[Federal Register Volume 63, Number 181 (Friday, September 18, 1998)]
[Rules and Regulations]
[Pages 49823-49837]
From the Federal Register Online via the Government Publishing Office [www.gpo.gov]
[FR Doc No: 98-24986]


=======================================================================
-----------------------------------------------------------------------

LIBRARY OF CONGRESS

Copyright Office

37 CFR Part 253

[Docket No. 96-6 CARP NCBRA]


Noncommercial Educational Broadcasting Compulsory License

AGENCY: Copyright Office, Library of Congress.

ACTION: Final rule and order.

-----------------------------------------------------------------------

SUMMARY: The Librarian of Congress, upon the recommendation of the 
Register of Copyrights, is announcing the rates and terms of the 
noncommercial educational broadcasting compulsory license for the use 
of music in the repertoires of the American Society of Composers, 
Authors and Publishers and Broadcast Music, Inc. by the Public 
Broadcasting Service, National Public Radio and other public 
broadcasting entities as defined in 37 CFR 253.2, for the period 1998-
2002. The Librarian is adopting

[[Page 49824]]

the determination of the Copyright Arbitration Royalty Panel (CARP).

EFFECTIVE DATE: January 1, 1998.

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, S.E., Washington, 
D.C. 20559-6000. It is also available on the Copyright Office's 
website: (http://lcweb.loc.gov/ copyright/carp).

FOR FURTHER INFORMATION CONTACT: David O. Carson, General Counsel, or 
William J. Roberts, Jr., Senior Attorney for Compulsory Licenses, P.O. 
Box 70977, Southwest Station, Washington, D.C. 20024. Telephone (202) 
707-8380.

SUPPLEMENTARY INFORMATION:

I. Background

    Section 118 of the Copyright Act, title 17 of the United States 
Code, creates a compulsory license for the public performance of 
published nondramatic musical works and published pictorial, graphic 
and sculptural works in connection with noncommercial broadcasting. 
Terms and rates for this compulsory license, applicable to parties who 
are not subject to privately negotiated licenses, are published in 37 
CFR part 253 and are subject to adjustment at five-year intervals. 17 
U.S.C. 118(c). As stipulated by the parties, the terms and rates 
adopted in today's order are effective for the period beginning January 
1, 1998. They will be effective through December 31, 2002.
    The noncommercial educational broadcasting compulsory license 
provides that copyright owners and public broadcasting entities may 
voluntarily negotiate licensing agreements at any time, and that such 
agreements will be given effect in lieu of any determination by the 
Librarian of Congress provided that copies of such agreements are filed 
with the Register of Copyrights within 30 days of their execution. 
Those parties not subject to a negotiated license must follow the terms 
and rates adopted through arbitration proceedings conducted under 
chapter 8 of the Copyright Act.
    The Library published a notice in the Federal Register requesting 
comments from interested parties as to the need of a CARP proceeding to 
adjust the section 118 terms and rates. 61 FR 54458 (October 18, 1996). 
After a protracted negotiation period, several parties submitted 
proposals for royalty fees and terms with respect to certain uses by 
public broadcasting entities of published musical works and published 
pictorial, graphic and sculptural works. The Library published these 
proposals in the Federal Register, in accordance with 37 CFR 251.63, 
and adopted them as final regulations effective January 1, 1998. See 63 
FR 2142 (January 14, 1998).
    Certain parties notified the Library that agreement could not be 
reached for the use of musical works and that a CARP would be required. 
The Library initiated a CARP proceeding on January 30, 1998, and the 
CARP delivered its report to the Librarian on July 22, 1998. Today's 
final rule and order adopts that report.

II. Parties to This Proceeding

    As noted above, certain parties could not reach agreement as to the 
proper adjustment of the royalty rates and terms for the use of musical 
works. The musical works at issue are those belonging to composers and 
publishers affiliated with the American Society of Composers, Authors 
and Publishers (ASCAP) and to composers and publishers affiliated with 
Broadcast Music, Inc. (BMI). The public broadcasting entities wishing 
to make use of these musical works are the Public Broadcasting Service 
(PBS), National Public Radio (NPR), and other public broadcasting 
entities as defined in 37 CFR 253.2.
    ASCAP and BMI are both performing rights societies which, among 
other things, license the nonexclusive right to perform publicly the 
copyrighted musical compositions of their respective members. ASCAP and 
BMI filed separate written direct cases in this proceeding, and each 
sought a separate royalty fee for the use of musical works within their 
respective catalogues.
    PBS is a non-profit membership corporation which, among other 
things, represents the interests of its member noncommercial 
educational broadcasting stations in rate setting and royalty 
distribution proceedings in the United States, Canada, and in Europe. 
NPR is a non-profit membership organization dedicated to the 
development of a diverse noncommercial educational radio programming 
service. PBS and NPR submitted a joint written direct case in this 
proceeding and are referred to in this final rule and order as ``Public 
Broadcasters.'' The Corporation for Public Broadcasting (CPB), which 
provides funding for both PBS and NPR, is also represented in this 
proceeding, though it is not a user of music.

III. Prior History of Section 118 Rate Adjustments

    Congress intended that the parties affected by the section 118 
compulsory license negotiate reasonable license rates and terms. If the 
parties could not agree, the Copyright Royalty Tribunal (CRT) was to 
establish rates and terms in 1978 and at five-year periods thereafter 
if necessary. In section 118, Congress gave the CRT no statutory 
criteria beyond ``reasonable'' but did say that the CRT could consider 
``the rates for comparable circumstances under voluntary license 
agreements negotiated as provided in paragraph (2).'' 17 U.S.C. 
118(b)(3).
    When Congress replaced the CRT with the current CARP system in 
1993, it did not make any substantive modifications to section 118 or 
to the ``reasonable terms and rates'' standard prescribed by section 
801. See Copyright Royalty Tribunal Reform Act of 1993, Public Law 103-
198, 107 Stat. 2304.
    For the initial license term of 1978-1982, the Public Broadcasters 
successfully negotiated a voluntary license with BMI that provided for 
a payment of $250,000 for the first year with certain possible 
adjustments for each of the succeeding four years. No agreement was 
reached for the use of ASCAP music by Public Broadcasters, and the CRT 
held a proceeding to establish rates and terms.
    To determine what constituted a ``reasonable'' rate for ASCAP, the 
CRT examined the section 118 legislative history and found directives 
that the rate should reflect the fair value of the copyrighted 
material, that copyright owners were not expected to subsidize public 
broadcasting, and that Congress felt that the growth of public 
broadcasting was in the public interest. See 43 FR 25068 (June 8, 1978) 
(citing S. Rep. No. 94-473, at 101 (1975); H.R. Rep. No. 94-1476, at 
118 (1976)). From its review of the legislative history, the CRT 
concluded that it had broad discretion based on the interests Congress 
had defined. 43 FR 25068 (June 8, 1978).
    The CRT then looked at a number of different formulas submitted by 
ASCAP and Public Broadcasters for calculating royalties and concluded 
that there was no one ideal solution within the framework of a 
statutory compulsory license. 43 FR 25069 (June 8, 1978). Based on what 
it had before it, the CRT then concluded that an annual payment of 
$1.25 million was a reasonable royalty fee. It also provided for an 
inflationary adjustment during the 1978-1982 period and explained that

[[Page 49825]]

the annual fee was not determined by application of a particular 
formula, but was ``approximately what would have been produced by the 
application of several formulas explored by this agency during its 
deliberations.'' Id.
    In adopting the annual fee, the CRT stated that its determination 
was made on the basis of the record before it, and stressed that 
``[w]hen this matter again comes before the CRT, the CRT will have the 
benefit of several years experience with this schedule. The CRT does 
not intend that the adoption of this schedule should preclude active 
consideration of alternative approaches in a future proceeding.'' Id. 
The CRT, however, never conducted another section 118 proceeding before 
its abolition in 1993, because voluntary licenses were negotiated for 
all subsequent periods. Today's decision is the first section 118 rate 
adjustment that has required a formal proceeding.

IV. Report of the Panel

    After six months of hearings and written submissions of ASCAP, BMI, 
and Public Broadcasters, the CARP delivered its report to the 
Librarian. The Panel determined that Public Broadcasters should pay an 
annual fee of $3,320,000 to ASCAP, and $2,123,000 to BMI, for the 
public performance of works containing ASCAP and BMI music, 
respectively. The Panel also stated that these annual fees should be 
paid in accordance with the terms attached as an appendix to its 
report.1 Costs of the proceeding (i.e. the arbitrators' 
fees) were assessed at one-third each to ASCAP, BMI, and Public 
Broadcasters.
---------------------------------------------------------------------------

    \1\ The parties to this proceeding stipulated to the terms of 
payment. Consequently, only the rates are in issue in this 
proceeding.
---------------------------------------------------------------------------

    In attempting to determine what constituted a ``reasonable'' fee 
for ASCAP and BMI, the Panel consulted the CRT decision described above 
and examined the same legislative history reviewed by the CRT. The 
Panel observed that while section 118 did not define the term 
``reasonable,'' the legislative history indicated that ``reasonable'' 
meant ``fair value,'' and that ``fair value'' was the functional 
equivalent of ``fair market value.'' Report at 9. The Panel noted that 
the parties also generally agreed that fair market value was the proper 
standard for determining rates, and that fair market value meant ``the 
price at which goods and services would change hands between a willing 
buyer and a willing seller neither being under a compulsion to buy or 
sell and both having reasonable knowledge of all material facts.'' Id. 
In the Panel's view, although the CRT called it ``fair value'' rather 
than ``fair market value,'' the rate determined for ASCAP in the 1978 
proceeding was a fair market value determination. Thus, with respect to 
ASCAP, the Panel was adjusting the fair market value of ASCAP music in 
1978 to its present fair market value and, for the first time, 
establishing the current fair market value of BMI music. Id. at 10-11.
    To fix the fair market value of ASCAP and BMI music, respectively, 
the Panel searched for some type of method or formula that would 
establish a benchmark to assist in establishing fair market value. 
ASCAP and BMI, while employing somewhat differing adjustment 
parameters, advocated using music licensing fees recently paid by 
commercial television and radio broadcasters as a benchmark for valuing 
the license fees that Public Broadcasters should pay under section 118. 
Public Broadcasters urged the Panel to set license fees based upon 
prior voluntary licensing agreements between Public Broadcasters and 
ASCAP and BMI.2 The Panel ultimately rejected each of the 
parties' approaches and adopted instead its own benchmark.
---------------------------------------------------------------------------

    \2\ As the Panel observed, these were the primary approaches 
advocated by the parties. They also advocated alternative approaches 
and variants of the primary approach. The Panel noted, however, 
citing examples, that the parties equivocated with respect to these 
alternatives and sometimes disavowed them entirely. Id. at 11-12.
---------------------------------------------------------------------------

