Mr. Jonathan Potter

May 15, 2002

Mr. Chairman, Senator Hatch, and Members of the Committee:

I am Jonathan Potter, Executive Director of the Digital Media Association ("DiMA"), which represents the interests of the online media industry. On behalf of nearly 40 DiMA member companies, including more than 20 companies that are webcasters or support webcasters, I sincerely thank you for inviting me to testify today concerning the future of the nascent Internet webcasting industry -- an industry imperiled by the February 20, 2002 sound recording royalty recommendation of a Copyright Arbitration Royalty Panel (hereinafter "CARP Report").

This hearing could not be more timely or important for the webcasting industry. In less than a week, the Librarian of Congress will decide whether the CARP Report will be affirmed or modified. It is no exaggeration, Mr. Chairman, to state that this decision holds in the balance the future of all of today's Internet webcasters and the nature of webcasting itself.

For webcasters large and small, large media companies or local entrepreneurs, the CARP-recommended royalty is astronomically high, and bears absolutely no rational relationship to traditional copyright licensing benchmarks - rates set in the marketplace or by ratemaking bodies for analogous rights, or the costs of the business that is using the specific rights at issue. In a factual vacuum the CARP-recommended royalty rates may seem tiny - $.0014 per song per listener for Internet only webcasters, $.0007 for broadcast radio retransmissions. But in context, the following examples illustrate that the recommended rates defy any reasonable assessment of webcasting or music economics:

? For the year 2001, Beethoven.com, a leading classical music webcaster based in Connecticut, would owe $48,720 in royalties, when its total revenue was only $33,500;
? Without accounting for any growth, Internet Radio, Inc., which operates ChoiceRadio.com and alldanzradio.com, anticipates 2002 sound recording royalties of $518,783, against gross revenue of about $100,000 and annual overhead of $240,000;
? For the six months ending March 31, 2002, Live365 of Foster City, California would owe $1,024,421 in sound recording royalties, against advertising revenues of $113,782 and gross revenues of $351,032.

And the fundamental unfairness goes well beyond independent webcasters:

? Radio Free Virgin, the Internet radio division of the Virgin Group (which includes the prominent Virgin Megastore record chain), had April 2002 revenue of $23,000, and combined license fees to ASCAP, BMI and SESAC of less than $1,000 -- but a CARP-proposed sound recording royalty rate of $67,000.

For webcasters of every size -- no matter the number of performances, or the size of the media company -- the CARP's webcasting royalty leads to an unsustainable economic result. Even for subsidiaries of large multimedia companies, the business model for webcasting must be capable of generating profits and standing on its own. The CARP's royalty rate ensures this will not be the case. The webcasters covered by the CARP Report are nonsubscription services that generate revenue streams primarily from advertising, to the extent that ad sales are achieved. Typical industry advertising rates simply cannot support the cost per listener, discussed below, that the CARP's recommended rate would require.

Let me put the CARP's proposed rates in further context. The CARP's recommended webcasting rate would require payments of between $12 and $15 per listener per month. (This results merely from multiplying: (i) the $.0014 per performance rate times (ii) 12-15 performances per hour, times (iii) 24 hours per day, times (iv) 30 days per month.)

Were these webcasting fees applied to today's broadcast radio industry -- which, of course, pays no performance royalties for performing sound recordings -- the broadcasters would be incurring in the aggregate, according to Arbitron, over $4 billion annually in sound recording royalty fees, correlating to more than 25 percent of their broadcast industry revenues. Compare this figure to the approximately $350-400 million the broadcasters pay for performances of the musical works embedded in the identical sound recordings under the combined 3-3.5 percent of revenue fees paid by them to ASCAP, BMI and SESAC.

Looking at the same facts from another direction, assume today's modest Internet webcasting listenership grows to equal the broadcast radio audience. At the CARP's recommended royalty rate, Internet webcasters would incur, as noted, over $4 billion in sound recording performance fees (per Arbitron). At in excess of $4 billion, the webcasters' total performance fees (for sound recordings and musical works) would stand in stark contrast to the $350-400 million in royalties paid by the broadcasters, against whom the Internet webcasters must compete.

Neither economics nor the law justify saddling Internet media companies with this impossible burden. This result cannot be what Congress had in mind when it created the statutory license. This result cannot be permitted to stand.

For thousands of independent webcasters, community broadcaster webcasters, and college webcasters, the reality is very simple: the CARP recommendation means the end of streaming. This is not a problem for only the small start-up, college or hobbyist webcaster, Mr. Chairman. Whether you are Yahoo! or Onion River Radio, whether you are a powerful network affiliate in New York City or WNCS in Montpelier, the CARP proposal is a rate that cannot be justified financially. Companies from the largest broadcasters to the smallest start-ups are likely to cease or significantly diminish webcasting if the CARP recommendation takes effect. Innovative entrepreneurial local tastemakers who were utilizing the new media to introduce new music and support local artists will cease to exist.

