Opinion
Civ. A. No. 77-325.
November 24, 1981.
David A. Drexler, William H. Sudell, Jr., and John F. Johnston, of Morris, Nichols, Arsht Tunnell, Wilmington, Del., for plaintiffs.
William J. Wier, Jr., Robert Jacobs, and Terry C. Seningen of Bader, Dorsey Kreshtool, Wilmington, Del., for defendants.
OPINION
Virtually all licensing of performing rights to musical compositions in the United States is conducted by two large organizations representing thousands of individual copyright owners, Broadcast Music, Inc. ("BMI") and the American Society of Composers, Authors, Publishers ("ASCAP"). In recently completed, multi-phased litigation, the Columbia Broadcasting System, Inc. ("CBS") raised ultimately unsuccessful antitrust challenges to the practices of BMI and ASCAP in licensing performing rights to music used by television networks. This lawsuit involves antitrust and copyright misuse challenges to BMI's practices in licensing music rights to small establishments, like nightclubs and bars, that provide live music.
Columbia Broadcasting System, Inc. v. ASCAP, 400 F.Supp. 737 (S.D.N.Y. 1975) (CBS II); rev'd and remanded 562 F.2d 130 (2d Cir. 1977) (CBS III); rev'd and remanded sub nom BMI v. Columbia Broadcasting System, Inc., 441 U.S. 1, 99 S.Ct. 1551, 60 L.Ed.2d 1 (1979) (CBS IV); on remand sub nom Columbia Broadcasting System, Inc. v. ASCAP, 607 F.2d 543 (2d Cir. 1979) (CBS V); 620 F.2d 930 (2d Cir. 1980) (CBS VI); cert. denied 450 U.S. 970, 101 S.Ct. 1491, 67 L.Ed.2d 621 (1981).
In 1977, BMI and several publisher affiliates initiated a copyright infringement action against Moor-Law, Inc., a Delaware corporation doing business as the Triple Nickel Saloon. In 1979, BMI initiated another infringement action under the amended Copyright Act. The two actions were consolidated. Triple Nickel raised the affirmative defense of copyright misuse, and counterclaimed for violations of Sections 1 and 2 of the Sherman Act and Section 3 of the Clayton Act. In earlier decisions, this Court granted summary judgment in favor of BMI on the copyright infringement claims, BMI v. Moor-Law, Inc., 484 F.Supp. 357 (D.Del. 1980), and postponed the resolution of the Triple Nickel's motion for class certification pending the resolution of the liability issue, BMI v. Moor-Law, Inc., CCH Trade Regulation Reporter Transfer Binder ¶ 63,472 (D.Del. 1980).
17 U.S.C. § 101, et seq. (1978).
The central issue at trial was the legality of BMI's use of a "blanket license" agreement. This agreement provides users like the Triple Nickel with access to all compositions within the BMI repertory, and bases the fee for that access not on the amount of BMI music used, but on the user's total entertainment expenses.
I. FACTUAL BACKGROUND.
Responding to the need of a large number of musical copyright holders to license the rights to their music to a large number of geographically diverse music users, ASCAP was founded in 1914 as the first musical performing rights organization in the United States. In 1939, radio users, angered by ASCAP's rising licensing fees, formed BMI as a competitor. Both organizations have grown over the years and together dominate the musical performing rights field today. ASCAP controls rights to approximately 3,000,000 compositions; BMI controls the rights to approximately 1,000,000 compositions and claims to have a greater share of the currently more popular songs. BMI has 40,000 writers and 16,000 publishers affiliated with it; it has 350 employees. There is also a smaller, less well-known performing rights organization called SESAC which controls the rights to approximately 150,000 compositions.
As chronicled in the recent CBS litigation, both ASCAP and BMI have been subject to antitrust litigation in the past. BMI's licensing activities have been subject to two consent decrees, United States v. BMI, 1940-1943 Trade Cases ¶ 56,096 (E.D. Wis. 1941); United States v. BMI, 1966 Trade Cases ¶ 71,941 (S.D.N.Y. 1966). As a result of the 1966 consent decree, BMI is required, inter alia, to make a per composition license, as well as its blanket license, available to all licensees and to allow direct licensing by individual copyright holders.
BMI licensees fall into two categories: broadcast and non-broadcast (or general). Included among the broadcast licensees are television networks like CBS, television stations, and radio stations. Included among the non-broadcast licensees are hotel and motels, dance studios, skating rinks, colleges, concert halls, and a sub-category called "GLAs." GLAs are small establishments like bars, nightclubs and restaurants that offer live music and are subject to BMI's General Licensing Agreement (GLA).
The Triple Nickel falls within the GLA category. Owned and operated by Mr. Robert Moor, the Triple Nickel has been doing business in Bear, Delaware, since September 1976. It offers primarily country and western music. Live music is performed nightly at the Triple Nickel by both nationally known musicians and by lesser known, often local, performers. It is open 365 nights a year for four hours each evening. Approximately 40 songs per evening are performed at the Triple Nickel.
Mr. Moor included among the Triple Nickel's recent performers Johnny Paycheck, Freddy Hart, Moe Bandy, Del Reeves and Margo Smith.
The "blanket license" offered by BMI to establishments like the Triple Nickel provides the licensee with the right to immediate, indemnified access to any and all songs in the BMI repertory. The GLA licensee is charged an annual fee for the license based on its annual entertainment expenses (i.e., the cost of hiring musicians). The rate schedule which currently includes a $75 minimum fee, is contained in BMI's standard GLA Agreement, which every licensee must sign. Edward Cramer, president of BMI, testified that the current standard Agreement, including its rate schedule, has remained substantially unchanged for about ten years. The Triple Nickel's fee for BMI's license based on its 1979 estimated entertainment expenses of $75,000 would have been $400. ASCAP offers the same full repertory blanket license to GLA licensees, but its fee is derived from a formula which includes the price of drinks served, seating capacity, and other variables. Mr. Moor testified that an ASCAP representative informed him that their licensing fee for the Triple Nickel would be approximately $600.