A. The ASCAP Approach

    According to the Panel, ASCAP's proposed use of commercial 
television and radio license fees was premised on several assumptions: 
(1) that commercial license fees represented fair market value of ASCAP 
music, whereas past agreements between ASCAP and Public Broadcasters 
did not; (2) that in recent years, Public Broadcasters have more 
closely resembled commercial broadcasters due to the rise in 
commercialization of public television and radio, fiscal success, 
sophistication, and size; (3) that after adjusting for music usage, the 
Public Broadcasters should pay the same proportion of their revenues as 
license fees as do commercial broadcasters; and (4) that ASCAP's 
proposed methodology takes into account any perceived differences 
between Public Broadcasters and commercial broadcasters by excluding 
from Public Broadcasters' revenues any revenues derived from government 
sources. Only ``private revenues,'' such as corporate underwriting and 
viewer/listener contributions, were considered under ASCAP's 
methodology because they, like commercial broadcasters' revenues, are 
audience sensitive. Id. at 13.
    ASCAP's witnesses testified that its methodology yielded an annual 
fee of $4,612,000 for television plus $3,370,000 for radio--a total of 
$7,982,000. Id. at 14. ASCAP also performed a confirmatory analysis of 
this fee by projecting forward the ASCAP fee adopted by the CRT. ASCAP 
first calculated the ratio of 1995 Public Broadcasters' private 
revenues 3 to the Public Broadcasters' 1978 private revenues 
and multiplied this figure by the 1978 fair market value fee set by the 
CRT. That result was then multiplied by the ratio of 1995 ASCAP music 
use by Public Broadcasters to the 1978 ASCAP music use by Public 
Broadcasters. This methodology generated total license fees for 1995 
for television and radio of $8,225,000, a figure that ASCAP asserted 
confirmed its primary methodology. Id. at 14-15.
---------------------------------------------------------------------------

    \3\ Public Broadcasters' 1995 revenues were the most recently 
available annual revenues to ASCAP at the time it filed its written 
direct case.
---------------------------------------------------------------------------

B. The BMI Approach

    According to the Panel, the BMI approach was quite similar to 
ASCAP's. However, in addition to examining Public Broadcasters' 
revenues and music use, BMI also examined Public Broadcasters' 
programming expenditures and audience size. BMI compared total 
revenues, programming expenditures, and audience size and determined 
that public television was 4% to 7% the size of commercial television, 
and that Public Broadcasters should therefore pay a music licensing fee 
between 4% and 7% of the fee that BMI anticipates commercial television 
will pay in 1997. BMI similarly concluded that public radio was 3% to 
4% the size of commercial radio in recent years. Id. at 15. However, 
BMI acknowledged that a one-third downward adjustment for music use by 
public radio stations as compared to commercial radio stations was 
necessary, yielding a total fee between 1% to 2% of the fees BMI 
anticipates will be paid by commercial radio in 1997. This methodology 
yielded a license fee for BMI for 1997 for public television between $4 
and $7 million and between $1 to $2 million for public radio. BMI 
recommended adopting the midpoint between these ranges, yielding $5.5 
million for public television and $1.395 million for public radio--a 
total of $6,895,000. Id. at 15-16.
    BMI also submitted that, regardless of its proposed fee, the Panel 
should not set a fee for BMI less than 42.5% of the

[[Page 49826]]

combined ASCAP and BMI fees. This argument was based on BMI's assertion 
that 42.5% of the total share of music on public television belonged to 
BMI. BMI had no data on its relative share of its music on public 
radio, but submitted that using BMI's music share on public television 
was a good proxy for music on public radio in the absence of any 
evidence on the relative shares of ASCAP and BMI music on public radio. 
Id. at 16-17.

C. Public Broadcasters

    Public Broadcasters argued that the best method for determining 
fair market value of ASCAP and BMI music was to use the 1992 negotiated 
licenses between Public Broadcasters and ASCAP and BMI as a benchmark, 
and then to adjust for any changed circumstances. Public Broadcasters 
asserted that this was the only method explicitly encouraged by the 
framers of section 118. Id. at 17.
    While conceding that there is no precise definition of ``changed 
circumstances'' since the 1992 voluntary agreements with ASCAP and BMI, 
Public Broadcasters asserted that changes in their programming 
expenditures and music use offered the best indicators of ``changed 
circumstances.'' Public Broadcasters performed an economic regression 
analysis with respect to programming expenditures and found a growth 
rate of 7.15% from 1992 through 1996. By mathematically increasing the 
combined ASCAP and BMI license fees payable under the 1992 agreements 
and determining that music use did not change during that time period, 
Public Broadcasters advocated a combined ASCAP/BMI license fee for both 
public television and radio of $4,040,000 per year. Id. at 18. Public 
Broadcasters then apportioned this fee between ASCAP and BMI based upon 
music usage and determined that BMI's share of music on public 
television was 38-40% of the total music usage. As did BMI, Public 
Broadcasters assumed that it was reasonable to use public television 
music usage as a proxy for public radio music usage. Id. at 19.

D. The Panel's Analysis

    After examining the parties' approaches, the Panel concluded that 
``[b]oth general approaches * * * suffer significant infirmities.'' Id. 
at 19. The Panel agreed with Public Broadcasters that previously 
negotiated licenses with ASCAP and BMI were logical starting points to 
determine fair market value, but concluded that the agreements from 
1982 through 1997 understate the fair market value of ASCAP and BMI 
music. The Panel also determined that, while licenses negotiated with 
similarly situated parties should be considered, ASCAP's and BMI's 
licenses with commercial broadcasters overstate the fair market value 
of music on public television and radio. Id. at 19-24. Instead, the 
Panel adopted its own methodology based upon the CRT's 1978 
determination, yielding an annual fee of $3,320,000 for ASCAP, and 
$2,123,000 for BMI.
    Because the Panel considered the voluntary license agreements that 
Public Broadcasters negotiated with ASCAP and BMI for the 1992-1997 
license period to be a logical starting point to determining fair 
market value, the Panel first considered Public Broadcasters' approach. 
The Panel was particularly impressed with the fact that the ASCAP 
license agreements contained ``no-precedent clauses'' which, in 
essence, are statements that the rates and terms prescribed in the 
agreement have no precedential value in any future negotiation or 
proceeding before a CARP. These no-precedent clauses were included in 
the voluntary agreements at the insistence of ASCAP. The Panel 
concluded that ``[t]his clause clearly evinces an attempt by ASCAP to 
protect itself from future tribunals which might be tempted to use the 
prior agreement as a benchmark for establishing fair market value. And 
such an attempt to protect itself is corroborative of ASCAP's genuine 
belief that the agreed rates were below fair market value.'' Id. at 22. 
The Panel made a similar finding with respect to ``nondisclosure'' 
clauses included in BMI's license agreements which forbade disclosure 
of the terms of the agreements to the public, including a CARP. Id. at 
22-23. The Panel also concluded that the ``huge disparity'' between 
recent ASCAP/BMI commercial license rates and the rates for Public 
Broadcasters under private agreements underscored that the prior 
agreements were not indicative of fair market value. Id. at 23. 
Therefore, the Panel rejected Public Broadcasters' approach.
    The Panel then focused on ASCAP and BMI's approach using commercial 
broadcaster license rates. The Panel rejected this approach because, 
while Public Broadcasters have become more ``commercial'' in recent 
years, ``significant differences remain which render the commercial 
benchmark suspect.'' Id. at 24. Commercial broadcasters raise revenues 
through advertising and audience share, whereas Public Broadcasters 
have no such mechanism:

    In the commercial context, audience share and advertising 
revenues are directly proportional and also tend to rise as 
programming costs rise--increased costs are passed through to the 
advertiser. No comparable mechanism exists for Public Broadcasters. 
Increased programming costs are not automatically accommodated 
through market forces. Contributions from government, business, and 
viewers remain voluntary. For these reasons, commercial rates almost 
certainly overstate fair market value to Public Broadcasters and, 
even restricting the revenue analysis to ``private revenues,'' as 
did ASCAP, does not fully reconcile the disparate economic models.

Id. at 24 (citations omitted).
    Having rejected both sides' approaches, the Panel fashioned its own 
benchmark for determining fair market value of ASCAP and BMI music. The 
Panel's methodology was based upon the fundamental assumption that the 
fee set by the CRT in 1978 was the fair market value of ASCAP music 
under the section 118 license as of that time. According to the Panel, 
that assumption was ``an eminently reasonable, and essentially 
uncontroverted, assumption. Indeed, this Panel is arguably bound by the 
1978 CRT determination of fair market value of the ASCAP license.'' Id. 
at 25. The Panel took the 1978 rate and ``trended [it] forward'' to 
1996 by adjusting for the change in Public Broadcasters' total revenues 
and the change in ASCAP's music share. This methodology yielded the 
fair market value of an ASCAP license to Public Broadcasters. The Panel 
then determined the fair market value of a BMI license to Public 
Broadcasters by applying its current music use share to the license fee 
generated for ASCAP for 1996. The Panel noted that its methodology was 
``similar to alternate analyses employed by both ASCAP and Public 
Broadcasters to demonstrate the reasonableness of their approaches.'' 
Id.
    To ``trend forward'' the CRT's 1978 ASCAP license fee to the 
present, the CARP divided that fee ($1,250,000) by Public Broadcasters' 
total 1978 revenues ($552,325,000) and multiplied the result by Public 
Broadcasters' total 1996 revenues ($1,955,726), resulting in a ``1996 
trended ASCAP license fee'' of $4,426,000, before adjusting the fee to 
take account of a decline in ASCAP's share of music usage. Id. at 26.
    The Panel determined that the change in Public Broadcasters' 
revenues from 1978 to 1996,4 along with changes in music 
share, were the best indicator of relevant changed circumstances which 
required an adjustment to the chosen benchmark. That is, Public 
Broadcasters would likely pay license fees that constitute the same 
proportion of their

[[Page 49827]]

total revenues as did the license fees that they paid in 1978, the last 
occasion in which they paid fair market rates. Id. at 27. The Panel did 
acknowledge there was ``no commonly accepted indicator that would allow 
a finder-of-fact to precisely adjust a fair market value benchmark to 
reflect relevant changed circumstances,'' noting that other factors, 
such as revenues, audience share, programming expenditures, and the 
Consumer Price Index have been used. Id. at 27-28.

    \4\ The most recent year for which data was available to the 
Panel. See footnote 7 infra.
---------------------------------------------------------------------------

    Of these, the Panel concludes that revenues is [sic] the best 
indicator of relevant changed circumstances because it incorporates 
the forementioned factors and others. Changes in audience share and 
programming expenditures are reflected in revenues. Changes in 
revenues over time also serve as a proxy for an inflation 
adjustment. While the CPI gauges inflation at the consumer level, 
revenues gauge inflation at the industry-specific level. 
Accordingly, in our analysis, an inflation adjustment from 1978 to 
1996 is obviated.

