Distribution of 1998 and 1999 Cable Royalty Funds

AGENCY:

Copyright Office, Library of Congress.

ACTION:

Final order.

SUMMARY:

The Librarian of Congress, upon the recommendation of Register of Copyrights, is accepting in full the determination of the Copyright Arbitration Royalty Panel and is announcing the final Phase I distribution of cable royalties for 1998 and 1999.

EFFECTIVE DATE:

January 26, 2004.

ADDRESSES:

The full text of the CARP's report to the Librarian of Congress is available for inspection and copying during normal business hours in the Office of the General Counsel, James Madison Memorial Building, Room LM-403, First and Independence Avenue, SE., Washington, DC 20559-6000.

FOR FURTHER INFORMATION CONTACT:

David O. Carson, General Counsel, or William J. Roberts, Jr., Senior Attorney, P.O. Box 70977, Southwest Station, Washington, DC 20024. Telephone: (202) 707-8380. Telefax: (202) 252-3423.

SUPPLEMENTARY INFORMATION:

Background

In 1976, Congress adopted a statutory license for cable television operators to enable them to clear the copyrights to over-the-air television and radio broadcast programming which they retransmit to their subscribers. Codified at 17 U.S.C. 111, the section 111 license allows cable operators to submit semiannual royalty payments, along with accompanying statements of account, to the Copyright Office for subsequent distribution to copyright owners of broadcast programming retransmitted by those cable operators. In order to determine how the collected royalties are to be distributed amongst the many copyright owners filing claims for them, the Copyright Office, under the auspices of the Librarian of Congress, conducts a distribution proceeding under chapter 8 of the Copyright Act. Distribution of cable license royalties are conducted in two phases. In Phase I, the royalties are divided among eight categories or groups of copyright owners that represent all of the kinds of copyrighted broadcast programming carried by cable systems: movies and syndicated television programs; sports programming; ² commercial broadcast programming; ³ religious broadcast programming; ⁴ public television broadcast programming; ⁵ Canadian broadcast programming; ⁶ public radio broadcast programming; ⁷ and music.⁸ In Phase II the money allotted each category is subdivided among the various copyright owners within that category. Today's proceeding is a Phase I proceeding for royalties collected from cable operators for the years 1998 and 1999.

The royalty payment scheme of the cable statutory license is technical, complex and, many would say, antiquated. The license places cable systems into three categories based upon the amount of money they receive from their subscribers for over-the-air broadcast stations. Small and medium-sized systems pay a flat fee. Large cable systems—whose royalty payments comprise the lion's share of the royalties to be distributed in this proceeding—pay a percentage of the gross receipts they receive from their subscribers for each distant over-the-air broadcast station they retransmit. How much they pay for each broadcast station depends upon how the carriage of that station would have been regulated by the Federal Communications Commission (“FCC”) in 1976, the year the current Copyright Act was enacted. The royalty scheme for large cable systems employs the statutory device known as the distant signal equivalent (“DSE”). Distant signals are determined in accordance with two sets of FCC regulations: the “must carry” rules for broadcast stations in effect on April 15, 1976, and a station's television market as currently defined by the FCC. 17 U.S.C. 111(f). A signal is distant for a particular cable system when that system would not have been required to carry the station under the FCC's 1976 “must carry” rules and the system is not located within the station's local market.

Large cable systems pay for carriage of distant signals based upon the number of DSE's they carry. The statute defines a DSE as “the value assigned to the secondary transmission of any nonnetwork television programming carried by a cable system in whole or in part beyond the local service area of the primary transmitter of such programming.” 17 U.S.C. 111(f). A DSE is computed by assigning a value of one to a distant independent broadcast station, and a value of one-quarter to distant noncommercial educational and network stations, which do have a certain amount of nonnetwork programming during a typical broadcast day. Large cable systems pay royalties based upon a sliding scale of percentages of their gross receipts depending upon the number of DSE's they incur. The greater the number of DSEs, the greater the total percentage of gross receipts and, consequently, the larger the total royalty payment. The monies collected under this payment scheme are received by the Copyright Office and identified as the Basic Fund.

The complexity of the royalty payment mechanism does not, however, end with the Basic Fund. As noted above, the operation of the cable license is intricately linked with how the FCC regulated the cable industry in 1976. The Commission restricted the number of distant signals that cable systems could carry in 1976 (the distant signal carriage rules), and required them to black-out programming contained on a distant signal where the local broadcaster had purchased the exclusive right to that programming (the syndicated exclusivity rules). However, in 1980, the Commission took a decidedly deregulatory stance towards the cable industry and eliminated the distant signal carriage rules and the syndicated exclusivity (“syndex”) rules. Malrite T.V. v. FCC, 652 F.2d 1140 (2d Cir. 1981), cert. denied sub. nom., National Football League, Inc. v. FCC, 454 U.S. 1143 (1982). Cable systems were now free to import as many distant signals as they desired without worry of restrictions.

Pursuant to its statutory authority and in reaction to the FCC's action, the Copyright Royalty Tribunal (“CRT”) initiated a rate adjustment proceeding for the cable license to compensate copyright owners for the loss of the distant signal carriage rules and the syndex rules. This rate adjustment proceeding produced two new rates applicable to large cable systems making section 111 royalty payments. 47 FR 52146 (November 19, 1982). The first, to compensate for the elimination of the distant signal carriage rules, was the adoption of a royalty rate of 3.75% of a cable system's gross receipts for carriage of each distant signal that would not have been previously permitted under the former distant signal carriage rules. Distant signal royalties which are paid at the 3.75%—known as the “penalty fee” in cable circles—are identified by the Copyright Office as the “3.75% Fund” and are separate from royalties placed in the Basic Fund.

The second rate adopted by the CRT, to compensate for the elimination of the syndex rules, is known as the syndex surcharge. Large cable operators must pay this additional fee when the programming appearing on a distant signal imported by a cable system would have been subject to black-out protection under the FCC's former syndex rules. Royalties comprising the syndex surcharge are identified by the Copyright Office as the “Syndex Fund” and are separate from royalties placed in the Basic Fund and the 3.75% Fund.

The royalties in these three funds—Basic, 3.75% and Syndex—are the royalties that are eligible for distribution to copyright owners of nonnetwork broadcast programming in a section 111 cable license distribution proceeding.

This Proceeding

On November 20, 2001, the Library of Congress opened Docket No. 2001-8 CARP CD98-99, a consolidated Phase I distribution proceeding for cable license royalties collected from cable operators for the years 1998 and 1999. Of the eight Phase I categories or “parties” filing Notices of Intent to Participate in this distribution proceeding, two parties—Devotional Claimants and NPR—settled with the others as to the amount of their distribution and voluntarily withdrew their claims. The Library turned to the task of scheduling a Copyright Arbitration Royalty Panel (“CARP”) proceeding for the remaining six parties and, after several requests for postponement from these parties, a final schedule was issued on October 28, 2002. Order in Docket No. 2001-8 CARP CD 98-99 (October 28, 2002). The six parties filed their written direct cases on December 2, 2002, and the Library conducted discovery and motions practice throughout the winter. On April 24, 2003, the Library convened the three-person CARP who conducted hearings on the written direct cases, received rebuttal testimony and considered each party's written proposed findings of fact and conclusions of law. The Panel reviewed and analyzed nearly 20,000 pages of testimony and issued a 94-page determination, complete with an appendix of the mathematical calculations performed by the CARP to arrive at the distribution percentages for each of the six parties for 1998 and 1999, and another appendix identifying all exhibits submitted during the proceeding and whether or not they were admitted into evidence. The CARP report represents six months of intensive work. Following is a summary.

The CARP Report

The six parties who litigated division of the 1998 and 1999 cable royalties have a long history in the distribution of section 111 royalties. When Congress created the cable license and the distribution process in the 1976 Copyright Act, it did not provide any criteria or guidelines for how the royalties should be divided amongst the various copyright owners. Consequently, in the first cable distribution proceeding for cable royalties collected in 1978, the Copyright Royalty Tribunal identified five factors that would guide its distribution decisions. The primary factors were: (1) The harm caused to copyright owners by distant retransmissions; (2) the benefit derived by cable systems from distant retransmissions; and (3) the marketplace value of the works retransmitted. 45 FR 63026, 63035 (September 23, 1980). The Tribunal also identified two secondary factors for consideration: (1) The quality of the retransmitted programs; and (2) time-related considerations. Id.

As the years passed and subsequent distribution years were litigated, the Tribunal refined these criteria. Time-related considerations were given little weight in dividing the royalty pool and in the 1989 distribution determination, the Tribunal announced that program quality would no longer be considered. 57 FR 15287, 153303 (April 27, 1992) (“[Q]uality will no longer be a criterion in the Tribunal's distribution because it conflicts with the First Amendment”). When the Tribunal was replaced by the CARP system, the first, and until this proceeding only, CARP to conduct a Phase I cable distribution chose to focus solely on the marketplace value criterion and exclude all the others. The current CARP has chosen to embrace relative marketplace value of the programming retransmitted as the sole criterion governing distribution of the 1998 and 1999 royalties because the previous CARP's decision on this point was upheld by the Librarian and on appeal, and all six parties in this proceeding accepted that relative marketplace value is the sole relevant criterion.