Mr. Chairman, no one - neither the creators nor the webcasters - benefits from artificially inflated rates that only a few webcasters can afford. Fewer webcasters means less diversity of music on the Internet, less airplay for up-and-coming artists or niche musical genres, and less royalty money for creators. A recent survey of customers of a major Internet music retailer highlighted the economic benefits of webcasting for recording artists, finding that more than two-thirds of respondents said that their music purchases over the last six months have been influenced by listening to Internet radio. Indeed, the importance of Internet radio for consumers and artists has been highlighted in newspapers and op-ed pages since the February 20 decision. Attached to my testimony are editorials from major newspapers and a small sampling of articles from dozens of newspapers and online media, from all across the country, recognizing that the inordinately high CARP-recommended royalty rates will stifle the promise of the webcasting medium and harm the interests of consumers and the vast majority of performing artists who receive little or no airplay on mainstream radio.

Let me be clear that DiMA and its member companies seek no subsidy here. We support fair compensation to copyright owners and direct payment to recording artists. If the royalty rates recommended by the CARP were similar to those paid by webcasters and broadcasters to songwriters and music publishers through the ASCAP, BMI and SESAC collectives, there would have been no outcry and this hearing would not be necessary. Indeed, had the recording industry agreed with DiMA during the voluntary negotiations that preceded the CARP to accept compensation equivalent to that paid to these songwriters, there would never have been a CARP at all. Unfortunately, there was no agreement among the industries; the CARP was necessary, extraordinarily expensive and frustrating; and the result - a royalty for record companies seven times higher than is paid to songwriters on a per-performance basis (and multiples more than that on a percent-of-revenue basis ) - defies economic logic and legal precedent.

The question for this Committee is: how did this result occur? If this Committee and the Congress intended the webcast sound recording statutory license to promote both the welfare of artists and the development of Internet media entrepreneurs, where did the process go awry, and what can be done now to reinvigorate both of Congress' goals?

Congress Intended that the CARP Should be a Safeguard Against Anticompetitive Conduct.

In the Digital Performance Right in Sound Recordings of 1995 ("DPRSRA"), when Congress enacted the first limited sound recording performance right in digital transmissions, Congress expressed its desire to protect copyright owners and artists, and to promote the development of new technologies. See, e.g., S. Rep. No. 104-128, 104th Cong. 1st Sess. at 15; 141 Congressional Record S.11945, 11961 (August 8, 1995) (remarks of Sen. Leahy).

When considering the DPRSRA, this Committee expressed justifiable concern that an unlimited exclusive performance right might unintentionally facilitate anticompetitive activity, such as the setting of monopoly prices for licensing music. After consulting with the Antitrust Division of the Department of Justice, this Committee, under the leadership of Senators Leahy, Hatch and Thurmond, incorporated into the DPRSRA several provisions that sought carefully to balance the need for efficient administration of licensing rights (which benefits licensors and licensees) against the need to protect against the formation of a recording industry licensing cartel that would have unlimited immunity from antitrust scrutiny. See Congressional Record S.11945, 11961-11963 (August 8, 1995) (remarks of Sen. Leahy).

The fabric of the DPRSRA wove together several essential threads of protection against potential anticompetitive licensing behavior. First, a statutory license so that eligible entities could not be denied blanket licenses to play music over new digital services. Second, a "most favored nations" provision ensuring that record companies offer the same terms and conditions to entities that are similarly situated to the labels' affiliates. Third, an antitrust exemption so that licensees could go to a single recording industry agent for "one-stop shopping" for performance rights. Fourth, the availability of a CARP as a backstop or safeguard on fees to be charged for compulsory licenses.

Integral to the CARP process, of course, are the standards and factors to be used by the arbitrators to determine the appropriate rates and terms for the statutory license. In the DPRSRA, Congress applied the standards set forth in the Copyright Act at 17 U.S.C. § 801(b)(1), that balance the rights of copyright owners and performers, the interests of the services and the paramount public interests:

(1) To make determinations concerning the adjustment of reasonable copyright royalty rates ... [which] shall be calculated to achieve the following objectives:

(A) To maximize the availability of creative works to the public;

(B) To afford the copyright owner a fair return for his creative work and the copyright user a fair income under existing economic conditions;

(C) To reflect the relative roles of the copyright owner and the copyright user in the product made available to the public with respect to relative creative contribution, technological contribution, capital investment, cost, risk, and contribution to the opening of new markets for creative expression and media for their communication;

(D) To minimize any disruptive impact on the structure of the industries involved and on generally prevailing industry practices.

This standard was applied in the first CARP under the DPRSRA, and yielded a percentage of revenue royalty rate to be applied to the three then-existing cable and satellite digital music services.

In 1998, Congress again addressed how to appropriately balance creative and new media interests, and clarified the applicability of the sound recording digital performance right to Internet webcasters. At that time, webcasting was in its embryonic stages and new business models were just beginning to develop. DiMA companies were appreciative of Congress' intent, and accepting of the new royalty obligation to the benefit of creators so long as its impact was mutually supportive of the new industry.

To ensure the appropriate balance between licensing efficiency and anticompetitive risk, Congress relied again upon the same three elements: a statutory license, an antitrust exemption, and a CARP safeguard. This time, however, at the RIAA's urging -- and without revisiting how this new standard dovetailed with RIAA's previously-secured antitrust exemption -- Congress adopted a different standard to be applied by the CARP to determine statutory license rates and terms:

In establishing rates and terms for transmissions by eligible nonsubscription services and new subscription services, the copyright arbitration royalty panel shall establish rates and terms that most clearly represent the rates and terms that would have been negotiated in the marketplace between a willing buyer and a willing seller. In determining such rates and terms, the copyright arbitration royalty panel shall base its decision on economic, competitive and programming information presented by the parties, including-

(i) whether use of the service may substitute for or may promote the sales of phonorecords or otherwise may interfere with or may enhance the sound recording copyright owner's other streams of revenue from its sound recordings; and

(ii) the relative roles of the copyright owner and the transmitting entity in the copyrighted work and the service made available to the public with respect to relative creative contribution, technological contribution, capital investment, cost, and risk.