Id. at 28 (citation omitted).
    The Panel also determined that it was more appropriate to use 
Public Broadcasters' total revenues, rather than examine only 
``private'' revenues, as advocated by ASCAP. There was no need to 
confine the analysis to private revenues, because the Panel did not 
accept ASCAP's use of commercial broadcasters' rates as the appropriate 
benchmark and because the Panel was concerned with Public Broadcasters' 
revenue trends (i.e., increases) over the relevant period, not with how 
the revenues were raised. Id. at 29.
    Finally, with respect to revenues, the Panel explained why it used 
Public Broadcasters' 1996 revenues and 1978 revenues in its formula. 
Using the 1996 revenue data was important because it was the most 
recent data available to the parties and yielded the most accurate fee 
for the 1998-2002 period. Id. at 30. The Panel also rejected Public 
Broadcasters' assertion that the Panel should use Public Broadcasters' 
1976 revenues, which were the most recent revenues available to the CRT 
when it set its fair market value fee in 1978. The Panel stated that 
the record did not necessarily support Public Broadcasters' assertion 
and noted that use of 1976 revenues would have actually yielded higher 
license fees. Id. at 31.
    The Panel then adjusted the figure produced by its revenue growth 
trending formula to account for changes in the relative share of ASCAP 
music used by Public Broadcasters in 1996 as compared to 1978. The 
Panel determined that ``the ASCAP share of total ASCAP/BMI music used 
by Public Broadcasters has declined from about 80%-83% in 1978 to about 
60%-61% in 1996, representing about a 25% decline in its music share.'' 
Id. at 32. Accordingly, the Panel made a 25% downward adjustment to the 
``1996 trended ASCAP license fee'' of $4,426,000, resulting in an ASCAP 
license fee of $3,320,000. Id. at 26. In order to determine this 
decline, the Panel was required to infer the proportion of music shares 
between ASCAP and BMI in 1978 because evidence of such music shares 
does not exist.5 The Panel made this inference based upon 
two significant pieces of record evidence.
---------------------------------------------------------------------------

    \5\ Evidence does exist, however, for the proportion of music 
shares for 1996.
---------------------------------------------------------------------------

    First, since 1982, both ASCAP's and BMI's negotiated fees with 
Public Broadcasters reflect relative shares of about 80%/20% of the 
music use of Public Broadcasters. While acknowledging that the 
voluntarily negotiated licenses were not indicative of fair market 
value, the Panel was ``persuaded that the consistent division of fees 
reflects the parties' perception of respective music use shares, as 
confirmed by data available to each party.'' Id. at 33. Absent more 
reliable information, the Panel presumed that the 80%/20% split that 
had prevailed since 1982 also existed in 1978. The Panel felt 
buttressed in this assumption because ``in its trending formula, ASCAP 
did not hesitate to use its music use data from 1990 as a proxy for 
1978.'' Id.
    Second, the Panel determined that the 80%/20% split in music share 
was corroborated by the fact that in 1978 the CRT adopted a $1,250,000 
annual fee while being aware that BMI had negotiated a $250,000 annual 
fee. The Panel concluded, ``presuming the CRT did not arbitrarily 
determine fees without regard to relative music share, we infer music 
use shares for ASCAP and BMI of 83% and 17%, respectively, for 1978.'' 
Id. at 33-34. The Panel then concluded that ASCAP's 1996 music share 
was 60%-61%, based upon an analysis presented by Public Broadcasters 
that it found ``more comprehensive and more reliable'' than BMI's 
analysis. ASCAP did not present a music share analysis. Id. at 32 n.42.
    The Panel then took the $3,320,000 ASCAP fee and used it to 
determine BMI's fee. The Panel concluded that BMI's music share 
increased from about 17%-20% in 1978 to about 38%-40% in 1996. 
Selecting 39% as the appropriate figure, the Panel concluded that BMI's 
share of the combined ASCAP/BMI fees must also be 39%. The Panel 
calculated BMI's license fee of $2,123,000 by ``[m]ultiplying the ASCAP 
license fee by .63934,'' which ``yields the mathematical equivalent of 
39% of the combined license fees of both ASCAP and BMI (39%  x  
[3,320,000 + 2,123,000] = $2,123,000).'' Id. at 27 n. 40.
    The Panel offered several reasons why it was appropriate to derive 
BMI's fair market value share solely on the basis of music share. The 
Panel rejected ASCAP's assertion that the music contained in ASCAP's 
repertory is intrinsically more valuable than the music in BMI's 
inventory, finding no credible evidence for such a distinction. Id. at 
35.
    The Panel also rejected ASCAP and BMI's argument that the type of 
methodology adopted by the Panel is impermissible as a matter of law 
because section 118 requires that separate fees be set for ASCAP and 
BMI that are based upon separate evaluations of their respective 
licenses. The Panel found no proscription in the statute, the 
legislative history, or the 1978 CRT decision for a methodology which 
yields a combined fee, after which the combined fee is divided between 
ASCAP and BMI. While the Panel must set separate rates for ASCAP and 
BMI, the obligation to do so was ``wholly distinct from the methodology 
we employ to determine those fees.'' Id. at 36.
    The Panel undertook a separate approach to confirm its results for 
BMI by using the rate prescribed by the 1978 BMI negotiated license as 
a fair market value benchmark for 1978. The 1978 agreement is the only 
BMI or ASCAP agreement that did not contain a ``no-precedent clause'' 
or ``nondisclosure clause.'' However, the Panel did not accept this 
figure as representative of fair market value because the circumstances 
surrounding the 1978 negotiation were not sufficiently explored. 
Instead, the Panel used the figure solely for corroborative purposes. 
Id. at 36-37.
    The Panel used the same methodology for BMI as it did for ASCAP, 
dividing the 1978 BMI license fee by the Public Broadcasters' total 
1978 revenues and multiplying the result by the Public Broadcasters' 
total 1996 revenues. After adjusting for the increase in BMI's music 
share between 1978 and 1996, the formula yielded a figure of 
$2,082,000, within 2% of the fee adopted by the Panel under its primary 
approach. The Panel noted that it could also ``generate the ASCAP fee 
from the BMI fee just as we previously generated the BMI fee from the 
ASCAP fee--with similarly confirming results.'' Id.

[[Page 49828]]

    In conclusion, the Panel stated that its methodology yielded what 
it believed to be the best result:

    In adopting this methodology, we are fully cognizant of the 
several assumptions and inferences required. While we defend these 
assumptions and inferences as eminently reasonable, we must 
recognize the potential for imprecision. Such is the hazard of rate-
setting based upon theoretical market replication. The methodologies 
advanced by the parties involve, we believe, less reasonable 
assumptions and inferences. We do not here advance a perfect 
methodology (none exists), merely the most reasonable and least 
assailable based upon the record before us.

Id. at 38 (citation omitted).

V. The Librarian's Scope of Review

    The Librarian of Congress has, in previous proceedings, discussed 
his scope of review of CARP reports. See, e.g., 63 FR 25394 (May 8, 
1998); 62 FR 55742 (October 28, 1997); 62 FR 6558 (February 12, 1997); 
61 FR 55653 (October 26, 1996). The scope of review adopted by the 
Librarian in these proceedings has been narrow: the Librarian will not 
reject the determination of a CARP unless its decision falls outside 
the ``zone of reasonableness'' that had been used by the courts to 
review decisions of the CRT. Recently, the U.S. Court of Appeals for 
the District of Columbia Circuit issued its first decision reviewing a 
decision of the Librarian under the CARP process, and articulated its 
standard of judicial review for the Librarian's CARP decisions. 
National Ass'n of Broadcasters v. Librarian of Congress, 146 F.3d 907 
(D.C. Cir. 1998) (NAB). The court's determination is the pronouncement 
on the judicial standard of review in CARP proceedings, and warrants a 
consideration by the Register and the Librarian as to what effect, if 
any, the decision has on their review of a CARP decision.
    NAB involved distribution of cable royalties for the 1990-1992 
period. In that proceeding, the Librarian adopted the determination of 
the CARP, with some modifications, and explained why the CARP did not 
act in an arbitrary manner, or contrary to the provisions of the 
Copyright Act, that would have required a rejection of its report. The 
court reviewed the Librarian's decision in accordance with 17 U.S.C. 
802(g), which provides that the court may only modify or vacate the 
Librarian's decision if it finds that he ``acted in an arbitrary 
manner.'' The court undertook a discussion of how its review of the 
Librarian's decision under the section 802(g) arbitrary standard was 
different from its review of CRT determinations under the arbitrary 
standard set forth in chapter 7 of title 5 of the United States Code 
(i.e., the Administrative Procedure Act).
    After a lengthy discussion of its prior review of CRT 
determinations, and the amendments made to title 17 by the Copyright 
Royalty Tribunal Reform Act of 1993 which eliminated the CRT and 
replaced it with the CARP system, the court determined that Congress 
did intend to change the scope of judicial review of the Librarian's 
CARP decisions:

    We conclude that our review of the Librarian's distribution 
decision under subsection 802(g) is significantly more circumscribed 
than the review we made of the Tribunal decisions under section 810. 
As a result, in applying the ``arbitrary manner'' standard set forth 
in subsection 802(g), we will set aside a royalty award only if we 
determine that the evidence before the Librarian compels a 
substantially different award. We will uphold a royalty award if the 
Librarian has offered a facially plausible explanation for it in 
terms of the record evidence. While the standard is an exceptionally 
deferential one, we think it is most consistent with the intent of 
the Congress as reflected in the language, structure and history of 
the 1993 Act.

146 F.3d at 918.
    Quite naturally, the principal focus of the NAB decision is on the 
court's review of the Librarian's decision, not the Librarian's review 
of the CARP determination. The court did state, however, that the word 
``arbitrary'' that appears in section 802(f) of the Copyright Act 
(which gives the court its review authority), and the word 
``arbitrary'' that appears in section 802(g) (which gives the Librarian 
his review authority) are ``not coextensive.'' Id. at 923. The court 
further noted that the difference ``is not a surprising administrative 
arrangement given the bifurcated review of royalty awards (first by the 
Librarian and then by this Court) and the deference to be accorded the 
Register's and the Librarian's expertise in royalty distribution.'' Id. 
But the court did not say how exacting the review of the CARP report by 
the Librarian and the Register should be.
    Although the NAB court does not elucidate the standard of review to 
be applied by the Librarian and the Register, it does imply a 
difference between that review and the court's. If the Librarian's CARP 
decisions are entitled to an unusually wide level of deference, then 
his level of scrutiny of a CARP's decision must be higher than that 
which the court will apply to his decision.
    The Register and the Librarian do not interpret the court's 
statements to mean that they must engage in a highly exacting review. 
The court did acknowledge that the CARP, not the Register or the 
Librarian, is the fact-finder in CARP proceedings and ``is in the best 
position to weigh evidence and gauge credibility.'' Id. at 923, n.13. 
Moreover, the court stated that the Librarian would act arbitrarily if 
``without explanation or adjustment, he adopted an award proposed by 
the Panel that was not supported by any evidence or that was based on 
evidence which could not reasonably be interpreted to support the 
award.'' Id. at 923. It must be remembered that section 802(f) provides 
that the Librarian shall adopt a CARP's determination unless he finds 
that it acted arbitrarily or contrary to the Copyright Act.
    The Register and the Librarian conclude that their scope of review 
as announced in prior decisions remains an appropriate standard. That 
is, the Register and the Librarian will review the decision of a CARP 
under the same ``arbitrary'' standard used by the courts to review 
decisions of the CRT. If the CARP determination falls within the ``zone 
of reasonableness,'' the Librarian will not disturb it. See National 
Cable Television Ass'n v. Copyright Royalty Tribunal, 724 F.2d 176, 182 
(D.C. Cir. 1983) (NCTA v. CRT). It necessarily follows that even when 
the Register and the Librarian would have reached conclusions different 
from the conclusions reached by the CARP, nevertheless they will not 
disturb the CARP's determination unless they conclude that it was 
arbitrary or contrary to law. This standard is higher than the court's 
review announced in NAB, yet is consistent with the provisions of 
section 802(f).

VI. Review of the CARP Report

    Section 251.55(a) of the Library's rules provides that ``[a]ny 
party to the proceeding may file with the Librarian of Congress a 
petition to modify or set aside the determination of a Copyright 
Arbitration Royalty Panel within 14 days of the Librarian's receipt of 
the panel's report of its determination.'' 37 CFR 251.55(a). Replies to 
petitions to modify are due 14 days after the filing of the petitions. 
37 CFR 251.55(b).
    The following parties filed petitions to modify: ASCAP, BMI, Public 
Broadcasters, and SESAC, Inc. (``SESAC''). Replies were filed by ASCAP, 
BMI, Public Broadcasters, and SESAC.
    ASCAP, BMI, and Public Broadcasters all attack the Panel's adopted 
methodology as arbitrary and contrary to law, and each urges the 
Librarian to substitute his determination based upon that party's 
respective rate proposals.

