respective agreements with ASCAP. CBS's fee schedule called for escalating payments from $8.0 million in 1981 to $9.8 million in 1985, while ABC's schedule prescribed uniform payments of $8.96 million per year. Based on this information, we believe that the CBS fee of $9.8 million is the proper base level upon which to predicate our analysis for both networks. Regardless of the reasons that motivated the different payment schedules, the data from this period reveal that ABC's revenues steadily increased from 1981 to 1985,^[63] while its music use experienced considerable year-to-year volatility, but eventually exhibited a slight increase from 1981 to 1985.^[64] Because these factors—gross revenue and music use—are critical in assessing the royalties for an ASCAP license, the flat payment schedule for ABC over the 1981-1985 period is not, in our estimation, a realistic reflection of the varying value of the blanket license to the network over that period. Rather, we believe it more consistent from the perspective of the rate-setting approach adopted here, and from the viewpoint of even-handed treatment among users, to take the CBS 1985 fee of $9.8 million as the appropriate base figure for both CBS and ABC.

2. Accounting For Changed Circumstances

Applicants' proposal seeks to adjust the base fees chosen to account for inflation, and for changes in the "use and value" of ASCAP music to ABC and CBS. The first modification is accomplished by indexing the base fees to the Consumer Price Index. The second modification involves two steps: 1) adjustments to account for year-to-year changes in applicants' use of music, and 2) adjustments to account for changes in the number of viewers reached by network programming. The issues raised by each of these alterations are addressed in turn.

With regard to applicants' use of the Consumer Price Index to account for inflation, our discussion so far has already indicated we consider that another factor—namely changes in the networks' gross revenues— compensates quite adequately for the diminution in the value of the dollar as well as a number of other key variables. As a measure of inflation, the gross revenue earned by a network offers a more industry-specific index than does CPI. Moreover, accounting for changes in gross revenue gives effect to other considerations, including competition from alternative entertainment media, changes in the relative economic fortunes of the networks, and variations in network viewership or audience share. In short, we think the movement in the revenues received by applicants provides a more accurate index of the overall changes in the economic conditions confronting the networks.

With regard to adjustments made to account for the annual changes in the audiences reached by network programming, we share ASCAP's misgivings concerning the necessity for such a modification. Applicants' reliance on "total household impressions"^[65] overlooks the essential element: the true value of audience share to the networks, in terms of the advertising revenues generated. Revenues are not dependent solely upon the overall number of households reached but rather based on the advertisers' evaluations of various demographic and temporal criteria and the their decisions to allocate resources accordingly. Tr. 503-05. Moreover, the prices advertisers are willing to pay vary over time so that it is possible for a network's revenues to increase, though audience share may actually decline.^[66] Thus network revenue offers a far more reliable yardstick with which to measure the overall value of programming to the network. From a broader perspective, the Court's preference for relying on gross revenue in its calculus already accounts for variations in viewership. Applicants' expert admitted as much at trial, agreeing with the Court that because advertising rates bear a relationship to audience share, a formula that factors in changes in gross revenue automatically takes into consideration the audience share held by a network. Tr. at 740-41. Consequently, including variations in viewership as a factor in addition to changes in a network's gross revenue would constitute double counting.^[67]

With regard to adjustments made to account for changes in the networks' use of music, ASCAP emphatically contests the validity of any approach that relates blanket license fees to changes in the amounts of music actually used. First, ASCAP notes that such an approach marks a substantial departure from previous experience. The fact that the parties have generally negotiated flat-fee arrangements, however, does not signify that they did not independently consider the amounts of both previous and expected music use in their fee deliberations, or that this Court should ignore applicants' claims concerning a significant decrease in their use of ASCAP music in arriving at a reasonable royalty.

Second, ASCAP emphasizes that actual music use only represents one advantage of the blanket license, and does not reflect the value of unlimited access to the substantial ASCAP repertory and blanket indemnification from infringement actions. Because the networks are bulk users, argues ASCAP, these latter advantages are of substantial value. As stated at the outset of this discussion, the Court readily acknowledges the significance of all the rights conveyed by the blanket license and the value of the efficiencies achieved thereby. However, because the Court believes it is valid to predicate our calculations upon the prior prices negotiated between the parties, it must accept that the perceived values of all advantages of the blanket license were reflected in the prices agreed upon. What remains to be determined is how these values have changed since the time the agreement in question was reached. In this context, we think it is appropriate to rely on the change in the amounts of music actually used as a measure of the change in the value to the network of all the rights conveyed by the blanket license. This decision is based on the belief that the "access" and "indemnification" components of the blanket license—comprising, essentially, an option to use with impunity as much ASCAP music as desired—must themselves be wholly dependent on the amount of music a licensee expects to use. Indeed, it is obvious that the option to use even an unlimited amount of ASCAP music has no value to an entity that uses no ASCAP music. It also seems clear that the value of the "option" feature of the blanket license is directly proportional to the amount of music the licensee uses. Thus, on a prospective fee-setting basis, one who expects to use large amounts of ASCAP music may be expected to accede to higher prices than those agreed to by one who expects to use lesser amounts. Thus, for the purposes of a rate-setting inquiry, we think it appropriate to assume that the value of the blanket license to a network will vary in direct proportion to the amounts of music used by that network, all other factors remaining equal.