The 1980 standard Agreement (PX-13) contains the following fee schedule:
BMI offers a similar blanket license to non-GLA licensees. For radio licensees, for example, the license fee is determined by a schedule geared to the station's advertising revenues; for concert licensees, the fee is based on concert revenues; and for the hotel-motel licensees the fee is based, like it is for the GLAs, on entertainment expenses. While BMI sets the terms of the GLA licensing agreement unilaterally, the terms and rates of the standard contract for the blanket license for other licensees like radio stations and hotel and motels are negotiated between BMI and the industry trade association.
Revenues from all licensees are combined into a general fund. In 1980, BMI's revenue from all sources was approximately $90 million. Despite the greater number of non-broadcast licensees, roughly 90% of BMI's annual licensing revenues comes from broadcast licensees.
Before distributing revenues to its affiliates, BMI deducts its total expenses. A substantial portion of total expenses are attributable to licensing costs. The costs for licensing depend on the type of licensees. In the broadcast licensing area, there are fewer licensees and they typically tend to be more knowledgeable about their obligations under the copyright law. Because there are trade associations which negotiate standard contract terms, acceptance of the terms of the contracts by individual licensees is easily obtained. On the other hand, in the non-broadcasting field there are more licensees, few trade associations, and much higher turnover. For GLAs, costs include locating the GLA establishment, notifying it of its obligations under the copyright laws, reaching an agreement, and, where appropriate, initiating infringement litigation. Because of the conditions of large numbers, large turnover, lack of awareness, and resistance to imposed contract terms, BMI's enforcement problems also tend to be high for GLAs. Therefore, although BMI does not regularly break down its overall licensing costs among the different categories of licensees, BMI officials believe that the expense of licensing is much greater for non-broadcast licensees, especially GLAs.
Shortly before trial, BMI conducted its first ever financial analysis of its GLA licensing operation. Comparing fiscal year 1979 GLA licensing fee revenues with estimated fiscal year 1979 GLA licensing costs, BMI concluded that GLA licensing costs exceeded licensing revenues by $46,000 (PX-41).
After overall expenses are deducted from general revenues, net revenues are distributed as royalties to BMI's writer and publisher affiliates in accordance with current payment rate schedules. The rate schedules differ depending on the type of broadcast media, but all schedules use a "multiplier" which awards affiliates proportionately based on the total number of performances of each song by broadcast licensees. The publisher and writer are each guaranteed a minimum amount ( e.g., a penny per play) for a single performance of the song on broadcast media. In order to determine whether and how frequently a song is played, BMI relies on various sampling methods depending on the type of broadcast media. The television networks submit logs containing a listing of all the music that the network uses. For local TV stations, a computer listing of the 90 weekly editions of TV guide is matched with cue sheets for particular programs and with a computer listing of all BMI music. For radio stations, BMI engages in a random sampling of 100 stations a week, each station submitting a log of the week's songs by title and writer for identification. Neither ASCAP nor BMI monitors performances of music in GLAs or other non-broadcast establishments for purposes of setting payment rates for affiliates.
The events leading up to this litigation are not atypical of the experiences of other GLA licensees who testified at trial. In January 1977, Mr. Moor, the owner of the Triple Nickel, received a form letter from BMI informing him of his obligation under the copyright statute and the necessity for obtaining a BMI license. The Triple Nickel was also visited by a BMI representative; Mr. Moor refused to enter into the BMI blanket license. After a second visit by the BMI representative, during which the performance of live songs at the Triple Nickel were logged, and two more letters, BMI initiated this copyright infringement action against the Triple Nickel in August 1977. In 1978, counsel for BMI and the Triple Nickel engaged in correspondence concerning the form of license. BMI offered a per piece license to Mr. Moor if he could identify the songs he wanted to use; Mr. Moor requested a list of all BMI compositions; BMI explained such a list was not available since BMI had a constantly changing repertory of one million songs (over 40,000 new songs being added annually), but invited Mr. Moor to New York to search the BMI library himself. This was the end of negotiations between the parties.
II. CHARACTERISTICS OF THE MARKET.
The relevant market in this case is the licensing of musical performing rights to GLA licensees. An examination of the characteristics of this unique market is essential to an evaluation of the Triple Nickel's antitrust and copyright misuse claims. Although the parties dispute the significance of some market characteristics, they are in basic agreement about many features of the market, including the applicability of the economic concepts of natural monopoly and public goods.
Both parties' experts agreed that this market has natural monopoly characteristics. Because there are thousands of individual copyright "sellers" seeking to deal with thousands of GLA buyers, the potential transaction costs are very high. Economies of scale exist as sellers band together to spread transaction costs of identical transactions over a larger group. Thus, some pooling of copyrights by individual copyright holders is a necessity in order to take advantage of the natural monopoly characteristics of the market.
In addition, both parties' experts agree that the goods in this market — the performing rights to the musical compositions — have the characteristics of "public goods". Public goods have two salient characteristics which operate in this market. First, unlike private goods ( e.g., apples), one can use a public good without leaving any less for others to consume. Once a musical composition is created, the marginal cost of additional consumption is zero. The second characteristic of public goods is that it is difficult to exclude persons who do not pay from using the good. The owners of private goods can withhold their goods from the market and release them only in return for payment; but, once a composer's song becomes known, he or she finds it difficult to prevent that good from being "stolen" by users. The enforcement problem resulting from this public good characteristic manifests itself in the GLA market through users who don't pay any licensing fee. During the course of this litigation, this has been labeled the "free rider" problem.