[[Page 49829]]

    SESAC filed a petition to modify for the limited purpose of 
challenging a certain statement made by the Panel in a footnote of its 
report regarding music use by Public Broadcasters.6
---------------------------------------------------------------------------

    \6\ SESAC objects to footnote 10 on page 6 of the Panel's report 
wherein the Panel states that ``[t]he repertory of the third 
performing rights organization, SESAC, not a party to this 
proceeding, comprises only about one-half of one percent of PBS's 
music use.'' The task of the Register and the Librarian in CARP 
proceedings is to review CARP decisions, not to make corrections or 
modifications to statements made by the Panel at the behest of 
nonparties. However, the Register and the Librarian note that the 
Panel's statement regarding the music share of SESAC, a nonparty, is 
patently obiter dicta, and has no precedential value in this 
proceeding or future section 118 proceedings. The better practice in 
future proceedings would be for the CARP to avoid making statements 
that might be interpreted as affecting the rights or status of a 
nonparty. The Register notes that the parties to this proceeding 
expressly did not object to SESAC's petition to modify.
---------------------------------------------------------------------------

VII. Review and Recommendation of the Register of Copyrights

    As discussed above, the parties to this proceeding submitted 
petitions to the Librarian to modify the Panel's determination based on 
their assertions that the Panel acted arbitrarily or contrary to the 
applicable provisions of the Copyright Act. These petitions have 
assisted the Register in identifying what evidence and issues in this 
proceeding require scrutiny. The law gives the Register the 
responsibility to make recommendations to the Librarian regarding the 
Panel's determination, 17 U.S.C. 802(f); and in doing so, she must 
conduct a thorough review.
    Prior to reviewing the Panel's report and the parties' objections, 
the Register makes two important observations. First, the Register's 
review is confined to what the Panel did, not what it could have done. 
As described above, ASCAP, BMI, and the Public Broadcasters each 
proposed their own methodology-- their own mathematical formula--for 
calculating the appropriate annual royalty fees for the 1998-2002 
period. The Panel, however, adopted its own methodology. It is this 
methodology that the Register will review to determine whether it is 
arbitrary or contrary to law as provided by section 802(f) of the 
Copyright Act. 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 this proceeding in the first instance, she would have chosen 
another methodology. Only if the Register determines that the Panel's 
methodology is, in whole or in part, arbitrary or contrary to the 
Copyright Act will she recommend another methodology. If one or more 
aspects of the Panel's methodology is flawed, yet the methodology as a 
whole withstands scrutiny, then the Register will recommend changes so 
that the Panel's approach conforms with section 802(f). If, and only 
if, the Panel's methodology is fundamentally flawed will the Register 
recommend that the Librarian reject the Panel's approach in its 
entirety and adopt a different methodology for fixing the section 118 
royalty fees. See 63 FR 25398-99 (May 8, 1998).
    Second, the Register embraces the proposition that rate adjustment 
proceedings are not precise applications of mathematical formulas which 
yield the ``right'' answer. The Panel acknowledged this by observing 
that its methodology is not perfect, but is ``merely the most 
reasonable and least assailable based upon the record.'' Report at 38. 
The courts have also acknowledged that rate adjustments in the 
compulsory license setting involve estimates and approximations. See 
NCTA v. CRT, 724 F.2d at 182 (``The Tribunal's work * * * necessarily 
involves estimates and approximations. There has never been any 
pretense that the CRT's rulings rest on precise mathematical 
calculations; it suffices that they lie within the `zone of 
reasonableness.' ''). Therefore, in reviewing the various aspects of 
the Panel's selected methodology in this proceeding, and as a whole, 
the Register will not recommend rejecting the Panel's conclusions 
unless they draw no support from the record and are based upon 
irrational estimates or approximations.

A. Objections of ASCAP and BMI

    ASCAP and BMI raise numerous objections to the Panel's 
methodologies and recommend that the Librarian adopt their respective 
approaches as the means of assessing fees in this proceeding. Because 
several of ASCAP's and BMI's objections overlap, they are addressed 
here in a single section.
    1. The 1978 CRT fee was not a fair market value fee. The Panel 
accepted the CRT's $1.25 million fee as representing the fair market 
value of ASCAP music in 1978. BMI disputes this and offers several 
reasons why it considers the 1978 fee not representative of fair market 
value. First, BMI notes that the approach advocated by ASCAP to the CRT 
in 1978 took the rates paid by commercial broadcasters and discounted 
them by a range of 20% to 50%. This, in BMI's opinion, demonstrates 
that ASCAP was offering Public Broadcasters a subsidy. BMI Petition to 
Modify at 22. Second, BMI notes that representatives of ASCAP stated in 
an article appearing after the 1978 decision that they wanted to give 
Public Broadcasters a discount for the first 1978-1982 licensing 
period. Id.
    Third, BMI notes that the CRT stated that it did ``not intend that 
the adoption of [the $1.25 million fee] should preclude active 
consideration of alternative approaches in a future proceeding.'' Id. 
at 23 (quoting 43 FR 25069). BMI suggests that this statement is 
evidence that the CRT considered its fee to be ``experimental,'' and, 
therefore, not fair market value. Id. at 23-24.
    BMI submits that the Panel should have engaged in its own 
independent analysis of whether the 1978 fee represented fair market 
value before accepting the CRT figure. Failure to do so is, in BMI's 
view, arbitrary action. BMI asserts that it would have submitted 
information to the Panel on the inappropriateness of using the 1978 fee 
as a benchmark, if it had known that the Panel would reject BMI's 
methodology in favor of using the 1978 fee. BMI, therefore, charges 
that it was denied the opportunity to rebut use of the 1978 fee, 
particularly since it was not a party to the 1978 proceeding.
Recommendation of the Register
    The Panel did not act arbitrarily in accepting the 1978 CRT fee as 
the fair market value of ASCAP music for that period. The CRT plainly 
acknowledged in 1978 that it was required to adopt a royalty fee that 
represented the ``fair value'' of ASCAP music, and stated that the 
$1.25 million fee was a ``reasonable'' fee that accomplished that task. 
43 FR 25068 (June 8, 1978). The anecdotal evidence offered by BMI as to 
ASCAP's intentions in 1978 is far from conclusive proof that the 1978 
fee was not fair market value, and was in fact a subsidy for Public 
Broadcasters. Furthermore, the Register is not persuaded that the CRT's 
statement that its fee did not ``preclude active consideration of 
alternative approaches in a future proceeding'' is evidence that the 
CRT was adopting a fee less than fair market value. Rather, the CRT 
seemed to be stating that there may, in the future, be better ways to 
calculate fair market value, but the fee adopted by the CRT was 
nevertheless the most representative of fair market value for that 
proceeding.
    Concluding that the CRT's fee was not the fair market value of 
ASCAP music in 1978, or insisting that the Panel should have conducted 
its own study as to what was the fair market value of ASCAP music in 
1978, would be dangerous precedent. Such an approach would encourage 
collateral attack on all previous decisions of the CRT and the CARPs. 
No future CARP could rely on

[[Page 49830]]

the determination of this Panel or any other in attempting to reach its 
fair market value assessment under section 118. This is not to say that 
a prior decision of the CRT or CARP cannot be questioned by future 
parties and, if clearly demonstrated to be in error, rejected by a 
CARP. Nor should a future CARP ever be required to base its evaluation 
of ``fair market value'' on a previous determination of fair market 
value by the CRT or a previous CARP. But the Register does not 
recommend declaring, based on unconvincing evidence, that this Panel 
acted arbitrarily in accepting the CRT's 1978 fee.
    The Register is also not persuaded that BMI has been denied an 
opportunity to challenge the validity of the 1978 CRT fee. It is true 
that BMI did not know, until the Panel released its decision, that the 
Panel would use the 1978 fee as a basis for adopting its current fee. 
However, that will virtually always be the case in a rate adjustment 
proceeding or distribution proceeding when a CARP utilizes its own 
methodology as opposed to one offered by the parties. The Register will 
not reject the methodology of a Panel simply because the parties were 
not presented with the opportunity, during the hearing phase, to 
criticize and attack the Panel's chosen methodology. To do otherwise 
would effectively preclude a Panel from adopting a methodology other 
than one proposed by the parties.
    Furthermore, the 1978 fee was very much a part of the record in 
this proceeding. The existence of the fee and the CRT decision adopting 
it were recognized and acknowledged by all parties to this proceeding, 
including BMI. ASCAP used the 1978 fee in its alternative methodology 
to verify the accuracy of its primary methodology. That BMI did not 
mount a serious evidentiary challenge to the accuracy of the fee is not 
due to lack of opportunity.
    2. The Panel incorrectly used Public Broadcasters' 1978 revenues, 
rather than their 1976 revenues. Both ASCAP and BMI make this 
accusation. In order to ``trend forward'' from the $1.25 million 1978 
ASCAP award, the Panel began with Public Broadcasters' 1978 annual 
revenues (the Panel's equation is fair market value in 1978 divided by 
1978 Public Broadcaster revenues, or $1.25 million/$552.325 million). 
Report at 26. ASCAP and BMI assert that use of Public Broadcasters' 
1978 revenues is flawed because the CRT did not have these revenue 
figures when it calculated the $1.25 million fee. Rather, the most 
recent figure available to the CRT was Public Broadcasters' 1976 
revenues, which were $412.2 million. ASCAP notes that because the Panel 
used 1978 revenues instead of 1976 revenues, the effective rate of the 
1978 rate is reduced, thereby devaluing the CRT's 1978 determination.
    The effective rate of the 1978 CRT decision is, according to ASCAP, 
expressed as a percentage relative to Public Broadcasters' revenues. 
ASCAP Petition to Modify at 6. The $1.25 million fee divided by $412.2 
million (the 1976 revenues) yields an effective rate of .303% of 
revenues. According to ASCAP, this means that the CRT in 1978 intended 
to give ASCAP a fee that represented .303% of Public Broadcasters' most 
recently known revenues (i.e., the 1976 revenues). By using the 1978 
revenues, the Panel reduced the effective rate to .22% ($1.25 million 
divided by $552.325 million), which is not what the CRT intended to 
award. Both ASCAP and BMI assert that the Panel should have used the 
1976 revenues and ``trended forward'' from there in order to maintain 
the effective rate of the CRT decision.
    BMI asserts that there is another reason for using the 1976 data. 
As was the case for the CRT, the Panel used data to set a royalty fee 
beginning in 1998 that was only as recent as 1996.7 The 
Panel's methodology takes account of only an 18-year period, 1978-1996. 
BMI submits that the Panel should have taken account of a 20-year 
period, 1976-1996, in order to obtain a more accurate trend and to make 
up for the lack of data for 1997 and 1998. BMI Petition to Modify at 
28.
---------------------------------------------------------------------------

    \7\ At the time of filing of written direct cases in this 
proceeding, ASCAP and BMI had data of Public Broadcasters' revenues 
only up to 1995. However, Public Broadcasters introduced their 1996 
revenues as part of their case. See Public Broadcasters Direct 
Exhibit 4.
---------------------------------------------------------------------------