Having decided that the relative marketplace value of broadcast programming retransmitted by cable systems during 1998 and 1999 will govern how the royalties will be divided among the six parties, the CARP considered how to evaluate it. Given that the cable license substitutes for marketplace negotiations in the buying and selling of broadcast programming, there is no real marketplace for those broadcast programs retransmitted by cable systems. Thus, the CARP determined that it must “ ‘simulate [relative] market valuation’ as if no compulsory license existed.” CARP Report at 10. Forecasting a hypothetical marketplace absent the existence of the cable license is a difficult task. The Panel concluded, after considering several options, that marketplace negotiations for broadcast programming would most likely occur between individual cable operators (or perhaps multiple system operators or a collective that they might form) and individual broadcast stations that would act as intermediaries for copyright owners and that would license all the copyrighted programming broadcast by each station. As a result of this conclusion, the Panel observed that cable system operators (or multiple system operators or a collective) would face a fixed quantity of distant broadcast station programming in the hypothetical marketplace. The supply curve for each type of programming (movies, sports, music, etc.) would remain vertical, meaning that the supply of programming would remain the same irrespective of the price. Because of this, the Panel determined that in “the hypothetical marketplace structure that we envisage [it is] the ‘demand side’ that will determine relative values of each type of programming.” Id. at 13 (footnote and citations omitted). This is an important conclusion of the CARP because it governs how the Panel evaluated each of the six parties' evidentiary submissions.

As with previous cable distribution proceedings, the two principal evidentiary offerings of the parties that attempt to determine the value of the six program categories are the Bortz survey and the Nielsen study. The Bortz survey, offered by the Joint Sports Claimants, is a statistical survey of a selected group of cable operators that asks those with programming responsibilities at the chosen cable systems what value they place on the six categories of programming involved in this proceeding. The responses to the inquiries posed by the survey are then distilled in an effort to attach the relative marketplace value to each program category. The Nielsen study, offered by Program Suppliers, takes a decidedly different approach by utilizing the data supplied by Nielsen Media Research measuring television viewing during 1998 and 1999. The purpose of the Nielsen study is to show the amount of viewing of distant signal programming by households and persons that are in the Nielsen People Meter sample. Both the Bortz survey and the Nielsen study have been used by the CRT and the prior cable distribution CARP in determining the division of cable royalties, although both have received criticisms as to methodology and application. See, e.g. 57 FR 15287 (April 27, 1992) (1989 cable distribution); 61 FR 55653 (October 28, 1996) (1990-92 cable distribution).

After considering both the Bortz survey and the Nielsen study and examining their results, the CARP arrived at a significant conclusion. Unlike the CRT and the CARP in prior proceedings, the Panel determined that the Bortz survey best projected the value of broadcast programming in the hypothetical marketplace whereas the Nielsen study “does not afford an independent basis for determining relative value.” CARP Report at 44. The Panel arrived at this conclusion because it determined that the Nielsen study did “not directly address the criterion of relevance to the Panel,” to wit: “[t]he value of distant signals to [cable system operators] * * * in attracting and retaining subscribers.” Id. at 38. “The Nielsen study reveals what viewers actually watched but nothing about whether those programs motivated them to subscribe or remain subscribed to cable.” Id. The Panel did not discard the Nielsen study completely, however, and found that it could be a useful tool in those circumstances when the Bortz survey could not be used.

Having chosen the Bortz survey as the most “robust” and reliably predictive model for determining value, the Panel considered its application to each of the six Phase I parties. With respect to Joint Sports Claimants, Program Suppliers and NAB, the Panel determined that “the Bortz survey is more reliable than any other methodology presented in this proceeding for determining the relative marketplace value of these three claimant groups” for the Basic Fund and the 3.75% Fund. Id. at 31. Consequently, these three parties received the royalty shares of the Basic Fund and the 3.75% Fund as determined by the Bortz survey, adjusted for the settlement distribution percentages of NPR and the Devotional Claimants.

The Bortz survey was not so “robust” with respect to PBS, Canadian Claimants and the Music Claimants. The Panel found that the Bortz survey undervalued PBS programming because it removed from its sample cable systems who carried a PBS station as their only distant signal and assigned a value of zero to PBS for those cable systems that carried commercial stations on a distant basis but not a PBS station. The “result is an exclusion of the category of cable operators that would be expected to give the highest relative value to a [PBS] distant signal,” and the “exclusion of the [PBS]-only systems artificially depresses the [PBS] Bortz score. A consistent application of the Bortz methodology would arguably mean that if a CSO carries a [PBS] signal as its only distant signal, all other categories should automatically be assigned zeroes.” Id. at 23. Despite these flaws, the Panel concluded that PBS's Bortz share of 3.2% for both 1998 and 1999 established a minimum or “floor” from which to determine PBS's net distribution percentages. The Panel then turned to PBS's principal evidentiary presentation as to its marketplace value—a study sponsored by Dr. Leland Johnson designed to show the number of subscribers receiving distant PBS signals during 1998 and 1999—and rejected it because it “attempt[s] to equate relative programming volume with relative programming value.” Id. at 56 (emphasis in original). Instead, the Panel accorded weight to a fee generation approach (considering the royalties paid by cable systems into the 1998 and 1999 Basic Funds for carriage of PBS distant signals) along with the Bortz results because unlike other program categories such as sports or movies, PBS signals are retransmitted by cable systems as discrete, intact distant signals containing only PBS programming. The Panel also examined PBS's claims of “changed circumstances” and found “no persuasive evidence that [PBS's] relative value has significantly either increased or decreased since 1990-92.” Id. at 69. As a result, the Panel awarded PBS the same distribution percentage for the 1998 and 1999 Basic Funds that it received in the 1990-92 cable distribution proceeding. PBS did not receive a percentage of the 3.75% Fund or the Syndex Fund because it does not participate in those funds.

The Bortz survey is not designed to include Canadian Claimants and Music Claimants. With respect to Canadian Claimants, the Panel adopted a combination of the fee generation approach and changed circumstances. The Panel mostly, though not completely, accepted Canadian Claimants' proposed fee generation approach and determined that there were no significant changed circumstances that would significantly impact their award. As a result, Canadian Claimants received the distribution percentages yielded by the fee generation approach for the Basic Fund and the 3.75% Fund, adjusted to yield for net awards. Canadian Claimants do not share in the Syndex Fund.

Finally, with respect to the Music Claimants, the Bortz survey was not relevant because it does not measure music as a category of programming, and the fee generation approach is not applicable. The Panel rejected Music Claimants' arguments for using the 4.5% settled distribution percentage from the 1990-1992 cable proceeding as the base measurement of the relative value because the settlement by its terms had no precedential value and does not reflect how cable system operators would value music. Instead, the Panel accepted the testimony of Joint Sports Claimants' witness Dr. George Schink, who estimated a range for Music Claimants' award by comparing the amounts that Music Claimants receive in licensing fees from broadcasters and cable networks with the total programming expenses of those broadcasters and cable networks, as establishing the minimum of an award (2.3%), and used the 4.5% settled award from the 1990-1992 proceeding as the maximum. The Panel selected an award of 4.0% as falling within this “zone of reasonableness” as applied to the Basic Fund, 3.75% Fund, and the Syndex Fund for both 1998 and 1999. The remaining 96% of the Syndex Fund was awarded to Program Suppliers, consistent with prior rulings of the CRT.

The final distribution percentages are as follows:

Petitions to Modify

As provided by the CARP rules, the parties to the proceeding were given 14 days to submit their petitions to modify the CARP report and an additional 14 days for a reply. Petitions to modify were received from Program Suppliers, PBS, Music Claimants and Canadian Claimants. Replies were submitted by all parties. Following is a synopsis of these petitions.

1. Program Suppliers

Program Suppliers received the largest reduction in their royalty award from the percentages set in the 1990-1992 distribution proceeding and, not surprisingly, therefore strongly contest the CARP's determination in this proceeding. Program Suppliers' arguments are made along three principal lines. First, they contend that the Panel improperly abandoned precedent by rejecting the Nielsen study and favoring the Bortz survey. Second, they charge that the Panel completely ignored compelling evidence presented by Program Suppliers regarding the relevance of viewing in determining program value. And third, Program Suppliers argue that rationales accepted by the Panel for setting the awards for PBS, Canadian Claimants and Music underscore the Panel's arbitrary decision making.