See Cirace, CBS v. ASCAP: An Economic Analysis of a Political Problem, 47 Fordham L.Rev. 277, 282 (1978).
Because the high transaction costs derived from natural monopoly characteristics are increased by the public good enforcement problem, very large performing rights organizations, like BMI and ASCAP, in which individual copyright holders pool their rights are necessary to achieve efficiency. The larger the organization, the more efficient it will be in reducing transaction costs; indeed, Triple Nickel's expert, Dr. Cirace, advocated one combined licensing operation as the most efficient means of operating in this market. The necessity for these large licensing organizations makes competition in the sense of many sellers competing against each other in the GLA market unrealistic.
The parties are also in agreement that the nature of the GLA market makes some kind of blanket license a necessity. As the Supreme Court observed in CBS IV:
Individual sales transactions in this industry are quite expensive, as would be individual monitoring and enforcement, especially in light of the resources of single composers. Indeed, as both the Court of Appeals and CBS recognize the costs are prohibitive for licenses with individual radio stations, nightclubs, and restaurants, . . . and it was in that milieu that the blanket license arose.CBS IV, 441 U.S. at 20, 99 S.Ct. at 1562.
Moreover, the Supreme Court's recognition in the CBS case, Id. at 20, 99 S.Ct. at 1562, that most users want "unplanned, rapid and indemnified access" to a wide range of compositions, is particularly apt in the GLA market. Testimony at trial made clear that GLA users typically do not know in advance what compositions will be performed nightly in their establishments and yet want the right to perform them instantaneously. The blanket license provides instantaneous access to any composition desired.
A corollary of the conclusion that the blanket license is a necessity in the GLA market is that the alternatives required by BMI's consent decree of direct licensing with individual copyright owners or of per piece licensing are unfeasible in this market. Again, the parties seem to agree on this point. Unlike the situation in CBS where large networks were interested in a relatively small number of compositions known in advance of performance, GLAs like the Triple Nickel are small establishments which lack the resources or the advance notice to contact copyright owners individually on a large scale. Likewise, because GLA owners rarely know in advance of performance the songs a band intends to play and because GLA bands often take audience requests, a prospective per piece license is unrealistic.
Finally, although the parties disagree over appropriate methods of pricing, they seem to agree that the natural market forces of supply and demand do not operate normally on pricing in this market. Because of the public good characteristic that the marginal cost of using a musical composition is zero, normal cost-based pricing is not feasible. The parties seem to agree that some form of pricing based on benefit conferred is appropriate. But, since as a practical matter a GLA needs a license from both ASCAP and BMI, the normal constraint on benefit pricing — alternative supply — does not operate in this market.
Thus, although large performing rights organizations are a necessity in this market, the result in the current market is that BMI can exercise substantial monopoly power over price. This monopoly power of the seller is particularly strong in a negotiating situation where there is no corresponding power on the buyer's side. Unlike the television network market where buyers like CBS exercise some monopsony power of their own, the buyers in the GLA market are weak and diffuse.
While normal competitive forces do not operate in this market, it is not true that BMI's price for its GLA license is unconstrained. Testimony at trial convinced me that the free rider problem does provide a significant constraint on the price BMI charges. The higher the price it charges, the greater the resistance of GLA users is likely to be, and, conversely, the lower the price, the lower the resistance will be. Since the free rider problem tends to make BMI's enforcement costs high and can, indeed, cause increased costs to more than consume increased revenue from a higher price, BMI considers this problem when setting a price.
III. TRIPLE NICKEL'S CONTENTIONS.
The Triple Nickel's primary challenge is to the current licensing agreement BMI offers to GLAs. It maintains that the offering of that license without providing a realistic alternative constitutes an illegal tie-in which is a per se violation of Section 1 of the Sherman Act. But regardless of the disposition of the tie-in claim, the Triple Nickel attacks the combination of full repertory blanket license and fee tied to entertainment expense as a general restraint of trade illegal under Section 1. It further argues that BMI's imposition on GLAs of this blanket license involving pooled copyrights priced in such a way that it includes other factors besides the amount of BMI music used constitutes copyright misuse.
In addition, the Triple Nickel alleges that BMI has engaged in two conspiracies in restraint of trade in violation of Section One. The first alleged conspiracy is a scheme between BMI and ASCAP to stabilize prices in the GLA market. The second alleged conspiracy is between BMI and its affiliates to institute "factually inadequate" lawsuits against GLAs.
Finally, Triple Nickel alleges that all of BMI's practices attacked under Section One also violate Section Two either as monopolization, attempted monopolization, or conspiracy to monopolize.
IV. THE PRIMARY SECTION ONE CLAIM. A. Per Se Or Rule Of Reason Analysis?
The Triple Nickel argues that BMI's current licensing practices coerce it into purchasing music which it does not wish to perform in order to secure access to music which it does wish to perform. Because of this "tie-in", the Triple Nickel maintains that this case is an appropriate one for per se rather than Rule of Reason analysis. As I indicated in my prior opinion, however, this issue was resolved against the Triple Nickel in CBS IV.
In CBS IV, the Supreme Court held that ASCAP's blanket license required Rule of Reason analysis and each of the reasons given for that conclusion support the utilization of that approach in this case. Here, as there, the market is sufficiently unusual and the Court's experience with the challenged practice sufficiently sparse that one cannot confidently characterize the defendants' conduct as "`plainly anticompetitive' and very likely without `redeeming virtue.'" 441 U.S. at 9, 99 S.Ct. at 1557.