Recommendation of the Register
    The Register determines that the Panel did not err in using Public 
Broadcasters' 1978 revenues, as opposed to 1976 revenues, as the basis 
of its trending methodology. If it could be conclusively demonstrated 
that the CRT used Public Broadcasters' revenues as the means of 
fashioning the $1.25 million 1976 fee, ASCAP and BMI's argument would 
be more persuasive. That is not, however, the case. Although the CRT 
``examined a number of formulas,'' it concluded ``there is no one 
formula that provides the ideal solution, especially when the 
determination must be made within the framework of a statutory 
compulsory license.'' 43 FR 25069 (June 8, 1978). Although the CRT had 
Public Broadcasters' 1976 revenues before it, it is unclear what, if 
any, use it made of the data. The CRT said nothing about the $1.25 
million fee representing a .303% effective rate of Public Broadcasters' 
revenues, nor is there any indication in the 1978 decision that the CRT 
was attempting to establish a fixed effective rate. ASCAP's argument 
presumes that the CRT did use a mathematical formula in adopting a fee, 
even though the decision suggests the contrary.
    What is clear is that the CRT determined that the $1.25 million fee 
was the fair market value of ASCAP music in 1978, even if it did use 
data from 1976. Id. The Panel reached the same conclusion by stating 
that ``the blanket license fee set by the CRT in 1978, for use of the 
ASCAP repertory by Public Broadcasters, reflects the fair market value 
of that license as of 1978.'' Report at 25 (emphasis added). If $1.25 
million represented fair market value in 1978, then it was reasonable 
for the Panel to begin its analysis using Public Broadcasters' revenues 
from that same year, whether or not the CRT had access to such data. 
The Panel stated that it felt ``comfortable'' doing this because Dr. 
Adam Jaffe, Public Broadcasters' economic expert, had taken a similar 
approach in a different context. Report at 31 (Dr. Jaffe's formula used 
the 1992-1997 voluntary agreements with ASCAP and adjusted for changed 
circumstances from 1992, even though the parties presumably negotiated 
the 1992 agreement using only 1991 data). The Register sees nothing in 
the record that indicates it was arbitrary to take this approach.
    BMI's argument that the Panel should have considered changes in 
revenues over a 20-year period, rather than 18 years, to account for 
the lack of information for 1998 Public Broadcasters' revenues, also 
has no merit. It will probably always be the case in a section 118 
proceeding that data regarding revenues will not be completely current. 
Use of the Public Broadcasters' 1998 revenues, or 1997 revenues for 
that matter, would yield a fair market value fee that might be even 
more accurate than the Panel's. However, that data was simply 
unavailable. The Panel could have considered a 20-year period as a 
rough means of adjusting for lack of 1998 data. The fact that it did 
not do so was not arbitrary.8
---------------------------------------------------------------------------

    \8\ Furthermore, the Register questions the perceived accuracy 
of starting with 1976 data as a means of compensating for lack of 
1998 data. The only thing this approach guarantees is a larger fee 
since it is known that Public Broadcasters' revenues were less in 
1976 than they were in 1978.
---------------------------------------------------------------------------

    3. The Panel did not provide for fee adjustments during the 1998-
2002 period. ASCAP argues that it was

[[Page 49831]]

arbitrary for the Panel not to provide for interim adjustments to the 
ASCAP fee for each year of the 1998-2002 license period. ASCAP notes 
that the CRT provided for annual adjustments for inflation through use 
of the Consumer Price Index (``CPI'') in its 1978 decision, and that 
the Panel should have, at a minimum, provided for similar adjustments. 
As an alternative to using the CPI, ASCAP recommends that the effective 
rate of the CRT's 1978 decision (.303% of Public Broadcasters' 1976 
revenues) be applied to Public Broadcasters' revenues for each year of 
the 1998-2002 period to determine an annual fee.
Recommendation of the Register
    The Panel considered whether to provide cost-of-living adjustments 
and expressly decided not to do so, concluding that ``[g]iven the 
inherent vagaries and imprecision of estimating fair market value in an 
imaginary marketplace, we are comfortable concluding that the rate 
yielded for 1996 reasonably approximates a fair market rate for the 
entire statutory period.'' Report at 31.
    The Register cannot say that the Panel's conclusion was arbitrary. 
The Panel recognized that the methodology it used to set the fees was 
based on ``several assumptions and inferences'' which, although 
``eminently reasonable'' created a ``potential for imprecision. Such is 
the hazard of rate-setting based upon theoretical market replication.'' 
Report at 38 (citing NAB, 146 F.3d at 932). The Panel admitted that it 
was not ``advanc[ing] a perfect methodology (none exists), merely the 
most reasonable and least assailable based upon the record before us.'' 
Id.
    The Panel also observed that the 1996 Public Broadcasters' revenue 
figures that it used in determining the fee may have been somewhat 
overstated due to changes in accounting procedures. Id. at 30. Based on 
this finding and the CARP's determination that use of revenues account 
for inflationary changes (id. at 28), the Register cannot say that the 
Panel was arbitrary or unreasonable in deciding not to provide for 
annual adjustments. In fact, the Panel's assessment that the 1996 
revenue figures may have been an overstatement only supports its 
conclusion that no annual adjustment was necessary.
    Certainly, the Panel could have required annual adjustments of 
ASCAP's fee based on annual changes in Public Broadcasters' revenues, 
as ASCAP now requests. But it was not required to do so, given the 
absence of record evidence compelling such a result.
    4. The Panel arbitrarily excluded Public Broadcasters' ancillary 
revenues from their calculation. ASCAP asserts that the Panel excluded 
without explanation $122 million in ``ancillary'' revenues earned by 
the Public Broadcasters in 1996. ``Ancillary'' revenues, according to 
ASCAP, are comprised largely of the sale of public broadcasting 
merchandise such as videos, audiotapes, toys and books. ASCAP submits 
that ancillary revenues must be included in the Panel's calculation 
because the Panel acknowledged that gross revenues of Public 
Broadcasters were the best indication of their ability to pay. 
According to ASCAP, Public Broadcasters' 1996 revenues should be 
$2,077,776,000, instead of the $1,955,726,000 figure used by the Panel. 
ASCAP Petition to Modify at 9.
Recommendation of the Register
    In discussing what comprised the Panel's determination of Public 
Broadcasters' 1996 revenues, the Panel stated that they were excluding 
``all `off balance sheet income' such as revenues derived from 
merchandising, licensing, and studio leasing.'' Report at 30 (citing 
ASCAP Direct Exhibit 301 and ASCAP's Proposed Findings of Fact and 
Conclusions of Law (PFFCL)). While a specific explanation for exclusion 
of such income would be desirable, the Register does not find the Panel 
acted arbitrarily. First, the Register does not agree with ASCAP's 
conclusion that the Panel was setting Public Broadcasters' 1996 
revenues as gross revenues from all sources. The Panel stated that it 
was using Public Broadcasters' total revenues, and cited CPB's fiscal 
year 1996 report for that figure. Report at 26. As ASCAP acknowledges, 
CPB does not include ancillary income in its calculation of annual 
revenues. ASCAP PFFCL at 39, para. 94. The total revenues figure, 
therefore, expressly did not include ancillary income.
    Second, the Register concludes that it was reasonable for the Panel 
to exclude ancillary income. Merchandising of toys, tapes and books, 
and leasing studio facilities to others, are not part of the business 
of broadcasting music on public broadcasting stations. CPB apparently 
acknowledges this point as well, excluding ancillary income from its 
report of Public Broadcasters' revenues because ancillary income does 
not form a basis for awarding grants to Public Broadcasters. Id. ASCAP 
has failed to demonstrate that Public Broadcasters' activities such as 
selling books and toys are so closely tied to broadcasting activities 
that their revenues must be included in broadcast revenues. See 
Transcript (Tr.) at 1722 (Boyle)(stating that off balance sheet items 
``may or may not be relevant'' in calculating Public Broadcasters' 
revenues).
    5. The Panel arbitrarily concluded that overall music use remained 
static since 1978. Both ASCAP and BMI argue that it was arbitrary for 
the Panel to conclude that overall music use remained relatively 
constant from 1978 to 1996, given the fact that there was no reliable 
music use data available until 1992. ASCAP asserts that ``[i]f there is 
no evidence to support an adjustment, the adjustment cannot be made, no 
matter how relevant it might be.'' ASCAP Petition to Modify at 14. Both 
ASCAP and BMI submit that the record, in fact, belies static music use, 
noting that there are many more public broadcasting stations, and 
consequently more programs broadcast, since 1976 and that the total 
volume of music use must therefore have increased substantially. BMI 
goes on to state that the record supports that, since 1992, use of BMI 
music has increased an average of 10% on public broadcasting stations, 
and that the Panel should have factored this into its analysis and 
awarded BMI a greater fee.
Recommendation of the Register
    As described above, the Panel's methodology ``trends forward'' the 
CRT's 1978 fee and adjusts for changes in the relative shares of ASCAP 
and BMI music used by Public Broadcasters since 1978. The Panel did, 
however, consider whether any change to the methodology was required to 
account for changes in overall music usage since 1978. Evaluating the 
scant evidence on the subject, the Panel concluded:

    We find the music analyses presented by Public Broadcasters to 
be the most comprehensive and reliable. No credible data is 
available with respect to any trend in overall music usage by Public 
Broadcasters since 1978. However, we accept Public Broadcasters' 
conclusion that overall music usage has remained constant in recent 
years. Given the dearth of empirical, or even anecdotal, evidence to 
the contrary, it is reasonable to presume that overall music usage 
by Public Broadcasters has remained substantially constant since 
1978. See ASCAP PFFCL 152 (``[T]here is no evidence in the record 
that total music use on the [Public Television and Public Radio] 
Stations has changed significantly since 1978.'')

Report at 31-32 (citations omitted).
    BMI and ASCAP attack the Panel's conclusion regarding music use, 
arguing, in essence, that the Panel is forbidden from fact-finding in 
the absence of thoroughly comprehensive

[[Page 49832]]

record evidence. The Register cannot accept ASCAP and BMI's argument in 
this instance. There is no question that record evidence of music use 
prior to 1992 would place the Panel's conclusion on firmer ground. 
Complete and comprehensive evidence will always increase the accuracy 
of CARP decisions, but it is often such evidence does not exist, or is 
not presented in a CARP proceeding. See, e.g., 62 FR 55757 (October 28, 
1997) (rejecting satellite carriers' argument that Panel decision must 
be rejected because satellite carriers had no access to evidence to 
rebut copyright owners' contentions). The Register believes that it is 
acceptable, given the inherent lack of precision of these proceedings, 
for a Panel to make reasonable inferences based on an examination of 
the best evidence available. The Panel's inference regarding music use 
satisfies this requirement.
    In drawing its inference, the Panel examined the best evidence it 
had available to it: the music use analyses of the parties from 1992-
1996. The Panel adopted Public Broadcasters' analysis as the ``most 
comprehensive and reliable.'' Report at 31. The Panel concluded that 
Public Broadcasters' analysis demonstrated that overall music use in 
recent years has remained relatively constant. The Register has no 
grounds to question this finding. See, 61 FR 55663 (October 28, 1996) 
(``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.'') Given that music use was 
static for a period of five years, the Panel reasonably inferred that 
this trend was predictive of music use from 1978 to 1991. The inference 
was backed by ASCAP's statement in its proposed findings that ``there 
is no evidence in the record that total music use on the Stations has 
changed significantly since 1978. Nor is there any evidence in the 
record that the Stations' broadcasts of ASCAP music over the same 
period have changed significantly either in quality or quantity.'' 
ASCAP PFFCL at 152, para.32. The five-year period, coupled with ASCAP's 
statement, provide sufficient support for the Panel's presumption 
regarding music use.
    Moreover, the Register does not find that ASCAP's and BMI's 
assertions regarding the increase in the number of public broadcasting 
stations and programs broadcast require rejection of the Panel's 
inference. Both ASCAP and BMI presume that there is a direct 
correlation between number of stations and broadcast hours and the 
amount of music used. This certainly is a reasonable conclusion, but it 
is not a necessary one. It could, for example, be the case that public 
broadcasting stations prior to 1992 used far greater amounts of music 
than do public broadcasting stations today. Public Broadcasters' 
evidence tends to support that conclusion. See Public Broadcasters 
PFFCL at 50-51, Paras. 112-113. In sum, the Register will not, in the 
absence of concrete evidence to the contrary, allow an inference drawn 
by a party to trump an inference drawn by a Panel.9
---------------------------------------------------------------------------

    \9\ Given that the Register accepts the Panel's determination 
that music use has not increased, the Register rejects BMI's request 
for an adjustment to account for a ten percent increase in its music 
use.
---------------------------------------------------------------------------

    6. The Panel's dependence on music share is irrelevant and 
unsupported by section 118. ASCAP submits that section 118 
uncontrovertedly provides that copyright owners of music are entitled 
to compensation for use of their music by Public Broadcasters. The 
Panel's reliance on music share as opposed to music use, ASCAP insists, 
is irrelevant because music share does not necessarily have any 
correlation to music use. Further, ASCAP submits that reliance on music 
share is contrary to section 118 because music share presumes that 
ASCAP and BMI music is interchangeable, whereas section 118 requires 
establishing separate royalty fees for both catalogues of music.
Recommendation of the Register
    The Register determines that the Panel's use of music shares to 
adjust for the amount of ASCAP and BMI music used on public 
broadcasting stations since 1978 is not contrary to section 118. The 
Panel addressed ASCAP's contention that its methodology was contrary to 
section 118 when it stated:

    [B]oth ASCAP and BMI argue that the type of methodology we 
advance here is impermissible, as a matter of law, because Section 
118 requires that separate fees be set for ASCAP and BMI that are 
based upon separate evaluations of their respective licenses. The 
legislative history behind Section 118, they argue, proscribes any 
methodology that yields a combined fee, after which the combined fee 
is divided between ASCAP and BMI. The Panel finds no support 
whatever for this position in the legislative history of Section 
118, the express language of the statute itself, or in the 1978 CRT 
decision cited by ASCAP. It is undisputed that the statute requires 
the Panel to set separate rates for ASCAP and BMI but that is an 
obligation wholly distinct from the methodology we employ to 
determine those fees.

Report at 35-36 (footnotes omitted) (citations omitted). The Register 
agrees.
    The Register also concludes that the Panel's use of music shares is 
not arbitrary. The Panel used music shares to gauge changed 
circumstances since 1978, determining that the amount of ASCAP music, 
relative to BMI music, had declined from 1978. This is wholly 
consistent with the Panel's adopted methodology, and is one of the 
mechanisms necessary to that analysis to account for changed 
circumstances.
    7. There is insufficient record evidence to support the Panel's 
inferential findings regarding music share. ASCAP and BMI argue that, 
assuming music share is relevant to the Panel's methodology, the 
absence of evidence for music shares prior to 1992 prevented the Panel 
from inferring the shares of ASCAP and BMI music on public broadcasting 
in 1978.
Recommendation of the Register
    For the reasons stated in A5, supra, the Register will not question 
a reasonable inference of the Panel provided that it draws support from 
the existing record. The Panel determined that the ratio of ASCAP to 
BMI music in 1978 was in the range of 80/20 to 83/17. Report at 32. The 
Panel based this determination on the fact that, since 1981, both ASCAP 
and BMI negotiated fees that consistently reflected that share of 
music. The Panel stated that ``we are persuaded that the consistent 
division of fees reflects the parties' perception of respective music 
use shares, as confirmed by data available to each party.'' Id. at 33.
    The Panel also presumed music shares from 1978 to 1981 were at the 
same ratio, in the absence of evidence to the contrary. The Panel 
reasoned that this presumption was corroborated by the fact that the 
CRT, in awarding ASCAP a $1.25 million fee in 1978, was aware that BMI 
had negotiated a $250,000 fee. The Panel also relied on the fact that 
ASCAP itself used 1990 music use data as a proxy for 1978 data. See 
ASCAP PFFCL at 116, para.266, n.6 (``Because reliable music use data 
were not available for 1978, ASCAP relied on music use data starting 
from 1990, the first ASCAP distribution survey year for which detailed 
information was readily retrievable. Thus, the trended fee assumes that 
music use on Stations did not change substantially from 1978 to 1990 
(and there is no evidence in the record to contradict that 
assumption.'')). The Register determines that these pieces of record 
evidence support the reasonableness of the Panel's presumptions 
regarding music share in 1978.
    ASCAP also argues that the Panel's split of approximately 80/20 is 
inaccurate because the Panel mistakenly assumed that ASCAP relied upon 
its music share as a basis for negotiating its

[[Page 49833]]

fee in 1982, 1987 and 1992, when in fact it did not. The record appears 
far from clear on this point, particularly since Public Broadcasters 
submit that music share was important to them in negotiating the 
licenses. See Tr. at 2619-21 (Jameson). It is clear that BMI used its 
relative music share in negotiating its licenses with Public 
Broadcasters. See, Tr. at 3389 (Berenson). In any event, the Register 
agrees with the Panel that it was the parties' perceptions as to their 
music shares during their negotiations that is relevant:

    It is important to note that whether the music use shares we 
have adopted are actually accurate is not critical to our analysis 
so long as the parties perceived them to be accurate at the time 
they negotiated the agreements. As we have repeatedly expressed 
herein, our task is to attempt to replicate the results of 
theoretical negotiations. If the parties were to use the 1978 
license fee as a benchmark, we have no doubt that the resulting fees 
from such negotiations would reflect the parties' perceived change 
in ASCAP's music share since 1978, just as they would reflect the 
parties' perceived change in Public Broadcasters' total revenues.

Report at 34.
    8. It was arbitrary for the Panel to infer music share on public 
radio when no evidence of music use on public radio was presented. 
ASCAP faults the Panel's use of music share on public television as a 
proxy for music share on public radio. ASCAP argues that the Panel's 
citation to the negotiated licenses' historical use of television music 
use data as a proxy for radio is inappropriate because the Panel 
determined that those agreements are not representative of fair market 
value. Further, ASCAP submits that there was no probative evidence 
adduced that ASCAP ever acquiesced to the use of television data as a 
proxy for radio data. ASCAP Petition to Modify at 19.
Recommendation of the Register
    The Register determines that the Panel's use of television data as 
a proxy for radio data is not arbitrary. The Panel's statement that 
Public Broadcasters and ASCAP and BMI used television music data as a 
proxy for radio data (since no party keeps track of music usage on 
public radio) was based on the testimony of Paula Jameson, Public 
Broadcasters' then general counsel, who participated in the fee 
negotiations. Tr. at 2621-23 (Jameson). Although ASCAP asserts that 
there is testimony to the contrary, the Register will not disturb the 
Panel's evaluation of testimony in the absence of compelling grounds to 
do so. See, NAB, 146 F.3d at 923, n.13 (``The Panel, as the initial 
factfinder, is in the best position to weigh evidence and gauge 
credibility'').
    9. The Panel made an arbitrary assumption that Public Broadcasters 
should pay the same rate of revenue now as they did in 1978 despite 
their increased commercialization. BMI charges the Panel with failure 
to include an adjustment in its methodology to account for Public 
Broadcasters' increased commercialization. BMI notes that the Panel did 
recognize the increased commercialization, and acknowledged that such 
commercialization might justify the need to narrow the divergence 
between fees paid by Public Broadcasters and commercial broadcasters, 
but then did not do anything about it. BMI submits that using Public 
Broadcasters' private revenues since 1978, as opposed to total 
revenues, ``is a reasonable way to take into account the increased 
commercialization of public broadcasting in setting a rate based on the 
1978 CRT fee.'' BMI Petition to Modify at 37.
Recommendation of the Register
    While the Panel did observe that Public Broadcasters have become 
more commercialized in recent years, and that such a convergence 
between public and commercial broadcasting ``may'' justify a narrowing 
of the gap between the fees paid by Public Broadcasters and commercial 
broadcasters, that observation does not compel an adjustment to the 
Panel's methodology. The Panel also concluded that significant 
differences between Public Broadcasters and commercial broadcasters 
remain. See Report at 24 (``Though corporate underwriting may 
superficially resemble advertising * * *, the relevant economics are 
quite different''). Indeed, these differences specifically led the 
Panel to reject commercial fees as the benchmark for setting Public 
Broadcasters' fees. Id.
    Moreover, the Panel expressly rejected the use of private revenues 
in its methodology as the means of accounting for increased Public 
Broadcasters' commercialization:

    [W]hen performing a trending analysis based upon the 1978 Public 
Broadcasters' rates, there is no need to restrict the analysis to 
private revenues because the methodology does not employ any data 
from the commercial context. In this instance, we need make no 
attempt to account for differences in the manner the two industries 
raise revenues. We need not massage the methodology to obtain an 
`apples to apples' comparison. Accordingly, total revenues, 
reflecting the true increase in Public Broadcasters' ability to pay 
license fees, is the more appropriate parameter.

Report at 29-30.
    There is ample testimony to support the Panel's determination that 
the economics of public broadcasting and commercial broadcasting are 
quite different. Written rebuttal testimony of Dr. Adam Jaffe at 14-17; 
Public Broadcasters Direct Exhibit 4. The Panel was, therefore, not 
compelled by the evidence to account for increased commercialization of 
Public Broadcasters in adopting their methodology, and it was not 
arbitrary to reject the use of private revenues as a means for 
adjusting for commercialization.
    10. The Librarian should announce that ASCAP and BMI may seek rate 
parity with commercial broadcasters in future section 118 proceedings. 
BMI submits that, assuming that the Librarian does not choose to adopt 
a methodology that bases Public Broadcasters' fee on what commercial 
broadcasters pay for music, the Librarian should declare that ``BMI is 
free to argue in a future CARP proceeding that Section 118 license fees 
should be set on the basis of a comparison to commercial broadcasting, 
under the facts and circumstances as they may develop in the future.'' 
BMI Petition to Modify at 58.
Recommendation of the Register
    The task of the Register, and the Librarian, in CARP proceedings is 
to review the decision of a CARP panel, not to make pronouncements or 
declarations as to the character or nature of future proceedings. The 
Register recommends that the Librarian not accept BMI's invitation. The 
Register notes, however, that parties to a future section 118 
proceeding, or any CARP proceeding for that matter, are free to submit 
any and all evidence they deem relevant to the rate adjustment or 
royalty distribution, as the case may be.
    11. The Panel erred in its allocation of costs among the parties. 
ASCAP submits that the Panel erred because it did not follow prior 
CARPs' allocation of costs 10 in rate adjustment 
proceedings, and did not articulate a reason for its deviation. ASCAP 
asserts that the Panel should not have treated PBS and NPR as a single 
party for cost purposes, and instead should have equally split costs 
between ASCAP and BMI on the one hand, and PBS and NPR on the other. 
According to ASCAP, ``[f]airness dictates an equal division of costs, 
which is consistent with prior

[[Page 49834]]

precedent and which imposes equal burdens of the proceeding on 
copyright owners and users.'' ASCAP Petition to Modify at 30.
---------------------------------------------------------------------------

    \10\ ``Allocation of costs'' in a CARP proceeding are the 
monthly charges of the arbitrators. The costs of the Copyright 
Office and the Librarian are part of their operating budgets, and 
are not a part of a CARP's allocation of costs.
---------------------------------------------------------------------------