Program Suppliers submit that the CARP abandoned the precedent established by the CRT and the 1990-1992 cable distribution CARP which accorded value to the Nielsen study. Citing 17 U.S.C. 802(c), which provides that a CARP “shall act on the basis of a fully documented written record, prior decisions of the Copyright Royalty Tribunal, prior copyright arbitration royalty panel determinations, and rulings by the Librarian of Congress * * *,” and Nat'l Ass'n of Broadcasters v. Copyright Royalty Tribunal, 772 F.2d 922, 932 (D.C. Cir. 1985), Program Suppliers argue that the CARP in this proceeding was required to begin with the distribution percentages from the 1990-1992 proceeding. Given that those numbers must be the starting point, the Panel could then “only depart from the existing allocation methodology where it either finds ‘changed circumstances’ or that the earlier methodology was wrong. It cannot, therefore, adopt ‘one or more methodologies that provide reliable estimates of current * * * relative valuations.” Program Suppliers' Petition to Modify at 9 (citing CARP Report at 14). Program Suppliers argue that the CARP has failed to find that changed circumstances warranted departure from the Nielsen study. To the contrary, the CRT as well as the 1990-1992 cable distribution CARP recognized the value of the Nielsen study. Program Suppliers admit that there have been criticisms of the Nielsen study in the past, but there have been criticisms of the Bortz survey as well. Program Suppliers assert that improvements were made in this proceeding to the Nielsen study and the Bortz survey, yet “the Panel recognizes, and even praises, the methodological improvements made to the Bortz Study, but maintains virtual silence regarding those made to the Nielsen Studies.” Id. at 11. Nevertheless, criticisms of the Bortz survey remain, which the Panel acknowledged, thereby precluding the Panel from accepting the survey wholesale. Precedent has long established that actual viewing to programming is relevant to programming value, and it is arbitrary for the Panel to conclude otherwise.

Program Suppliers charge that the CARP ignored the compelling evidence that it submitted relevant to marketplace value. Contrary to the CARP's conclusion that cable operators only care about signing up and keeping subscribers and not about what they watch, Program Suppliers state that they presented considerable evidence demonstrating that cable operators do care about what their subscribers watch and will pay more for programming that receives high Nielsen viewing numbers. Program Suppliers argue that evidence from the cable network marketplace demonstrates that viewing plays a critical role in determining the licensing fees paid by cable systems for these networks, yet the CARP completely ignored this evidence. They contend that the witness testimony they presented demonstrating the importance of viewing to establishing licensing fees is not even discussed by the Panel, underscoring the arbitrary nature of their decision making.

Program Suppliers also charge that inconsistent treatment of similarly situated parties highlights the arbitrary nature of the Panel's approach. For example, the Panel relied on a fee generation approach in determining Canadian Claimants' award, but did not use it for similarly situated PBS. With respect to NAB, whose award nearly doubled from the 1990-1992 proceeding despite the fact that its Bortz numbers did not change substantially from that proceeding to the present one, “the Panel relied on the Nielsen viewing data to justify increasing NAB's share but ignored viewing when making other parties' allocations.” Id. at 49. Likewise, the Panel announced that Dr. Gregory Rosston's regression analysis was useful in corroborating the results of the Bortz survey but did not analyze whether that same regression analysis corroborated the results of the Nielsen study.

Finally, Program Suppliers allege that Music Claimants' distribution percentages for 1998 and 1999 are arbitrary and should be no more than 2.3%—the floor to the zone of reasonableness taken from the study done by Dr. George Schink. “[T]he Panel articulated no reasoning or determinations of fact in its findings regarding Dr. Schink's license fee analysis that indicated a lack of reliability in the results.” Id. at 53. Furthermore, the Panel never articulated a precise reason as to why it chose the distribution figure (4%) that it did.

2. PBS

Although PBS has asked for an award of 12% of the Basic Fund for 1998 and 1999, the CARP gave it the same award it received in the 1990-1992 cable distribution proceeding. PBS offers two principal arguments as to why the Panel's determination with respect to PBS is arbitrary and must be set aside. First, PBS submits that the Panel's logic is internally inconsistent. Second, the Panel acted arbitrarily by nearly doubling NAB's award from the 1990-1992 distribution proceeding while holding PBS's award constant. PBS then offers an evidentiary basis for the Librarian to increase its award.

PBS submits that the Panel's logic is internally inconsistent in two fundamental ways. First, after examining the Bortz survey and determining that it was inherently biased in its results against PBS (and therefore could only be used to establish the minimum award for PBS), the Panel then relied on those biased results to dismiss other methodologies for determining PBS's award. The Panel dismissed the quadrupling in PBS's Nielsen viewing share and the near doubling in PBS's subscriber instances share from 1992 to 1998 by pointing to the lack of increase in PBS's Bortz share during that same period. “The biases in the Bortz results that made them unusable in determining [PBS's] share also make them unusable as a measure of changed circumstances * * *.” PBS Petition to Modify at 6.

Second, PBS asserts that the Panel stated that it would rely on the Nielsen viewing data to assess PBS's changed circumstances since the 1990-1992 distribution proceeding, but then failed to do so.

[T]he Panel did not do what it said it would do. Contrary to its own express statement, the Panel did not “rel[y] upon the Nielsen study” to assess changed circumstances as to [PBS]. The Panel did not adhere to its own statement that “Nielsen studies can serve as a tool for assessing changed circumstances whenever the Bortz survey cannot be used.” To the contrary, the Panel completely disregarded and did not rely on the Nielsen viewing study as to [PBS] despite its own express ruling that the Bortz survey could not be used as to [PBS]. . . . The Panel's reasoning thus failed to adhere to the logical framework that it had established in the opinion.

Id. at 9 (emphasis in original; citations omitted).

PBS also charges that the Panel used NAB's increase in viewing share from the 1990-1992 distribution proceeding as corroboration that its award should nearly double from the prior proceeding, but then refused to use PBS's quadrupled viewing share as grounds to increase PBS's award from the prior proceeding. PBS contends that the Panel's refusal to credit its increased viewing share because its Bortz survey numbers had not significantly increased from 1992 to 1998 is wholly illogical when the Panel had already determined the Bortz survey was inherently biased against PBS. If such “major bias” in the Bortz survey numbers for PBS was not present in the 1990-1992 proceeding but is present in this proceeding, then PBS's award from the prior proceeding relative to its Bortz share at the time must go up in this proceeding given the increase in its Bortz share in this proceeding. “In short, both [PBS] and NAB experienced sizeable increases in their “true” Bortz shares and Nielsen viewing shares between 1990-92 and 1998-99, yet the Panel decided to nearly double NAB's award while holding [PBS's] award constant.” Id. at 12.

3. Canadian Claimants

The Canadian Claimants submit that the CARP made a mathematical miscalculation in Appendix B of its report that creates a computational side effect and results in a loss of its Basic Fund award. Specifically, Canadian Claimants argue that they should receive the share yielded by the fee generation approach adopted by the Panel reduced only for net awards to Music, the Devotional Claimants, and NPR, and not the net share awarded to PBS.

The CARP's award to Canadian Claimants is part of a four-step process. First, the Panel adopted the Bortz shares of Program Suppliers, Joint Sports Claimants and NAB and adjusted them to equal 100%. Next, the Panel focused on Canadian Claimants using the fee generation approach and determined the amount of the Basic Fund for 1998 and 1999 that was generated by cable systems paying for distant Canadian signals. Within the percentage for each year, the Panel identified the amount of fees attributable to Canadian Claimants' programming, Program Suppliers' programming and Joint Sports Claimants' programming based upon a survey presented by Dr. Debra Ringold. Since Dr. Ringold did not analyze the fees generated by the other parties in this proceeding, the Panel excluded them and adjusted her numbers to equal 100%. Third, the Panel took the adjusted Canadian numbers and added them to the Bortz-generated numbers for Program Suppliers, JSC and NAB, and adjusted those to 100%. Finally, the Panel combined the numbers for these four parties with the net awards determined for PBS, Devotional Claimants and NPR and adjusted them so all final distribution percentages would equal 100%.

The Panel's approach, according to Canadian Claimants, is flawed in several respects. First, Canadian Claimants charge that the combination process in step four should not have included PBS since, unlike the other categories, PBS programming does not appear on Canadian signals. Including PBS programming is inconsistent with the fee generation approach that the Panel said it was using. Second, by combining Canadian Claimants' fee generated numbers in step three with the Bortz numbers of Program Suppliers, JSC and NAB, the effect of the adjustment in step four is not the same for Canadian Claimants as it is for Program Suppliers, JSC and NAB. In step one, the Panel adjusted the Bortz numbers for Program Suppliers, JSC and NAB to equal 100% which meant they received a “bump up” in their actual numbers. Canadian Claimants received no such increase, meaning that when the Music, Devotional and PBS awards are deducted in step four, Canadians bear a higher pro rata loss to their Basic Fund award than do Program Suppliers, JSC, and NAB. “The effect of the Panel's approach is that the [Canadian Claimants] give[ ] up more of [their] initial award towards the ‘net’ claimants than does (sic) NAB, PS, or JSC, even though based on the rational (sic) behind the fee gen approach—the [Canadian Claimants] should give up none of its award to [PBS].” Canadian Claimants' Petition to Modify at 8. What the Panel should have done, according to Canadian Claimants, was to combine the Program Suppliers', JSC's, NAB's, Canadian Claimants' and PBS's awards before deducting the net awards to Music and Devotional Claimants.

4. Music Claimants

In determining the award to the Music Claimants, the CARP placed enough evidentiary weight on a study conducted by Sports Claimants' witness Dr. George Schink to use his distribution percentage as a “floor” in establishing the zone of reasonableness for Music Claimants' distribution percentage. Music Claimants argue that the CARP should have disregarded his testimony altogether. Additionally, Music Claimants charge that the Panel failed to give proper weight to the study it presented concerning music use from 1991/1992 to 1998/1999 and the witnesses it presented regarding increases in the use of music on broadcast programming from 1983 through 1999.