Accordingly, I hold that Rule of Reason analysis is appropriate in this case and turn to the evidence in order to assess the actual and potential adverse impacts of BMI's practices on the relevant market, the positive contributions of those practices in that market, and the effects of the alternative practices which Triple Nickel believes would involve less restraint on competition. See generally American Motor Inns, Inc. v. Holiday Inns, Inc., 521 F.2d 1230, 1246-47 (1975).
B. Restraints Of Trade
The first and most obvious restraint of trade which arises from BMI's current licensing practices is the one focused on in the CBS case, the elimination of price competition among those whose compositions are in its pool of music. There are additional anticompetitive effects, however, of the kind reflected in the leading "block booking" cases upon which the Triple Nickel so heavily relies.
This package selling arrangement results in the sellers selling more "product" than they would in a competitive market. In a competitive market, some sellers would sell more songs than others depending upon the popularity of their compositions. In this market, however, sellers pool their songs and sell them on an all-or-nothing basis. By thus banding together, rather than competing against each other, the sellers are able to lever the selling power of one copyright to the selling power of others and thus enlarge the monopoly of all. As a result, less popular sellers sell more than they otherwise would. As the Supreme Court has observed in a motion picture context:
Where a high quality film greatly desired is licensed only if an inferior one is taken, the latter borrows quality from the former and strengthens its monopoly by drawing on the other. The practice tends to equalize rather than differentiate the reward for the individual copyrights.United States v. Paramount Pictures, Inc., 334 U.S. 131, 158, 68 S.Ct. 915, 930, 92 L.Ed. 1260 (1948).
In addition to rewarding inferior copyrights, all copyrights will tend to achieve greater sales by package selling than they would without such an arrangement. Even the most popular songs are not wanted by all GLAs. Thus, any songwriter is likely to increase his sales by combining his songs in a package with all other songs, since some buyers who would not otherwise purchase his songs must now do so in order to get the songs that they do desire.
In addition to the anticompetitive effects on individual copyright sellers, there are corresponding effects on the copyright buyers. BMI's current licensing practices deprive GLA purchasers of the control which they would have in a competitive market over the amount of business they conduct with a particular source. In a competitive market, for example, a GLA would have the ability to reduce its purchases from a particular source in response to a price rise. Where the only feasible access to BMI's music is through an all-or-nothing license which bases the price paid on factors other than the quantity of BMI music performed, however, a GLA is unable to reduce its total obligation to BMI through control of the quantity of BMI music performed.
Finally, the all-or-nothing BMI license may erect barriers to entry to potential competition from other musical performing rights organizations. GLA proprietors testified that the combined costs of BMI and ASCAP all-or-nothing licenses were significant for them, and that the arrival of a third licensing organization, SESAC, increased their resistance to licensing proposals further. When the only option the seller offers is all-or-nothing, the buyer presumably wants to deal with as few sellers as possible. In such a situation, because the price the buyer is charged is not related to the quantity used, even if its total quantity of music used remains the same, it may face a higher total price with each additional seller, rather than a reallocation of total price among different sellers. One would expect that the burden of this increased resistance to licensing would not be visited equally on all sellers in the market. The buyers have a greater incentive to resist a new and smaller organization who controls a smaller percentage of the compositions desired and whose willingness and ability to enforce its rights through expensive litigation has not been as forcefully demonstrated in the marketplace. Since the willingness of most GLA's to take a license without litigation is crucial to success in this market, the increased resistance engendered by the absence of a relationship between price and quantity may constitute a barrier to entry. The significance of that barrier is impossible to assess, however, on the basis of the record in this case.
C. Justifications
BMI asserts, and the Triple Nickel does not dispute, that BMI's blanket license has redeeming virtues. As earlier noted, the parties agree that performing rights societies and their blanket licenses reduce transaction costs which would otherwise be prohibitive. BMI's blanket license thus has a pro-competitive effect in the sense that there would be no market if individual GLAs were left to negotiate with individual copyright holders.
Moreover, Triple Nickel acknowledges that some form of blanket license is necessary to meet the needs of GLAs. The "unplanned, rapid and indemnified access" which GLAs need can only be afforded by a form of licensing which does not require the identification of particular musical compositions in advance of their performance.
BMI justifies the "full repertory" blanket license both on the ground that it is the best means of meeting the needs of virtually all GLA buyers and on the ground that a less than full repertory license system, if feasible at all, would be significantly more expensive to administer than a full repertory one. Since GLA users tend to use a large number and wide variety of musical compositions, their needs can generally be most completely served with access to as large a repertory as possible. Because the marginal cost of additional consumption is zero, it is no less expensive for BMI to provide access to a more limited repertory of songs than to its full repertory of a million songs. Indeed, as discussed hereafter, a limited repertory license would undoubtedly be more expensive for BMI to provide because, among other things, of the additional cost of monitoring the holders of such licenses to ascertain if their use of BMI compositions exceeds the scope of their licenses.
Finally, BMI justifies pricing based on the GLA user's total entertainment expense as a convenient, inexpensive and reliable way of providing a rough measure of the benefit conferred by the music rights licensed. The only information necessary to administer the present pricing system is a GLA's annual figures for its entertainment expenses. Since these figures must be kept for tax purposes in any event, there is virtually no cost involved in gathering this data and it is likely to be reliable.
D. The Triple Nickel's Alternatives.
Triple Nickel acknowledges that there is no feasible way to create and maintain price competition between copyright holders in the GLA market. It asserts, however, that there are alternative forms of licensing which BMI could offer in addition to its current full repertory license which would involve less restraint on the GLA's freedom of choice and on the entry of additional performing right societies in the market. The Triple Nickel's first suggestion is that BMI offer a less-than-full repertory, or "mini" blanket, license as an alternative to its full repertory license and its per piece license. Second, the Triple Nickel urges that a license be offered that bases cost on a retroactive determination of the quantity of BMI music played each year and a "proxy", per performance price.