Recommendation of the Register
    Section 802(c) of the Copyright Act provides that ``[i]n ratemaking 
proceedings, the parties to the proceedings shall bear the entire cost 
thereof in such manner and proportion as the arbitration panels shall 
direct.'' 17 U.S.C. 802(c). ASCAP's request raises the question whether 
a cost allocation decision of a CARP is reviewable by the Librarian 
under section 802(f).
    Section 802(f) of the Copyright Act is the source of the 
Librarian's review authority of CARP decisions. It provides in 
pertinent part that ``[w]ithin 60 days after receiving the report of a 
copyright arbitration royalty panel under subsection (e), the Librarian 
of Congress, upon the recommendation of the Register of Copyrights, 
shall adopt or reject the determination of the arbitration panel.'' 17 
U.S.C. 802(f). While the ``determination'' of the Panel is not defined 
in subsection (f), subsection (e) describes a CARP delivering ``a 
report'' of ``its determination concerning the royalty fee or 
distribution of royalty fees, as the case may be.'' 17 U.S.C. 802(e). 
It thus appears that the Library's review authority extends only to a 
Panel's decision on the merits of a ratemaking or distribution 
proceeding--i.e., the actual setting of rates or allocation of 
royalties. Is this review authority broad enough to encompass a Panel's 
allocation of costs under subsection 802(c)?
    The Register concludes that it is not. A plain reading of the 
statute limits the Librarian's review to the substance of the 
proceeding--the setting of rates or distribution of royalties--
contained in the Panel's report, and does not include allocation of the 
arbitrators' costs among the parties to the proceeding. The fact that 
the Panel's decision on costs was also contained in its report on the 
merits of the proceeding does not change the result. Allocation of 
costs has no bearing on the Panel's resolution on the merits of the 
proceeding. Furthermore, the Panel in this case could have just as 
easily issued a separate order allocating costs, and was not required 
to include such a decision in its report to the Librarian. The 
Librarian's jurisdiction should not depend on where the CARP announces 
its allocation of costs.
    Even if the Librarian had authority to review the Panel's 
allocation of costs, the Register would not recommend that the 
Librarian reject the Panel's allocation of one-third paid by ASCAP, 
one-third paid by BMI, and one-third paid by Public Broadcasters. The 
statute plainly gives the arbitrators broad discretion in allocating 
costs. 18 U.S.C. 802(c) (costs shall be allocated ``in such manner and 
proportion as the arbitration panels shall direct''). The Register is 
also not persuaded that the language of subsection (c) that requires a 
CARP to act on the basis of ``prior copyright arbitration royalty panel 
determinations'' applies to allocation of costs. This provision is 
directed to ``determinations'' of CARPs--i.e. their decisions as to 
rates and royalty distributions.
    The Panel concluded, for purposes of cost allocation, that ``ASCAP, 
BMI, and Public Broadcasters constitute three separate parties.'' 
Report at 39. It reached its conclusion based ``on the totality of 
circumstances including the 1978 CRT decision, the history of 
negotiations between the parties, and the manner in which the parties 
proceeded herein.'' Id. The Register believes that the CARP--and not 
the Register or the Librarian--is in the best position to evaluate 
these factors and apportion the costs. The Register, therefore, 
recommends that the Librarian not review or reject the Panel's 
allocation of costs.

B. Objections of Public Broadcasters

    Public Broadcasters fault the Panel for rejecting use of prior 
negotiated agreements as the benchmark for setting ASCAP's and BMI's 
fees. In support of this position, Public Broadcasters offer the 
following three arguments.
    1. The Panel violated section 118 by setting fair market value 
rates in the context of hypothetical free marketplace negotiations, as 
opposed to within the confines of section 118. Public Broadcasters do 
not challenge the Panel's evaluation of the meaning of fair market 
value--the price that a willing buyer and willing seller would 
negotiate--but they do contest the setting in which the Panel 
determined fair market value. The Panel stated:

    In the present context, a determination of fair market value 
requires the Panel to find the rate that Public Broadcasters would 
pay to ASCAP and to BMI for the purchase of their blanket licenses, 
for the current statutory period, in a hypothetical free market, in 
the absence of the Section 118 compulsory license.

Report at 9-10 (second emphasis added). Public Broadcasters charge that 
it was legal error for the Panel to determine fair market value outside 
the context of section 118, and that the Panel was required to take 
into account the purposes of section 118 in setting rates. Public 
Broadcasters Petition to Modify at 9-10 (citing the Librarian's recent 
section 114 rate proceeding for the proposition that reasonable rates 
are not the same as marketplace rates and that a statutory rate need 
not mirror a freely negotiated rate). This ``fundamental error,'' 
according to Public Broadcasters, incorrectly led the Panel to reject 
prior negotiated agreements under section 118 as the benchmark for 
setting rates in this proceeding.
Recommendation of the Register
    The Register determines that the Panel did not act contrary to 
section 118 by seeking to determine what rates the parties would 
negotiate in free, open marketplace negotiations, as opposed to within 
the context of section 118. Public Broadcasters attempt to create the 
notion that there are two kinds of fair market values: one negotiated 
in the context of the open marketplace, and another within the 
``particularized context of section 118.'' Public Broadcasters Petition 
to Modify at 9. The Copyright Act makes no such distinctions. The only 
provision for adjusting section 118 rates is contained in section 
801(b)(1), which provides that a CARP shall set ``reasonable'' rates 
for section 118. Unlike other compulsory licenses, section 118 does not 
contain any criteria or prescriptions to be considered in adjusting 
rates, other than a direction that a Panel may consider negotiated 
agreements. See, e.g., 17 U.S.C. 119(c)(3)(B) (fair market value rates 
established with consideration of certain types of evidence); 17 U.S.C. 
801(b)(1) (sections 114, 115 and 116 compulsory license rates adjusted 
to achieve specified objectives). Moreover, it is difficult to 
understand how a license negotiated under the constraints of a 
compulsory license, where the licensor has no choice but to license, 
could truly reflect ``fair market value.'' The Panel was, therefore, 
not required to consider fair market value confined to the context of 
section 118.15
---------------------------------------------------------------------------

    \15\ If this were the requirement, the only evidence in a 
section 118 rate adjustment proceeding presumably would be the 
agreements previously negotiated by the parties for the section 118 
license. This is, obviously, precisely what the Public Broadcasters 
wanted the Panel to consider. However, if fair market value within 
the section 118 license were the standard, Congress presumably would 
not have provided that a CARP ``may'' consider negotiated 
agreements, but rather would have mandated such a consideration. See 
17 U.S.C. 118(b)(3).
---------------------------------------------------------------------------

    Public Broadcasters' citation to the section 114 rate adjustment 
proceeding is also inapposite. Section 801(b)(1) of the Copyright Act 
prescribes that section 114 rates are to be adjusted to achieve four 
specific objectives. Because

[[Page 49835]]

section 114 rates must be observant of those objectives, they need not 
be market rates. See 63 FR 25409 (May 8, 1998). Such is not the case 
with section 118.
    2. The Panel's erroneous analysis of the no-precedent and 
nondisclosure clauses of the voluntary agreements led the Panel 
improperly to reject the agreements as the benchmark. Public 
Broadcasters argue that the Panel improperly used the no-precedent 
clause in the ASCAP agreement, and the nondisclosure clause in the BMI 
agreement, as grounds for rejecting the previously negotiated 
agreements between ASCAP/BMI and the Public Broadcasters as the 
benchmark for adjusting rates in this proceeding. Because Public 
Broadcasters assert that fair market value rates must be determined in 
the context of section 118 (see supra), Public Broadcasters assert that 
the ASCAP no-precedent clause and the BMI nondisclosure clause have no 
relevance to the rates the parties would have negotiated; and it was, 
therefore, illogical for the Panel to conclude that the existence of 
these clauses was evidence that the voluntary agreements understated 
fair market value.
Recommendation of the Register
    The Register determines that the Panel's analysis of the no-
precedent and nondisclosure clauses of the ASCAP and BMI agreements was 
not arbitrary or contrary to the provisions of the Copyright Act. 
First, as discussed above, the Register rejects the position that the 
Panel was required to set fair market value rates confined to the 
context of section 118 negotiations. The Panel was, therefore, not 
bound to accept the prior negotiated agreements as the only evidence of 
fair market value.
    Second, Public Broadcasters misperceive the significance of the no-
precedent and nondisclosure clauses as they affected the Panel's 
decision to reject the negotiated agreements as the benchmark for fair 
market value. The Panel did not use these clauses as the only evidence 
that the negotiated agreements were not representative of fair market 
value. Rather, the Panel stated:

    The Panel does not here find that the mere existence of a no-
precedent clause renders prior agreements unacceptable as benchmarks 
per se. Rather, after considering the totality of the circumstances, 
we find the no-precedent clause effectively corroborates ASCAP's 
assertion that it voluntarily subsidized Public Broadcasters in the 
past and now declines to continue such subsidization.

Report at 22 (footnote omitted). The record contains other evidence to 
support ASCAP's contention that the negotiated agreements were a 
subsidization to Public Broadcasters. See ASCAP's PFFCL at 126-130, 
Paras. 287-297. Because the Panel's rejection of prior agreements with 
ASCAP is supported by the evidence, the Register cannot disturb it.
    The same can be said for BMI's nondisclosure clause. The Panel 
found that the presence of the clause in the negotiated agreements was 
to prevent use of below-market rates as a benchmark for setting future 
rates, and that ``[n]o other plausible explanation has been offered by 
Public Broadcasters'' as to the existence of the clause. The record 
also contains evidence, aside from the nondisclosure clause, that 
supports the conclusion that BMI considered the negotiated license to 
contain below market rates. See BMI PFFCL at 67-73, Paras. 183-194. The 
Panel's determination is, therefore, neither arbitrary nor contrary to 
the statute.
    3. The Panel improperly relied upon the disparity between the rates 
paid by public broadcasters and commercial broadcasters for ASCAP and 
BMI music as evidence that the voluntary agreements represented a 
subsidy to Public Broadcasters. As further evidence that ASCAP and BMI 
had been voluntarily subsidizing Public Broadcasters in the negotiated 
agreements, the Panel cited the magnitude of the fee disparity that 
existed between public and commercial broadcasters. Public Broadcasters 
assert that the fact that commercial broadcasters pay considerably 
higher fees than public broadcasters is not evidence of a 
subsidization. Rather, it is demonstrative evidence that different 
users of the same goods and services can value such goods and services 
differently. Public Broadcasters also argue that the Panel ``gave undue 
weight'' to the testimony of one of BMI's witnesses in refuting Public 
Broadcasters' contention regarding the lack of probity of the fee 
disparity. Public Broadcasters Petition to Modify at 19.
Recommendation of the Register
    The Panel expressly addressed Public Broadcasters' contention of 
the lack of probity of the fee disparity:

    Public Broadcasters have not, or can not, cite any factual bases 
which might account for the huge disparity between recent ASCAP/BMI 
commercial rates and the rates for Public Broadcasters under prior 
agreements (even after adjusting commercial rates based upon various 
parameters). Public Broadcasters merely offer the general, but 
unhelpful, observation that ``[t]he differences in rates is 
accounted for by the fact that commercial and non-commercial 
broadcasters operate in separate and distinct markets.'' If, for 
example, evidence had been adduced demonstrating that Public 
Broadcasters pay less than commercial broadcasters for other music-
related programming expenses (such as radio disk jockeys, musicians, 
producers, writers, directors, or other equipment operators), the 
Panel might feel more comfortable accepting the heavily discounted 
music license fees as fair market rates. Virtually no such evidence 
was adduced. To the contrary, it appears that Public Broadcasters 
pay rates competitive with commercial broadcasters for other music-
related programming costs such as composers' ``up front fees.'' Tr. 
1636 [testimony of BMI witness Michael Bacon]. As discussed, infra, 
the Panel is cognizant that commercial and non-commercial 
broadcasters do, in fact, operate under different economic models 
and one should not be surprised that these models yield somewhat 
different results, including differences in fair market rates. It is 
the magnitude of the disparity that causes the Panel to further 
question whether the rates negotiated under prior agreements truly 
constituted fair market rates. We have concluded they do not.