Music Claimants' main bone of contention with Dr. Schink's study is that he did not tailor it to the “unique characteristics of the distant signal market.” Music Claimants' Petition to Modify at 6. Instead, he used data concerning music licensing fees in the broadcast television industry that included television networks and local stations, both of which are not relevant under the section 111 license. According to Music Claimants, the network music licensing data dramatically and unfairly lowers their distribution percentages for 1998 and 1999. Moreover, Dr. Schink's study also varies considerably from the approach adopted by the Copyright Royalty Tribunal in the 1978 and 1979 distribution proceedings—comparing music licensing fees to broadcast television expenditures—which did exclude network licensing data. The CARP failed to “explain adequately why, after some twenty years, it has become appropriate to use Network data to determine Music's share in a market in which Network programming is not compensable.” Id. at 10.

Music Claimants also charge that the CARP acted arbitrarily by failing to recognize that music licensing fees are often paid on an interim basis while litigation in a rate court is pending and therefore do not reflect marketplace value. Dr. Schink should have used the fees that result from rate court proceedings, which he did not. The CARP did not determine this aspect of Dr. Schink's testimony to be defective because interim fees “might well exceed final fees.” Id. at 11, citing CARP Report at 87 n.58 (emphasis in original). Music Claimants submit that this conclusion is erroneous and not supported by the record. Further, Dr. Schink's study did not present any 1999 data. In sum, his entire study should have been disregarded.

Music Claimants also assert that the CARP failed to accord any weight to the testimony it presented regarding increased music use which is contrary to precedent from the 1983 distribution proceeding, the last litigated music award. “[T]he value of music is, at least in significant part, determined by the density of use [and] is consistent with the uncontradicted evidence before the CARP in this proceeding of how music license fees are set in the marketplace.” Id. at 15.

Scope of the Librarian's Review

Section 802(f) of the Copyright Act directs the Librarian of Congress, on the recommendation of the Register of Copyrights, to either accept the determination of a CARP or, if he rejects it, to substitute his own determination after a full examination of the record created in the proceeding. 17 U.S.C. 802(f). The Librarian can only reject a CARP's determination if he finds that it is arbitrary or contrary to one or more provisions of the Copyright Act. Id.

The standard of review of a CARP determination by the Librarian has been thoroughly discussed in prior proceedings for both royalty distributions and rate adjustments and will not be repeated here. See Distribution of 1990-92 Cable Royalty Funds, 61 FR 55653 (October 28, 1996); Rate Adjustment for the Satellite Carrier Compulsory License, 62 FR 55742 (October 28, 1997); Distribution of 1993-97 Cable Royalty Funds, 66 FR 66433 (December 26, 2001); Determination of Rates and Terms for the Digital Performance Right in Sound Recordings and Ephemeral Recordings, 67 FR 45240 (July 8, 2002). Suffice to say, the scope of review is limited and is highly deferential to the panel members who serve as factfinders in a proceeding and are in the best position to judge the credibility of testimony and weigh the evidence. The Librarian will “not second guess a CARP's balance and consideration of the evidence, unless its decision runs completely counter to the evidence presented to it.” 62 FR 55742, 55757 (October 28, 1997), citing 61 FR 55653 (October 28, 1996) (1990-92 Cable Royalty Fund Distribution Proceeding). Even if the Register and the Librarian would have reached different conclusions, the determination of the CARP will stand if it is not arbitrary or contrary to the Copyright Act. 63 FR 49823, 49828 (September 18, 1998) (Noncommercial Broadcasting Rate Adjustment Proceeding). In sum, if a CARP's determination falls within a “zone of reasonableness” the Librarian will not disturb it. National Cable Television Ass'n v. Copyright Royalty Tribunal, 734 F.2d 176, 182 (D.C. Cir. 1983).

The Program Suppliers' Award

1. The CARP's Approach

For almost 25 years, the distant signal viewing study (the Nielsen study) presented by the Program Suppliers has been credited by the CRT and the CARPs in determining royalty distributions in cable proceedings. In the early cable proceedings, the Nielsen study was the premier piece of evidence used to determine distributions. The CARP in this proceeding, however, noted an historical trend that has significantly decreased the preeminence of the Nielsen study. CARP Report at 33 (“Over the years, however, the CRT placed less reliance on the Nielsen study”). Indeed, it remarked that in the 1990-92 cable distribution “[f]or the first time, the Bortz survey was given greater weight than the Nielsen study.” Id. As a result of this observation, its construct of the hypothetical marketplace and its thorough examination of the Nielsen study and Bortz survey, “the Panel conclude[d] that the Nielsen study provides relevant viewing information but, as tacitly conceded by the [Program Suppliers] for the first time, without a means of translating viewing shares to value, the study does not afford an independent basis for determining relative value.” Id. at 44.

The devaluation of the Nielsen study is a result of the Panel's consideration of the hypothetical marketplace. In deciding how to determine the relative marketplace value, the only relevant criterion, of the six programming categories in this proceeding, the Panel hypothesized how the distant signal marketplace for cable operators would function in the absence of the section 111 license. The Panel concluded that in the traditional supply and demand paradigm, the supply side facing cable operators (i.e., the amount of distant broadcast programming available) is fixed, meaning that the supply of programming remains the same irrespective of the price. As a result of this, it is the demand side (i.e., cable operators) that will determine the relative value of programming. Consequently, evidence that demonstrated how cable operators valued each program category was, in the Panel's view, the best evidence of marketplace value.

After considering both the Bortz survey and the Nielsen study, the Panel concluded that the Bortz survey best measured the value of programming. The Nielsen study was not useful because it measured the wrong thing.

[T]he Nielsen study does not directly address the criterion of relevance to the Panel. The value of distant signals to [cable system operators] is in attracting and retaining subscribers, and not contributing to supplemental advertising revenue. Because the Nielsen study “fails to measure the value of the retransmitted programming in terms of its ability to attract and retain subscribers,” it cannot be used to measure directly relative value to [cable system operators]. The Nielsen study reveals what viewers actually watched but nothing about whether those programs motivated them to subscribe or remain subscribed to cable.

Id. at 38 (citations omitted). The Panel observed that apparently Program Suppliers themselves did not believe that raw Nielsen viewing data was determinative of marketplace value since they offered the testimony of Dr. Arthur Gruen who performed an “avidity” adjustment in an effort to show how a sample demographic of 18 to 49 year olds favored certain types of programs over others. The Panel analyzed Dr. Gruen's avidity adjustments and concluded that, due to conceptual and methodological flaws, it failed to provide the needed conversion from raw Nielsen viewing numbers to relative value.

However, unlike the Nielsen study, the Panel found the Bortz survey to be “an extremely robust (powerful and reliably predictive) model for determining [the] relative value” of Program Suppliers, Joint Sports Claimants and NAB for both the Basic Funds and the 3.75% Funds. Id. at 31. First, the survey addressed the correct question in the Panel's view: What is the relative value of different programming categories to cable operators? Second, the Panel considered and rejected the three conceptual limitations of the Bortz survey expressed by the 1990-92 CARP Panel. The Panel determined that the relative brevity of the interviews conducted by Bortz Media with cable system programmers did not seriously jeopardize the results or skew them in favor of one or more parties. The concern that the Bortz survey only measures the attitudes of cable system programmers rather than the actual behavior of cable systems was alleviated by the regression analyses conducted by Dr. Gregory Rosston which corroborated the Bortz survey results. And the concern that the Bortz survey did not take into account the supply side of programming in the supply and demand equation was not problematic because the Panel determined that the demand side of the equation dictated marketplace value. Finally, the Panel rejected the contention that the removal of broadcast superstation WTBS from the Bortz survey should have resulted in a considerable change in Bortz numbers from the 1990-92 proceeding thereby undermining the validity of the survey.

2. Program Suppliers' Arguments

Program Suppliers offer a host of arguments in opposition to the CARP's report, criticizing the Panel's awards to all parties with the exception of the Canadian Claimants. The heart of Program Suppliers' Petition to Modify is a fierce attack on the Panel's decision to accept the Bortz survey as a better determinative of marketplace value than the Nielsen study. Program Suppliers offer several reasons why the Panel's decision is arbitrary.

First, Program Suppliers charge that the Panel improperly abandoned long-established precedent that recognizes the Nielsen study to be indicative of the marketplace value of programming. According to Program Suppliers, the Panel only could deviate from precedent if it found changed circumstances or new evidence in this proceeding and neither of those conditions existed. Second, Program Suppliers argue that the Panel's determination to consider the marketplace value of distant broadcast signal programming from cable systems' perspective is contrary to precedent and the legislative intent of section 111.

Third, Program Suppliers submit that the Panel was wholly precluded from relying on the Bortz survey because of the short duration of the interviews conducted by Bortz Media, the attitudinal nature of the survey, the lack of the supply side perspective and the miscategorization of programs. Finally, Program Suppliers charge that the Panel simply ignored much of the testimony presented by its witnesses and improperly discredited Dr. Gruen's adjustments to the raw Nielsen data.