1. The "Mini" Blanket License.
The alternative form of license that was the primary focus of Triple Nickel's trial testimony and post-trial briefing was a limited repertory license based on a category of music, like "country and western". It also proposed a limited repertory license based on a family of music, like country, bluegrass, folk and gospel. In either event, the suggestion is that the price would bear the same relationship to the price for a full repertory license which the BMI compositions in the covered categories bears to the compositions in BMI's total repertory. For pricing purposes, BMI would categorize a limited sample of its songs in order to determine what percentage of BMI music fell into each field of music and this category percentage would be periodically adjusted through sampling of the new songs that BMI constantly adds to its repertory.
The first problem with the Triple Nickel's suggestion is that the "mini" blanket licenses which it proposes would not meet its own needs. The music performed at the Triple Nickel includes compositions that have gained popularity in a variety of performance styles. Mr. William Ivey, Director of the Country Music Foundation and an expert in the identification of the styles of musical performance, testified that only about two-thirds of the music performed at the Triple Nickel had a country and western performance history, and, therefore, even if it were possible to categorize music based on the primary style in which such music has historically been performed, approximately one-third of the compositions performed at the Triple Nickel would fall outside the country and western category. Indeed, the Triple Nickel admits that the music it plays includes rock and roll, folk, bluegrass, and gospel, as well as country and western. It has suggested no "mini" blanket license which would authorize it to play all that it is in fact using.
More fundamentally, however, it is simply not feasible to categorize music for licensing purposes into such performance style labels. As Mr. Ivey explained, the method of categorizing musical compositions proffered by the Triple Nickel is in reality a categorization not of the music but the style in which the music is performed. Since there is insufficient information residing in the composition itself (that is, the score and lyrics) to determine the performance style, to categorize a song based on score and lyrics alone would be arbitrary. Moreover, any such categorization based on a study of performance history would be subject to disagreement. Frequently a musician will take a song historically popular in one style and perform it in a different style. And it is also common for a song which has gained popularity in one performance style to subsequently gain popularity in another style. For these reasons, a "mini" blanket licensing system would create intolerable uncertainty in the marketplace.
In addition, even if it were assumed for purposes of argument that there are GLA users whose licensing needs would be satisfied by a limited category license, obviously no one would opt for that form of license if the fee for it equalled or exceeded the fee for the full blanket license. Perhaps in recognition of this fact, Triple Nickel suggests that the Court decree a proportionately reduced price for "mini" blanket licenses. It tenders no persuasive rationale for doing so, however. The benefit derived from the use of BMI music will not vary materially depending upon whether a user is performing twenty different BMI songs from different categories each night or twenty songs from the same category. Moreover, the "mini" blanket license would necessarily cost more to administer than the full repertory license because the "mini" license would involve categorization costs, additional policing costs, and the costs of resolving, and of attempting to avoid, disputes over the scope of the license. For these reasons, there is no basis for an injunction requiring a proportionately reduced fee for a "mini" license. As the experts acknowledged, however, in the absence of such a decree, there is no reason to believe that BMI's charge for a "mini" blanket license would be less than its charge for a full repertory one.
In examining the alternatives that an antitrust claimant has proposed, the Third Circuit has described the comparative standard normally applied under rule of reason analysis as follows:
[T]he test is not whether the defendant deployed the least restrictive alternative. Rather the issue is whether the restriction actually implemented is `fairly necessary' in the circumstances of the particular case, or whether the restriction `exceeds the outer limits of restraint reasonably necessary to protect the defendant.'American Motor Inns, supra, 521 F.2d at 1248-49. The record in this case convincingly demonstrates that the full repertory blanket license system is "fairly necessary" to serve the relevant market and that "mini" blanket licenses are not a practical alternative.
2. Alternative Pricing.
The primary problem which the Triple Nickel has with the current practices of BMI is not with the form of its licenses but rather with the prices that it charges. It proposes alternative pricing schemes with features that are purportedly designed to overcome several restraints which it attributes to BMI's current pricing arrangement. First, the Triple Nickel objects to the current practice because it is not related to the amount of BMI music actually performed by the licensee. It proposes a use-based pricing approach. Second, the Triple Nickel argues that a price based on total entertainment expense illegally extends BMI's monopoly power because it charges for factors other than the actual benefit the GLA user derives from the blanket license. It proposes a pricing formula which utilizes the number of performances of BMI music, the number of listeners per performance, and a per performance or "per use" price. Finally, the Triple Nickel seeks to solve the problem caused by the absence of competitive restraints on BMI's pricing by requiring it to charge a "proxy", per use price.
In place of the current licensing agreement which fails to account for actual use, the Triple Nickel proposes that the price of a full repertory blanket license be based on the amount of BMI music actually performed, measured after the fact. It concedes that measuring use daily throughout the year would be impractical, so it has suggested a sampling system under which songs used at a GLA establishment would be logged by the licensee over a short period of time. Dr. Lamb, a computer and statistics expert, suggested that a sample of 150 songs would be sufficiently reliable. BMI would then use its computer to determine what percentage of the GLA's total song consumption belongs to BMI. This percentage and an estimate of the GLA's total performances during the year would provide the number of BMI performances for which it would be billed.
The first problem with this proposal for a per use price is a conceptual one. It fails to account for a significant part of the benefit provided by the BMI license. The inability of GLAs to identify in advance the compositions which will be played means that they need the right to perform all BMI music which the musician may ultimately select. Thus, with the full repertory blanket license the GLA purchases more than the right to play the BMI compositions which are actually performed; he also purchases immediate and unlimited access to BMI's entire repertory. As the Supreme Court observed in CBS IV, this makes the blanket license akin to a single product.