In establishing such rates and terms, the copyright arbitration royalty panel may consider the rates and terms for comparable types of digital audio transmission services and comparable circumstances under voluntary license agreements negotiated under subparagraph (A).

17 U.S.C. § 114(f)(2)(B). Unfortunately, the legislative history offers no explanation or any reasons why Congress adopted this different standard, and little guidance as to how the standard is to be applied:

The test applicable to establishing rates and terms is what a willing buyer and willing seller would have arrived at in marketplace negotiations. In making that determination, the copyright arbitration royalty panel shall consider economic, competitive and programming information presented by the parties including, but not limited to, the factors set forth in clauses (i) and (ii).

H.R. Rep. No. 105-796, 105th Cong., 2d Sess. at 86.

Given the care that Congress undertook in 1995 to prevent anticompetitive licensing conduct by the recording industry (including several written consultations with the Antitrust Division of the Department of Justice), Congress surely could not have intended that this new standard be more susceptible to anticompetitive behavior in the collective licensing process. Unfortunately, as documented by the CARP Report, that is precisely what occurred. Even more unfortunately, the CARP erroneously believed that its flexibility was so limited by the new standard, that its decision ultimately relied only on benchmarks from agreements that the arbitrators noted, throughout their decision, were tainted by RIAA's willful and intentional anticompetitive behavior.

Regrettably, Mr. Chairman, the February 20 CARP Report, the first decision issued under this new "willing buyer/willing seller" rule, is a case study of the potential for cartelized activity under the revised standard, as the RIAA successfully pursued the establishment of monopoly prices as the statutory license rate.

But the fundamental problem for DiMA and its members is not just the insupportably high rates recommended by the arbitrators in this arbitration. Those rates are but the symptom. The central flaw is how the "willing buyer/willing seller" standard to determine the statutory license rates and terms has been manipulated by the RIAA in order to extract monopoly rates, and was construed by the arbitrators as requiring them to ignore compelling evidence that would have led to a far more reasonable and justifiable result. As explained below, DiMA urges this Committee and the Congress to consider rectifying these essential flaws in the CARP report:

1. The arbitrators erroneously interpreted the "willing buyer/willing seller" standard. They invented a prototypical "willing buyer" in the abstract, without reference to the ability of actual buyers to shoulder the rates they proposed. They incorrectly viewed the "willing seller" and the "marketplace" as an exclusive seller in a noncompetitive market. This caused the Panel to adopt rates based on what the arbitrators conceded were "above market" benchmarks obtained as a result of RIAA's single-seller market power and its "take-it-or-leave-it" licensing approach. The Panel's flawed understanding of the standard caused them to reject the most compelling and analogous evidence in the case - the rates paid by thousands of radio stations and webcasters to composers, lyricists and publishers to perform their copyrighted music.

2. Despite having chastised RIAA for its deliberate attempts to set above-market prices by licensing only a selected universe of unsophisticated licensees willing to pay such prices because of their peculiar circumstances, the arbitrators' erroneous interpretation of the legal standard led them to rely for rate-setting benchmarks exclusively upon the rates contained in the very agreements obtained by RIAA's deliberate effort to set artificially-inflated rates, including most prominently certain rates set forth in a single agreement between the RIAA and Yahoo! Inexplicably, the Panel admitted that the very incremental per-performance rates it relied upon, that were drawn for the Yahoo! agreement, did not genuinely reflect reasonable rates for Internet-only (or webcasting) services. Indeed, the Panel conceded that the rates in the Yahoo agreement were not "marketplace" rates and that they were designed by RIAA to maximize their legal arguments before the CARP (and by Yahoo to reduce the actual royalties they would pay upon their unique webcasting business model). The unfortunate, perhaps inevitable consequence of these errors is that the arbitrators, mistakenly believing they were compelled to rely on the Yahoo license, recommended royalty rates for Internet webcasting that will essentially bankrupt today's Internet webcasting industry - rates that not even Yahoo would accept.

3. Equally important is that the arbitrators failed to adopt an alternative rate on a percentage-of-revenue model that would enable webcasters to build a critical mass of listeners, attract advertising and related revenue, and thereby build a solid foundation for their future business.

4. Finally, for the benefit of future CARP proceedings (and although it will not assist webcasters in the current proceeding), in light of the documented efforts by RIAA to abuse the "willing buyer/willing seller" standard, Congress should carefully consider whether that standard should be amended or rescinded in favor of the pre-existing section 801(b) standards for setting sound recording performance royalty rates and terms.

RIAA's Anticompetitive Licensing Practices, as Exposed by the February 20 CARP Report, Torpedoed Legitimate Marketplace Negotiations.

As I noted previously, Mr. Chairman, DiMA initially sought to negotiate an industry-to-industry settlement with the RIAA in 1999. This effort collapsed when the RIAA refused to discuss any rate that approached the types of reasonable, established royalties that were being paid to perform musical compositions. The reasons for that collapse may be gleaned from the CARP Report, which describes how RIAA, concurrent with those negotiations, devised a scheme to artificially inflate the royalty rates, with the intention of creating "precedent" for above-market rates before the CARP.

The Report recounts how the RIAA Negotiating Committee attempted to manipulate the standard so as to establish a high benchmark to be used as "willing buyer/willing seller" precedent for the CARP. See CARP Report at 48-51. Per the direction of the RIAA Negotiating Committee (comprised of a senior executive of each of the five major recording companies), the RIAA intentionally adopted a "take-it-or-leave-it" licensing approach, with a small segment of the webcaster community consisting almost entirely of small, impecunious and inexperienced webcasters, to set an artificial, above-market rate at what the RIAA Negotiating Committee called its "sweet spot." Id. RIAA "meticulously crafted" confidentiality clauses that would prohibit any licensee from disclosing the rates and terms to other parties, yet reserved RIAA's right to use the agreements as it wished before the CARP. Id. at 50. Indeed, the CARP found that the RIAA "devoted extraordinary efforts and incurred substantial transactional costs" to negotiate these agreements only with "minor" webcasters "that promised very little actual payment of royalties" in return -- "sacrificial conduct mak[ing] economic sense only if calculated to set a high benchmark to be later imposed upon the much larger constellation of services." Id. at 50-51. The Panel further determined that RIAA's effort to justify this conduct "lacked credibility." Id. at 50.

DiMA did its best to communicate to webcasters that they did not have to deal with RIAA or be victimized by their predatory scheme in order to continue webcasting, and that they instead could avail themselves of the Congressionally-mandated statutory license under the CARP result. But, as the CARP Report acknowledged, a segment of the webcasting industry had particular needs or desires that required an RIAA license, even at artificially-inflated rates. For those webcasters, DiMA's advice was of no avail.

The CARP Procedures Exploited by the RIAA

RIAA's calculated efforts to rig the CARP rate were aided, as discussed below, by the unduly limiting and unique rules of CARP proceedings which are in dire need of reform. Those rules require that the parties' entire main case be presented in writing by a date certain, and without any advance warning or notice to the other parties identifying their rate-setting model. Here, having first secured a legislative change to a "willing buyer/willing seller" standard, RIAA proceeded to obtain licenses from its self-selected group of two dozen webcasters -- all but one of whom RIAA conceded were "unsophisticated" -- and posited simplistically as follows:

? the RIAA, armed with its antitrust exemption, was the only "seller" of the brand new Internet sound recording performance rights at issue;
? the RIAA, as such, knew that the only "buyers" were the 2-dozen companies that RIAA licensed (pursuant to their scheme described above); and,
? speaking for itself and its licensee-"buyers", RIAA thus assured the CARP that both RIAA, as the seller, and the licensees with whom it had done deals, were "willing" sellers and buyers and that these deals were the only licenses ever secured covering this new right.

Having thus manipulated the market (as the CARP Panel later found), the RIAA then set about to take advantage of the CARP's uniquely-limiting procedural rules. Unlike litigation in federal and state courts, the CARP rules effectively limit "discovery" to documents specifically referenced in a party's witness statements. There is no ability in a CARP to ask interrogatories, take depositions or subpoena testimony from anyone; and normal evidentiary rules (for example, those prohibiting hearsay) do not apply. Finally, irrespective of the complexity of the case or number of parties, CARP cases by statute must be concluded (decision and all) within 180 days, an extremely truncated timetable for what was called the "mother of all CARPs."

Knowing all this, RIAA set out, in calculating fashion, to ensure that every one of its license agreements with webcasters prohibited the licensee from disclosing the rates, terms and surrounding circumstances of their licenses to anyone, while nonetheless permitting RIAA alone to use the agreements affirmatively, as it pleased, in the CARP proceedings. These one-way confidentiality provisions, combined with the highly restrictive CARP discovery rules, prevented webcasters from getting substantive information about the circumstances surrounding these agreements, and thus created a remarkably unjust litigation scenario. RIAA proceeded to provide the CARP with a one-sided, selective and distorted factual landscape -- and then argued that the CARP Panel was required by the "willing buyer/willing seller" standard to establish rates on a par with those set forth in the licenses which RIAA previously had secured (and that RIAA alone was able to testify about).

In these circumstances, it is almost miraculous that the webcasters -- via a combination of cross-examination and successfully urging the CARP Panel to request further disclosures from the RIAA, as well as testimony from RIAA licensees -- were able to demonstrate how much RIAA sought to manipulate the CARP rate. But, as noted above, it was small solace for webcasters to be able to demonstrate RIAA's decidedly anticompetitive intent and licensing approach, insofar as the Panel still relied exclusively, in establishing its royalty rates, on the tainted license agreements that RIAA had secured .

The fact that the rates recommended by the CARP were considerably less than those secured by RIAA from its self-selected licensees may have led the Panel to believe that some measure of justice was being done. But even awarding RIAA 40% of the per performance royalty it requested will cripple the webcasting industry at its inception.

DiMA requests this Committee and Congress to closely scrutinize these aspects of the CARP Report and their resulting implications for the above-market royalty rates adopted by the CARP. This is particularly important at this critical time, because two more arbitrations are to be conducted within a few months from now: the first CARP for new subscription webcasters, and the second CARP for these very same nonsubscription webcasters -- both to be decided under the new "willing buyer/willing seller" standard.

The Arbitrators Misinterpreted the "Willing Buyer/Willing Seller" Standard.

Even a cursory reading of the CARP Report reveals that the "willing buyer/willing seller" standard was not correctly applied by the Panel. The CARP and the parties agreed that this standard envisioned a hypothetical negotiation between a willing buyer and willing seller in the marketplace. However, the CARP Report misinterprets three key elements of the test: what is the nature of the "willing buyer"; what is the nature of the "willing seller;" and whether evidence from the most analogous market, i.e., licenses for performing musical works, are relevant to the willing buyer/willing seller test for performing sound recordings. Unless the Librarian of Congress remedies the CARP's errors, we believe that the misinterpretations of these three critical elements of the standard require Congressional clarification.

The Definition of a "Willing Buyer"

Congress should make clear that the hypothetical "willing buyers" should reflect actual economic conditions of typical buyers within the buyers' industry. The CARP Panel agreed that the "willing buyers" in this equation are the webcasters. However, the Panel erroneously viewed these hypothetical buyers purely in the abstract, without any attempt to factor in the general economic condition of the buyers, or the impact that these economic conditions might have on their willingness or ability to pay royalties at a particular level.

Obviously, industry economics influence how "willing" or able a hypothetical buyer would be to agree to particular rates and terms. Actual economic conditions were particularly relevant in light of the specific CARP findings that most of the webcasters that had agreed to pay royalties to the RIAA under voluntary licenses, in fact, could not afford the rates they had agreed to. The Panel found that only three, perhaps four, of the 26 webcasters that signed such licenses with the RIAA paid any significant royalties at all. Id. at 52-54. For these very sorts of reasons, as discussed below, courts and the Department of Justice have refused to allow copyright licensing collectives to rely upon rates in license agreements from nascent industries. Yet, the Panel dismissed any relevance of industry economics in a brief footnote, contending that the considerable detailed record evidence of webcasting economics and the operation of webcast services would not materially aid its determination of what royalty rates willing buyers and willing sellers would actually agree to for the licenses at issue. Id. At 36 n. 23. It is no wonder that the rates hypothesized by the CARP, set without any reference to the business conditions of actual webcasters, will bankrupt the actual buyers.

The Definition of a "Willing Seller"

It similarly should be clarified that the "willing seller" must be a willing seller in a competitive market, not a market of single-source monopoly sellers. The CARP correctly rejected RIAA as an appropriate model for the "willing seller," inasmuch as they conceded that a single concentrated licensing entity was a monopolist rather than a willing seller. Yet, paradoxically, after properly labeling the RIAA a single seller with undue market power, the Panel nonetheless relied exclusively on benchmarks from a marketplace in which the RIAA acted as a single-seller collective.

In the end, the Panel erroneously simulated negotiations in a non-competitive market, which not surprisingly resulted in the CARP assessing an anticompetitive -- monopoly -- rate. It is readily apparent from the Panel's Report, which established royalty rates drawn exclusively from agreements that the Panel found were at "above market," "artificially high" rates, that the Panel did not view its task as requiring it to set rates that would approximate those set in a competitive "willing buyer/willing seller" marketplace.

If Congress truly intended that the "willing buyer/willing seller" test should result in competitive market rates (and absent this intention the CARP would serve no purpose), then it is clear that the CARP erred by analyzing only a non-competitive single-seller market as the basis of determining willing buyer-willing seller prices. A truly competitive market that avoids cartel pricing, of concern to Congress, requires competition among licensors, not licensees. Predictably, this faulty analysis resulted in an artificially elevated rate as reflected in the CARP Report.

The Scope of the Relevant Marketplace

The Report further errs in narrowly defining the market so as to reject the single most relevant evidence presented to the CARP, namely, the long history of payments to ASCAP, BMI and SESAC for the right to perform the musical works -- the notes and lyrics -- embodied in the sound recordings. Each is an essential copyright right necessary to perform music online.

As the Copyright Office found in a prior proceeding, no credible evidence suggests that a license to perform sound recordings is inherently worth more than a license to perform the musical works. To the contrary, in light of the promotional value of airplay and the higher revenue derived by sound recording copyright owners from the sale of sound recordings, there are persuasive economic arguments for a lower license rate. In prior proceedings, the CARP and the Copyright Office used musical work performance licenses as an appropriate analogy for the sound recording performance licenses at issue.

Moreover, the CARP record contained overwhelming evidence from analogous marketplaces -- where musical works and sound recordings are licensed at the same time for the identical use -- demonstrating that sound recording licenses are no more valuable than musical composition licenses. For example, the voluminous license data ($20 million in license transactions covering more than 700 songs) for transactions involving the reproduction rights associated with the incorporation of songs into movies and TV programs demonstrated the overwhelming extent to which producers value licenses to use sound recordings and musical compositions equally (with publishers and record companies alike typically seeking and obtaining most-favored-nations treatment against each other's rates). Similarly, the evidence from international markets -- where there is a long history of concurrent sound recording performance and musical work performance rights payments made by radio broadcasters to both record company and publisher collectives -- demonstrates that sound recording performances typically are compensated at rates lower than or equal to the rates paid for performance of musical works.

Yet, the CARP Panel amazingly chose to express "no opinion" on the relative value of musical works performances to sound recording performances and thus completely ignored all of this evidence. It did not even look to musical composition performance rates as a "sanity check" on the rates that the Panel proposed for the webcasting industry.

The CARP's reliance on the RIAA's negotiated deals with a handful of webcasters, while failing to consider the evidence of musical works performance rates, was particularly prejudicial and erroneous, in light of legal precedent admonishing courts to take particular care when evaluating agreements entered into in new and evolving industries, where there is little market experience in valuing the rights at issue. For example, the ASCAP and BMI rate courts consistently have rejected the use of voluntary agreements entered into at the inception of an industry, because such agreements generally reflect extraneous factors such as the desire to avoid or settle litigation, and rarely are indicative of a competitive market outcome. See, e.g., United States v. ASCAP; Application of Showtime/The Movie Channel Inc., 912 F.2d 563, 579-82 (2d Cir. 1989). Indeed, such considerations underlie provisions in the most recent ASCAP consent decree, which preclude ASCAP from relying upon agreements entered into during the first five years of licensing music users in a new industry as evidence of a reasonable final royalty fee. See United States v. ASCAP, No. 41-1395 at 13-14 (S.D.N.Y. June 11, 2001). See Memorandum of the United States in Support of the Joint Motion to Enter Second Amended Final Judgment, at 35, United States v. ASCAP ("[new] music users are fragmented, inexperienced, lack the resources [to litigate over rates] and are willing to acquiesce to fees requiring payment of a high percentage of their revenue because they have little if any revenue").

Consequently, by ignoring long-established music performance license rates, and by relying on inherently unreliable sound recording licenses engineered by RIAA to yield above-market rates, the Panel set a rate for the performance of sound recordings that is seven times higher (on a per-performance basis) than the standard industry rates historically paid for analogous rights.

A Percentage of Revenue Alternative for Commercial Webcasters is Necessary.

In enacting the limited digital sound recording performance right in 1995, and in extending that right to cover Internet webcasters in 1998, Congress explained its unequivocal intention to promote the growth of new technologies and music delivery methods for the benefit of consumers. Under the statutory license, Internet webcasting has become a vibrant opportunity even for entrepreneurial companies, whereby sites like Onion River Radio, Beethoven.com, SOMA-FM, WOLF-FM and others can get a foot in the door and provide compelling programming alternatives to commercial radio.

Another reason fueling the entrepreneurial webcasting opportunity is its relatively low start-up costs in comparison with terrestrial, cable or satellite broadcasting. Notwithstanding, Internet services encounter the "chicken-and-egg" problem I alluded to previously in my testimony. To attract any advertising revenues, a service must attract substantial numbers of listeners; to attract sufficient revenues to support an ongoing webcast venture, a service must have extraordinary numbers of listeners. Yet, unlike broadcasting, in which the upfront costs are high but the marginal cost of adding each new listener is zero, for webcasting, the upfront costs are lower, but adding each new listener adds additional costs in terms of bandwidth and server capacity.

Given these fundamental economic realities, it is easy to see how the per-performance/per-listener royalty can prejudice Internet webcasters, whether they are entrepreneurial start-up ventures or well-funded media companies. If a webcaster must pay high royalties, regardless of revenue, then the royalty scheme creates substantial new barriers to entry and success of smaller marketplace competitors.

The traditional marketplace licensors for blanket performance licenses in the music industry, ASCAP, BMI and SESAC, all recognize this economic fact of life. Their licenses always have offered a minimum fee with a percentage of revenue option. Under their licenses, the rights holders always are guaranteed a fair return via the minimum fee. Then, through the percentage of revenue fee, the more successful a webcaster becomes, the more revenue it will have, and the greater will be the royalties that it pays. In the interim, however, the performance license itself does not become an insurmountable barrier to entry.

While percentage-of-revenue licenses cannot always be justified as the sole formula for assessing music license fees, a revenue-based fee option thus has demonstrable virtues -- particularly in emerging industries. By rejecting a percentage of revenue option, the CARP Report has perhaps inadvertently dealt a death blow to entrepreneurship in webcasting. Consistent with Congress' policy to promote the growth of Internet webcasting, it is critical that action is taken now to ensure that a minimum fee and percentage of revenue option remains available to webcasters.

The CARP Report, to Great Harm, Erroneously Based the Industry-Wide Rate
on a Single Agreement with a Unique Company.

A great deal of attention has been paid in the CARP proceeding and the media to the deal between RIAA and a single, unique webcaster -- Yahoo! The reason why it has garnered so much attention is because the CARP, after eliminating all other evidence under an erroneous interpretation of the "willing buyer/willing seller" standard, had ONLY the Yahoo agreement left to consider. Yet, the facts set forth in the CARP Report demonstrate why even this agreement was an unreliable measure for an industry-wide rate, and why the CARP's erroneous and undue reliance upon this single agreement resulted in exorbitant and insupportable rates for the webcast industry at large.

First, as the CARP Panel observed, the RIAA entered into negotiations with Yahoo for the same reasons as their negotiations with the other 25 licensed entities: to create a precedent that would allow RIAA to manipulate the CARP rates. Reciprocally, as the CARP admitted, Yahoo was quite willing to permit itself to be manipulated in this manner, as long as its unique business objectives were met. Id. at 64-66. On its face, the rates and terms of such a license cannot satisfy the public interests in assuring that the statutory license is fair for all market participants.

Second, while the CARP believed that, given Yahoo's stature, the parties were negotiating on a level playing field, id. at 61, the CARP failed to recognize that Yahoo was playing a different game from the rest of the webcasting industry. The overwhelming majority of Yahoo's webcast streaming performances were retransmissions of radio station signals. Id. at 61. Internet-only programming on Yahoo was, at that time, a very small percent of its webcast offerings. By contrast, the vast majority of Internet webcasters are exclusively streaming Internet-only channels. Naturally, Yahoo was concerned primarily with obtaining the lowest possible total fee. After negotiating an upfront lump sum fee that covered the vast majority of the performances it made, Yahoo (as any rational business) naturally would have cared a lot more about the incremental rate stated for its radio retransmission performances than the rate stated for Internet-only performances. As the CARP conceded, Yahoo had little reason to resist RIAA's attempts to artificially inflate the incremental Internet-only rate, so long as the broadcast retransmissions rate was acceptable, and the "blended rate" resulting from the mix of broadcasts and Internet-only royalty payments also was acceptable. Id. at 63-65.

Consequently, Yahoo cared least about what concerned the rest of the industry most; and it was plainly erroneous for the CARP Panel to use as a benchmark an "artificially" inflated rate that even the Panel conceded was intentionally manipulated to be unreasonably high.

Third, as the CARP Panel notes, Yahoo recognized that by entering into a voluntary agreement before the CARP determination, they would be saving substantial amounts of money in CARP litigation costs. Id. at 68. And, by settling Yahoo would save "opportunity costs" by allowing key managers to devote time and resources to pursuing business deals rather than the arbitration. Id. at 68 and n. 46. The net result was that the costs for Yahoo to participate in the CARP could reasonably be expected to exceed their potential royalty obligations under the proposed deal with RIAA -- a fact that would motivate any rational economic actor to proceed with a deal even where the royalties were excessive under any objective measurement.

Fourth, perhaps because the CARP Panel erroneously believed that webcast economics was irrelevant, see Report at 36 n. 23, the Report fails even to mention that at the time of the Yahoo-RIAA agreement, Yahoo was not a "benchmark" type of webcaster that could accurately serve as a proxy for an industry "willing buyer." Virtually unique among Internet companies, Yahoo was a highly profitable public company with sizable revenues and millions of dollars of cash in the bank. Similarly unique among webcasters, Yahoo focused its branding on being an Internet "portal," not a webcaster.

In sum, the Yahoo agreement embodies artificially-inflated rates entered into by a sui generis, diversified Internet company in order to avoid prohibitively expensive arbitration costs, and to allow its managers to focus instead on attracting new revenue. It defies credulity to suggest that this single agreement somehow serves as an appropriate benchmark or proxy for appropriate rates to be paid by all webcasters. Indeed, the Panel's approach seems all the more bizarre and egregious in light of the fact, as noted above, that there existed a benchmark performance rate that all webcasters and broadcasters pay on an equivalent basis - the rates in the blanket licenses offered by ASCAP, BMI and SESAC for the performance of musical works.

There is No Basis to Set Disparate and Prejudicial Rates and Terms for Internet-Only Webcasts over Broadcast Retransmissions.

Fairness suffered two additional casualties in the CARP Report.

First, the Panel proposed a rate for Internet-only transmissions that is double the proposed rate for webcast retransmissions of terrestrial radio broadcasts. It is significant that neither RIAA nor the webcasters asked for disparate rates. Instead, the Panel was led to this aberrant result based upon yet another misapplication of the terms of the Yahoo-RIAA agreement. But their purported rationale (i.e., the promotional value of radio airplay) holds no water in this instance. The arbitrators merely assumed that "this factor likely was considered by RIAA and Yahoo!, and is evidently reflected in the resulting difference between [radio retransmission] and [Internet-only] rates." Id. at 75; emphasis added. Notably, however, the Report cites no actual record evidence to support these assumptions and, as noted above, the evidence is to the contrary.

Second, the irrationality of the CARP Panel recommendations goes beyond issues of the rate itself. The Panel illogically applied the inflated Internet-only per-performance rate even to partial performances of any duration. The CARP's report makes webcasters liable for full payments even when a song is discontinued a moment after it starts streaming (whether due to technological glitches or listener action). Thus, the Panel Report would impose punitive costs on webcasters for simple conveniences enjoyed for free within the broadcast radio world, wherein a listener can shift from one station to the next if he/she does not wish to listen to a certain song, or encounters signal static or interference.

The "Willing Buyer/Willing Seller" Standard Should be Revisited.

With the benefit of 20/20 hindsight, perhaps it seems obvious why the "willing buyer/willing seller" standard does not and cannot work in a new, highly concentrated market dominated by a very few powerful players who hold monopoly rights over vast catalogs of the world's most popular sound recordings. Unfortunately, the willing buyer/willing seller standard adopted in the DMCA creates perverse incentives to manipulate the marketplace, so as to set the CARP rates at far above fair market value.

In this regard, it bears noting that industry has come before Congress to intervene in the CARP process twice in the last three years, to set aright royalty determinations that bore no reasonable relationship to the economics of the industries in which they were levied. DiMA cannot help but remind the Committee that each of these disproportionate results occurred in arbitrations that applied the "willing buyer/willing seller" standard.

We urge Congress to ensure that such attempts to manipulate the CARP standards and processes cannot recur. DiMA has suggested above a number of concrete ways in which Congress can clarify its intentions in the existing "willing buyer/willing seller" standard, so as to rectify the unjust results of this most recent CARP, and to re-set the economic foundations for the webcasting industry. Nevertheless, in light of the experiences of webcasters and the satellite television industry, perhaps the unavoidable truth of the matter is that the "willing buyer/willing seller" standard simply does not work in the context of new and emerging industries. Although it will not help webcasters for this CARP, DiMA respectfully submits that Congress should seriously consider whether it is advisable and necessary to rescind the willing buyer/willing seller standard and replace it with the balanced section 801(b)(1) factors applicable to past sound recording performance rights CARP arbitrations.

Procedural Deficiencies in the CARP Process Should be Rectified.

Congress intended the CARP arbitration process to be a swift, relatively inexpensive and reasonably fair way to adopt and adjust copyright royalty rates. Webcasters' experiences in this CARP suggest that certain procedural safeguards are necessary to ensure the fairness of the process.

Discovery

As noted above, current CARP discovery rules limit the parties to discovery only of documents that specifically underlie the written testimony submitted by the parties. The unfortunate consequence of this limitation is that the CARP and the parties may be denied access to information that would be extremely valuable in setting a fair and reasonable rate. We therefore recommend that Congress consider adopting limited-in-time and limited-in-scope "federal"-type discovery provisions. This would avoid the pitfalls of "trial-by-ambush" strategies of the nature employed by RIAA in the recent CARP without delaying the CARP litigation process.

Similarly, preparation of the webcasters' cases in this proceeding were substantially hampered by the intentional insertion by RIAA of confidentiality clauses that restricted the licensees from discussing the terms of their agreements with the services' counsel. Therefore, at minimum, we recommend that at a date no later than 30 days before all written cases are due, parties should be required to disclose the identity of any license that it intends to submit in support of its written case; to affirmatively state in such disclosure the name and last known address and telephone number of the persons who negotiated the licenses for both licensor and licensee; and, for purposes of use in the CARP proceeding, to specifically release all such persons from any confidentiality obligations that otherwise may have been imposed under the agreements.

Further, as a means of streamlining the issues for hearing, we suggest that the parties be permitted to serve Requests for Admission to other parties, as set forth in the Federal Rules of Civil Procedure.

Subpoena Power

We note that much valuable evidence was introduced during the rebuttal phase in this CARP proceeding from webcasters that had signed licenses with the RIAA, including Yahoo, after all the RIAA's licensees were asked by the Panel to testify (at webcasters' urging). Had these licensees not stepped up to the plate, the arbitrators would have been denied access to record evidence that became essential to their decision; yet this last-phase and limited testimony was hardly sufficient to protect webcasters from the distorted and one-sided evidentiary record that RIAA was able to develop by manipulating the existing CARP rules, as discussed above. We recommend that the arbitrators, like other administrative agency judges, be authorized to issue subpoenas and permit CARP participants to seek to enforce such subpoenas in the federal courts.

Webcasters Need Relief Now.

As I stated at the beginning of my testimony, a few days from now DiMA and all Internet radio services will learn of the decision of the Librarian of Congress. If the recommendations in the CARP Report are upheld, hundreds, perhaps thousands, of Internet webcasters and broadcasters will shut off their programming, rather than continue incurring extraordinarily high royalty liabilities that they manifestly cannot pay, and as a matter of fairness no webcaster should be required to pay. I have summarized in my testimony several of the most significant flaws in the CARP Report. We hope that the compelling arguments made by the broadcasters and webcasters to the Copyright Office will win the day, and that reason and economic rationality ultimately will prevail. Should this not be the case, we hope that Congress will consider carefully, and rapidly promulgate, the clarifying statements recommended in our testimony. These clarifications will guide the Librarian of Congress and, potentially, the Court of Appeals, in the proper application of the current standards, and will restore greater confidence in the checks and balances built into the current CARP process.

Whatever the result for this proceeding, DiMA respectfully submits that the CARP Report in this proceeding, and the negotiations chicanery undertaken by the RIAA, prove beyond question that the "willing buyer/willing seller" standard must change. There can be no hypothetical "willing seller" in a non-competitive market where there are no alternative sources for the same licensed works. In such a market, as shown in the CARP Report, the marketplace can too easily be manipulated by monopoly sellers so as to produce an artificial above-market rate, concocted solely as precedent for the CARP. We submit that Congress should act promptly to rescind the willing buyer/willing seller standard, to reinstate the section 801(b) factors as the relevant touchstone for the CARP determination and, to paraphrase that great rock philosopher Pete Townshend, to ensure that we don't get fooled again.

Thank you again for your attention and your consideration. I would be pleased to answer any questions the Committee may have.