3. Recommendation of the Register

a. The role of Precedent With Respect to the Nielsen Study

Section 802(c) of the Copyright Act states that CARPs “shall act on the basis of * * * prior decisions of the Copyright Royalty Tribunal, prior copyright arbitration panel determinations, and rulings by the Librarian * * *” 17 U.S.C. 802(c). The concept of “precedent” therefore plays an important role in CARP proceedings. The CARP in this proceeding recognized that, devoting a lengthy discussion to it, and acknowledged that it “must accord precedential value to prior awards.” CARP Report at 13. Nonetheless, the Panel observed that prior decisions are not cast in stone and can be varied from when there are (1) changed circumstances from a prior proceeding or; (2) evidence on the record before it that requires prior conclusions to be modified regardless of whether there are changed circumstances. Id. at 14.

The Register agrees with the Panel's analysis of the role of precedent. As we stated in the 1990-92 cable distribution proceeding, while a Panel must take account of precedent it “may deviate from it if the Panel provides a reasoned explanation of its decision to vary from precedent.* * * It would make little sense to require the CARPs to apply Tribunal [and CARP] precedent in all circumstances, and allow no deviation, especially in the area of determining the relevant factors for distributing royalties.” 61 FR 55653, 55659 (October 28, 1996).

The Register disagrees with Program Suppliers' assertion that the CARP abandoned wholesale the role of the Nielsen study without adequate explanation. To the contrary, the Panel plainly articulated that the Copyright Royalty Tribunal placed less and less reliance on the importance of the Nielsen study over time and correctly observed that the CARP in the 1990-92 proceeding could not quantify the Nielsen data as evidence of market value. See 1990-92 Cable Royalty Distribution Proceeding, CARP Report at 44. It is the view of the Register that Program Suppliers overstate the precedential value of the Nielsen study. An examination of prior Phase I cable royalty distributions reveals that it is difficult, if not impossible, to determine precisely what evidentiary weight was given the Nielsen studies. It is clear, however, that the role of the Nielsen study, almost preeminent in the beginning, has eroded considerably through the years. See 47 FR 9879, 9892 (March 8, 1982) (1979 royalty distribution); 48 FR 9552, 9564 (March 7, 1983) (1980 royalty distribution); 51 FR 12792, 12808 (April 15, 1986) (1983 royalty distribution); 57 FR 15286, 15300 (April 27, 1992) (1989 royalty distribution). The Panel in this proceeding did nothing more than continue this trend and did so with a full explanation of its reasons.

Furthermore, the Panel did not completely disregard the Nielsen study. The Panel observed that “the Nielsen study provides relevant viewing information,” and held that it can “serve as a tool for assessing changed circumstances whenever the Bortz survey cannot be used.” CARP Report at 44 (footnote omitted). The Panel also noted that while raw Nielsen data is not indicative of marketplace value, it might be converted into such evidence through proper adjustments. That Dr. Gruen's adjustments failed to make that conversion does not rule out the possibility that it could be made appropriately in the future. Clearly, the rejection of the Gruen testimony does not amount to wholesale abandonment of the Nielsen study.

Finally, the Nielsen study in the record in this proceeding is not like the Nielsen study in prior proceedings. Contrary to Program Suppliers' assertion, there are changed circumstances from prior proceedings and this Nielsen study as adjusted by Dr. Gruen is arguably new evidence. The Panel thoroughly examined it and more than adequately explained its reasons why it did not find this Nielsen study to be persuasive evidence of marketplace value. Consequently, it is the Register's view that the Panel was not arbitrary in its application of precedent in this proceeding.

b. The hypothetical marketplace

To assist in determining the relative marketplace value of programming in this proceeding, the CARP posited a hypothetical marketplace in which no statutory license exists and examined the factors that would likely control the valuation of programming. Applying traditional supply and demand analysis to the hypothetical marketplace, the Panel determined, based on record testimony, that the supply side of distant broadcast programming would remain fixed. Written Rebuttal Testimony of Dr. Andrew Joskow at 8. Because the supply of programming in such a market would remain fixed, value would be determined by the buyer side, i.e., cable operators purchasing distant broadcast signals. According to the Panel, programming is significant to cable operators for its ability to attract and retain subscribers. In the Program Suppliers' view, this description of the hypothetical marketplace is fundamentally flawed, produces absurd results, and must be rejected. The Register does not agree.

While this is the first cable distribution CARP to describe in detail its construct for determining marketplace value, it is not the first time the economic factors comprising the discussion of the hypothetical marketplace have been addressed. The Bortz survey, a longtime mainstay of cable distribution proceedings, has always attempted to quantify how cable operators would buy programming in a marketplace in which the cable license did not exist. By deeming the Bortz survey as relevant to the value of distant signal programming, the 1990-92 cable distribution CARP and the CRT were necessarily accepting the assumptions of its construct. Neither the prior CARP nor the Tribunal ever concluded that the Bortz survey operated from false assumptions or asked the wrong questions. It therefore cannot be said that the CARP in this proceeding manufactured an economic theory out of thin air. While Program Suppliers may disagree with the Panel's consideration of the hypothetical marketplace and in particular its conclusion that it is the perspective of cable operators that best determines how much different categories of programming would be worth, the Panel's actions are based on prior decisions.

The Register also recommends rejection of Program Suppliers' contention that determining marketplace value from cable operators' perspective runs counter to the legislative intent of the cable license. While it is accurate to observe that the section 111 license is intended to compensate copyright owners for the use of their works, Program Suppliers erroneously assert that the use of copyrighted works must be determined by their viewing. Other methods may, and have, been appropriately employed. As the CRT has stated “viewing per se [does] not necessarily correspond to marketplace value.” 57 FR 15286, 15301 (April 27, 1992). The Panel's decision to give greater weight to methodologies that quantify marketplace value other than from the perspective of viewing is not contrary to legislative intent.

c. Consideration of the Bortz Survey

Program Suppliers contend that the Bortz survey should have been rejected outright by the Panel because of four fundamental flaws: the interviews Bortz Media conducted with cable operator programmers were too short; the Bortz survey measures attitudes about programming and not actual behavior in the buying of programming; the survey fails to consider the supply side of distant broadcast programming; and the survey contains numerous program miscategorizations that render its results useless. For the reasons described below, none of these arguments preclude the Panel from accepting the results of the Bortz survey.

1. Short duration of interviews. The CARP in this proceeding addressed the criticism of the Bortz survey leveled by the 1990-92 cable distribution CARP that the interviews conducted by Bortz Media with cable system programmers were too short to be accurate and concluded that “[t]hough the interviews are relatively brief, the Panel does not believe the execution of the survey seriously jeopardizes the integrity of the Bortz survey results.” CARP Report at 20. This conclusion is specifically grounded by the Panel in record evidence. See, id. (testimony of witnesses Egan, Crandall, Fuller and Allen). When a CARP's determination with respect to a particular point is grounded in record evidence, the Register will not second guess it. 67 FR 45239, 45253 (July 8, 2002) (“Where such determinations are based on testimony and evidence found in the record, the Register and the Librarian must accept the Panel's weighing of the evidence and its determination * * * ”).

2. Attitudes v. behavior. Another criticism of the Bortz survey by the 1990-92 cable distribution CARP was that the Bortz survey measured the attitudes of cable system programmers as opposed to their actual behavior in purchasing distant broadcast signals. The Panel in this proceeding, however, concluded that such a criticism was not valid, stating that “uncontroverted testimony and years of research indicate rather conclusively that constant sum methodology, as utilized in the Bortz survey, is highly predictive of actual marketplace behavior.” Id. at 21. This statement is based on the testimony of Dr. Debra Ringold, a Canadian Claimants' witness who testified on the use of constant sum methodologies. In addition, the regression analysis conducted by Dr. Rosston, which did measure actual behavior, corroborated the results of the Bortz survey. Because the CARP's determination is record based, there are no grounds to disturb it.

3. The supply side perspective. Regarding the 1990-92 CARP's criticism of the lack of a supply side perspective, the CARP in this proceeding acknowledged that while the Bortz survey does not take into consideration the supply side of the supply and demand paradigm, the supply side perspective was not important because the Panel determined that in the hypothetical marketplace it was considering, the supply of distant broadcast programming is fixed and therefore does not determine the value of the programming (programming is determined from the demand side, i.e., the cable system side). As discussed above, the Panel's discussion of the hypothetical marketplace is not arbitrary. Further, its conclusion that the supply side of distant broadcast programming remains fixed is based on record testimony. See Written Rebuttal Testimony of Dr. Andrew Joskow at 8.

4. Program miscategorization. Unlike its first three criticisms of the Bortz survey, program miscategorization was not identified by the 1990-92 cable distribution CARP as a potential limitation to the accuracy or usefulness of the Bortz survey. Program miscategorization, according to Program Suppliers, is the failure by cable system programmers to accurately identify the correct program categories (syndicated series and movies, sports, devotional programming, etc.) for individual programs when completing their Bortz Media surveys. Program Suppliers point to the testimony of JSC witness Michael Egan who, though he could not remember having completed a Bortz Media survey in the past, was questioned by Arbitrator Michael Young as to how he would categorize certain types of programs. Egan Tr. at 1334. Program Suppliers categorize two of his responses as incorrect thereby conclusively demonstrating, in Program Suppliers' view, that miscategorization of programs by respondents to Bortz Media surveys is considerable and invalidates the results.