The major difficulty with this proposed per use licensing feature, however, is a practical one, the increased costs associated with the sampling. Initially, there would be the expense of educating GLAs with respect to the necessity of taking the sample and the proper method of performing that task. Regardless of the effectiveness of this educational process, it is to be expected that some will fail to sample and others will sample improperly. Following up to secure a suitable sample from these GLAs will also involve expense. Even with respect to those GLAs who submit what appear to be appropriate samples there would have to be some system for at least spot checking their accuracy. Finally there would be the expense of determining which compositions listed on the logs are in the BMI repertory. While computers would materially aid this process, incomplete information on the logs would require human attention. Many songs have the same title, for example, and whether BMI is the copyright holder can often be determined only by reference to the composer's name. The expense associated with each of these operations would be additional expense since BMI currently has no need to sample or monitor the music performed by its GLA licensees. No attempt has been made to estimate the additional cost of a per use system but one can say with confidence that it would be substantial.
In addition to the disadvantage of increased costs, I am unable to say on this record that institution of per use pricing would materially effect the restraints which have been identified in the market. Such a system would not provide a constraint on BMI's pricing by giving a GLA control over the quantity of BMI music used and paid for. GLAs do not know which compositions belong to which organization and providing complete lists of BMI's repertory to educate them would require the equivalent of a two thousand page phonebook, with frequent periodic supplements to keep it current. And, even if the GLA owner were motivated enough to purchase and work with such a cumbersome publication, it is the musician, and not the GLA owner, who actually decides which compositions will be performed.
It is, of course, true that under a per use pricing scheme a GLA could reduce its total consumption of music in order to reduce the amount paid during the period covered by the license. But so long as the unit price in each new license is controlled by the performing rights societies, nothing in this record demonstrates that the per use pricing system by itself would serve as a constraint on BMI's power over price.
It is only in the area of the potential barrier to entry that I perceive the possibility of a beneficial effect from a per use pricing approach. Under such a system, if and when a new performing rights society attempts to enter this market, presumably by gaining rights to compositions which would otherwise belong to existing organizations, its licensing fees would be reflected in correspondingly reduced charges by the existing organizations and, as noted above, one would expect this fact to lessen GLA resistance to that entry. As discussed in more detail in Section VII, however, one cannot tell from this record the significance of that barrier or, indeed, whether such an entry would be feasible under either pricing scheme.
Given the additional cost of per use licensing and the fact that the only potential benefit from its adoption is a highly speculative one, I decline to require its adoption by BMI.
In combination with per use pricing, the Triple Nickel asks the Court to impose a "proxy price" to be charged for each performance of a BMI composition. This request brings us to the heart of this matter. The core problem here is, of course, that normal competition does not and cannot operate in this market to establish the price which GLAs will pay for the right to perform music. The Triple Nickel insists that this means the Court must step in to set an appropriate price. I decline to do so for several reasons.
First, the Triple Nickel has not suggested an adequate rationale for adopting any of its suggested means of determining that price. While price regulation pursuant to legislative directive may well be the ultimate answer to the absence of price competition in this market, the Triple Nickel's terse suggestion in its brief that the GLA price be keyed to the $100 "blanket price" for the performance of recorded music in bars and clubs suggested by House Resolution 1805 was not supported by any evidence or rationale. The other proposal suggested by Triple Nickel in its post-trial briefing is that the Court look to the price charged by BMI to WAMS, a local Wilmington A.M. radio station, for the right to perform BMI music. The proposal does have the virtue that this proxy price is one set in a market where there is negotiating power on both sides. But those negotiations occur in a materially different market and the Triple Nickel has not even attempted to articulate an explanation of why those negotiations would provide a rational basis for pricing in the GLA market. Indeed, the market most like the GLA market in which prices are nevertheless set by negotiations between parties having market power is the hotel-motel owner's market. BMI negotiates rates in this market with a national trade association. The Triple Nickel does not rely on those rates, however, because they would appear to result in a higher price than it is now paying.
Triple Nickel apparently favors the WAMS approach because it is said to be a better measure of the benefit conferred by the use of a BMI copyright than BMI's current rate based on total entertainment expense. Total entertainment expense is said to "add-on factors", like the popularity of the musician playing the song, that are not directly derived from the product, i.e., the musical composition provided by BMI. The WAMS proxy price formula, on the other hand, is said to be geared to the number of listeners enjoying the music. I am not convinced, however, that requiring BMI's price to be based on audience size will produce a significantly more competitive price or alleviate any of the restraints on commerce which have been identified as inherent in BMI's current licensing practices.
But there is a more fundamental problem with the "proxy price" approach. This is not a case where the Court is being asked to change market practices so that competition will function more effectively in the future. Rather, the Court is being asked to make up for the absence of competition in the relevant market by tying the prices in that market to prices in a different market where competition does exist. Such action would, I believe, be both unprecedented and unwise. Markets are unique as well as continually changing. Frankly, I am not quite sure how one would go about finding a market which would supply a price approximating the price competition would currently produce in another market. But if one were fortunate enough to find one, it would be unrealistic to assume that the markets involved would maintain their current relationship for very long.
There are two other possible remedies for the unique problem which the Triple Nickel here brings to the Court: concentration of market power on the buyer's side of the market of the kind which has occurred in the hotel and motel market, or some form of continuing regulation which will fix and modify prices in light of the present and future conditions in this market. The first remedy I am obviously unable to prescribe. The second is one that this Court is not suited to administer. If Congress sees fit to occupy this field, I am confident it can make a more rational assignment of the task. As Judge Kaufman recently observed in a Section Two context, "[J]udicial oversight of pricing policies would place the courts in a role akin to that of a public regulatory commission. We would be wise to decline that function unless Congress clearly bestows it upon us." Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263, 294 (2d Cir. 1979).
V. THE COPYRIGHT MISUSE CLAIM.
Regardless of whether various aspects of BMI's current pricing system constitute restraints of trade objectionable under the antitrust laws, they are also subject to scrutiny under the copyright misuse doctrine. I conclude, however, that the Triple Nickel's case fares no better under copyright misuse analysis.The Triple Nickel maintains that BMI's practice of basing its license fee on entertainment expense unlawfully extends the scope of its lawful copyright monopolies. As already noted, it objects to entertainment expense because it contains "add-on" factors; it includes the popularity of the musician and it reflects not only the contribution of the music in BMI's pool, but that of all music performed. In support of its argument it relies primarily upon Zenith Radio Corp. v. Hazeltine Research, Inc., 395 U.S. 100, 89 S.Ct. 1562, 23 L.Ed.2d 129 (1969).
In the Zenith case, Zenith asserted a patent misuse defense against an infringement claim. The contested licensing agreement allowed Zenith to use all of Hazeltine's radio and television patents for a fee based on a percentage of Zenith's total sales. The court concluded that such an arrangement might constitute patent misuse, distinguishing a similar earlier case, Automatic Radio Mfg. Co. v. Hazeltine Research, 339 U.S. 827, 70 S.Ct. 894, 94 L.Ed. 1312 (1950), in which it had found nothing "inherent" in a patent arrangement based on a percentage of total sales which illegally extended the monopoly conferred by the patent. On remand, the District Court in Zenith was to be guided by the Court's statement that "If convenience of the parties rather than patent power dictates the total sales royalty provision, there is no misuse of the patents and no forbidden conditions attached to the license." Zenith, 395 U.S. at 139, 89 S.Ct. at 1585.
Under the Zenith approach, the relevant issue here is whether the feature of BMI's licensing which bases the fee on a percentage of entertainment expense rather than on a per use price times the number of BMI songs played is a product of the convenience of the parties or of a BMI effort to extend its monopolies beyond their legal boundaries. The evidence on this issue weighs in BMI's favor. The record establishes that the percentage of entertainment expense approach is a simple and convenient one for both parties because it utilizes reliable data collected and maintained for other purposes. It further establishes that a pricing system based on a per use price times the number of BMI songs played would require the gathering of additional information which would be substantially more costly while substantially more susceptible to manipulation. Further, we know that in another market in which BMI customers have substantial bargaining power, the hotel-motel market, the negotiated licenses provide for fees based on entertainment expense. Finally, the record fails to persuade me that as a practical matter BMI's current approach enables it to abuse its monopoly power more easily than would be the case if it determined a per unit price and charged on the basis of BMI plays. Accordingly, I perceive no motive for BMI coercing use of the former in lieu of the latter. Under these circumstances, it is more likely than not that the percentage of entertainment expense approach arose and exists as a matter of the convenience of the parties.
VI. THE SECTION ONE CONSPIRACY CLAIMS.
The Triple Nickel also asserts that BMI has engaged in two conspiracies, apparently arguing that both alleged conspiracies constitute restraints of trade violating Section One of the Sherman Act. From the evidence presented at trial, I find insufficient basis to conclude that either conspiracy exists.
First it is alleged that BMI and ASCAP have agreed to stabilize prices and preclude competition in the GLA market. Proof of an express agreement is not required to find conspiracy under the Sherman Act. Interstate Circuit v. United States, 306 U.S. 208, 59 S.Ct. 467, 83 L.Ed. 610 (1934); Sweeney Sons, Inc. v. Texaco, 637 F.2d 105, 111 (3d Cir. 1980). Consciously parallel business behavior combined with other circumstantial evidence may support an inference of an agreement. Schoenkopf v. Brown Williamson Tobacco Corp., 637 F.2d 205 (3d Cir. 1980). To prove consciously parallel behavior, however, an antitrust claimant must show not only that the alleged co-conspirators' behavior was parallel and that they were aware of each other practices, but also that their awareness was an element in their decisional process. Schoenkopf, 637 F.2d at 208. The Triple Nickel alleges that BMI and ASCAP engage in and are aware of similar practices such as offering a full repertory blanket license which does not depend on music used and failing to log GLA music for purposes of distributing royalties to affiliates. Assuming awareness of these and other similar practices, there is insufficient evidence to persuade me that this awareness affected BMI's decisionmaking process.
Even if consciously parallel behavior had been established, however, in order to prove an antitrust violation the Triple Nickel would still have to prove that (1) BMI acted in a manner which, in the absence of a conspiracy, would be contrary to its economic interests and (2) that BMI had a motive to enter such an agreement. Schoenkopf, 637 F.2d at 208. Yet, there is a legitimate business justification for every BMI action the Triple Nickel attacks. For example, although BMI currently may be losing money in its GLA operation, BMI's explanation of the "snowball effect" reason for licensing GLAs at a loss is at least as plausible as any anticompetitive purpose. And BMI letters to GLA users explaining that they are also likely to require an ASCAP license are a legitimate response to inquiries whether a BMI license alone is all that is needed. Further, I find no record support for the suggestion that the alleged agreement to preclude competition in the GLA market is motivated by common interest in permitting resources to be concentrated in the more lucrative broadcast licensing market.
The second alleged conspiracy is an illegal agreement between BMI and its affiliates to initiate "factually inadequate" copyright infringement actions in order to coerce GLAs into accepting blanket licenses. This "conspiracy" claim is premised upon the allegation that BMI brings infringement suits based upon insufficient evidence. Not only has the Triple Nickel failed to prove that BMI engages in such a practice, but there is also no evidence in this record from which a trier of fact could infer an agreement, tacit or otherwise, between BMI and its affiliates pertaining to the circumstances under which copyright infringement will be initiated.
VII. SECTION TWO CLAIMS.
The Triple Nickel also challenges as violations of Section Two of the Sherman Act all of BMI's practices discussed above in the context of Section One and copyright misuse. In order to establish monopolization in violation of Section Two, an antitrust claimant must establish not only monopoly power, but also conduct by the monopolist which prevents or impedes competition. See generally Berkey Photo, Inc. v. Eastman Kodak Co., 603 F.2d 263, 272-275. The parties did not directly address whether BMI's percentage share of the performing rights market established the requisite monopoly power. It nevertheless seems clear that BMI's position as one of two dominant sellers in a market where buyers must deal with both sellers gives it the requisite economic power — i.e., the power to control price — to constitute monopoly power under Section Two. I conclude, however, that the Triple Nickel has failed to prove that any of the challenged BMI practices were either motivated by a purpose of, or had the effect of, impeding potentially competing performing rights organizations. Therefore, Triple Nickel's monopolization claim must fail.
The suggestion that BMI's method of pricing its GLA license may erect barriers to entry by competing performing rights organizations because it is not related to the quantity of BMI music used has already been mentioned in the Section One discussion. As noted in that discussion, however, there is no economic evidence in this record to assess the potential significance of such a barrier if it exists.
Moreover, not only was there an absence of economic evidence of anticompetitive effects of BMI's license practices on potentially competing organizations, but the thrust of the economic testimony that was presented was that this market required very large sellers. The economic experts seemed to agree that because of the unique transaction costs in this market, organizations the size of BMI were a necessity. Indeed, as noted earlier, the Triple Nickel's expert, Dr. Cirace, urged that the only rational structure for this industry would be one involving a single licensing operation. Given the inherent barriers to entry by other organizations that this testimony suggests, it would be inappropriate to conclude that the speculative barriers created by non-used-based pricing or any other feature of BMI's license constitute anticompetitive effects.
The alleged conspiracy between BMI and ASCAP discussed above as a possible Section One violation in restraint of trade might also constitute a Section Two conspiracy to monopolize. Even if there was sufficient evidence of the existence of such a conspiracy (which, as discussed above, there is not), a Section Two conspiracy would also require a specific intent to monopolize. See American Tobacco Co. v. United States, 328 U.S. 781, 809, 66 S.Ct. 1125, 1138, 90 L.Ed. 1575 (1946). I have found no evidence in this record of a specific intent to monopolize by BMI. Finally, even if any of the acts alleged to be part of the Section One conspiracies are viewed as unilateral action engaged in by BMI, there is no basis here for conclusion that BMI has abused its monopoly power. With respect to all of these alleged acts, I have concluded either that there is an absence of anticompetitive effect and purpose (e.g., collecting license fees at all in a GLA market where current costs slightly exceed revenues); or that there is insufficient evidence to conclude that the practice exists (e.g., "factually inadequate" lawsuits).
VIII. CONCLUSION.
In defending against a copyright infringement action, the Triple Nickel has contended that a number of licensing practices of BMI in the GLA market — primarily its use of a full repertory blanket license in combination with a price based on total entertainment expenses — are illegal. It has challenged these practices as violations of Sections One and Two of the Sherman Act and as copyright misuse. I have concluded that it has failed to prove any of these claims. Therefore, judgment will be entered in favor of BMI on its copyright infringement claim and against the Triple Nickel on its antitrust counterclaims.
SUBMIT ORDER.
SCHEDULE "A" — RATE OF ANNUAL FEES [*] BRACKET OF ANNUAL ANNUAL ENTERTAINMENT COSTS LICENSE FEE _____________________ ___________ Less than $ 5,000.00 $ 75.00 5,000.00 to 7,999.99 90.00 8,000.00 to 9,999.99 100.00 10,000.00 to 14,999.99 120.00 15,000.00 to 24,999.99 180.00 25,000.00 to 34,999.99 240.00 35,000.00 to 49,999.99 300.00 50,000.00 to 64,999.99 325.00 65,000.00 to 79,999.99 400.00 80,000.00 to 99,999.99 500.00 100,000.00 to 119,999.99 600.00 $120,000.00 to $139,999.99 $ 700.00 140,000.00 to 159,999.99 800.00 160,000.00 to 179,999.99 900.00 180,000.00 to 199,999.99 1,000.00 200,000.00 to 249,999.99 1,250.00 250,000.00 to 299,999.99 1,300.00 300,000.00 to 349,999.99 1,400.00 350,000.00 to 399,999.99 1,500.00 400,000.00 to 449,999.99 1,600.00 450,000.00 and over 1,700.00 [*] "Entertainment Costs" shall mean all expenditures of every kind and nature (whether in money or any other form of consideration) made by LICENSEE, or on LICENSEE'S behalf, for the services of musicians and all other entertainers actually present and performing at LICENSEE'S premises. The services of LICENSEE as a musician or entertainer shall be included in "Entertainment Costs" at the prevailing rate for such services in the community. Included in "Entertainment Costs" is the agreed value of room and board where LICENSEE is obligated to provide accommodations to musicians and entertainers as part of the consideration for such entertainment services. The agreed value of room and board shall be deemed to be one-half of the prevailing rate charged to guests for similar accommodations.See Cirace, CBS v. ASCAP: An Economic Analysis of a Political Problem, 47 Fordham L.Rev. 277, 282 (1978).