Report at 23 (citation omitted).
    The Register concludes that the Panel's explanation of its 
consideration of the fee disparity is well-articulated and reasonable, 
and is not arbitrary or contrary to the Copyright Act. And, as the 
Register has made clear on several occasions, absent compelling 
evidence to the contrary, the Register will not disapprove the weight 
accorded by a CARP to the testimony of a witness. See, e.g. 62 FR 55757 
(October 28, 1997).

C. Conclusion

    Having fully analyzed the record in this proceeding and considered 
the contentions of the parties, the Register recommends that the 
Librarian of Congress adopt the rates and terms for the use of ASCAP 
and BMI music by Public Broadcasters as set forth in the CARP's report.

Order of the Librarian

    Having duly considered the recommendation of the Register of 
Copyrights regarding the report of the Copyright Arbitration Royalty 
Panel in the matter of adjustment of the royalty rates and terms for 
the noncommercial educational broadcasting compulsory license, 17 
U.S.C. 118, the Librarian of Congress fully endorses and adopts her 
recommendation to accept the Panel's decision. 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, and 
amending the rules of

[[Page 49836]]

the Library and the Copyright Office, announcing new royalty rates and 
terms for the section 118 compulsory license.

List of Subjects in 37 CFR Part 253

    Copyright, Music, Radio, Television.

Final Regulation

    In consideration of the foregoing, the Library of Congress amends 
part 253 of 37 CFR as follows:

PART 253--USE OF CERTAIN COPYRIGHTED WORKS IN CONNECTION WITH 
NONCOMMERCIAL EDUCATIONAL BROADCASTING

    1. The authority citation for part 253 continues to read as 
follows:

    Authority: 17 U.S.C. 118, 801(b)(1) and 803.

    2. Section 253.3 is added to read as follows:


Sec. 253.3  Performance of musical compositions in the repertory of 
ASCAP and BMI by PBS and NPR and other public broadcasting entities 
engaged in the activities set forth in 17 U.S.C. 118(d).

    (a) Scope. This section shall apply to the performance during a 
period beginning January 1, 1998, and ending on December 31, 2002, by 
the Public Broadcasting Service (PBS), National Public Radio (NPR) and 
other public broadcasting entities (as defined in Sec. 253.2) engaged 
in the activities set forth in 17 U.S.C. 118(d) of copyrighted 
published nondramatic musical compositions in the repertory of the 
American Society of Composers, Authors and Publishers (ASCAP) and 
Broadcast Music, Inc. (BMI), except for public broadcasting entities 
covered by Secs. 253.5 and 253.6.
    (b) Royalty rates. The following annual royalty rates shall apply 
to the performance of published nondramatic musical compositions within 
the scope of this section: $3,320,000 to ASCAP, and $2,123,000 to BMI.
    (c) Payment of royalties. The royalty payments specified in 
paragraph (b) of this section shall be made in two equal payments on 
July 31 and December 31 of each calendar year, except for 1998, in 
which year the royalty payments shall also be made in two equal 
installments, the first of which shall be made within thirty (30) days 
from the date the Librarian of Congress renders his decision in In the 
Matter of Adjustment of the Rates for Noncommercial Educational 
Broadcasting Compulsory License, Docket No. 96-6 CARP NCBRA, and the 
second of which shall be made on December 31, 1998, subject to 17 
U.S.C. 802(g).
    (d) Identification of stations. PBS, NPR and/or the Corporation for 
Public Broadcasting (CPB) shall annually for the years 1999-2002, by 
not later than January 31 of each such calendar year, and in 1998, 
within thirty (30) days of the date the Librarian of Congress renders 
the decision in In the Matter of Adjustment of the Rates for 
Noncommercial Educational Broadcasting Compulsory License, Docket No. 
96-6 CARP NCBRA, furnish to ASCAP and BMI a complete list of all public 
broadcasting entities within the scope of this section, as of January 1 
of that calendar year. Such lists shall include:
    (1) A list of all public broadcasting entities operating as 
television broadcast stations that are associated with PBS (``PBS 
Stations''), and the PBS licensee with which each PBS Station is 
associated (``PBS Licensees''), identifying which PBS Licensees are 
Single Feed Licensees and which are Multiple Feed Licensees, and which 
PBS Stations or groups of stations are Independently Programmed 
Stations, as those terms are defined in paragraph (e)(2) of this 
section;
    (2) A list of all public broadcasting entities operating as 
television broadcast stations that are not associated with PBS (``Non-
PBS Stations'');
    (3) A list of all public broadcasting entities operating as radio 
broadcast stations that are associated with NPR (``NPR Stations''), 
which list shall designate which NPR Stations have six (6) or more 
full-time employees and which NPR Stations repeat one hundred (100) 
percent of the programming of another NPR Station; and
    (4) A list of all public broadcasting entities operating as radio 
broadcast stations that are not associated with NPR (``Non-NPR 
Stations''), which list shall designate which Non-NPR Stations have six 
(6) or more full-time employees.
    (5) For purposes of this section, Non-PBS Stations and Non-NPR 
Stations shall include, but not be limited to, public broadcasting 
entities operating as television and radio broadcast stations which 
receive or are eligible to receive general operational support from CPB 
pursuant to the Public Broadcasting Act of 1967, as amended.
    (e) Records of use. (1) PBS and NPR shall maintain and, within 
thirty-one (31) days after the end of each calendar quarter, furnish to 
ASCAP and BMI copies of their standard cue sheets listing the 
nondramatic performances of musical compositions on PBS and NPR 
programs during the preceding quarter (including to the extent such 
information is reasonably obtainable by PBS and NPR the title, author, 
publisher, type of use, and manner of performance thereof). PBS and NPR 
will make a good faith effort to obtain the information to be listed on 
such cue sheets. In addition, to the extent the information is 
reasonably obtainable, PBS shall furnish to ASCAP and BMI the PBS 
programming feed schedules including, but not limited to, the PBS 
National Programming Service schedule. PBS and NPR shall make a good 
faith expeditious effort to provide the data discussed in this 
paragraph in electronic format where possible.
    (2) PBS Licensees shall furnish to ASCAP and BMI, upon request and 
designation of ASCAP and BMI, music use reports listing all musical 
compositions broadcast by a particular PBS Station owned by such PBS 
Licensee showing the title, author, and publisher of each composition, 
to the extent such information is reasonably obtainable; provided, 
however, that PBS Licensees shall not be responsible for providing cue 
sheets for programs for which cue sheets have already been provided by 
PBS to ASCAP and BMI. PBS Licensees will make a good faith effort to 
obtain the information to be listed on such music use reports. In the 
case where a PBS Licensee operates only one (1) or more PBS Stations 
each of which broadcasts simultaneously or on a delayed basis all or at 
least eighty-five (85) percent of the same programming (a ``Single Feed 
Licensee''), that Single Feed Licensee will not be obligated to furnish 
music use reports to either ASCAP or to BMI for more than one of its 
PBS Stations in each calendar year. In the case where a PBS Licensee 
operates two (2) or more PBS Stations which do not broadcast all or at 
least eighty-five (85) percent of the same programming on a 
simultaneous or delayed basis (a ``Multiple Feed Licensee''), that 
Multiple Feed Licensee may be required to furnish a music use report 
for each PBS Station or group of stations which broadcasts less than 
eighty-five (85) percent of the same programming as that aired by any 
other PBS Station or group of stations operated by that Multiple Feed 
Licensee (such station or group of stations being referred to as an 
``Independently Programmed Station'') in each calendar year. In each 
calendar year, ASCAP and BMI shall each be limited to requesting music 
use reports from PBS Licensees covering a total number of PBS Stations 
equal to no more than fifty (50) percent of the total of the number of 
PBS Single Feed Licensees plus the number of Independently Programmed 
Stations operated by Multiple Feed Licensees;

[[Page 49837]]

provided, however, that ASCAP and BMI shall be entitled to receive 
music use reports covering not less than ninety (90) PBS Stations in 
any given calendar year. Subject to the limitations set forth above, 
PBS Stations shall be obligated to furnish to ASCAP and BMI such music 
use reports for each station for a period of no more than seven days in 
each calendar year.
    (3) Non-PBS Stations shall furnish to ASCAP and BMI, upon request 
and designation of ASCAP and BMI, music use reports listing all musical 
compositions broadcast by such Non-PBS Stations showing the title, 
author and publisher of each composition, to the extent such 
information is reasonably obtainable. Non-PBS Stations will make a good 
faith effort to obtain the information to be listed on such music use 
reports. In each calendar year, ASCAP and BMI shall each be limited to 
requesting music use reports from no more than fifty (50) percent of 
Non-PBS Stations. Subject to the limitations set forth above, Non-PBS 
Stations shall be obligated to furnish to ASCAP and BMI such music use 
reports for each station for a period of no more than seven days in 
each calendar year.
    (4) NPR Stations which have six (6) or more full-time employees 
shall furnish to ASCAP and BMI, upon request and designation of ASCAP 
and BMI, music use reports listing all musical compositions broadcast 
by such NPR Station showing the title, author or and publisher of each 
composition, to the extent such information is reasonably obtainable; 
provided, however, that NPR Stations shall not be responsible for 
providing cue sheets for programs for which cue sheets have already 
been provided by NPR to ASCAP and BMI. NPR Stations will make a good 
faith effort to obtain the information to be listed on such music use 
reports. In each calendar year, ASCAP and BMI shall each be limited to 
requesting music use reports from no more than fifty (50) percent of 
NPR Stations which have six (6) or more full-time employees. 
Notwithstanding the foregoing, if the number of NPR Stations with six 
(6) or more employees (from which ASCAP and BMI shall initially 
designate and request reports) falls below twenty-five (25) percent of 
the total number of all NPR Stations, then ASCAP and BMI may each 
request reports from additional NPR Stations, regardless of the number 
of employees, so that ASCAP and BMI shall each be entitled to receive 
music use reports from not less than twenty-five (25) percent of all 
NPR Stations. NPR Stations shall be obligated to furnish music use 
reports for each station for a period of up to one week in each 
calendar year to ASCAP and BMI.
    (5) Non-NPR Stations which have six (6) or more full-time employees 
shall furnish to ASCAP and BMI, upon request and designation of ASCAP 
and BMI, music use reports listing all musical compositions broadcast 
by such Non-NPR Station showing the title, author and publisher of each 
composition, to the extent such information is reasonably obtainable. 
Non-NPR Stations will make a good faith effort to obtain the 
information to be listed on such music use reports. In each calendar 
year, ASCAP and BMI shall each be limited to requesting music use 
reports from no more than fifty (50) percent of the Non-NPR Stations 
which have six (6) or more full-time employees. Notwithstanding the 
foregoing, if the number of Non-NPR Stations with six (6) or more 
employees (from which ASCAP and BMI shall initially designate and 
request reports) falls below twenty-five (25) percent of the total 
number of all Non-NPR Stations, then ASCAP and BMI may each request 
reports from additional Non-NPR Stations, regardless of the number of 
employees, so that ASCAP and BMI shall each be entitled to receive 
music use reports from not less than twenty-five (25) percent of all 
Non-NPR Stations. Non-NPR Stations shall be obligated to furnish music 
use reports for each station for a period of up to one week in each 
calendar year to ASCAP and BMI.

    So Ordered.
James H. Billington,
The Librarian of Congress.

    Dated: September 17, 1998.

    So Recommended.
Marybeth Peters,
Register of Copyrights.
[FR Doc. 98-24986 Filed 9-17-98; 8:45 am]
BILLING CODE 1410-33-P