The Panel did not specifically address the matter of miscategorization of specific programs, apparently determining that it was not an impairment to the results yielded by the Bortz survey. This is not surprising for two reasons. First, the Panel was not presented with evidence that demonstrated sufficiently widespread miscategorization of programs by Bortz Media respondents that would likely affect the survey results. Mr. Egan's responses to Arbitrator Young reflect only how he might respond and were offered by someone who could not recall if he had ever completed a Bortz Media survey. Second, and more importantly, the Bortz Media surveys do not question cable operators as to individual programs, but rather question them as to the value they attach to categories of programs. See Trautman Tr. at 324-25 (Respondent are “not thinking about each and every program that is aired on that signal. They are thinking about the general categories of program.”). If Program Suppliers pointed to evidence that demonstrated that Bortz Media respondents misapprehended entire categories of programs when assigning them value, then the Panel might have been required to address such contentions. That is not the case here, and consequently the Panel did not act arbitrarily in considering the evidence presented regarding program miscategorization.

d. Consideration of the Nielsen Study

Program Suppliers contend that the CARP improperly ignored the weight to be given the Nielsen study contrary to precedent, unfairly criticized Dr. Gruen's adjustments to the raw Nielsen viewing data, and ignored most of the evidence that Program Suppliers put forth regarding the marketplace value of distant broadcast signal programming. None of these contentions require rejection of the CARP Report.

The role of precedent in CARP proceedings is discussed above. There is no requirement that automatic weight must be assigned to the Nielsen study. The Panel is required to examine the evidence on the record before it and may deviate from what the CRT or prior CARPs have done provided that it provides a reasoned explanation. This CARP did provide a reasoned and detailed explanation as to why the Bortz survey was more persuasive evidence of marketplace value than the Nielsen study. The Panel did not “abandon” the Nielsen study but instead continued a trend from prior decisions that placed less and less reliance on the weight to be accorded the Nielsen study. That Nielsen is less persuasive than Bortz is undoubtedly upsetting to Program Suppliers, but that result is supported by the evidence. Whether the Register or the Librarian might have attached greater evidentiary weight to the Nielsen study is irrelevant where the Panel's weighing of the evidence is supported by the record.

The Nielsen study presented in this proceeding is also not the same as in prior proceedings. This Nielsen study contains the adjustments performed by Dr. Gruen in an effort to convert raw viewing data into direct evidence of marketplace value. In performing his adjustments, Dr. Gruen focused on the viewing data for the 18-49 age demographic because he believed that this age group of cable subscribers was the most likely to buy the new ancillary and digital services offered by cable systems. Gruen Written Direct Testimony at 16-22. The Panel disagreed with Dr. Gruen's testimony on this point, agreeing instead with the testimony presented by several other witnesses that additional demographic categories are relevant. Once again, the CARP is in the best position to weigh the testimony of witnesses, and neither the Register nor the Librarian should second guess it. 62 FR 55742, 55757 (October 28, 1997). The Panel also disagreed with the mechanics of Dr. Gruen's avidity adjustment which attempted to show the loyalty of viewers to particular types of programs as an indication of their marketplace value. The Panel found the avidity adjustment to be flawed “both conceptually and methodologically” and rejected it based on its own analysis and the testimony of other witnesses. CARP Report at 42. There is nothing arbitrary about the Panel's approach or its conclusions.

Finally, Program Suppliers argue that the Panel ignored altogether the evidence they presented in this proceeding on marketplace value and evaded its responsibility to evaluate the testimony of each of their witnesses in the Report. Program Suppliers point to the following statement of the CARP as evidence of arbitrary decision making:

[I]n this Report the Panel attempts to articulate only the principal grounds upon which our determinations are based. Of course, at arriving at these determinations, the Panel has carefully reviewed and considered all of the parties' evidence and arguments. To the extent this Report comports with a particular contention of a party, we accept that contention. To the extent that it does not, we reject that contention.

CARP Report at 7. The Register rejects Program Suppliers' contention that a CARP must articulate its consideration of every piece of evidence presented to it. To the contrary, the Copyright Act requires that the Panel set forth the facts it found relevant to its determination, not all the facts that were presented to it. 17 U.S.C. 802(e). Indeed, the cases cited by Program Suppliers in its Petition to Modify, Permian Basin Area Rate Cases, 390 U.S. 747 (1968), City of New York v. FCC, 814 F.2d 720 (D.C. Cir. 1987), and Motor Vehicle Mfrs. Ass'n et al. v. State Farm Mutual, 463 U.S. (1983), require that a decision-making body must consider the pertinent factors and the important aspects of the problem it is facing, not that it consider and resolve (much less articulate) all the evidence presented to it. The CARP in this proceeding fulfilled its obligation by carefully and precisely describing its rationale for preferring the Bortz survey over the Nielsen study and did not arbitrarily disregard relevant evidence.

The PBS's Award

1. The CARP's Approach

PBS requested a distribution of 12% of the Basic Fund for the 1998 and the 1999 cable royalties. PBS Proposed Findings of Fact and Conclusions of Law at 138-139. In support of its claim, PBS attempted to demonstrate to the CARP that circumstances had changed considerably in its favor from the 1990-1992 CARP proceeding wherein it received 5.5% of the Basic Funds for those three years. PBS presented a study conducted by Dr. Leland Johnson which attempted to show a relationship between the relative number of “distant subscriber instances” to PBS signals and the relative marketplace value of the programming carried on those signals. Dr. Johnson's original study sought to compare the number of distant subscriber instances of PBS programming in 1989 with those in 1999 but later adjusted his study to focus on observations for 1998 and 1999 without reliance on changes from earlier periods. Dr. Johnson concluded that if it is assumed that cable operators valued all distant subscriber instances equally, PBS would be entitled to an award of royalties equal to its share of distant subscriber instances. Id. at 4; Tr. 9196 (Johnson).

The CARP rejected Dr. Johnson's studies:

Both subscriber instances studies offered by Dr. Johnson suffer from the same fundamental infirmity-they attempt to equate relative programming volume with relative programming value. Furthermore, Dr. Johnson's fundamental premise that [PBS] signals are at a level of “parity” with other signals is contradicted by substantial record evidence, including the Rosston regression analyses. * * *

We view Dr. Johnson's change in subscriber instances theory as relatively unuseful because it is based on a measure of time, not value.

CARP Report at 56-57 (emphasis in original). Instead, the CARP looked to alternative methods to establish PBS's distribution awards. It considered the Bortz survey numbers for PBS but, unlike for Program Suppliers, JSC and NAB, found some methodological flaws that disadvantaged PBS. Specifically, it found that PBS programming was undervalued in the Bortz survey because cable systems that carried PBS as their only distant signal were removed from the survey and because cable systems that did not carry any PBS stations on a distant basis automatically assigned a zero value for PBS programming. Id. at 22-23. The CARP therefore determined that PBS's Bortz number of 3.2% for 1998 and 1999 established the “floor” to a PBS award and that the value of PBS programming “is somewhere above 3.2%.” Id. at 26. The CARP then examined the royalty fees actually paid by cable operators in 1998 and 1999 for distant PBS signals—the fee generation approach—and attributed “some weight [to it], along with the Bortz floor and changed circumstances,” in determining PBS's award. Id. at 64. The Panel then considered the evidence regarding changed circumstances from the 1990-92 CARP proceeding and concluded that “there is no persuasive evidence that [PBS's] relative value has significantly either increased or decreased since 1990-92.” Id. at 69. Consequently, the Panel awarded PBS the same distribution percentage it received for 1991 and 1992 from the 1990-92 proceeding for both 1998 and 1999.

2. PBS's Arguments

PBS finds three fundamental errors with the CARP report: it uses discredited evidence to refute Dr. Johnson's studies; it treats PBS differently from NAB; and it violates precedent by placing “some weight” on the fee generation method.

PBS's discredited evidence argument is centered on the Panel's analysis and use of the Bortz survey with respect to PBS. The Panel correctly determined, in PBS's view, that the Bortz survey results were inherently biased against PBS and understated the value of PBS programming. However, “in flat contradiction of its own ruling that the Bortz results were “inherently biased” and could not be used to value [PBS], the Panel then relied on those very same Bortz results to dismiss the relevance of the dramatic four-fold increase in [PBS's] viewing share.” PBS Petition to Modify at 3. Specifically, PBS points to the Panel's consideration of changed circumstances for PBS from 1990-92 to this proceeding wherein the Panel observed that while PBS's distant subscriber instances share had gone up, its Bortz survey share remained the same, in contrast to NAB whose distant subscriber instances share and Bortz survey share had both gone up. CARP Report at 66. PBS charges that it was illogical and inconsistent for the Panel to make this observation, particularly where the Panel had previously concluded that the Bortz survey was more biased against PBS during the 1998-99 period than it was during the 1990-92 period. Id. at 22-23. PBS also submits that the Panel failed to consider PBS's Nielsen viewing data at all despite the fact that it had ruled that the “Nielsen studies can serve as a tool for assessing changed circumstances whenever the Bortz survey can not be used.” Id. at 44.

PBS argues that the Panel treated PBS disparately relative to NAB. Specifically, the Panel found that the increase of NAB's viewing share from 8 percent to 14.7 percent between the 1990-92 and 1998-99 proceedings “was apparently perceived as increased value by [cable operators] as confirmed by their responses to the Bortz study,” which also reflected significant increases. However, “[i]n sharp contrast to its treatment of NAB, the Panel found that the quadrupling of [PBS's] viewing share did not establish any increase in [PBS's] relative value.” PBS Petition to Modify at 11. “Such “disparate treatment of similarly situated parties” is a classic example of arbitrary action that demands a remedy.” Id. at 12.