The alternative posed by ASCAP—to disregard changes in music use in calculating a reasonable usage fee—would oblige us to ignore entirely the claim that evolving conditions have transformed the value to the networks of the rights contained in the blanket license. We find this alternative unacceptable. The blanket license is, after all, a music performance license; to ignore variations in the amounts of music actually performed, would be to engage in a fee calculation inconsonant with reality.

In measuring the changes in the amounts of music used by the networks, applicants' analysis excluded music-use data for the so-called "other" category, comprising music used in commercials and network promotional and public service announcements. The grounds for this exclusion are rooted in the testimony of ASCAP's own economist, which revealed that the data within this category was plagued with problems of a sort that made it unreliable for the purposes of trends analysis. Tr. at 824-42. ASCAP attempts to portray the problems as primarily timing related, so that comparing averages, rather than year-to-year statistics eliminates the variances. The data, however, belie this explanation. The statistics reveal astronomical leaps in use credits exhibited in the "other" category between 1981 and 1991, particularly during the later years.^[68]See DX Q. These surges are aberrational as compared with use credits for the remaining music categories. ASCAP offered no evidence to rationalize these marked discrepancies in use-credit data, or to explain increases in the order of magnitude presented by the statistics reported within the "other" category. The reluctance to rely on ASCAP's data as offered does not stem from concerns as to the credibility of the statistics contained in the later years, but rather from doubts as to the dependability of data for the earlier years; as noted by the Court during trial, the problem is not fear of overstatement in the later periods, but understatement in the earlier periods due to the possibilities of under-reporting and lags in obtaining network compliance. Tr. at 836-42. ASCAP remains unable to verify that the exceptional escalation in reported "other" use credits was the result of increased music use and not a product of changes in network reporting patterns and/or increased diligence in ASCAP's monitoring during the later years. Tr. 838-42. Nor does taking averages eliminate the problem. The weight of the "other" use credits in the later years defeats the desired smoothing effect of taking an average: comparing the average of ABC's "other" use credits over the 1981-1985 period to a similar average over the 1986-91 period shows an increase of nearly 350%. See JX Q. The incongruities in CBS "other" use credit data are less striking, but no less apparent. It is clear that inclusion of statistics from the "other" category will distort any analysis of trends in network music use. Consequently, we agree with applicants that use-credits data for the "other" category is better excluded from an approach that seeks to account for the changes in network music use.^[69]

Applicants also advocate the exclusion of 1991 use credit data because the numbers reported are out of line with prior years. App.Br. at 63; Tr. at 721-22, 145-47. See DX Q. With regard to CBS, it appears appropriate to substitute the higher 1990 music use data in place of the numbers reported in 1991. ASCAP represents that CBS 1991 data are incomplete—a fact applicants do not dispute—and also that the 1990 statistics are "closer to the norm;" Tr. at 147, 727-28. We find this to be sufficient ground to use the 1990 data for both 1990 and 1991. With regard to ABC, the excuse of incomplete data is not available to justify exclusion of the 1991 statistics. True, the 1991 numbers do appear to be significantly and inexplicably higher than those of prior years. See DX Q; Tr. 145-47, 721-22. However, testimony from witnesses presented by both parties reveals that music use data are occasionally marked by timing problems caused by lags in reporting, so that data from one year appears "bunched" with data in another year. Tr. 823-42, 721-22. Thus while ABC's use data show an aberrational leap in 1991, they also show an equally inexplicable dive in 1988.^[70] JX Q. Excluding the peaks without excluding the troughs would be unfair. On average, however, the numbers as reported do indicate a balance and, indeed, this is an instance where taking an average over the 1986 to 1991 period would smooth out the year-to-year variances.^[71]

In analyzing the networks' counter-proposal, the discussion so far has underscored the defects the Court finds in applicants' methodology. The Court agrees that a calculus that begins with a base fee deemed fair at the time entered, and adjusts that fee to account for the appropriate changes in circumstance, provides a suitable approach with which to arrive at a reasonable royalty. As previously stated, we believe that the fees paid in 1985, rather than 1980, offer the proper base from which to begin. With regard to the appropriate changes in circumstance for which to account, we agree with applicants as to the propriety of adjusting for changes in network music use, but disagree as to the propriety of indexing fees both to the Consumer Price Index and to variations in viewership when the changes in network revenue provides a straightforward way by which to account for both inflation and changes in audience share.

C. Calculating A Reasonable Royalty

As indicated, the approach adopted by the Court takes the fees paid by each network in 1985 and adjusts it to account for changes in applicants' gross revenues and use of ASCAP music. The change in gross revenue not only compensates for inflation but reflects changes in the overall value of programming to the network, for example due to changes in viewership. The change in applicants' use of ASCAP music provides a measure of the varying value of the music contained in the programming broadcast.

Not surprisingly, the parties differ in their views as to how this approach should be implemented. Applicants, for the most part, advocate that changes in gross revenue and music used should be based on a comparison of the actual data in the years being compared: to set fees for 1991, applicants would adjust the 1985 base fee for the percentage change in revenue and music use between 1985 and 1991.^[72] ASCAP advocates the use of multi-year averages to compare the changes in revenue and music use: to set fees for 1991, ASCAP would adjust the 1985 base fee by the percentage changes in revenue and music use between the average over the 1981-1985 period and the average over the 1986-1991 period. ASCAP prefers the use of averages to smooth out fluctuations in the year-to-year data that may otherwise distort the analysis.

After studying the data,^[73] we are convinced that the use of averages is necessary to provide meaningful comparison. The raw data do exhibit sizable year-to-year variances which could potentially skew the computation, particularly if the base year against which changes are measured is statistically aberrational. For example, CBS's 1985 gross revenue figure is larger than the figure from the previous year and the figures from each of the subsequent three years. DX PPP. Fluctuations are especially evident in music use data; as already discussed, the ABC 1988 and 1991 numbers are both outliers, and the 1991 CBS numbers are incomplete, requiring the substitution of 1990 CBS data for 1991. The reasons for the year-toyear fluctuations in data were explained by the testimony of more than one witness. With regard to the use-credit statistics, as already noted, timing problems caused by lags in reporting result in noticeable anomalies. With respect to gross revenue, while no doubts are raised as to the accuracy of the networks' accounting, the revenue figures in any particular year may yet be aberrational due to the incidence of unusual programming.^[74] For these reasons, we think it appropriate to use averages rather than statistics for individual years. Since the 1985 fee is used as the base rate, a comparison of the average revenue and average music use between the 1981-1985 period and the 1986-1991 period offers a suitable procedure to account for changing trends and thus carry forward the 1985 fees into the present period.

The approach outlined may be implemented as follows. We start with the 1985 fee and adjust for the percentage changes in average revenue and in average music use from the 1981-1985 period to the 1986-1991 period. For CBS, the analysis yields the following results. Average music-use credits for CBS during the 1986-1991 period were [redacted] of the average use credits for the 1981-1985 period.^[75] Average gross revenues for CBS during the 1986-1991 period were [redacted] of the average gross revenues for the 1981-1985 period. Applying these percentages to the CBS base fee of $9.8 million yields a figure of $9.75 million. We believe this to be the reasonable royalty due ASCAP from CBS for 1991.

For ABC the calculations proceed as follows. Average music-use credits for ABC during the 1986-1991 period were [redacted] of the average use credits for the 1981-1985 period.^[76] Average gross revenues for ABC during the 1986-1991 period were [redacted] of the average gross revenues for the 1981-1985 period. Applying these percentages to the ABC base fee of $9.8 million yields a figure of $10.47 million. Since ABC remains open with regard to final fees for a period dating back to 1986, we think this figure represents the reasonable fee due ASCAP from ABC for each of the years 1986 to 1991.

It is interesting (and reassuring) to note that the 1991 royalties for ABC and CBS computed by the formula approved above each represent [redacted] of their gross revenues for that year, so that as between these two respective companies, there can be no serious question of disparate treatment. We also note the substantial similarity between the rate yielded by the present analysis for ABC and CBS and the 0.44% rate paid by NBC in 1992.

With regard to the fees due ASCAP for 1992 and 1993, the Court is faced with two alternatives. We could leave the parties free to calculate their respective fees, based on the formula prescribed herein, once music use and revenue data become available for these years. This would keep open the final fee determination for 1992 and 1993 until well into 1994. Alternatively, the Court could establish the prices determined above to be the final fees for 1992 and 1993. This bears the obvious flaw of assessing a fee derived from an analysis of changing trends in revenue and music use over a period where the data for the pertinent time span are currently unavailable.

Despite the evident problems with the latter approach, we believe it best to set now the fees for 1992 and 1993 at the dollar levels arrived at under the above analysis. First, this has the manifest advantage of achieving finality, rather than leaving the determination of final fees open until well into the next year. Second, establishing the fees presently obviates the need for further scrutiny of statistical data; we have some apprehension about adopting a fee-setting formula whose implementation is dependent on data the parties produce in the future. This would risk leaving the parties—and perhaps the Court—open to addressing disputes as to the reliability of particular data for the purposes of trends analysis. The degree of scrutiny given to statistical data in the present proceeding —the resolution of the difficulties posed by the use-credits data in the "other" category, the deficiencies in the CBS 1991 use-credit statistics, and the deviations in ABC data for 1988 and 1991—underscores the concern that the parties will perforce engage in similar polemics over the reporting of future data. Third, we have some confidence that the prices derived from the present analysis provide a fair approximation of the royalties reasonably due over the next two years. The trends analyses undertaken spanned a decade of data—from 1981 to 1991—to assess the effect of changing conditions on the networks, and to appraise variations in the value of the rights at issue. Our aim was not to pin-point prices precisely, accounting for each variant on a year-to-year basis, but rather to arrive at a reasonable approximation of the fair royalties due after giving consideration to the broad changes purported to have made the fees previously negotiated outdated measures of fair value. We believe that accounting for the effect of relevant changes over the last six years on the fees negotiated at the start of that period provides a good indication of the rates reasonable for the immediate future. For these reasons we find that the figures derived above—$10.47 million for ABC and $9.75 million for CBS—are also appropriate levels for the fees due ASCAP in 1992 and 1993.

D. Interest

ASCAP's proposal sought interest from applicants in the amount its recommended fees exceeded applicants' previous interim fee payments. To the extent its proposal endeavored to replicate the NBC agreement, ASCAP contended that interest be computed from April 30, 1992, "the date on which NBC paid the fees it agreed to pay in excess of its interim payments." ASCAP Contentions ¶ 3.

We believe a provision for interest is appropriate. The rationale for assessing interest from the date of the 1992 NBC agreement is not applicable, however, since the Court declined to adopt the NBC rates as reasonable for ABC and CBS. Rather, it appears more logical to assess interest from the period covered by each interim payment. Interest should be computed at a rate of 6% per annum on a simple interest basis.

With respect to ABC, the parties inform the Court that they have agreed on the computation of interest due ASCAP, in the manner and at the rate provided above, for the period beginning January 1, 1986 and ending June 30, 1993. The agreed upon computations are as follows: the additional fees due (in excess of ABC's interim payments) total $5,025,000, and the interest due on these additional amounts totals $1,143,187. Thus ABC owes ASCAP an aggregate additional amount of $6,168,187 in fees and interest to cover the period January 1, 1986 to June 30, 1993. In addition, ABC will owe ASCAP the balance of the $10.47 million per annum fee prescribed by this Opinion and Order for the period July 1, 1993 to December 31, 1993.

With respect to CBS, since the network's interim payments exceed the amount of the final fees assessed herein, we believe it appropriate that ASCAP pay CBS interest in the manner and at the rate provided above. The parties have also agreed on the computation of interest due CBS from ASCAP for the period beginning January 1, 1991 and ending June 30, 1993. The agreed upon computations are as follows: the excess payments made by CBS total $125,000 and the interest on these excess amounts totals $9,688. Thus ASCAP owes CBS an aggregate amount of $134,688 for the period January 1, 1991 to June 30, 1993. CBS will owe ASCAP the balance of the $9.75 million per annum fee prescribed by this Opinion and Order for the period July 1, 1993 to December 31, 1993.

CONCLUSION

For the foregoing reasons, the royalties due ASCAP are finalized as follows. ABC's blanket license fee for the period January 1, 1986 to December 31, 1993 shall be $10.47 million per annum. ABC shall pay ASCAP interest on the amount by which this final fee exceeds the interim payments made by the network on a 6% per annum simple interest basis. This will result in a total payment (net of the interim payments made and interest due) of $6,168,187 from ABC to ASCAP for the period ending June 30, 1993. ABC will owe ASCAP the balance of the fee provided for herein for the period July 1, 1993 to December 31, 1993.

CBS's blanket license fee for the period January 1, 1991 to December 31, 1993 shall be $9.75 million per annum. ASCAP shall pay CBS interest on the amount by which the interim payments made by the network exceed the final fee on a 6% per annum simple interest basis. This will result in a total payment of $134,688 from ASCAP to CBS for the period ending June 30, 1993. CBS will owe ASCAP the balance of the fee provided for herein for the period July 1, 1993 to December 31, 1993.

ASCAP should submit to the Court a proposed Judgment Order for settlement upon ten days notice to applicants.

SO ORDERED

NOTES

[1] Applicants initially also sought a separate fee determination for a per-program license, permitting the non-exclusive right to perform, in unlimited fashion, any music contained in the ASCAP repertory on the particular program for which the license is issued. Prior to trial, however, applicants withdrew their application for determination of per-program license fees, without prejudice to their right to secure such licenses on reasonable terms for succeeding license periods.

[2] The designation "SF" refers to Stipulated Facts. Hereinafter, "AX" will be used to refer to Applicants' exhibits, "DX" for ASCAP exhibits, "JX" for joint exhibits, and "Tr." for the trial transcript. Applicants' brief will be referred to as "App. Br.", and their reply brief as "App. Rep. Br." ASCAP's brief will be referred to as "ASCAP Br.", and its reply brief as "ASCAP Rep. Br."

[3] For a brief history of the evolution of the Consent Decree see, United States v. ASCAP, 782 F. Supp. 778, 782-85 (S.D.N.Y.1991).

[4] The Consent Decree, Section VII(B), also requires ASCAP to make a per-program license available to users.

[5] The affiliates earn revenues primarily through the sale of commercial announcements in and around locally-originated programs, syndicated programs and network programs. They also receive payments from the networks as compensation for airing network programming. SF ¶ 9.

[6] For programs produced by the networks themselves, music selection is made by the individual networks and may include pre-existing compositions, works obtained from music libraries or production companies, or music composed specifically for the program. SF ¶ 15.

[7] For the period 1970-1980, CBS had heretofore paid interim fees in the amount $4.32 million per year. JX 21.

[8] The 1981 CBS agreement prescribed fees for each of the years 1970-1980. In 1985, however, CBS received a most-favored-nations rebate of $3.64 million, attributable to the 1977-1980 period. This amount was not allocated to specific years, leaving the total final fees paid for the 1977-1980 period at $18.84 million (an average of $4.71 million per year). See SF 21.

[9] The CBS 1981-1985 fees ran as follows: $8.0 million for 1981; $8.5 million for 1982; $9.0 million for 1983; $9.5 million for 1984; $9.8 million for 1985. JX 21.

[10] It was this agreement that triggered the most-favored-nations rebate of $3.64 million to CBS, thus resulting in equal fees for CBS and ABC for the 1977-1980 period.

[11] The CBS 1986-1990 fees ran as follows: $9.8 million for 1986; $10.2 million for 1987; $10.7 million for 1988; $11.0 million for 1989; $11.3 million for 1990. JX 22.

[12] Until this agreement, NBC had not paid final license fees to ASCAP since the third quarter of 1976. The interim fees NBC had paid ASCAP between 1976-1990 totalled roughly $15 million less than the sums paid as final fees by CBS for that period. Tr. at 126, 257.

[13] Gross network revenue was defined as the total amount payable to NBC for the broadcast of commercial announcements, net of agency commissions. JX 32.

[14] NBC broadcast the Olympic games in 1992, earning extraordinary revenues for that year, but also facing expenses from the games that potentially exceeded the revenues derived. Consequently, in order to avoid saddling NBC with ASCAP fees calculated from an excessively high gross revenue base, it was agreed that 1992 revenues would be ignored entirely, and instead an average of 1991 and 1993 revenues would be substituted as the appropriate revenue base from which to derive 1992 fees. Tr. 99-102.

[15] The trial court's opinion in Showtime is published in the Federal Reporter as an appendix to the Second Circuit's opinion. ASCAP v. Showtime, 912 F.2d at 572-98.

[16] See BMI v. CBS, 441 U.S. at 19, n. 32, 99 S.Ct. at 1562, n. 32: "It takes an organization of rather large size to monitor most or all uses and to deal with users on behalf of the composers. Moreover, it is inefficient to have too many such organizations duplicating each other's monitoring of use." (citing Twentieth Century Music Corp. v. Aiken, 422 U.S. 151, 162, 95 S. Ct. 2040, 2047, 45 L. Ed. 2d 84 (1975)).

[17] The striking feature of the blanket license is that it lowers ASCAP's monitoring cost and therefore may be cheaper to offer than an alternative performing license that grants the user something less than blanket immunity.

[18] Consistent with the provisions of the 1992 NBC agreement, ASCAP's proposal calculates 1992 fees for each of the networks as a percentage of the average network gross revenue for 1991 and 1993. Of course, the purported rationale behind that provision of the NBC contract, the surge in NBC revenues and costs in 1992 due to the Olympic Games, does not apply to ABC and CBS. See note 14, supra.

[19] Consistent with the NBC agreement, gross revenue is taken to mean broadcast revenue, net of agency commissions.

[20] SF ¶ 3, 4, 6.

[21] SF ¶ 8-11.

[22] JX 4, 5; AX 7.

[23] Using the consumer price index as a measure of inflation, with the average of CPI for 1982 through 1984 as the base.

[24] CBS's fixed fees, when expressed as a percentage of revenue, yielded a lower rate in each of the years 1990 to 1981 than the 0.44% rate presently proposed, with the exception of one year, 1983, where the fee amounted to [redacted] of gross revenue. DX D.

[25] The Court calculated, from ASCAP exhibit M, the average of the proposed fees for 1986 through 1991, as expressed in real dollars, and compared it to the average of the fees, as expressed in real dollars, paid during the years 1981 through 1985. ASCAP preferred a comparison that used 1981 as the base year to show that the proposed fee for each of the years 1986 through 1991 is lower than the 1981 fee in real dollar terms. However, the same does not hold true when comparing the fees ASCAP proposes to the fees paid in the years 1983 through 1985. See DX M. The difference may be explained by the fact that ABC's total fees for 1981 through 1985 were apportioned evenly across each of those years, rather than on an escalating scale, and thus the 1981 fees are bound to appear higher in real dollar terms. In short, taking an average over the two periods provides a more instructive comparison.

[26] ASCAP points out ABC's revenue figures through 1984 include "other" revenue—revenue not derived from the sale of broadcast time— while the figures for 1985 to 1991 do not. If the "other" revenue were excluded from the pre-1985 figures for consistent comparison, the revenue figures would be lower and thus the percentage of revenue rate yielded by the fixed fees for the 1981 to 1985 period would be higher.

[27] This range reflects two internal computations performed at NBC. One was based on the rate used by the Court in Showtime—prime plus three percent—resulting in interest in the order of $37 to $45 million, depending on the compounding method employed. The other was based on the flat prime rate, resulting in interest in the order of $25 to $29 million. Tr. at 257. That a base amount of $15 million could generate interest levels in the magnitude of $45 million seems, at a glance, to be the product of fairly generous interest rates applied on a compound rather than simple interest basis. Were the matter to be litigated, the prospects of interest exposure of these magnitudes may have affected NBC's view of its total liability; however, from the perspective of negotiating the 1992 agreement, the claim that NBC considered itself offsetting a $45 million dollar exposure must be viewed with skepticism when the facts reveal that ASCAP's significantly lower estimates were communicated to NBC. Tr. at 255, 267.

[28] Mr. Cotton's testimony reveals that the lowest amount ASCAP was willing to accept as interest was $9 million. Tr. at 267.

[29] The percentage-of-revenue approach represents a departure from the practice of the prior 20 years. Tr. at 216. During fee negotiations, NBC appears to have expressed its willingness to discuss a percentage of revenue format, as "an additional carrot to try to get settlement discussions going." Tr. at 271-72. NBC maintains it deemed the revenue-based approach undesirable and inappropriate, but offered to acquiesce in the hopes of reaching a resolution acceptable to the network in terms of its exposure from the open period. Tr. 271-75.

[30] ASCAP explains that NBC made it clear that it would seek to pay less than CBS if the matter were litigated. ASCAP Br. at 14; AX 6.

[31] That ASCAP's interest claim "rested almost entirely" on the Showtime decision does not alter the reality that this Opinion was the only litigated final fee decision to date; NBC did not have the benefit of a judicial precedent on the issue to the contrary. ASCAP apparently requested an interest award in that proceeding, as it does in the present proceeding.

[32] Specifically, the NBC agreement prescribed fees in amount of approximately $2.2 million more than what CBS paid for the period 1976 to 1990, a 1991 fee that matched CBS's 1991 payment, and a percentage of revenue formula for 1991 and 1992. SF ¶ 21.

[33] A premium of $4 million a year to ASCAP for 1991 and 1992, would result in a total premium of $16 million in music licensing expense, if it is assumed that an equivalent payment to BMI also constitutes a $4 million premium to that society. It hardly appears a bargain for NBC to agree to these sums, when the estimate that ASCAP itself brought to the table to initiate negotiations was a demand for $15 to $17 million in interest, and when ASCAP indicated it may accept $9 million lump sum to settle the interest issue. Of course, if NBC realistically believed it faced exposure in the order of $37 to $45 million were the matter litigated, then the numbers proffered by applicants in their post-trial submissions may appear more plausible.

[34] Questions as to the impact of the 1992 agreement on NBC's fees to BMI arose during cross-examination of Mr. Cotton. Tr. at 287-302. The testimony does not offer a full picture as to how NBC viewed its BMI exposure relative to the considerable weight it placed on avoiding an ASCAP agreement authorizing a lump sum payment to ASCAP designated as interest.

[35] ASCAP submits that NBC willingly pursued the percentage-of-revenue approach, while the network maintains it consistently preferred the flat fee arrangements of previous years, but offered to relent to resolve the interest question. Irrespective of the degree to which NBC may have been amenable to the revenue based method, it is apparent that ASCAP favored the formula and endorsed its application for NBC, as it does here for ABC and CBS.

[36] Applicants attempt to explain away this glaring fact by attributing the difference in pre- and post-1981 fees to the allocation or distribution constraints facing ASCAP. Tr. at 712-14. However, applicants' expert failed to explain to any degree of satisfaction why an agreement that spans an extended period and spawns artificially inflated fees would then see the premiums produced allocated to one particular interval of the entire period, at the expense of another interval. See Tr. 712-14.

[37] Notably, Mr. Vradenburg never testified that he had Mr. Poltrack's projections in mind, much less relied on them in arriving at the fee levels to which he was willing to accede for music performance licensing. Nor did he suggest that favorable forecasts fostered a generous disposition at CBS towards the operating costs the network was willing to bear into 1990.

[38] Both Mr. Poltrack and Mr. Vradenburg testified that they were aware that other predictions were considerably more pessimistic those contained in The Road to 1990. Tr. 436, 35-53.

[39] One of the advantages of the blanket license is that it obviates the need for individual negotiations and thus eliminates the difficulties of dealing with hold-outs.

[40] The reference is to an ASCAP distribution practice known as the "spot adjustment factor." SF 17; Tr. 155-57. Since 1967, a percentage of the license fees collected by ASCAP from local television stations has been allocated, for distribution purposes, to ASCAP members whose works have been performed by the networks. This percentage represents an estimation by ASCAP's independent survey experts of the percentage of local station license fees derived from the revenue generated by commercials in or adjacent to network programs. The practice is permitted by Court Order, see JX 2 at 2, and is not challenged here. Nor is it really clear what effect the practice would have on the cost of source and direct licensing were specific terms and prices to be precisely negotiated (Mr. Vradenburg's testimony indicates that CBS's negotiations were only over options to obtain source and direct licenses, with prices to be determined later, when CBS chose to exercise its options).

[41] ASCAP goes to great lengths to reveal the "curiously incomplete" nature of the evidence presented. Applicants failed to call as a witness the individual who primarily conducted the source and direct licensing efforts. Instead, the testimony offered, and the documents submitted, largely provide second-hand accounts of the results reached. Missing from these accounts is an explication of the types of resistance encountered or of the nature of the concerns expressed by the publishers and program packagers. Moreover, much of the correspondence submitted as exhibits post-dates the 1985 CBS agreement and therefore is not pertinent as to the impact of these efforts on the terms of the 1985 contract.

[42] Even Mr. Vradenburg's testimony does not manifest that the fees agreed upon were unreasonable:

"We were looking for a two-to-three-year license so that we could continue our efforts, and ASCAP quoted us a blanket license fee which at least in our estimation was unwarranted.... It had significant jumps in the fees. We went back to ASCAP and sought lower fees. They would agree to lower fees only if we extended the license five years. So we ended up agreeing to a five year license from '86 to '90.

Tr. at 425.

[43] The blanket license agreements between ASCAP and the networks do not contain carve-out features under which the network would receive credit against its blanket license fees for the compositions it was able to access through source or direct licensing.

[44] Despite the assault applicants make on all agreements entered into with ASCAP, the fee proposal applicants advance also admits the necessity of finding a prior agreement from which to commence analysis.

[45] See App.Br. at 16, n. 10: "A fairly priced per-program license would facilitate resort to source and direct licensing."

[46] Cf. Showtime, 912 F.2d at 570 (opportunity to resort to rate court would have little meaning if that court were "obliged to set a `reasonable' fee solely or even primarily on the basis of fees ASCAP had successfully obtained from other users"). (emphasis added).

The Second Circuit also noted that the Disney rate was rejected as a benchmark for fees due from Showtime "because this rate was agreed to at an early stage of the Disney Channel's existence and thus reflected considerations not pertinent to arms-length bargaining between ASCAP and an established cable service like SMC." Showtime, 912 F.2d at 567. No such comparison can be made in the instant case; as ASCAP points out, ABC and CBS have been giants in the television industry for nearly half a century.

[47] While Mr. Vradenburg testified that CBS viewed the Court to be in an "impossible position," in rate-setting in having to either ratify prior agreements or undertake a "rather difficult" analysis, neither he nor other network executives testified to any previous aversion on the networks' part toward this Court based on fears of bias. Tr. at 424.

[48] The issue was ultimately litigated and decided in the cable suppliers' favor. See United States v. ASCAP (Application of Turner Broadcasting System, Inc., 782 F. Supp. 778 (S.D.N.Y.1991), aff'd, 956 F.2d 21 (2d Cir.), cert. denied, ___ U.S. ___, 112 S. Ct. 1950, 118 L. Ed. 2d 554 (1992).

[49] CBS's fee to BMI in 1981 was 75% of the fee paid to ASCAP; the 1985 BMI fee was 85% of the ASCAP fee. For 1992, CBS agreed to pay BMI 100% of the final fee the network pays ASCAP. JX 84, 87; SF ¶ 21.

[50] See JX Q. The "other" category comprises music in commercial announcements, promotional announcements and public service announcements. Tr. 138.

[51] ASCAP also maintains that gauging the fluctuations in the amount of music used by a network does not provide a full measure of the value of a blanket license; beyond the music actually used, the license offers the additional rights of unlimited access to the entire ASCAP repertory and blanket indemnification against infringement actions. The degree to which reliance on changes in the amounts of music used by the networks provides an adequate measure of the full value of a blanket license is also a matter to be addressed later in this opinion. The fact that there may be other valuable components to the blanket license does not diminish the reality that actual music use remains an indispensable factor in evaluating the overall worth of the license to the network.

[52] Applicants note that while music may enhance the dramatic presentation of any program, it can hardly be said to be the essential element behind the success of most programs and thus the driving force behind a network's revenues. See Tr. at 580-592; see also, Tr. at 238, 275-6, 540-48, 560-62. The networks' decisions as to the types of programming likely to attract viewers and, in turn, to attract advertising revenues are generally unaffected by the musical content of the programs. Tr. at 588-90. Moreover the testimony of network finance executives suggests that the sizable sources of revenue in the recent past have come from types of programming that depend little on music, such as sports, news and reality-based programming. Tr. at 540-55, 612-31, 274-76. Indeed, the very fact that ASCAP agreed to structure NBC's 1992 fees based on an average of 1991 and 1993 revenues—in recognition of the fact that the surge in revenues due to the broadcast of the 1992 Olympic Games, which relied little on ASCAP music and which turned little profit for the network, would distort the fees due ASCAP—serves as an acknowledgement of this reality. Tr. at 274-76.

[53] Such arrangements bear a marked distinction, applicants argue, from the regime instituted under an ASCAP blanket license. In other intellectual property cases, creators receive payment based on the success of the creation to which they have personally contributed. Where a patent licensee generates revenues from the use of the patented product or process, it may be appropriate to index the fees owed to the creator to the revenues generated by the specific use of that product. Similarly, where an author receives royalties based on sales, such royalties reflect a percentage of the publisher's gross from the sale of that author's book, not a percentage share of the publisher's total receipts from all publishing endeavors. App.Rep.Br. at 29, n. 26. The same is true for scriptwriters and directors; residuals are paid as a percentage of the revenues resulting from the exhibition of the shows on which they worked, not as a share of the revenues collected in connection with the distribution of all other films. In each of these cases, a percentage-of-revenue formula occasions compensation based on the success of the creations to which the licensor has directly contributed. In contrast, the percentage-of-revenue formula for the blanket license gives ASCAP a share of the network's revenue from all programs, including those programs that contain no ASCAP music. Because the formula takes no account of music use, it would provide a windfall for ASCAP in situations where large revenues are generated from programs that use no music. Cf. Consent Decree, Sec. VII(A) (enjoining ASCAP from demanding fee based on a percentage of income from programs that contain no ASCAP music).

[54] ASCAP's allusion to its experience with radio stations, which have historically acquiesced to revenue-based agreements, also seems inapposite. Tr. at 40-42. As applicants note, the radio broadcasters' traditionally extensive dependence on feature music raises considerations as to the propriety of revenue-based licenses not applicable to the television networks' circumstance. As to ASCAP's reference to the experience of local stations, (Tr. at 42), we are aware that it can hardly be claimed to offer an enthusiastic endorsement of the revenue-based approach. See, United States v. ASCAP; Application of Buffalo Broadcasting Co., Inc., et al., Slip. Op. at 18-20, (S.D.N.Y. February, 17, 1987) (interim fee opinion).

[55] It appears to the Court that fee disputes and the resultant open periods occur when the networks refuse to assent to ASCAP's proposed fee levels; irrespective of whether those levels were generated by a percentage of revenue formula or by a flat fee arrangement, it seems the networks' resistance would remain equally resolved. Indeed the very existence of this litigation and the networks' emphatic opposition to a revenue-based formula underscores this point. And ASCAP's own economist, Dr. Boyle, allowed that the total dollars to be paid ASCAP were ultimately the crucial consideration, regardless of the method employed in generating those fee levels. Tr. at 124.

[56] At trial, applicants presented considerable testimony concerning the proliferation of competitive forces facing the networks; the accessibility of VCR's, the growth of independent stations, the emergence of the Fox Broadcasting Company, and the spread of cable television present powerful challenges to the once uncontested primacy of network television. Tr. 449-61, 480-88 (testimony of Alan Wurtzel, Senior Vice-President for ABC). To the extent these alternative sources of home entertainment diminish the perceived ability of the networks to reach households, and thus reduce their attractiveness from the perspective of advertisers, the networks' gross revenues provide a yardstick by which to measure the impact of these forces on the financial performance of the network.

[57] ASCAP maintains that the networks' prospects are not as dire as applicants claim for purposes of this proceeding. ASCAP Br. at 24-26. The networks are still the most efficient and effective way for advertisers to reach households. Tr. at 500. With regard to applicants' net income positions, ASCAP points out that ABC has posted healthy profits since 1989, while CBS's recent losses were due in large part to over-spending on sports licensing, including write-downs for future years. Tr. at 549-53.

[58] We are aware that applicants raise the same objection to the use of gross revenue—that it bears little relation to the networks' use of ASCAP music. Indeed the concerns expressed formed the basis for our rejection of a formula based solely on revenues. See note 52 above and accompanying text. An approach that accounts for fluctuations in music use should address the concerns applicants raised. The observation made here regarding profits arises with respect to different considerations as the discussion indicates.

[59] In fact, taking into account all other costs of operation, which the reliance on net profits would entail, may actually cut against applicants' interests. If net profits have declined due to substantial increases in overall network expense, this provides greater support for ASCAP's contention that its fees should not be singled out for reduction, than it does for applicants' contention that ASCAP should be penalized for the networks' deteriorating bottom-line performance.

[60] ASCAP also notes that applicants never advanced this position previously, when their profits were "vaulting ahead." ASCAP Rep.Br. at 4.

[61] Applicants speak in terms of finding a prior agreement which "could be said most closely to approximate competitive market fee levels." App.Br. at 59. As our analysis at outset of this discussion indicates, we prefer to eschew such analysis because we believe it involves such nebulous concepts as to be of little assistance in the Court's assigned task.

[62] The Supreme Court denied certiorari in the CBS v. ASCAP on March 2, 1981, see 450 U.S. 970, 101 S. Ct. 1491, 67 L. Ed. 2d 621 (1981), and denied CBS's petition for rehearing on April 6, 1981, see 450 U.S. 1050, 101 S. Ct. 1772, 68 L. Ed. 2d 247 (1981). The 1981 CBS agreement was joined with ASCAP in June, 1981. JX 21.

[63] ABC's 1981 gross revenue (broadcast revenue net of agency commissions) in 1981 was [redacted]; its gross revenue in 1985 was [redacted]. DX OOO, AX 151.

[64] ABC's total use credits (excluding use credits in the disputed "other" category) for 1981 were 7,501; its use credits for 1985 were 7,926.

[65] Total household impressions reflect, on an annual basis, the average number of U.S. television households viewing network television per minute, across a total network broadcast day, multiplied by the number of minutes of network programming transmitted that year. The viewership data on which applicants rely was compiled by A.C. Nielson Co. App.Br. at 64, n. 36; Tr. at 717-24.

[66] ASCAP notes that despite reaching fewer viewers, the networks have been able to maintain, or increase their advertising revenues.

[67] ASCAP also points to a technical problem posed by any analysis that seeks to rely on changes in network audience share. Evidently, A.C. Nielson Co. altered its measurement technique in 1987, raising questions as to the comparability of pre- and post-1987 Nielson data. Tr. at 478, 461-475. Whether the drop in network audience share is due, in part, to the adoption of new measurement mechanisms need not be evaluated, given the methodological reasons why the Court chooses to avoid audience share statistics entirely.

[68] For example, comparing ABC's 1991 "other" use credits to those of 1985 shows an increase of nearly 900%. Taking an average for ABC for 1990 and 1991 and comparing it to the average of the 1981 to 1985 period shows an increase of over 700%. JX Q.

[69] ASCAP argues that its music use data bear strong indicia of reliability: the vigilance of ASCAP members, whose income depends on ASCAP's music use credits data provides a "built-in check" of the statistics reported. ASCAP Br. at 30; Tr. at 841-42. However, as the Court observed during trial, ASCAP's members cannot be presumed to be eternally vigilant; moreover, they surely rely for the most part on ASCAP's diligent watchfulness to protect their interests. Tr. at 842.

[70] The numbers reported in 1991 are 25% higher than the 1986-1991 average; the numbers reported in 1988 are 22% lower than this same 1986-1991 average. See DX Q.

[71] The situation presented by the 1991 music use data must be distinguished from the dilemma posed by the data in the "other" category. In the latter case, there appeared a systematic bias towards larger use credits reported in later years, and the discrepancies exhibited were of a striking order of magnitude. The same cannot be said of the 1991 music use data.

[72] Applicants advocate the use of averages with respect to the ABC music use data between 1986 to 1991, based on the fact that the numbers reported for ABC in 1991 are aberrationally higher than those of previous years.

[73] The data for music use are taken from ASCAP's exhibit Q. The data for network gross revenue are taken from Applicants' exhibit 129 (CBS) and Applicants' exhibit 151 (ABC). Gross network revenue is assumed to mean gross broadcast revenue net of agency commissions, as defined in the 1992 NBC agreement. The revenue data are also contained in ASCAP exhibit PPP (CBS) and ASCAP exhibit OOO (ABC).

[74] For example, NBC experienced extraordinary revenues in 1992 due to the broadcast of the Olympic games. Similarly, ABC experienced a year in which it broadcast football playoff games from two different seasons as well as the Super Bowl, sizably increasing its revenues over that of previous years. Tr. at 615-16.

[75] The calculations were made excluding data from the "other" category and using 1990 use credits data for both 1990 and 1991. See DX Q.

[76] As in the case of CBS, use credits from the "other" category were excluded. Unlike CBS, however, ABC's 1991 use credit numbers were taken as represented. See DX Q.