Finally, PBS submits that the Panel's decision to afford “some weight” to the fee generation approach is “contrary to 20 years of precedent, logic, and the record in this case-all of which established that “fees generated” are not a proper measure of market value.” Id. at 13.

3. Recommendation of the Register

Unlike the awards to Program Suppliers, Joint Sports Claimants, NAB, and Canadian Claimants which where determined by use of a particular distribution methodology, the award to PBS was accomplished through consideration of a number of factors: the Bortz survey alone to establish a floor of 3.2%; “some weight” attributed to the fee generation approach which implied an award of 3.9%; and an examination of PBS's changed circumstances from the 1990-92 proceeding (wherein it received 5.49125%) to 1998-99. PBS asserts in its first argument, described above, that once the Panel used the Bortz survey to establish the floor value of PBS's award, it was precluded from considering any aspects of the survey in evaluating the changed circumstances from the 1990-92 to 1998-99 proceedings. The Register disagrees with this argument and concludes that it does not render the CARP decision arbitrary.

Contrary to PBS's assertion that the Panel did not consider PBS's Nielsen viewing shares after stating earlier in its report that it would do so, the Panel plainly observed that PBS's and NAB's Nielsen viewing shares (and their share of distant subscriber instances) had “dramatically” increased from 1990-92 to 1998-99. CARP Report at 66. The Panel then attempted to determine why this might have happened. It resolved that these increases were due to the elimination of superstations WTBS and WWOR from the cable royalty funds which accounted for a large portion of the viewing shares attributable to Program Suppliers. Id. at 66. The windfall to NAB and PBS in viewing shares did not, of course, automatically mean that the value of PBS's and NAB's programs went up as well since the Panel expressly concluded that viewing shares (and distant subscriber instances) do not measure program value. The Panel then noted that while both NAB's and PBS's viewing numbers (and distant subscriber instances) went up, only NAB showed a concomitant increase in its Bortz share between 1992 and 1998, while PTV did not: NAB's Bortz share increased 19% from 1992 to 1998 while PBS's went down from 3.0% in 1992 to 2.9% in 1998 and 1999. Id. Had the Panel stopped here and concluded that the value of NAB had gone up while the value of PBS programming remained the same, then PBS's argument that the Panel improperly used the Bortz survey might be persuasive. But the Panel did not stop there and undertook an examination of why PBS's Bortz numbers did not track the same type of path as NAB's given the increased viewing shares and distant subscriber instances to both. The Panel considered the two flaws in the Bortz survey for PBS-elimination of cable systems carrying only a distant PBS station and zero value to PBS programming for cable systems not carrying a distant PBS station—and determined that they did not by themselves explain the lack of a PBS Bortz survey increase. Id. (“While lack of increase in [PBS's] Bortz share might be explained partially by the elimination of [PBS]—only systems from the survey (which had a real impact for the first time in 1998), that factor certainly can not explain it fully”); id. at 66 n.36 (“The other anti-[PBS] bias (assignment of automatic zeroes) should not differentially affect the studies for either period.”). The Panel then went on to consider other factors that might explain PBS's lack of an increase in Bortz share from 1992 to 1998 such as fierce competition from cable “look-alike” networks and increased carriage of distant PBS signals due to FCC-mandated must-carry rules as opposed to an increase in value of distant PBS stations to cable operators. These considerations led the Panel to conclude that “despite th[e] relative growth of [PBS] [in Nielsen viewing share and distant subscriber instances share], constancy in the raw Bortz shares from 1992 to 1998 likely reflects the net marketplace impact of all these circumstances.” Id. at 68 (footnote omitted). This conclusion is grounded in record evidence, and the Register will not recommend that it be disturbed. See, 62 FR 55742, 55749 (October 28, 1997)(“Because this conclusion is grounded in the record, it is not arbitrary.”)

The Register also recommends that PBS's argument that it is being treated disparately vis-a-vis NAB is not persuasive. PBS creates the misperception that the Panel used NAB's doubling in Nielsen viewing share from 1990-92 to 1998-99 as the justification for increasing NAB's award. This is incorrect. NAB received its award based solely on the shares it received in the Bortz survey, as corroborated by the Rosston regression analysis. See CARP Report at 50-51. It was only after the Panel firmly concluded that the Bortz survey was the methodology to determine NAB's share that it made the statement that NAB's doubling in Nielsen viewing share “was apparently perceived as increased value by [cable system operators] as confirmed by their responses in the Bortz study.” Id. at 51. This anecdotal observation merely confirmed what the Panel already determined: NAB would receive its Bortz survey shares. PBS's Nielsen viewing share was considered by the Panel but it, like the Bortz survey, did not play a decisive role in determining PBS's award. PBS and NAB are not similarly situated parties; consequently, the Panel did not treat them disparately.

Finally, the Register concludes the Panel's affording “some weight” to PBS's fee generation numbers does not fly in the face of 20 years of precedent, logic and the record. The Panel duly noted that the CRT previously took a dim view of using the fee generation method, but did use it to exclude PBS from sharing in the 3.75% fund and used it in the 1989 cable royalty distribution proceeding to reduce PBS's award. Further, the 1990-92 CARP expressly used the fee generation approach in determining the Canadian Claimants' award, a point which PBS reluctantly admits. While PBS adamantly opposes using the fee generation method for itself and others, there does exist precedent for using it. Furthermore, the Panel addressed and rejected PBS's testimony as to why the fee generation method was not appropriate, determining that while it is true that fees generated do not measure the absolute value of programming, it does not mean that they are not capable of measuring the relative value of programming between the claimant groups. Id. at 63-64. Nevertheless, the Panel elected not to accord full weight to the fee generation approach with respect to PBS; this clearly was within its discretion. See, Nat'l Ass'n of Broadcasters v. Librarian of Congress, 146 F.3d 907, 923 n.13 (The CARP is in the best position to weigh evidence and gauge credibility).

In sum, the Panel's treatment of PBS comports with its stated approach for determining a party's award that cannot be derived through application of a particular distribution methodology: examine that party's changed circumstances from its 1990-92 distribution award by examining the available record evidence. CARP Report at 16. There is nothing arbitrary to the approach or its application to PBS in this proceeding.

The Canadian Claimants' Award

1. The CARP's Approach

The Canadian Claimants requested the following distribution percentages: for 1998, 2.25479% of the Basic Fund and 0.17332% of the 3.75% Fund; for 1999, 2.48141% of the Basic Fund and 0.43023% of the 3.75% Fund. The Canadian Claimants principally rely on a “fee generation” approach—the section 111 royalties paid by cable operators for distant retransmission of Canadian signals—although they cite changed circumstances to corroborate the substantial increase requested from their 1990-92 distribution percentages. Through an analysis of the volume of Canadian programming contained on Canadian broadcast signals and application of a constant sum survey, similar to the Bortz study, the Canadian Claimants' requested distribution percentages are based on their conclusion that approximately 70% of all programming contained on Canadian broadcast signals belongs to them; thus, they request 70% of the fees generated by Canadian signals.

The CARP generally accepted the Canadian Claimants' fee generation approach with some exceptions. Since there are no Bortz survey results for Canadian programming, the CARP used the award adopted in the 1990-92 proceeding as a reference point since it, too, was based on the fee generation approach. The Panel did not find any changed circumstances that merited an increase in the Canadian Claimants' award, other than the fact that Canadian signals generated substantially more revenues in 1998-99 than they did in 1990-92. As a result, Canadian Claimants received their fee generated award.

2. The Canadian Claimants' Arguments

The Canadian Claimants do not dispute the fee generation approach utilized by the CARP. Rather, they dispute the way in which their award was incorporated into the CARP's mathematical approach for establishing final distribution percentages. As discussed earlier in this Order, the CARP was cognizant that each party's distribution award could not be determined in a vacuum. Since different distribution methodologies were being employed to determine awards, adjustments must be made so that all awards when aggregated would equal the total royalty pools available. The CARP's mathematical approach to make all awards equal 100% of the funds is detailed in Appendix A of its report. Canadian Claimants' objection comes with respect to how its award was adjusted to account for the “net” award to PBS, the Music Claimants and the Devotional Claimants.

The gravamen of the Canadian Claimants' petition to modify is this: its award should have been combined with Program Suppliers, JSC, NAB and PBS before adjusting for the “net” awards to Music Claimants and Devotional Claimants. The Canadian Claimants submit that such result is fair for the following reasons. First, since the Panel adopted a fee generation approach for Canadian Claimants, they should receive precisely the percentages due them under that approach. The Panel's approach robs them of their full fee generation share and is contrary to the methodology the Panel stated that it was employing. Second, the fact that PBS received a “net” award from the Panel is unfair to Canadian Claimants particularly where there is no PBS programming on Canadian broadcast signals. Third, the Panel's mathematical approach described in Appendix A of its report took the Bortz survey results of Program Suppliers, JSC and NAB and adjusted them up to 100% before applying a pro rata reduction to those awards to account for the “net” awards to PBS, Music Claimants and Devotional Claimants. Canadian Claimants are forced to share in the pro rata reduction to account for the “net” awards, but did not share in the upward adjustment enjoyed by Program Suppliers, JSC and NAB.

3. Recommendation of the Register

In the 1990-92 cable distribution proceeding, the Librarian was called upon to make a “mathematical adjustment” to the distribution percentages of the Canadian Claimants for the 1991 and 1992 3.75% Funds. In that proceeding, the CARP intended to award Canadian Claimants its fee generation percentage of the 3.75% Funds, just as this CARP has intended to do. However, in the 1990-92 proceeding, the CARP failed to account for the fact that there are other program categories represented in the 3.75% royalties generated by distant Canadian broadcast stations. This omission, which the Panel later admitted was an error, necessitated a mathematical adjustment to the Canadian Claimants' 3.75% awards for 1991 and 1992 to account for the two other program categories (Program Suppliers and JSC) represented on Canadian signals. As a result, Canadian Claimants' distribution percentages for the two funds decreased slightly. See 61 FR at 55663.

In this proceeding, another “mathematical adjustment” is requested—this time in Canadian Claimants' favor. Unlike the previous proceeding, however, no adjustment is required here. The Register concludes that the Panel did not act arbitrarily in choosing the method that it did to reconcile all awards to equal 100% of the royalty pools. Some method of reconciliation was necessary because the Panel did not employ the same distribution methodology for all parties. Three of the parties—PBS, Music Claimants and Devotional Claimants—received “net” distribution awards because the Panel was unable to adopt a specific distribution methodology to calculate their awards. CARP Report at 69 n. 42. The remaining parties' shares were derived by use of particular methodologies and their shares were reduced pro rata to account for the net awards. While the methodology-based parties do surrender a portion of their award to account for the others, it was not impermissible for the Panel to do this. It is true that the Panel could have chosen not to give a “net” award to either PBS or Music Claimants (or both) and made a pro rata reduction in those awards as well when it accounted for the entire distribution. Canadian Claimants submit that such an approach is particularly applicable to PBS since there is no public television programming contained on Canadian broadcast signals. But while the Panel could have adopted this approach, it was not compelled to do so. A decisionmaker's choices between a number of reasonable alternatives cannot be considered arbitrary. Georgia Indus. Group v. FERC, 137 F. 3d 1358, 1364 (D.C. Cir. 1998). “The Register will not consider what the Panel could have done or what a party asserts it should have done, even if, had she heard th[e] proceeding in the first instance, she would have chosen another methodology.” 63 FR 49823, 49829 (September 18, 1998).

The Music Claimants' Award

1. The CARP's Approach

Of all the awards made in this proceeding, it appears that the Panel was most troubled in establishing an award for the Music Claimants because of a lack of reliable evidence upon which to base the distribution. The Music Claimants did not participate in the 1990-92 distribution proceeding, instead settling for 4.5% of all three Funds. In this proceeding, they requested an award of 5.0% of each of the Basic Fund, the 3.75% Fund and the Syndex Fund. The Music Claimants' request for an increase is premised upon a music use study that purports to show an 11% increase in the use of music on distant signals between 1991-92 and 1998-99.

The CARP found the music use study to be unpersuasive and of no value. Instead, the CARP considered the study presented by Joint Sports Claimants' witness Dr. George Schink who compared the amounts of licensing fees that Music Claimants receive from broadcasters and cable networks outside of the statutory licensing scheme with the total programming expenses of those broadcasters and cable networks. Based on his study of broadcasters and cable networks, Dr. Schink concluded that the Music Claimants' 1998-99 share should be no higher than 2.33%. The Panel used this figure to establish the floor to the zone of reasonableness to fixing the Music Claimants' award (similar to the way in which the Panel used PBS's Bortz survey share to establish the floor for its award) but did not accept it fully because the study included fees paid by television networks who are not compensated under the section 111 licensing scheme. The Panel then looked to the last litigated net award for Music Claimants from the 1983 distribution proceeding—4.5%—and used that figure to establish the ceiling to the zone of reasonableness for the Music Claimants' award. The Panel then concluded that 4.0% of each of the three Funds was the appropriate distribution percentage.

2. The Music Claimants' Arguments

The Music Claimants argue that the CARP failed to properly consider the evidence they presented in this proceeding and should have wholly discarded the testimony of Dr. Schink. With respect to the music use study they presented, Music Claimants argue that “[t]he CARP gave insufficient weight to the testimony of ASCAP's Chief Economist, Dr. Peter Boyle, and BMI's witness, Frank Krupit, concerning the value of the [music use] study.” Music Claimants Petition to Modify at 5. Music Claimants also charge that the CARP improperly gave no weight to the testimony of three of their witnesses who testified that the use of music in broadcast programming had dramatically increased from 1983 through 1999. Music Claimants also charge that the CARP ignored established precedent that music use is the way to determine the marketplace value of music.

With respect to Dr. Schink's study, Music Claimants charge that it is fatally flawed for three reasons. First, his inclusion of non-compensable network programming artificially depressed Music Claimants' distribution percentage. Second, his calculation was based in part on interim music licensing fees that do not reliably reflect the market value of music in the relevant years; and third, he presented no data for 1999. As a result of these flaws, and coupled with the fact that Dr. Schink's testimony was not presented until the rebuttal phase of this proceeding, Music Claimants submit that his testimony should have been completely disregarded.

3. Recommendation of the Register

Music Claimants' arguments in their Petition to Modify all suffer from the same flaw: they ask the Librarian to re-weigh the evidence. As we have made clear in this proceeding and others, the Librarian will not second guess a CARP and recast the evidence. “[T]he Librarian's scope of review is very narrow. This limited scope certainly does not extend to reconsideration of the relative weight to be accorded particular evidence, and the Librarian will not second guess a CARP's balance and consideration of the evidence, unless its decision runs completely counter to the evidence presented to it.” 61 FR at 55663 (October 28, 1996). The CARP, not the Register or the Librarian, “is in the best position to weigh evidence and gauge credibility.” NAB v. Librarian of Congress, 146 F.3d at 923 n.13 (D.C. Cir. 1998). Only if a CARP acts in complete contravention of the evidence and with no rational basis is the Librarian forced to reconsider the evidence. That is not the case here.

If the CARP in this proceeding had fully credited Dr. Schink's study and used it as the basis for determining Music Claimants' award, then Music Claimants' protestations might require intervention by the Librarian. But the Panel did not fully credit Dr. Schink's study, as Music Claimants reluctantly admit, and acknowledged the very flaws in the study that Music Claimants discuss in their Petition to Modify. See CARP Report at 84-87. Although the Panel explained its reservations about the Schink analysis, it found the study to be useful enough in establishing the minimum to Music Claimants' award. The Panel was well within its discretion to use the Schink study in this fashion.

Likewise, the CARP was well within its discretion to discount the testimony of three of Music Claimants' witnesses: ASCAP's Seth Saltzman, television and film critic Jeffrey Lyons and music composer W.G. “Snuffy” Walden. The testimony of these witnesses centered on their personal observations regarding a perceived increase in the use of music, particularly theme music, on broadcast television programming in recent years. Music Claimants submit that because the testimony of these witnesses was (in their opinion) unrebutted by other testimony, the CARP was compelled to accord it weight. This is not correct. The CARP is vested with discretion to gauge the credibility of witnesses, NAB v. Librarian of Congress, 146 F.3d at 923 n.13, regardless of whether other parties put forward other witnesses to specifically rebut it. The Panel stated that the testimony of these witnesses was “anecdotal and subjective opinion,” and that “[a]bsent quantitative corroboration, the Panel is unable to credit significantly this evidence.” CARP Report at 75 n.46. This determination is within the discretion of the Panel and is not arbitrary. See, also 62 FR 55742, 55751 (October 28, 1997) (satellite royalty rate adjustment).

Finally, the Register does not agree with Music Claimants' contention that music use is the only way to determine the market value of music, and that “[t]he CARP ruled, contrary to all applicable music licensing precedent and without adequate explanation, that changes in music use were not relevant to the establishment of Music's award for 1998-99.” Music Claimants' Petition to Modify at 5. This is not what the Panel said. Rather, the Panel found that Music Claimants' music use study failed to accurately demonstrate an increase in the use of music from the relevant starting point of 1983 (the time of the last litigated Music award) to 1998-99 because the data relied upon by Music Claimants “is too incomplete to provide reliable estimates.” CARP Report at 82. The Panel did not say that music use was irrelevant; it accepted Dr. Schink's criticisms of the Music Claimants' study. That the CARP did not use data that focused on music use is not a rejection of music use per se; rather it was a rejection of the evidence of music use presented by Music Claimants.

Order of the Librarian of Congress

Having duly considered the recommendation of the Register of Copyrights regarding the report of the Copyright Arbitration Royalty Panel in the Phase I distribution of the 1998 and 1999 cable royalty funds, the Librarian of Congress adopts her recommendation to accept in full the Panel's determination. For the reasons stated in the Register's recommendation, the Librarian is exercising his authority under 17 U.S.C. 802(f) and is issuing this order setting forth the distribution of royalties. After deducting National Public Radio's 0.18% share for each year per its agreement with the other parties to this proceeding, it is ordered that the 1998 and 1999 cable royalties shall be distributed according to the following percentages: