Copyright Royalty Tribunal

NOTICES
 
COPYRIGHT ROYALTY TRIBUNAL
 
(Docket No. CRT. 91-2-89CD)
 
1989 Cable Royalty Distribution Proceeding
 
Monday, April 27, 1992
 
AGENCY: Copyright Royalty Tribunal.

 *15287 ACTION: Notice of final determination.

 SUMMARY: The Copyright Royalty Tribunal announces the adoption of its final determination in the proceeding concerning the distribution to certain copyright owners of royalty fees paid by cable systems for secondary transmissions during 1989.

 FOR FURTHER INFORMATION CONTACT: Robert Cassler, General Counsel, Copyright Royalty Tribunal, 1825 Connecticut Avenue NW., suite 918, Washington, DC 20009 (202-606-4400).

 SUPPLEMENTARY INFORMATION:

 Authority

  Section 111(d)(3) of the Copyright Act authorizes the Copyright Royalty Tribunal each year to distribute to certain copyright owners royalty fees paid by cable systems. The eligible copyright owners are those owners whose works were broadcast on television or radio stations and retransmitted by cable systems in a distant market.

 This Proceeding

  In this proceeding, the Tribunal takes up the distribution of the royalty fees deposited by cable operators for the calendar year 1989. In accordance with past procedure, the Tribunal resolved that the 1989 distribution proceeding would be conducted in two phases. In Phase I, the Tribunal would determine the allocation of cable royalties among various program categories of claimants.

 The Phase I categories were: Program Suppliers, Sports, U.S. Noncommercial Television (PBS), U.S. Commercial Television (NAB), Music, the Devotional Claimants, the Canadian Claimants, Noncommercial Radio (NPR), and Commercial Radio.

 In this proceeding, there was a controversy among five Phase I categories: Program Suppliers, Sports, NAB, PBS, and the Devotional Claimants. The claimants within the remainder of the Phase I categories settled for the same entitlement as they received in the 1983 cable royalty distribution.

 In Phase II, the Tribunal determines the allocation of cable royalties to individual claimants within a category. There were no disputes in Phase II. Consequently, the only controversy to be heard by the Tribunal was the Phase I controversy among Program Suppliers, Sports, NAB, PBS and the Devotional Claimants.

 The Claimants

  699 individual or joint claims were filed for the 1989 cable royalty fund. The Copyright Act provides that claimants may agree among themselves as to the proportionate division of compulsory licensing fees. Accordingly, the claimants coalesced into eight claimant groups.

 Motion Picture Association of America (MPAA). MPAA is the major claimant in the Program Suppliers' category. MPAA is a trade association which represents 104 producers and distributors of syndicated series, movies and television specials. Cooper, p. 2.

 The Joint Sports Claimants (JSC). The Joint Sports Claimants consist of Major League Baseball, the National Basketball Association, the National Hockey League, and the National Collegiate Athletic Association. Vincent, p. 1.

 Public Broadcasting Service (PBS). PBS is part of the U.S. noncommercial television claimant group consisting of PBS and 340 claimant member television and television satellite stations, and 85 producers of public television programs. PTV Ex. JL-2.

 National Association of Broadcasters (NAB). NAB is a trade association which represents U.S. commercial television and radio broadcasters for the programs they produce. Phase I Direct Case Statement of NAB.

 The Music Claimants (Music). The Music Claimants consist of three performing rights societies: The American Society of Composers, Authors and Publishers (ASCAP), Broadcast Music, Inc. (BMI), and SESAC, Inc. (SESAC). Phase I Direct Case Statement of Music (withdrawn).

 The Devotional Claimants. The Devotional Claimants consist of the Christian Broadcasting Network, Inc. (CBN), Old Time Gospel Hour (OTGH), Christian Television Corporation, Inc. (CTC), Heritage Ministries, Inc. (Heritage), Oral Roberts Evangelistic Association (OREA), In Touch Ministries (ITM), and Coral Ridge Ministries Media Outreach Division (CRM). Several Notices of Intent to Participate.

 The Canadian Claimants. The Canadian Claimants represent Canadian programs broadcast by Canadian television stations. The group consists of Canadian Broadcasting Corporation (CBC), CFAC-TV, Lethbridge Television, Ltd; CFAC-TV, Calgary Television Ltd; COGECO, Inc.; Glen-Warren Productions; Niagara Television, Ltd; Tele-Metropole, Inc.; TV-Ontario; Danger Bay Productions, Inc.; New Wilderness Productions, Inc.; Owl-TV; Nelvana Ltd & NPL Ltd; RSL/Alliance Entertainment Corp.; Settler Film Prods. Inc.; Spectrafilm Inc.; Sullivan Films; Societe de Radio-Television du Quebec. CC Ex. CDN-A (withdrawn).

 National Public Radio (NPR). NPR represents NPR and 170 member noncommercial radio stations. Joint Claim, NPR, filed July 31, 1990.

 Background and Chronology

  On February 20, 1991, the Tribunal published a notice directing all claimants to inform the Tribunal by April 19, 1991 whether any controversies existed with regard to the distribution of the 1989 cable royalty fees. 56 FR 6841.

 The Tribunal was informed that a Phase I controversy existed, and controversies existed in several Phase II categories. Accordingly, the Tribunal declared the commencement of the 1989 cable royalty distribution proceeding, effective April 26, 1991. 56 FR 19352. In the same notice, the Tribunal asked the claimants to comment by May 10, 1991 how much of the 1989 cable fund could be distributed while retaining sufficient funds to satisfy all controversies. Id.

 Based on the comments received, the Tribunal determined that it could distribute 70% of the 1989 cable fund, and ordered a 70% distribution to all parties to be made on May 23, 1991. Order, dated May 14, 1991.

 In a separate Order, the Tribunal scheduled the filing of Phase I direct cases for July 12, 1991. Order, dated May 14, 1991. NAB, PBS, the Canadian Claimants and NPR filed motions to extend the time to file direct cases. Their motions were granted, and the Tribunal extended the time for filing to August 16, 1991. Order, dated June 21, 1991.

 On August 16, 1991, the Phase I parties filed their direct cases. IN their filings, NAB and Music reported a complete settlement of the Commercial Radio category to be based on the 1983 allocation decision of the Tribunal.

 Similarly, the Program Suppliers and Music announced a settlement of that portion of the 1989 cable fund derived from the syndicated exclusivity surcharge (syndex) and moved the Tribunal to declare that no controversies existed concerning syndex. This motion was opposed by PBS which had filed testimony asserting a claim to the syndex royalties. The Tribunal declined to render a declaratory ruling and ordered that PBS be afforded an opportunity to give testimony on the issue. Order, dated October 28, 1991.

 On August 26, 1991, NPR reported to the Tribunal that it had reached a settlement of its Phase I, and as a consequence, it withdrew its direct case. The settlement stipulated an award of *15288 0.18% of the total fund, the same percentage NPR has received since the 1983 proceeding.

 Presentation of the Phase I direct cases began on September 12, 1991. On September 10, 1991, the Canadian Claimants reported that they had reached a settlement, and withdrew their direct case. The settlement stipulated an award of 0.75% of the basic fund and 0.25% of the 3.75% royalties, the same percentage the Canadian Claimants have received since the 1983 proceeding.

 Consequently, the Tribunal ordered a full distribution of NPR's and the Canadian Claimants' share of the 1989 fund which was made October 10, 1991.

 On October 15, 1991, the Music Claimants reported a settlement of their Phase I claim, and withdrew their direct case. Music settled for 4.5% of the basic, 3.75% and syndex funds, the same percentage the Music Claimants have received since the 1983 proceeding. The Music Claimants received the remaining protion of their share of the 1989 fund on November 7, 1991.

 Hearing of the direct cases of the five remaining Phase I claimants, Program Suppliers, Sports, NAB, PBS, and Devotional Claimants, concluded on November 5, 1991 after 25 days of hearings.

 The Phase I rebuttal cases were filed November 19, 1991. The hearing of the rebuttal cases commenced on December 3, 1991 and concluded on December 17, 1991 after 10 days of hearings. However, an additional rebuttal hearing was scheduled for January 14, 1992.

 On January 22, 1992, the Tribunal closed the Phase I record. The parties filed Proposed Findings of Fact and Conclusions of Law on January 24, 1992. Reply findings were filed February 3, 1992.

 Structure and Criteria of Tribunal Analysis

  In accordance with past procedure, the Tribunal took evidence based on the criteria established by the Tribunal in the 1978 proceeding:

 (a) The harm caused to copyright owners by secondary transmissions of their copyrighted works by cable systems;

 (b) The benefit cable systems derive from the secondary transmission of certain copyrighted works;

 (c) The marketplace value of the works transmitted; and to a secondary degree;

 (d) The quality of copyrighted program material; and

 (e) Time-related considerations.

 The Tribunal has made Phase I cable royalty distributions subsequent to fully litigated proceedings for calendar years 1978, 1979, 1980 and 1983. For 1982, there was a partially litigated Phase I proceeding. In all other years, the parties reached a negotiated settlement.

 Consequently, the Tribunal approaches this decision based on the cumulative experience and expertise gained from those earlier decisions. The questions considered by the Tribunal in the 1989 proceeding were: Have there been any factual changes since 1983 which justify a change in the awards previously made? Has any party presented better evidence to entitlement than in the past? Should the Tribunal continue the basis upon which it has made its distribution, or should it adopt a new basis?

 While the five Phase I parties made various showings concerning changed circumstances or improved evidence, the heart of the 1989 case concerned whether the Tribunal should continue to base its allocation of cable royalties on the Nielsen study of distant signal viewing or should it be based on the Bortz survey of the importance cable operations give to various program categories.

 The difficulty of the problem was outlined in the 1983 determination. As we stated there, the Tribunal's goal in allocating the fund among various program types is to "simulate market valuation." The achievement of this goal is frustrated by the fact that cable operators do not obtain distant signal programming on a program-by-program basis. The operator "purchases" by the compulsory license entire broadcast signals consisting of a variety of program types. Operators must take the distant signal as is or not at all. Therefore the Tribunal must perform a judgment that does not occur in the distant signal marketplace. It assigns relative values among program types; the cable operator does not.

 Program Suppliers have offered a special Nielsen study of the viewing by cable subscribers of distant signals as the best method by which to determine relative value among program types, on the theory that a program's value is determined by how many people watch it. The Tribunal has traditionally agreed with Program Suppliers, and in past decisions, the Tribunal has used the Nielsen study as its "starting off point."

 The four other parties in Phase I disagree with using the Nielsen study as the starting point. They have argued that ratings are only relevant in an industry like broadcasting which is driven by advertising. However, the cable industry's goal in importing distant signals is not to increase advertising; cable operators cannot insert their own advertising in distant signal. It is to attract and retain subscribers. Consequently, the cable operator's interest in distant signals would be more in the nature of adding diverse programs to their range of offerings or responding to particular segments of their market than in responding to raw viewing data.

 These four parties support the Bortz survey commissioned by the Joint Sports Claimants as the most relevant evidence. That survey asked cable operators to assign the relative value of various program types in terms of how much of a fixed budget each operator would spend on each program type.

 The 1989 cable distribution proceeding was primarily marked by arguments as to which is the best indicator of the market value of the Phase I program categories: the Nielsen study of distant signal viewing, or the Bortz survey of cable operators' budget preferences.

 Findings of Fact

 Program Suppliers

 The Nielsen Study

  As part of its case for several proceedings, Program Suppliers has commissioned a special Nielsen study to establish its entitlement and the entitlement of other claimants. The study measures the viewership in cable households of distant signal nonnetwork programming. Cooper, p. 4.

 For this proceeding, Program Suppliers selected to be included in the Nielsen study U.S. commercial television stations which reached 80,000 subscribers of Form 3 cable systems during the first half year of 1989. Stations which had been available to an average of at least 80,000 subscribers during 1986 to 1988, but not in the first half of 1989, were also included. PS Ex. MEK-8, p. A-6.

 Program Suppliers also included in the study noncommercial educational stations which reached 100,000 subscribers on Form 3 cable systems based on averaging the two accounting periods for 1989. Id.

 A total of 146 stations met Program Suppliers' criteria: 74 network-affiliates, 52 independents, and 20 noncommercial educational stations. Cooper, p. 7; NAB Ex. 39X; Tr. 652-653.

 Once the sampel stations were selected, the study was based on data compiled by the A.C. Nielsen Company from Nielsen Station Index (NSI) diaries distributed on a random basis to television households throughout the U.S. Tr. 402. For its February, 1989 study, Nielsen mailed 213,786 diaries to a sample of designated television households. Usable diaries were *15289 received from 105,556 households, of which approximately 58,000 subscribed to cable service (Form 1, 2, and 3 cable systems included). Thus, over the course of a four-cycle study, diaries were returned and tabulated from over 200,000 cable households. Cooper, p. 5; JSC Ex. 50RX; Tr. 5599.

 The Nielsen study measured six four-week periods, or "cycles." During 1989, these cycles encompassed: January 5-February 1, February 2-March 1, April 27- May 24, July 6-August 2, September 28-October 25, and November 2-November 29. PS Ex. MEK-8, p. A-67.

 For the four principal cycles, February, May, July, and November, all 146 stations were measured, except for WWSB, Sarasota which was not measured during the July cycle. Cooper, p. 7. 19 stations were studied for five cycles. 78 stations were studied for all six cycles.Id.

 Although the 146 stations did not represent the total number of stations that were carried in distant markets by cable systems, they represented 96.2% of the fees paid by Form 3 cable systems in 1989. Id.

 The cable household viewing hours found by the Nielsen study were broken down to five of the nine Phase I categories: Program Suppliers, sports, NAB, PBS and the Devotional Claimants. The study did not measure Music, Canadian Claimants' programs, noncommercial radio or commercial radio.

 The four-cycle data, as initially presented by Program Suppliers showed the following household viewing hours:

 
 -------------------------------------------------
                                          Percent
 -------------------------------------------------
Program Suppliers ....... 2,087,360,979 ... 80.471
Sports .................... 328,391,280 ... 12.660
NAB ....................... 118,268,531 .... 4.559
BS ......................... 53,472,341 .... 2.061
Devotional Claimants ........ 5,589,086 ..... .215
Other ......................... 832,513 ..... .032
                         -------------- PS Ex. 1.
 -------------------------------------------------
 

 The six-cycle data, as initially presented by Program Suppliers showed the following household viewing hours:

 
 -----------------------------------------------
                                         Percent
 -----------------------------------------------
Program Suppliers ....... 3,166,907,013 . 83.858
Sports .................... 370,960,051 .. 9.823
NAB ....................... 156,386,226 .. 4.141
PBS ........................ 73,096,288 .. 1.936
Devotional Claimants ........ 8,243,619 ... .218
Other ......................... 926,773 ... .025
                         -------------- Id.
 -----------------------------------------------
 

 Program Suppliers asserted that the 1989 study was an improvement over the 1983 study because the 146 stations it studied represented an increase of 29 stations over the 117 stations it studied in 1983, including 18 additional network-affiliates, 7 additional independents, and 4 additional PBS stations. Cooper, pp. 6-7; NAB Ex. 39X; Tr. 652-653.

 Program Suppliers also eliminated its 1983 categories of Minor Sports and Specialty Stations, and reassigned the programs contained therein to their appropriate Phase I categories. Kessler, pp. 12-13. In addition, Program Suppliers reassigned from the Other category to the Program Suppliers category syndicated parades and political programs. id., pp. 11-12.

 Viewing-to-Time Ratios

  Program Suppliers provided data to show that its share of viewing is greater than its share of the actual time its programs are on the air on the belief that the viewing-to-time ratio is "indicative of the relative popularity and value" of a program category. Cooper, pp. 9-10; Tr. 515.

 In the four-cycle data, as initially presented by Program Suppliers, movies' share of viewing is 27.267% while its share of time is 11.487%, or a ratio of about 2.4:1. Television series' share of viewing is 53.204% while is share of time is 52.25%, or a ratio of 1.01:1. Sports' share of viewing is 12.66% while its share of time is 1.203%, or a ratio of about 10.5:1. Tr. 516. The other categories, NAB, PBS and Devotionals, were below a ratio of 1:1.

 However, NAB's cross-examination demonstrated that a program that had one hour of time on a superstation would more likely result in a higher viewer-to-time ratio than a program that had one hour of time on a regional station, because a superstation reaches many more viewers. Therefore, the viewing-to-time ratio could be more a function of the access to viewership, than the intensity of the viewers, and would unfairly affect a program category like NAB which has most of its shows on regional stations. Tr. 581-598.

 Harm

  Jack Valenti, President of MPAA, testified on the harm to program syndicators of distant signal retransmission by cable operators. The license fees paid by the networks to the creators of first run television programs are less than what it costs to produce them. An average 30-minute sit-com costs $687,000 per episode to produce in 1989-90, yet the average license fee from the networks was $477,000 per episode. An average 60-minute drama cost $1.24 million per episode, yet the average license fee from the networks was $940,000 per episode. Valenti, p. 4.

 Similarly, movie production requires substantial up-front costs. The average production cost for a feature film in 1989 was $23.5 million. Advertising costs averaged $7.8 million. Print costs averaged $1.4 million. id.

 Owners of television series and movies depend on revenues obtained from subsequent distribution for recovery of costs plus profit. Broadcast stations that purchase syndicated series and movies for subsequent exhibitions can experience audience decline for a program as a result of cable systems importing distant signals. Lower audiences for broadcasters lead to less favorable terms for licensing these programs. Id., pp. 4-5.

 Mr. Valenti offered no statistical data on the degree to which program suppliers are harmed by distant signal importation, but referred to a general finding by the FCC that substantial audience diversion occurs from distant signal importation. Id., p. 5.

 Claim

  Based on the above proffered showing, Program Suppliers claim 70.25% of the basic fund, 74% of the 3.75% fund, and 95.5% of the syndex fund. PS Prop. Findings, p. 129.

 Criticisms of the Nielsen Study Methodology

  Program miscategorizations. NAB asserted that there were programs that Program Suppliers categorized as "syndicated," but should have been categorized as "locally-produced," and assigned to NAB. Ducey Reb., p. 1. Program Suppliers accepted these recategorizations. Tr. 4300.

 Accordingly, the four-cycle data, as corrected by NAB, increased NAB's Nielsen share to 6.076%, and lowered Program Suppliers' share to 78.954%. NAB Ex. 51R. The six-cycle data, as corrected by NAB, increased NAB's share to 5.783% and decreased Program Suppliers' share to 82.215%. NAB Ex. 50R.

 Exclusion of qualifying stations. Although MPAA intended the Nielsen study to include all commercial stations which reached 80,000 Form 3 cable system subscribers during the first half year of 1989, 22 stations who met the criterion at the time the sample was drawn were not included. NAB Ex. 38X.

 MPAA witness Allen Cooper explained that one station, WCDC, was not included because it is a satellite station of WTEN and Nielsen does not provide separate data for WCDC. Tr. 637. WPCB, a religious specialty station, and KSCI and KTMD, two foreign-language specialty stations, were not *15290 included by Mr. Cooper because measurements of those stations by Nielsen are reported by daypart only. They do not have sufficient audience to be reportable on a program-by-program basis. Tr. 705, 709-710. Other stations were excluded by Mr. Cooper based on his judgment concerning the ultimate number of subscribers that would be found to have been reached by the station in question. Tr. 644- 645. However, Mr. Cooper could not explain why other stations, like WNBC, were excluded, because in his judgment, they should have been included. Tr. 638.

 Similarly, although MPAA intended to include in its study all noncommercial stations which reached 100,000 Form 3 cable subscribers during the first half of 1989, one station, KCTS, Seattle, Washington, which had an average reach of 163,000 subscribers during 1989, was excluded. Tr. 194. Five other PBS stations would have been included in the study if the selection of PBS stations were made at the same standard as commercial stations. Tr. 195-196.

 Sample stations not representative of universe. The sample of independent, network-affiliated and educational stations chosen by MPAA did not reflect the universe of distantly-carried television stations. The 52 independents accounted for 94.4% of all independent station Form 3 cable subscriber incidents, while the 74 network stations accounted for only 59.5% of all network-affiliated Form 3 cable subscriber incidents. The 20 educational stations accounted for 61.2% of all educational Form 3 cable subscriber incidents. NAB Ex. 40X.

 Responding to the criticism, Program Suppliers furnished an adjustment of the six-cycle percentages which brought independent, network-affiliated and educational stations to equal weight in the study. This adjustment included NAB's recategorizations. The adjustment was stratified to take into account any differences in station type (i.e., independent, network, educational) and station size (i.e., the number of households reached). Frankel, p. 7. The results of the stratified adjustments were:

 
Program Suppliers ...................... 81.56%
Sports .................................. 9.05
NAB ..................................... 6.38
PBS ..................................... 2.72
Devotional Claimants .................... 0.26
Other ................................... 0.03
                        PS Ex. MF/1R (adjusted)
 

 Underrepresentation of baseball in 6-cycle data. Major League Baseball is broadcast on a nonnetwork basis for seven months during the year, including six months of the regular season and one month pre-season, or about 58.3% of the year. Tr. 507. Baseball playoffs and the World Series are in October, but those are network telecasts, ineligible for cable royalty compensation. Tr. 508.

 In the four-cycle data, baseball is measured in two of the four sweep periods, or 50% of the time. Id. In the six-cycle data, baseball is measured in two of the six sweep periods, plus three days of a third period. PS. Ex. 1. That amounts to a 35% representation of baseball (59 days divided by 168 days) in the six-cycle data, resulting in an underrepresentation of baseball.

 Questioned estimating techniques for January and October cycles. January and October viewing data, the fifth and sixth cycles of the Nielsen study, are based not on diaries from the 200 designated market areas (DMAs) of the country, but on diaries from 18 markets and 23 markets, respectively. Tr. 5621. These 18-23 markets represent 35 to 40 percent of the viewing audience. Consequently, Nielsen must estimate the viewing in the remaining 60 percent of the country. Tr. 5624.

 To do this, Nielsen first determines the local viewing to a particular show (in January or October). Then to construct the distant viewing, it refers back to the next earliest four-cycle sweep period (except July which Nielsen does not consider a representative period), and for that same time period, it ascertains the ratio of distant-to-local viewing that occurred during the same daypart. Nielsen then applies the ratio to the January or October show, and determines the distant viewing. Tr. 5626-5636.

 For example, if the ratio was 15,000 distant viewers to every 85,000 local viewers for the earlier show shown in the comparable daypart, and the show in January or October was viewed by 42,500 households in the local market, Nielsen would estimate that 7,500 households viewed it in distant markets. Tr. 5642- 5643.

 Paul Lindstrom, Vice President of Nielsen Home Video Index, could not guarantee that the show in October would be the same show in the comparable daypart in May. A movie might be shown on a Saturday afternoon in October, while a baseball game might occupy that same time slot in May. Tr. 5632.

 Questioned inclusion of 3.75% viewing. PBS witness Nat Katzman asserted that the Nielsen study unfairly weighed against PBS, because it included viewing of distant stations that were paid for at the 3.75% rate. If the Nielsen percentages were to be the starting point for an allocation of the basic fund, he reasoned, the 3.75% viewing should be excluded. PTV Ex. NK-2, pp. 3-4.

 In the six-cycle data, 395,517,463 viewing hours were attributed to 3.75% viewing. Id., p. 7. Subtracting those hours from the six-cycle data would increase PBS' percentage share from 1.936% to 2.162%.

 Criticisms of the Validity of the Nielsen Study.

 Sports witness Dr. Leonard Reid, a doctor of communications research, testified that, in his opinion, the Nielsen study does not measure actual behavior because it depends on individuals completing diaries, which are often reconstructed from memory. Viewers may forget which programs they watched; they may make honest reporting mistakes; and they may enter false viewing choices to reflect more "socially acceptable" viewing. Furthermore, viewers have no stake in the measuring process. Reid, p. 15. Dr. Reid noted that there are now more viewing options on a cable system--30 to 40 channels--so the task of filling out diaries has become more complicated. Tr. 1428.

 Dr. Reid also noted the problem of nonresponse error. Diaries are typically completed by half the households sampled, and in his opinion, those who fill out and return diaries tend to be different from nonrespondents in significant ways. Reid, pp. 14-15.

 In rebuttal to Dr. Reid's criticisms, Mr. Lindstrom asserted:

 (1) There is a difference in the effect of short-term recall and long-term recall. Long-term recall, such as recalling viewing for the past week, may be inaccurate, but short-term recall, generally is not. Mr. Lindstrom believe that the diaries are generally filled out at the same time or within a brief period of actual viewing. Tr. 5559-60.

 (2) Nielsen has taken steps to alleviate the confusion caused by the multiple cable channel offerings. The 1989 diaries included listings of all channels and all channel numbers on the subscribers' system. Further, Nielsen codes each of 10,000 cable headends, so that their editors can correctly credit viewing entries even if the subscriber made a mistake. Tr. 5561.

 (3) Nielsen recognizes that the nonrespondent differs from the respondent and weights the data accordingly. Tr. 5558.

 Joint Sports presented Mr. William S. Rubens, consultant to NBC, ABC, and CBS, on Nielsen procedures, and former member of the Electronics Media *15291 Ratings Council (EMRC) Board which audits and accredits audience measurement services. Mr. Rubens observed that MPAA's Nielsen study does not use the best available data. While the study relies on NSI diary data alone for all markets, Nielsen itself has been moving to meter its markets. By 1989, NSI meters were available in 13 markets which provided viewing estimates for more than 30% of the nation's households. Mr. Rubens, pp. 1, 4-5; Tr. 4483- 86. In metered markets, Nielsen only uses the diaries for demographic purposes. The meters are used to measure viewing. Tr. 4448. However, Mr. Lindstrom observed that mixing meter data with diary data could invalidly alter the percentage viewing shares of each Phase I program type. Tr. 5584-85.

 Mr. Rubens noted that although the NSI diaries have been subject to industry scrutiny, the special MPAA-designed study has not, and because it's not the kind of data advertisers or broadcasters would use, it would not be subject to EMRC accreditation. Tr. 4431-32. However, Mr. Rubens conceded that special research he conducted for NBC similarly did not seek or receive EMRC accreditation. Tr. 4478-79.

 Mr. Rubens also faulted the MPAA-Nielsen study for not providing information on the distant market samples, so that one could determine whether the sampling plan was adequate. Certain distant viewing to individual stations in the study could have such low audiences that the error rate could be quite high. No standard error estimates were provided by Nielsen. Rubens, pp. 6-7; Tr. 4460-61.

 Finally, Mr. Rubens disagreed that all cable houshold viewing hours have equal value, as implied by the Nielsen study. From Mr. Rubens' industry experience, advertisers would pay significantly different amounts for the same ratings, depending on such factors as age, sex, educational and income levels of the viewers, the nature of the program, and the time of its viewing. Rubens, p. 8.

 Diary errors. Sports presented Dorothy Stein, president of the Market & Audience Research Group which reviews and analyzes thousands of Nielsen and Arbitron audience diaries for clients. Stein, pp. 1-2. Stein believes that diaries do not report actual behavior, but only estimates of reported behavior, because diary-keeping is subject to many potential errors. Id., p. 4.

 Ms. Stein believes the Nielsen sample is subject to nonresponse error, meaning that nonrespondents are different that respondents. Younger persons, blacks and Hispanics, are mostly underrepresented. Although Nielsen weights its results for this problem, Ms. Stein does not believe that corrects the problem. Tr. 4347-49.

 Ms. Stein observed that diaries may contain illegible or confusing entries. In such cases, Nielsen applies certain "editing rules." Stein, p. 8. Although these rules are designed to save viewing credit for those who deserve it, nonetheless the editing rules can result in the improper crediting of viewing to programs that were not in fact viewed, or a failure to credit programs that were viewed. Id.; Tr. 4359-63.

 Ms. Stein believes there is inevitably processing error when so much data must be processed. Nielsen keypunch operators process, at a minimum, 50 million diary entries each year. Id.; Tr. 4372.

 In recent years, Ms. Stein notes trends that make it more difficult to obtain accurate data. Cable systems offer many more channels, and often switch their placement, adding to the diary-keeper's confusion. Tr. 4340-41; 4353-55. Consequently, the diary itself is more complicated than in earlier years. Tr. 4355. In addition, the use of remote control leads to channel-switching, and it is unclear that the diary-keeper is noting the predominantly watched channel for each quarter hour. Tr. 4350-52. In rebuttal, Mr. Lindstrom asserted that channel-switching resulted in a "wash" as far as program percentages were concerned. Tr. 5586-87.

 Ms. Stein was asked whether these data problems affect one of the Phase I program categories more than the other. It was her opinion that "normative" behavior, the programs the household usually watches, would not suffer from these problems as much as "nonnormative" behavior, the programs the household doesn't normally watch. Prime-time network affiliates get reported more than independent stations. If the diary-keeper is the wife, then children's programs and the husband's sports-watching might not get reported as much. The phenomenon of "nonnormative behavior" might affect Spanish-language programming or PBS or out-of-town local news. Tr. 4383-84; 4410. Ms. Stein had no hard data to offer on this point. Tr. 4384. Mr. Lindstrom also observed that independent stations tend to be underreported. Tr. 5584. Criticism of the Relevance of the Nielsen Study.

 PBS witness Nat Katzman argued that subscriber viewing is not a fair indicator of the value of programming to cable operators. In his opinion, cable operators are interested in carrying channels whose presence will encourage households to subscribe to cable. Hence, there are "niche" services that succeed, not because of substantial viewers, but because they offer a type of programming cable operators perceive as adding value to their overall package. Katzman, p. 2. Examples of such niche services are MTV and the Family Channel, delivered to about 90% of all cable households, but with average audiences of less than 0.6%. Headline News Network reaches 76% of cable subscribers, but its average audience is only 1 in 250 households. The Discovery Channel reaches 89% of cable subscribers, but its average audience is less than 1 in 200 households. Id., p. 3.

 Questioned Compliance With CRT Orders and Rules

  Access to diaries. On September 5, 1991, the Tribunal granted the request of the Joint Sports Claimants for access to the Nielsen viewing diaries. The Tribunal stated that it "expects that MPAA will make arrangements with Nielsen to permit JSC promptly to inspect the material as long as reasonably necessary, and to make copies (minus any identifying information), without the three-week advanced notice mentioned in Nielsen's procedures." Order, dated Sept. 5, 1991.

 Program Suppliers transmitted to Nielsen the Tribunal's Order the same day. PS Opposition to Motion to Strike, filed Oct. 2, 1991. On September 23, 1991, JSC filed a motion to strike the Nielsen study. JSC stated that it could not accept Nielsen's requirement that the over 200,000 diaries be copied by hand. JSC argued that it was necessary that it be provided the opportunity to make photocopies. JSC also questioned other Nielsen requirements--that the diaries be transported from one location to another and that their inspection be supervised by Nielsen employees--because of what it believed was the burden, time and substantial cost. JSC' Motion to Strike, filed Sept. 23, 1991.

 Program Suppliers stated that they believed they had obtained from Nielsen adequate arrangements--waiver of the three-week waiting period, allowing inspections to last more than one day--and they agreed to stipulate to the accuracy of hand-copied information. PS' Opposition to Motion to Strike, filed Oct. 2, 1991.

 The Tribunal ruled that it would not strike the Nielsen study, but that argument could be made as to how much weight should be accorded the study given the access restrictions placed by Nielsen. Order, dated Nov. 6, 1991.

 *15292 When asked to explain why the Nielsen company would not allow photocopying of the diaries, Mr. Lindstrom said, "The question would come down to people would love to be able to go in and photocopy the diaries because then they would be able to take that and generate out whatever type of special analysis they like. We're talking about an implication of millions of dollars on that precedent * * *." Tr. 5679-80.

 Mr. Lindstrom stated that his company policy of not allowing photocopying extended to resisting court subpoenas, and Nielsen would only allow it if they were legally obliged to comply with a subpoena. Tr. 5680.

 Confidence intervals. When Program Suppliers filed their direct case, the Nielsen study did not include any estimate of standard errors. At hearing, a request was made by NAB for the confidence intervals surrounding the Nielsen study. Tr. 299.

 On September 19, 1991, Program Suppliers stated that Nielsen was in the process of computing the confidence intervals. Tr. 618. On November 4, 1991 Program Suppliers informed the Tribunal that Nielsen stated that it would supply the requested information by the week of November 18. The Tribunal thereupon ordered the production of the confidence intervals by November 18. Order, dated Nov. 4, 1991.

 On December 20, 1991, one month after the November deadline, Program Suppliers transmitted a letter from Nielsen containing the standard errors for four categories: total programming, syndicated programming, movies, and other programming. Letter, dated December 20, 1991. The data were as follows:

 
 ------------------------------------------------------------------------------
        Category Percent 95% confidence Relative error
                            allocation interval (percent) (percent)
 ------------------------------------------------------------------------------
Total Programming .................. 100 NA ................... NA
Syndicated Programming .............. 55 .............. -0.096 .......... 0.17
Non-network movies .................. 28 .............. -0.098 .......... 0.35
Other programming ................... 16 .............. -0.084 .......... 0.53
 ------------------------------------------------------------------------------
 

 The results were based on the viewing numbers obtained for the 146 sample stations with 6-cycle data. PL-1RX. The confidence intervals do not reflect what the percentage shares would be if all distantly-carried television stations were studied. Tr. 5765.

 When asked by the Tribunal why the production of the confidence intervals took so long, Mr. Lindstrom explained, "What it required us to do was to go back and regenerate the data from the actual home-by-home information in order to try and produce estimates of the correlation of viewing and the variances, etc. * * *. We had to recreate all of the routines to process this information in our stat research department, and then go through, generate all the information again, and virtually run this many, many times. Data had to be produced individually for each program type, for each measurement period." Tr. 5575-76.

 In Sports' rebuttal, Mr. Rubens asserted that the confidence intervals for all of the 146 stations combined were not the relevant data. The appropriate procedure would have been to calculate the standard errors for each program category on a station-by-station basis. He noted that the sample size of some distantly-carried stations was very small, and calculating the standard errors for all of the stations collectively masks the size of the errors for the stations individually. Affid. of Rubens, pp. 2-3.

 Joint Sports Claimants

  The Bortz survey. In each of the fully litigated cable distribution proceedings, the Joint Sports Claimants have presented surveys designed to measure the value to cable operators of sports and other programming. 51 FR 12795.

 In this proceeding, JSC retained Paul Bortz, President of Bortz & Co. (Bortz), to conduct its survey. JSC Ex. 1, p. 1. Bortz chose to study all Form 3 cable systems. A stratified random sampling approach was used, with stratification based on royalty payments during the second half of 1988. Id., p. 23. The choice to stratify was based on a belief that a purely random sample would overrepresent the low royalty paying systems. Tr. 825. The sample design included four strata of royalty classes--$2190-31,699, $31,700-109,999, $110,000-$299,999 and $300,000 or more. In the highest stratum, there were 35 cable systems and Bortz chose to do a complete census and survey all 35 systems. In the second stratum, there were 160 systems, and Bortz chose to survey 50%, or 81 systems. In the third stratum, there were 528 systems, and 66 of them were surveyed, or 12.5%. In the fourth stratum, there were 1,254 systems and 62 of them, or 5% were surveyed. Id., pp. 23-24; Tr. 831.

 As a result, of the 1,977 Form 3 systems, 244 were chosen for surveying. Seven of these systems were discarded due to lack of complete signal data, or uncertain status as a cable system, and thus the sample size was 237 systems. Id., p. 30.

 Interviews were completed with 198 respondents from the 237 systems in the sample. The key question in the survey concerning program valuation was answered by 187 respondents, for a response rate of 79%. Id., p. 2. The 187 systems accounted for 25% of the royalties paid in the second half of 1989. Tr. 911.

 Interviewers were not told the name of the client or given any information, other than that on the survey form, regarding the purpose of the survey. Id. Interviews took place from December 4, 1989 to March 8, 1990. Id. 60% of the interviews took place in 1989; 40% took place in 1990. Tr. 829. Interviewers were instructed to ask to speak with the cable system manager, but if he or she were unavailable, to ask to speak with the person most familiar with programming carried by the system. JSC. Ex. 1, p. 32.

 In each interview, the cable system employee was informed which distant signals his or her cable system carried during 1988, and then asked whether the system had added or dropped any distant signals during 1989. Id. p. 33.

 Reminding the interviewee which distant signals his or her system carried in 1989, the interviewer then asked, "Assume you have a fixed dollar amount to spend on the nonnetwork programming carried on these stations; in other words, a programming budget. Please think in terms of what percentage, if any, of the fixed dollar amount you should spend for each type of programming?" Id. p. 35. The results of this question were:

 
 ------------------------------------------------------------------------------
                 Category Percent Confidence interval
                                              valuation (percent)
 ------------------------------------------------------------------------------
Live professional and college sports ............... 34.2 ............... -2.5
Movies ............................................. 31.2 ............... -2.0
Syndicated shows and series ........................ 16.9 ............... -1.5
News and public affairs ............................ 11.8 ............... -1.3
Devotional/religious ................................ 4.3 ............... -0.9
PBS, educational and other public
  television ........................................ 1.3 ............... -0.6
Canadian ............................................ 0.2 ............... -0.2
                                             ------------ ( super1 )
 ------------------------------------------------------------------------------
Id., p.28.
 

 Mr. Bortz modified the 1989 survey in an effort to meet the concerns expressed by the Tribunal in the 1983 decision. First, the 1983 survey was conducted in March, 1985, about 15 months after the end of 1983. In contrast, the 1989 survey was conducted 60% in 1989 and 40% in the first ten weeks of 1990. JSC Ex. 1, p. 6.

 Second, the 1983 survey asked cable operators to allocate 100 percent of the *15293 total value of distant signal programming among program types. The 1989 survey changed the key question. It asked the cable operator to allocate a fixed budget for distant signals among program types. Mr. Bortz did this on the basis that the survey question would more closely resemble a function the cable operator performs. Id., p. 7. Third, Mr. Bortz added two more Phase I categories to the 1989 survey: Devotional and Canadian programming. Id.

 However, despite earlier stated Tribunal concerns, Mr. Bortz continued his practice of not offering the respondent definitions of each category, because of Mr. Bortz' belief that the respondent was sufficiently sophisticated to understand what was being asked. Id. Mr. Bortz also continued his practice of not asking the cable operator about the value of PBS or Canadian programming if the system did not in fact carry PBS and/or Canadian programming during 1989. In Mr. Bortz' view, this was the correct approach since no portion of a programming budget would have been spent on a signal that a system did not carry. Id.

 Testimony supporting the validity of the Bortz survey. Dr. Robert Crandall of the Brookings Institution testified that although the Bortz survey measured the relative total value of each program type, rather than the relative marketplace value of each program type, its results were still valid. Total value is related to marginal value by the price sensitivity or elasticity of cable operator demand for each program type. If the price sensitivity is the same for all program types, the relative total values will be equal to the relative marginal values, and Dr. Crandall stated that he was unaware of any evidence suggesting that the price sensitivity of cable operator demand varied among program types. Therefore, he concluded that the Bortz study provided the best available measure of relative marketplace value of the distant signal program categories. Id., pp. 6-7.

 Dr. Crandall also concluded that the Bortz study would be a valid measure of marketplace value if the cable operator were faced with an all-or-nothing choice for each program type. If each Phase I group were to bargain collectively with cable operators, the maximum license fees each could obtain would equal the total value of each program type, as reflected in the Bortz study. Id., p. 7.

 Dr. Leonard Reid testified that Bortz' use of the constant sum technique was a valid and well-accepted research tool because it reveals relative comparative judgment of items, it eliminates consistent positive, negative or neutral response patterns, and it yields substantial information that predicts how the respondent would likely act in making choices. Reid, pp. 1, 3-4. Dr. Reid conceded that variables, such as price, convenience and availability, might alter a respondent's actual behavior, but reiterated that a constant sum survey has a relationship to actual behavior, given all the variables. Tr. 1603-05.

 Dr. Samuel Book, President of Malarkey-Taylor Research, testified that the fact that the 1989 Bortz survey was conducted at the end of 1989 and the beginning of 1990 cures the survey of its recall problem. Book, pp. 1, 3. He also disagreed with the criticism that the constant sum exercise is not something that cable operators do in reality. Dr. Book asserted that cable operators are frequently called upon to assess the value of program types when deciding to carry a new service or drop an existing service. Dr. Book testified that the modification of the survey to have the cable operators allocate a fixed 1989 program budget made the 1989 survey a more realistic exercise than the 1983 survey. Id., pp. 3-4. Finally, it was Dr. Book's opinion that the brevity of the exercise did not tend to diminish the certainty of the results, because he felt that the questions were clear, the survey was narrowly focused, and the respondents were knowledgeable cable system executives. Id., p. 4.

 Harm. Francis Vincent, Commissioner of Baseball, testified that the compulsory license deprives baseball of control over the distribution of its product, and that baseball receives under the compulsory license significantly less compensation than it would in the free marketplace. Vincent, p. 5.

 Mr. Vincent conceded that baseball is reimbursed between $9 and $10 million by WTBS each year for the adverse effect importing baseball games on WTBS has on local events. Tr. 1715. Another $10 million is paid by other superstations for the same reason. Tr. 1730. However, Mr. Vincent considered these payments far less than compensatory payments. He used baseball's contract with ESPN which is for 175 games at $90 million as the reasonable comparison, and payments less than $90 million in his opinion from the superstations do not adequately compensate baseball. Tr. 1716.

 Benefit. Mr. Bortz sponsored an exhibit of ads that were in three trade publications, Multichannel News, Cablevision, and Cable Marketing. These ads were targeted to operators to induce them to carry a particular superstation, and sports programming on those stations is heavily featured in those ads. JSC Ex. 4; Tr. 879.

 Marketplace value. Mr. Vincent testified concerning what he regarded was the special marketplace value of sports. He noted that sports programming is live, serves to distinguish the programming of a given channel, and represents an ingrained part of American culture. Vincent, pp. 3-4.

 David Stern, the Commissioner of the National Basketball Association, testified that the increased price for NBA telecast rights from 1983 to 1989 indicated a increased marketplace value for sports. Test of David Stern. The following table shows the increase:

 
                     NBA National Rights Payments
 ----------------------------------------------------------------------
  Season Network/station Maximum No. of games Rights payments
 ----------------------------------------------------------------------
1982-1983 .. USA/ESPN ............................ 100 ..... $5,500,000
1983-1984 .. USA/ESPN ............................ 100 ...... 5,500,000
1984-1985 .. WTBS ................................. 76 ...... 9,000,000
1985-1986 .. WTBS ................................. 75 ..... 11,000,000
1986-1987 .. WTBS ................................. 79 ..... 12,000,000
1987-1988 .. WTBS ................................. 81 ..... 13,000,000
1988-1989 .. WTBS ................................. 81 ..... 23,000,000
1989-1990 .. TNT .................................. 81 ..... 27,000,000
              ---------------- --------------------- Id., p. 2
 ----------------------------------------------------------------------
 

 During 1989, the NBA and TNT agreed to a four-season package of NBA games, 80 games per season, for $57 million in 1990-1991, $66 million in 1991- 1992, $74 million in 1992-1993 and $78 million in 1993-1994. Id. Similarly, in 1989, the NBA and NBC negotiated a four-season package of NBA games to begin in 1990 for $601 million. Id., p. 3.

 Robert Wussler, former president of WTBS, and Roger Werner, former CEO of ESPN, discussed the marketplace value of sports in analogous markets. Both testified that in the cable industry, whose primary goal is to attract and retain subscribers, licensing fees for a type of program may be quite different than the amount of viewing that program generates. Wussler, p. 3; Werner, pp. 2-3.

 Robert Wussler testified that in December 1989, Turner Broadcasting agreed to pay Columbia Pictures Television approximately $22.15 million for the right to televise 1,000 feature films during 1990-1998 on WTBS and TNT. Assuming up to 16 runs per title, that amounted to up to 32,000 hours of *15294 programming. Tr. 1858-59. Whereas, at about the same time, Turner Broadcasting agreed to pay the NBA a total of $275 million during 1990-1994 to televise up to 320 NBA games (about 800 hours). Id., p. 3. On a per hour basis, the TNT movie package cost $690 an hour, while the NBA contract cost $340,000 per hour. Tr. 1812-15, 1859.

 Mr. Wussler also noted that at WTBS in 1989, he could obtain syndicated series such as "The Beverly Hillbillies," and "Petticoat Junction" for under $5,000 per half-hour episode, with the right to broadcast them an unlimited or very high number of times. Tr. 1812.

 Mr. Wussler testified that while sports constituted 8 1/2 % of the time on WTBS, it accounted for about 30-35% of the advertising revenues for WTBS. Major sports programming accounted for a range of 9 to 17 percent of the viewing on WTBS. PS. Ex. 1, p. 638; (13.4% of six-cycle diary-based viewing) Cooper Reb., p. 3.

 Mr. Wussler conceded that Turner made major deals for movies. During 1985-86, Turner paid $1.5 billion for ownership of the MGM-UA library consisting of 3,700 titles. Tr. 1830-31. In 1989, TNT planned to spend $75 million to produce 24 made-for-cable movies. Tr. 1834-35. Mr. Wussler noted that when one owns a movie, it includes all rights--unlimited runs, domestic syndication, international syndication, home video rights--but when one acquires a license for a sports game, it is limited to one run, no ability to syndicate, domestic or international, and no home video. Tr. 1861.

 Roger Werner testified that while baseball telecasts on ESPN represented less than 6% of ESPN's schedule and 15 percent of the share of total viewing generated by ESPN in 1990, the payment made by ESPN for those telecasts amounted to more than 30 percent of ESPN's total program budget for 1990. Werner, p. 2; Tr. 1488. In 1989, ESPN contracted to carry 175 games of Major League Baseball for four years beginning in 1990 for $100 million each year. Tr. 1487.

 Mr. Werner testified that although USA Network (which is primarily movies and series) delivered average audience 20-25% higher than those delivered by ESPN, nonetheless, cable operators paid 32 cents per subscriber per month for ESPN and only 18 cents per subscriber per month for USA. Id., p. 3. Mr. Werner was unaware of any cable operator that uses viewing for budget allocations. Tr. 1493.

 The reason for this disparity, according to Mr. Werner is that cable systems take into account audience demographics, the importance of uniqueness or exclusivity to a program's perceived value, and the blue-chip brand image of established sports. Id.

 Changed circumstances. Dr. Peter Lemieux performed an analysis of the 1989 cable royalty fund based on information compiled by Cable Data Corporation. JSC Ex. 3.

 Of the 12,493 cable systems that submitted statements of account for the second half of 1989, 7,997 were Form 1 systems, 2,464 were Form 2 systems, and 2,032 were Form 3 systems. Id., p. 2.

 Because Form 3 systems list which distant signals they import in the statement of account and Form 1 and 2 systems do not, and because they account for 98% of the fund, Dr. Lemieux studied Form 3 systems. The signals Form 3 systems imported were analyzed as to instances of carriage and fees generated.

 
                             Instances of Carriage
 ------------------------------------------------------------------------------
     Type of signal 1983 2nd half Percent 1989 2nd half Percent
 ------------------------------------------------------------------------------
Original Superstations ............. 2,201 ..... 39.3 .......... 3,367 ... 47.0
Newer Superstations .................. 300 ...... 5.4 ............ 347 .... 4.8
Other U.S. Independents ............ 1,007 ..... 18.0 .......... 1,228 ... 17.1
U.S. Network Affiliates ............ 1,538 ..... 27.4 .......... 1,638 ... 22.8
Noncommercial Stations ............... 427 ...... 7.6 ............ 488 .... 6.8
Canadian Stations .................... 114 ...... 2.0 ............ 102 .... 1.4
Mexican Stations ...................... 19 ...... 0.3 .............. 3 ... 0.04
                                       Id.; 51 FR 12797; PS Ex. 24X
 ------------------------------------------------------------------------------
 

 Dr. Lemieux defined the original superstations as WTBS, Atlanta, WGN, Chicago and WWOR, New Jersey. He defined the newer superstations as WPIX, New York, WSBK, Boston, KTLA, Los Angeles, and KTVT, Dallas-Fort Worth. All seven are superstations by virtue of being satellite-delivered. JSC Ex. 3, pp. 5-6.

 
                          Basic Fees Generated
 ------------------------------------------------------------------------
     Type of signal 1983 percent 2nd half 1989 percent 2nd half
 ------------------------------------------------------------------------
Original superstations ...................... 53.3 ................. 67.2
Newer superstations .......................... 9.0 .................. 8.1
Other U.S. independents ..................... 24.1 ................. 15.7
U.S. network affiliates ...................... 7.6 .................. 4.9
Noncommercial stations ....................... 2.4 .................. 2.0
Canadian stations ............................ 3.2 .................. 2.1
Mexican stations ............................. 0.3 ................. 0.02
                            ---------------------- Id., p. 9
 ------------------------------------------------------------------------
 

 The fees generation analysis only studies fees derived from the basic fund, and not 3.75% stations or the syndicated exclusivity surcharge. Id.

 Placing emphasis on the seven superstations, Dr. Lemieux noted that each superstation telecast the games of a major league baseball team in 1989. WGN, WWOR, KTLA and KTVT were the flagship stations for a team in the National Basketball Association, while WTBS carried the Atlanta Hawks and other NBA games including divisional playoffs. WWOR and WSBK carried National Hockey League games, and WTBS, WGN and WWOR carried college football and basketball games. Id., p. 6.

 Dr. Lemieux noted that collectively the seven superstations represent more than half the instances of carriage and generated three-quarters of all basic royalties paid by cable operators, and attributed their growing popularity to their sports offerings. Id.

 Claim. Based on the above proffered showing, the Joint Sports Claimants claim 30% of the basic fund, and 33% of the 3.75% fund. JSC Prop. Findings, p. 52.

 Criticism of the Methodology of the Bortz survey

  Alleged PBS undercount. PBS witness Dr. Robert Peterson, professor of marketing, asserted that the Bortz survey underestimated PBS' value to cable operators in several respects. First, only 14.4% of the cable systems participating in the survey carried PBS stations, while 21.4% of all Form 3 cable systems carried PBS stations in 1989. To correct for this underrepresentation, Dr. Peterson proposed multiplying PBS' percentage by 1.5%. Peterson, p. 2.

 Second, when a cable system was not carrying a PBS station, the cable operator was not asked to assign PBS any value. To correct for this, Dr. Peterson proposed multiplying PBS' percentage by 1.2. Id., pp. 2-3.

 Third, about 56% of the respondents that carried PBS (15 out of 27) worked for cable systems that were owned by multiple systems operators that had an ownership interest in WTBS. Believing that this created "respondent bias," and noting that WTBS "affiliated" cable systems gave PBS a 7.7% allocation, while systems not "affiliated" with WTBS gave PBS a 12.3% allocation, Dr. Peterson proposed multiplying PBS' percentage by 1.6%. Id., pp. 3-4.

 Fourth, there were 27 responding cable systems that carried a PBS station. Dr. Peterson believes this was too small a sample to be reliable, and that external corrections using results from the Bortz survey of 1986 and 1990 were *15295 appropriate because it would give a sample size (89) that was more reliable. Based on other years' data, Dr. Peterson proposed multiplying PBS' percentage by 1.4. Id., p. 5.

 Since these alleged defects were independent of each other, Dr. Peterson believed that his multipliers should be multiplied sequentially, with a final proposed adjustment for PBS of 5.2%.

 Last, PBS witness Nat Katzman proposed a further adjustment of the proposed 5.2% for PBS. In his view, because cable operators did not pay the 3.75% rate for PBS stations, and because, he believes, the cable operators made their responses to the Bortz surveyors about payments they would have made to the total royalty fund, the 3.75% payments should be backed out of the calculation and PBS' percentages should be adjusted by a factor of 1.3, yielding a proposed result of 6.8%. Katzman, pp. 11-12. However, Mr. Bortz stated there was nothing in his survey questions to suggest that respondents were to consider their 1989 copyright payment as the fixed budget which they were allocating. Tr. 949.

 Alleged Program Suppliers undercount. Program Suppliers' witness Dr. Stanley Besen, Senior Economist of the Rand Corporation, argued that the splitting of the Program Suppliers category into two separate categories in the Bortz survey--movies and syndicated shows and series--adversely affected the total sum allocated by cable operators to Program Suppliers. Besen, pp. 9-10. Dr. Besen stated, from an economic viewpoint, that the maximum amount that a cable operator is willing to pay (that is, the reservation price) for all programs within a category consisting of program types A and B combined will be greater than the sum of the reservation prices for program types A and B when considered separately. Id. This is so whenever program choices are substitutable. Respondents will consider the second category as having diminishing marginal value and will rate it less. Tr. 4689-90. It was Dr. Besen's opinion that the percentage for Program Suppliers would have to be greater than 48.1%, which represents the sum of its two categories, movies (31.2%) and syndicated shows and series (16.9%).

 Criticism of the Validity of the Bortz Survey.

  Program Suppliers witness Dr. Alan Rubin, director of the Communications Research Center at Kent State University, asserted that some of the interviewed cable system employees were not the appropriate respondent. 56.6% were system or general managers, 30.3% were marketing directors or managers, 2% were programming directors or managers. Tr. 5327. The remaining 11.1% were categorized as office managers, technicians, plant supervisors, custom service directors and others. Tr. 5274-75. It was these 11.1% that Dr. Rubin believed had insufficient familiarity with either programming or budgeting to give informed responses. Id. Further, although respondents were screened for their familiarity with programming, there was no screening question concerning familiarity with budgeting, which was the key question in the Bortz survey. JSC Ex. 1, p. 7.

 Dr. Rubin questioned the number of call backs that the Bortz surveyors made to attain their 79% response rate. An excessive number of call backs raises the question whether the respondents answered voluntarily or only answered to get the caller off the phone. Dr. Rubin noted that about 1/3 of the respondents were called eight or more times while he believed that standard research procedures specifies a limit to the number of call backs (he suggested three). Rubin, pp. 2-3; Tr. 5361. Dr. Rubin conceded, however, that he did not know whether the numerous call backs he complained of were due to busy signals, no answers, and/or answering machine recordings which would not have affected the state of mind of the respondent. Tr. 5361-64.

 Dr. Rubin argued that the program categories established by Bortz did not directly comport with the program categories as defined by the Tribunal. Rubin, pp. 3-4. Dr. Rubin believed the category labels should have been augmented with descriptions of familiar programs in each category. Tr. 5323. Program Suppliers argued that the lack of more detailed explanations cost them a number of programs that they believe the typical respondent assigned to other categories, such as "CNN Headline News," (syndicated, but probably thought of as "news and public affairs"), wrestling and auto racing (often recorded syndicated series, but probably thought of as "sports"). Tr. 974, 2496; PS Prop. Findings, pp. 68-69. However, NAB witness Richard Ducey believed that respondents had a "dominant impression" of what each category contained, and any misimpressions were likely to be a "wash." Tr. 2522-24.

 Dr. Rubin believed that the brief telephone interviews (5-10 minutes to answer 11 questions), instead of face-to-face interviews, provided no assurance of predictive validity. Rubin, p. 13; Tr. 936. Dr. Rubin believed that the constant sum technique should allow the respondent an opportunity to visualize his or her responses, to reconsider, and to reconcile or reorder priorities upon reflection. Rubin, p. 5. Dr. Rubin considered face-to-face interviewing important when the task being asked of the respondent is a complex one, and he considered that a budget allocation among five to seven categories is complex. Tr. 5422-23.

 Dr. Rubin questioned whether a constant sum survey is capable of being used for statistical analysis for three reasons: First, the cable respondents gave ipsative, or dependent responses, where the value assigned to one program category necessarily took away from the value that could be assigned to a different category. Dr. Rubin doubted whether confidence intervals, which depend on assumptions such as normal distribution, could be applied to such statistics. Rubin, pp. 8-9; Tr. 5300.

 Second, a constant sum question may overrate or underrate likes or dislikes, or may induce a respondent to create differences where none exist. Rubin, p. 9. Third, if some respondents felt unwilling to accord any category a zero value, then without a true zero, it cannot be said that something is four times as great as something else. All that can be said is that one thing is higher than another. Id., Tr. 5303-06.

 Last, Dr. Rubin believed that the Bortz survey required the respondent to recall all the distant-signal programming of 1989, and to recall what would have been budgeted a year earlier. He noted that long-term recall is faulty. Rubin, pp. 11-13. Program Suppliers urged the Tribunal to find that asking cable operators what they would have budgeted for programs they acquired over the past year is an unreal exercise; budgets are normally projected for future expenses. Tr. 1049; PS Prop. Findings, p. 83.

 Criticism of the Relevance of the Bortz Survey

  Dr. Besen questioned the relevance of the key question in the Bortz survey. He observed that the question asking operators to apportion a fixed budget contained no information about the prices at which these program types would be available. He believed that the operators could only make meaningful judgments if they were apprised of the prevailing prices. Besen, p. 7.

 Dr. Besen also believed that the concept of a fixed budget is incorrect, because businesses will add to their budget if they believe by doing so they will add to their revenues. Id., p. 8.

 Mr. Bortz conceded the validity of the observation about price, by observing *15296 that when a cable operator who has a certain budget allocation in mind goes into the marketplace, he or she may encounter surprises--some programs being cheaper, others being more expensive--so that the marketplace result will be different than the buyer's initial allocation. However, Mr. Bortz believed that his survey is as close to the marketplace value of the programs as he could get. Tr. 959-962.

 Dr. Besen disagreed that the Bortz survey bore any relationship to marketplace value. The Bortz survey measures the total value to cable operators of all programs in a given category (the marketplace value plus the consumer surplus). A survey demonstrating the relative total value of seven different program types would not demonstrate the relative marketplace value of those program types except where the demand curves for all program types are linear and the demand elasticities are the same for all program types at the equilibrium prices. Besen, pp. 4-5. Dr. Besen contends that these conditions are stringent and that there exists no evidence to support that these conditions exist. Id., pp. 18-19.

 Dr. Besen conceded that relative total value would equal relative marketplace value if cable operators were offered an entire program category as a package on an "all or nothing" basis. However, Dr. Besen asserted that such an arrangement would be contrary to the antitrust laws, and therefore, it was not something that could be practically considered. Id., pp. 19-21.

 When asked what would have been the relevant question for Bortz to have asked, Dr. Besen replied that marketplace value is the marginal value of one additional program multiplied by the quantity, so he would ask the operator, "What would it be worth to you to have one additional program in this category?" Then he would multiply the answer by the number of programs in the category. Tr. 4701-02.

 Criticism of Testimony on Marketplace Value

  Program Suppliers witness Dr. John Woodbury, consultant on regulatory economics, questioned Sports' assertions that the rise in the price of NBA telecast rights represented an increase in the marketplace value of sports, in general. Reb. Test. of John Woodbury.

 Dr. Woodbury questioned whether the increase of 500% in the price of the NBA's national cable contract on a per game basis between 1983 and 1989 was representative of all other sports. He asserted that in the same time period, contracts for the NFL rose by 71%, baseball by 125%, college football by 60%, that contracts for college basketball were either flat or declining and that the NHL was unable to contract for national rights. Id., pp. 4-7. Combining all sports together, and weighting them for their representation in the Nielsen viewing data, Dr. Woodbury found an increase in the price per game of sports in the range of 141-171%. Woodbury, Attach. 1, p. 29. This compares to an increase of 162% in the price broadcast stations paid for non-sports programming from 1983 to 1989. Id., Attach. 1, p. 33.

 Dr. Woodbury also believed that the increase in value of sports to a cable network, like TNT or ESPN, does not necessarily mean an increase in the value to cable operators because cable networks have two streams of revenues: charges on a per subscriber basis to cable operators, and advertising. Advertising on cable rose 355% from 1983 to 1989, and therefore, Dr. Woodbury believes that it's likely that a substantial portion of the rise in the value of sports was due to the second stream of revenue--advertising--and not to the value ascribed by cable operators. Id., pp. 14-15. Mr. Werner confirmed that when ESPN negotiated its contract with Baseball, it intended to recapture its cost through advertising sales, and not to impose higher charges on the cable operator. Tr. 1508.

 Dr. Woodbury also noted that national sports contracts to cable networks include a premium for getting the better games, leaving the games with less national interest for local broadcast. Id., p. 12. Mr. Werner confirmed that ESPN's contract with Baseball allowed it to choose, within scheduling limits, the better games, and that this was more attractive than a package of solely local games. Tr. 1545, 1547-48. Mr. Stern confirmed that the NBA's national contract with WTBS/TNT allowed WTBS/TNT to pick the more attractive games. Tr. 1763.

 Dr. Woodbury conceded he obtained his industry data from media reports and made no attempt to verify the data. Tr. 4922-27. Dr. Woodbury conceded that he did not include in his analysis contracts that were negotiated by the NBA, the NFL and Major League Baseball in 1989 for future years, nor did he take into account a Baseball national cable contract and a NHL contract that were in existence in 1983 or a NHL contract that existed in 1989 for regional cable coverage. Tr. 4971-5026.

 Commercial Television Claimants (NAB)

  Bortz survey. Richard Ducey, Senior Vice President of the NAB Research and Informational Group, asserted that the best evidence for quantifying the value of station-produced programming is the Bortz survey which assigned a value of 11.8% to NAB. Ducey, p. 28.

 Dr. John Robinson, professor of Sociology at the University of Maryland, supported the Bortz survey as one that was carefully conducted and followed accepted principles of survey research design. Dr. Robinson particularly noted the high response rate and the carefully drawn probability sampling frame. Robinson, pp. 2-3. Dr. Robinson would not characterize the Bortz survey as an "attitudinal" survey (that is, one measuring likes vs. dislikes), because the cable operators were asked about the value of the programs, not how much they liked them. Id., p.4.

 Benefit. NAB presented two cable operators, Robert Davies and Philip Viener, to testify on the benefit of station-produced programming to cable systems. Test. of Robert Davies; Test. of Philip Viener.

 Robert Davies, who worked at the Naples, Florida cable system during 1986 to 1988, reported that because of subscriber complaint, the cable system, which had dropped WGN, had to reinstate it. Complaints were generally about the loss of Chicago news and Chicago baseball games. Davies, pp. 1-2. Mr. Davies estimated the number of complaints and the number of disconnects at about 50, but it could have been up to 200. Tr. 2760-61, 2780. Mr. Davies stated that as a cable operator, he didn't go by viewing data. He went by what people wanted to watch, and if people were willing to buy a package with those particular programs, then so be it. Tr. 2784.

 Mr. Davies also worked as a consultant and regional manager for Mid North 1 Cablesystems in Minnesota. Davies, p. 1. He has observed that cable systems in North Central Minnesota import KMSP from Minneapolis-St. Paul because of the importance of state news from the capitol and because 10% of the cable subscribers in that area are summer residents from Minneapolis-St. Paul. Davies, p. 2.

 Philip Viener served as system manager and then regional marketing director for Continental Cablevision of Virginia from 1980 and 1984, a company that operated seven cable systems in East Central Virginia. Viener, pp. 1-2. During his employment, those seven systems chose to import WTTG-TV from Washington, D.C. in part because of the 10 p.m. local news from Washington and for the "Redskins Playbook," a station-produced program. Id., p. 4.

 Mr. Viener also stated that viewing data would have little relevance on an *15297 operator's evaluation of which distant signals to carry and promote. Viener, p. 5. Mr. Viener supported the 11.8% value for news and public affairs programs in the Bortz survey as an accurate reflection of how operators and subscribers in central Virginia would feel. Tr. 2815-16.

 Marketplace value. Richard Ducey assertd that station-produced programming on superstations and on other distant signals has general audience interest regardless of where the audience lives, and also has special regional appeal. Ducey, p.4. Station-produced programming also is unique. Because it is only available on that station, it is programming that cannot be obtained anywhere else. Id., Tr. 2371.

 Mr. Ducey also asserted that station-produced programs are diverse. Aside from regularly scheduled news, they include: agricultural shows, children's programs, comedy, consumer affairs, crime, cultural programs, documentaries, education, health, holiday specials, interview shows, magazines, ethnic programs, music, public affairs, religious programs, senior citizen shows, sports and pre/post game shows, talk shows, youth shows, and travel programs. NAB Ex. 20.

 Changed circumstances. NAB has been given credit by the Tribunal in past proceedings for the regional appeal of its programs. NAB submitted data purporting to show that industry and demographic changes from 1983 to 1989 have placed more of the cable viewing public within the regional area of the local broadcaster, thereby making the appeal of station-produced programs more relevant to more cable audience. Ducey, pp. 18.19; NAB Exs. 13-23.

 In 1983, 81.9% of the instances of distant cable carriage of non-superstations were within 150 miles of the broadcast station. In 1989, that figure rose to 86.5%. NAB Ex. 14. In 1983, 88.8% of the instances of distant cable carriage of non-superstations were either within 150 miles of the broadcast station or within the same state. In 1989, that figure rose to 93.2%. NAB Ex. 16.

 NAB also purported to show that there was a greater percentage of population in those close-in distant carriages, thereby increasing the appeal of station-produced programming. Ducey, p. 19; Test. of John Elkins.

 The combined population of the six largest cities in the North and East declined by 2.9% from 1980 to 1990, while the population bordering the metropolitan areas of these cities rose by 6.3%. NAB Ex. 21. Similarly, the combined population of the six largest cities in the South and West rose by 11%, but the population in counties bordering their metropolitan areas rose by 26%. NAB Ex. 22. Nationwide, in 1980, 40.1% of the population lived in central cities and 59.9% lived outside central cities. By 1990, 35.4% lived in central cities, and 64.6% lived outside. NAB Ex. 18.

 Claim. Based on its above proffered showing, NAB claims 9% of the basic fund and 9.5% of the 3.75% fund. NAB Prop. Findings, para. 93.

 Noncommercial Television Claimants (PBS)

  Bortz survey. PBS supported the Bortz survey as more relevant than the Nielsen study, and PBS argued that the Bortz survey should be adjusted to yield a result of 6.8% for public television. PBS Prop. Findings, paras. 45-53.

 Cume viewing data. John Fuller, Director of Research for PBS, argued that the Nielsen study is not the relevant measurement for PBS. PBS does not design its programs to attract the largest audience for each show. It targets its programs for particular tastes and age groups (especially children ages 2-5, 5- 8, etc.) which assure low ratings. Fuller, pp. 14-15.

 Mr. Fuller offered cume viewing data as more relevant for PBS. During February-March 1989, Nielsen found that 83% of U.S. cable households watched public television an average of 7 1/2 hours per households. Fuller, p.28. Mr. Fuller conceded that the MPAA-Nielsen study does show the sum total of all PBS distant signal viewing. Tr. 3142. Mr. Fuller had no knowledge of what the cume data would be for viewing distant signal PBS stations. Tr. 3238.

 Instances of carriage. While the overall instances of distant carriage of PBS stations in 1989 were 6.7%, when the 3.75% signals are subtracted, the instances of carriage of basic signals were 7.4%. Larson, pp. 2-3. In 1983, the overall instances of PBS carriage were 7.6%, but excluding the 3.75% signals, they were 8.06%. 51 FR 12811; Tr. 2058. In both 1983 and 1989, half of all Form 3 cable systems that imported a PBS signal carried no local PBS station. Larson, p. 3.

 Harm. Five of the 150 largest television markets in the U.S. have no local PBS station--Santa Barbara, Cal; Tyler, Tex.; Bakerfield, Cal.; Beaumont, Tex.; and Rockford, Ill.--yet each has a cable system that imports a distant PBS signal. Lawson, pp. 37-38. PBS Executive Vice President Jennifer Lawson considers it highly likely that distant signal importation has inhibited the development of local stations in these communities. Id., p. 39. PBS is helped by additional stations because they bring new revenues from local fundraising, and local corporate and foundation support. Tr. 3895. The loss of new stations is not offset by viewer contribution from distant markets because may viewers believe part of their cable fees go to support these stations. Tr. 3881. Ms. Lawson had no knowledge of marketing efforts to encourage distant viewer's support. Tr. 3896-97.

 Marketplace value and benefit. John Fuller offered several reasons why PBS programs have special marketplace value and benefit for cable operators. PBS programs are relatively commercial-free. Typical hour programs on PBS have about 57 minutes of program content, while commercial hour programs range from 46 to 48 minutes of program content. Id., p.4. Mr. Fuller conceded that PBS runs pledge drives, 16 days in March, 9 days in August and 9 days in December, but spread out over the entire year, it comes to an additional 40 seconds per hour of non-program content. Tr. 3067-68.

 There is little direct duplication of programs when a cable system carries more than one PBS station. A survey of 30 cable systems yielded a result of an average of 15% ( -7.7%) of head-to-head duplication. Tr. 3087- 91. However, if the sme hour program was started a half-hour later, that was not calculated as duplication. Tr. 3171-72.

 As one measure of viewer intensity, Mr. Fuller noted that viewers' home taping of PBS programs accounted for 5.8% of all taping. This is higher than its nationwide Nielsen rating share of 3.4%. Fuller, p.26. Mr. Fuller had no knowledge of what the statistic would be for taping of distant signal television. Tr. 3236.

 Benefit. Sharon Rockefeller, President and CEO of WETA-TV, Washington, D.C., testified to the benefit of PBS' children's programming which parents prefer their children to watch over commercial children's programs. Rockefeller, pp. 3-5. Ms. Rockefeller also noted the benefit of the news and public affairs programming of PBS which is more in-depth than commercial news, and is without commercial interruption. Id., pp. 5-8.

 Ron Hull, Associate General Manager of the Nebraska Educational Television Network, testified that public television stations are of particular benefit to the smaller towns and rural areas that do not have the cultural institutions of big cities. Hull, p. 4.

 Marsha Leopard, a schoolteacher and mother in Chillicothe, Mo., watches KCPT-TV, a distant PBS signal from Kansas City, Mo. She believes the children's programming on PBS is better *15298 than any children's programming available on commercial channels. She also uses PBS programs in her classroom. Leopard, pp. 1-5; Tr. 3753.

 Quality. Dr. John Carey, communications researcher, asserted that one measure of quality can be drawn from the opinions of the nation's critics and awards organizations. Carey, pp. 1-2. The New York Times, The Wall Street Journal, and the Christian Science Monitor have recommended public television programs to their viewers more often than programs from any other source. In 1989, The Wall Street Journal gave PBS shows 157 recommendations, compared to 38 for NBC, 25 for HBO and 4 for all nonnetwork commercial programs combined. PTV Ex. JC-3, pp. 8-9.

 The New York Times gave 58% of its "excellent" reviews to PBS shows. The next runner up, CBS, received 9%. Id., p. 13. Public television won 64 national Emmys in 1988-1990, while all of nonnetwork commercial television won 44 Emmys. Id., Table 11.

 In a 1989-90 Consumers Union study, respondents were asked to evaluate program quality on 44 different cable services. PBS tied for second with a 76 rating. The other distant signals ranked as follows: WTBS-62 (tied for 13th), WGN-61 (tied for 16th), WPIX and WWOR-56 (tied for 29th), and Fox Network-55 (35th). PTV Ex. 4X; Katzman, p. 4.

 Ms. Rockefeller noted the quality of PBS's children and public affairs programs. Rockefeller, pp. 2-8. Ms. Rockefeller defined quality as "excellence", (emotional) impact on the viewer, high production values, if possible, adequate (and accurate) research." Tr. 2885-86. She also noted that through focus groups and research, "quality" is the word that comes up when public television is discussed. Tr. 2887.

 Mr. Hull said that quality is what the collective minds of previous generations tell us is quality. Tr. 2893. He also noted the quality is a combination of every element of a production, giving special mention to the writing and the attention to detail and accuracy. Tr. 2986-87.

 Arthur Unger, television critic for The Christian Science Monitor for 15 years, defined quality as consisting of three components--does it entertain? does it stimulate? does it educate? Tr. 3370-71.

 Former Senator Charles MCc. Mathias characterized quality as "merit, in a cultural or an artistic or an educational sense." Tr. 3601.

 Peggy Charren, President of Action for Children's Television, would use five questions to assess quality: 1. Does the program have something to say, or sell? 2. Would you submit it for an award? 3. Would you list it in defense of an FCC license challenge? 4. Would parents tape it for their children? 5. Would you mention it by name if your child's teacher asked you what you do for a living? Charren, pp. 3-5.

 Syndex fund. PBS claims a portion of the syndex fund. PBS Prop. Findings, paras. 103, 110-112. During 1989, Form 3 cable systems paid a total of $405,388 in syndicated exclusivity surcharges for the improtation of PBS signals. Larson, p. 6. The Copyright Office has interpreted the surcharge as applying to imported PBS stations and the statement of account so indicates. PTV Exs. TL-1, TL-2. During 1989, PBS stations broadcast syndicated programs such as "Zoobilee Zoo," "Captain Kangaroo," "World of Survival," and "Nightly Business Report." Lawson, p. 8, 18, 28; Tr. 3289.

 Claim. Based on its above proffered showing, PBS claims 8.5% of the basic fund, and a share of the syndex fund in the same proportion vis-a-vis Program Suppliers and the Music Claimants as the Tribunal ultimately decides is its proportional share of the basic fund. PBS Prop. Findings, para. 112.

 The Devotional Claimants

  The Bortz survey. The Devotional Claimants supported the Bortz survey which assigned a value of 4.3% to the Devotional Claimants, and the most relevant evidence for the Tribunal to consider in terms of benefit and marketplace value. DC Ex. 2, pp. 1-2; JSC Ex. 1.

 Glyn Wooldridge, Chairman, Mathematics Department of Liberty University, testified that the Bortz survey had internal validity in that the survey was well-designed, well-executed, and the respondents understood the questions. Mr. Wooldridge also believed the survey had external validity because rigorous steps to create a random sample were used and a high response rate was achieved. DC Ex. 3, pp. 3-4.

 Time plus fee generation study. Tom Larson, President of Cable Data Corporation, performed an analysis of the 127 commercial television stations in the six-cycle Nielsen study to ascertain what royalties might be considered, on a time basis, to have derived from devotional programming. On those 127 stations, devotional programming occupied 5.073% of the quarter-hours of all programming. If one assumes all hours to be of equal value, the 5.073% of devotional programming generated 1.857% of all the basic and 3.75% fees paid by cable systems to carry those 127 stations. DC Ex. 6, p. 1; DC Ex. 7, p. 1; Tr. 2192. The discrepancy between 5.073% and 1.857% was explained that the stations having more hours of devotional programming were carried by fewer distant cable systems. Tr. 2190.

 In 1983, the Devotional Claimants calculated a share of 4.63% for the first half of 1983, and 4.36% for the second half of 1983. 51 FR 12802. Devotional witness Dr. David Clark attributed the decline from 1983 to 1989 to programming changes on WWOR, a superstation. Devotional programming on WWOR in 1983 amounted to 10% of its schedule. In 1989, it comprised less than a quarter of one percent. Tr. 21283-2284.

 The Devotional share of the fees generated would have been smaller than the 1.857% claimed if the 20 PBS stations in the Nielsen study had been included. Tr. 5212-13.

 Specialty stations. Specialty stations have at least one-third religious, one-third foreign language, or one-third automated programming. The total fees generated by specialty stations in 1989 were $1,297,548. DC Ex. 9, p. 1. Religious specialty stations generated $559,856 in fees in 1989. Id., p. 9.

 Harm. Pat Robertson, Chairman and CEO of the Christian Broadcasting Network, Inc., testified that devotional programming is harmed by distant signal importation because it siphons off audience from local signal viewing, and the local broadcaster might consider that the local devotional program had diminished value as a lead-in to the commercial program to follow. The local broadcaster would then feel he or she would have to charge the devotional program more to be carried, or deny the devotional program a better time slot. Mr. Robertson said the harm was not major, but it was still there. Tr. 2092-93.

 Benefit. The Tribunal observed in the 1983 decision that a programming category with small viewing cannot be given credit for being a benefit to cable systems unless it is shown the avidity of the small number of viewers for such programming. 51 FR 12808.

 Mr. Robertson testified that the 4.3% Bortz survey result for devotional programming was the best objective evidence of avidity. In addition, he has personally experienced the intensity of this viewership from the calls and letters, especially the thousands of phone calls from persons in distress. DC Ex. 1, pp. 3-4.

 Myrtle Huggings, a cable subscriber in Arroyo Grande, California, watches The 700 Club on KTLA from Los Angeles, a distant signal. The 700 Club accounts for half of all her television viewing. She is an avid viewer and religious *15299 programming is the reason she subscribes to cable. DC Ex. 11./

 Marketplace value. Mr. Robertson testified that the reason "The 700 Club" pays to be on some 200 broadcast stations is because he and others do not want commercials to be run in the middle of the religious programs. He must therefore rely on listener sponsorship, rather than commercials. Tr. 2092.

 In the cable market, however, TCI, the country's largest MSO, made an investment in The Family Channel for 16% ownership. Tr. 2087. CBN has a contract over a 10-year period with TCI to carry The Family Channel at increasing subscriber rates each year. Tr. 2096. Currently, The Family Channel charges cable operators five cents per subscriber per month. Tr. 2276. The Family Channel carries "The 700 Club," the Rev. Charles Stanley, the Rev. D. James Kennedy, Dr. Ben Hayden, the Rev. Jerry Falwell and "Superbook." Tr. 2108, 2343-44. Mr. Robertson conceded that The Family Channel carries many movies and syndicated series besides religious programs. Tr. 2107-08.

 Claim. Based on its above proffered showing, the Devotional Claimants claim 4% of the basic fund, and 4% of the 3.75% fund. DC Prop. Findings, para. 55.

 Conclusions of Law

 The Basic Fund

  In the parties' efforts to offer the Tribunal a basis upon which to simulate the 1989 cable distant-signal marketplace, two main pieces of evidence were presented: The Nielsen study of distant signal viewing, and the Bortz survey of cable operator budget choices.

 As in the 1983 proceeding, the Tribunal found both surveys essentially valid and relevant, but neither survey is perfect and neither survey provides the entire answer to the question of what each party should rightfully receive.

 Before proceeding to discuss the relative weight to be accorded the Nielsen study and the Bortz survey, it is first necessary to make methodological adjustments to the results of both surveys to assure fairness to all Phase I parties.

 The Nielsen study--methodological fairness. Two methodological errors were acknowledged by the Program Suppliers. The Nielsen study miscategorized some programs which belonged to NAB, and it underrepresented network and education stations to the detriment of NAB and PBS.

 Accepting these criticisms, Program Suppliers performed a stratified adjustment of the six-cycle data, but not the four-cycle data, yielding this result: Program Suppliers--81.56%; Sports--9.05%, NAB--6.38%; PBS--2.72%, Devotionals--0.26% and Other--0.03%.

 The Tribunal believes this was the proper adjustment. However, we believe that the six-cycle data were too flawed to be used. The Tribunal had criticized Program Suppliers in past proceedings for using only four-cycle data. We were concerned that those were national "sweep" periods that, because of the special programming that aired during that time, might distort the results. This was only a concern. It was not proved, and we looked to additional data to confirm or dispel the concern. Yet, the two additional "partial sweep" periods--January and October--lacked validity.

 First, adding the two extra periods resulted in an underrepresentation of the baseball season. Further, because these were partial sweeps, only measuring 18 to 23 markets, rough estimating techniques were required to approximate the other 180 markets. This involved looking at comparable daypart ratings from one of the earlier four-cycle periods, without regard to whether the earlier program was the same type as the program being measured. While these estimating techniques may be the best that can be done under the circumstances, it nonetheless rendered the two additional cycles unusable for Tribunal purposes.

 As a result, the Tribunal considered only four-cycle data in this proceeding. Since Program Suppliers did not perform a stratified adjustment for four-cycle data, the Tribunal assumed the same ratio of changes that resulted in the six-cycle data would occur in the four-cycle data, and performed its own adjustment. For example, NAB went from 5.783 to 6.38 in the six-cycle data after adjusting for underrepresentation. Assuming the same ratio of improvement in the four-cycle data gave NAB a result of 6.703% (6.38 divided by 5.783 equals 6.703 divided by 6.076). This gave the following results: Program Suppliers--78.325%; Sports--11.664%, NAB--6.703%, PBS--2.896%, Devotionals-- 0.256%, Other--0.038%.

 A further adjustment needed to be made for PBS. PBS argued that in using viewing data to determine the basic fund, the viewing of 3.75% stations should be excluded. The Tribunal agrees. PBS performed a calculation for six-cycle data, finding that 395,517,463 viewing hours of 3.75% stations should have been excluded, increasing its pre-adjusted share from 1.936% to 2.162%. Assuming the same ratio for its four-cycle percentage, PBS' four-cycle pre-adjusted share rose from 2.061% to 2.302%.

 Assuming PBS' four-cycle pre-adjusted share was not 2.302%, then the four-cycle stratified adjustment for underrepresentation became as follows: Program Suppliers--78.060%; Sports--11.625%; NAB--6.680%; PBS--3.235%; Devotionals-- 0.255%; Other--0.038%.

 The last adjustment to be made was to reduce all percentages to account for the shares that Music and the Canadian Claimants, two groups unmeasured by Nielsen, received in the basic fund. What follows is the Nielsen percentages the Tribunal relied to on make its 1983 and 1989 decisions:

 
 -------------------------------------------------------
                         1983 (percent) 1989 (percent)
 -------------------------------------------------------
Program suppliers ............... 72.542 ........ 74.166
Sports ........................... 9.511 ........ 11.045
NAB .............................. 6.879 ......... 6.347
Music ............................ 4.500 ......... 4.500
PBS .............................. 4.380 ......... 3.074
Canadian claimants ............... 0.750 ......... 0.750
Devotional claimants ............. 0.618 ......... 0.243
Specialty stations ............... 0.494 -------------
Minor sports ..................... 0.162 -------------
Other ................. --------------- ......... 0.036
 -------------------------------------------------------
 

 The Tribunal notes that any time an adjustment is made, there is necessarily an adding of one estimate onto another, removing us further and further from certainty. The Tribunal relied on the above percentages only by a preponderance of the evidence, and did not accord them any of the seemingly "solid" reliance that statistics going to the third decimal place tend psychologically to instill.

 We realize the enormous undertaking of conducting this special Nielsen study, but it is hoped that in future proceedings, fewer adjustments will have to be made.

 The Bortz survey--methodological fairness. PBS offered five adjustments to the Bortz survey. The Tribunal accepts the first two. The underrepresentation of PBS in the survey required an upward adjustment to 2.0%.

 The lack of opportunity for those cable systems that did not carry a distant PBS signal was another flaw. As we said in the 1983 decision, the fact that a cable system did not carry a PBS signal only meant that the actual price was too high. There could have been some lesser price they were willing to spend. Therefore, Bortz practice to accord PBS an automatic zero underrepresented PBS. However, the Tribunal appreciates that the survey was designed to refer the respondent to actual carriage during 1989, and asking them to assess how much they might *15300 have spent on PBS would have caused confusion. Therefore, the design of the survey is not faulted, but an adjustment nonetheless needed to be made. The Tribunal accepts Dr. Peterson's multiplier of 1.2, which increased PBS' percentage to 2.4%.

 However, the Tribunal rejects the other proposed adjustments. The adjustment based on the alleged bias of employees who work for cable systems in which WTBS has some minority ownership interest is wholly speculative.

 The adjustment that was requested because the sample size of 27 was too small has some merit, but the proposed solution to use responses from other years is not acceptable. The Tribunal is considering only those programs broadcast during 1989, and the 1989 Bortz survey was careful to remind its respondents of that. Mixing in other years would dilute the answer to the question this proceeding is addressing. However, because the sample size was small, the Tribunal has considered that 2.4% might well be the mean of a much more wide-ranging confidence interval than if the sample size had been greater.

 The adjustment that was requested to back out the 3.75% portion of the total fund in calculating PBS' share is also rejected. Mr. Bortz stated there was nothing in his survey to suggest that respondents were considering their 1989 copyright payment as the fixed budget they were allocating. Therefore, the very premise of this proposed adjustment was unproved.

 Regarding Program Suppliers, the Tribunal accepts Dr. Besen's observation that because the Bortz survey asked about total value, splitting Program Suppliers into two categories--movies and syndicated series--had, as a matter of economics, an adverse effect on their percentage allocation. Summing up, the Tribunal is adjusting PBS' share to 2.4%, bearing in mind a higher confidence interval for PBS, and it is also bearing in mind that the Program suppliers' percentage should be somewhat higher than the sum of the movie category and the syndicated series category combined.

 The following is an adjusted 1989 Bortz survey, reduced proportionally for Music which was not measured:

 
 ------------------------------
                        Percent
 ------------------------------
Program suppliers ....... 45.53
Sports .................. 32.37
NAB ..................... 11.17
Music .................... 4.50
Devotional claimants ..... 4.07
PBS ...................... 2.30
Canadian claimants ....... 0.19
 ------------------------------
 

 Comparing this to the 1983 Bortz survey is difficult. The Devotional and Canadian Claimants were not included. However, if Music, the Devotional Claimants and the Canadian Claimants are assigned the same values they received in 1989, then the 1983 Bortz percentages would look like this:

 
 ------------------------------
                        Percent
 ------------------------------
Program suppliers ....... 44.23
Sports .................. 32.72
NAB ..................... 10.97
Music .................... 4.50
Devotional claimants ..... 4.07
PBS ...................... 3.40
Canadian Claimants ........ 019
 ------------------------------
 

 Having now considered the methodological fairness of both surveys, we now turn to the question of the validity of each survey.

 The Nielsen study--validity. We first note that Program Suppliers have taken appropriate steps to strengthen the validity and reliability of the Nielsen study. 29 additional stations were studied in 1989, and Program Suppliers reassigned programs from Minor Sports, Specialty Stations and Other to their proper Phase I categories.

 However, in choosing the stations to be studied, it appears that some were excluded even though they met the objective threshold established by MPAA itself. While MPAA's witness was able to explain some exclusions, others could not be explained. Also, PBS stations were held to a more difficult cut-off standard than commercial stations. For reliability, the Tribunal needs to be assured that no station has been excluded without clear justification.

 More criticism than ever was lodged by the other parties as to Nielsen's own procedures in developing the underlying data for the MPAA-Nielsen study. Dr. Reid and Ms. Stein testified at length about the multitude of errors that can occur using the diary method. These difficulties seem to have grown from 1983 to 1989 because of the dynamic changes in the cable industry. An explosion of channel capacity and the use of remote control has made it more difficult than ever for the diary-keeper to be accurate. Ms. Stein stated that these difficulties affected "nonnormative behavior" the most--the lesser-watched programs. As shown below, this observation was taken into account when the Tribunal considered the proper allocation for each category.

 Another changed circumstance affecting the Nielsen study was the changeover from diary-keeping to metering. By 1989, 30% of the national audience was being measured by meter. However, Mr. Lindstrom stated that it was invalid to mix metered viewing with diary viewing. We accept Mr. Lindstrom's statement. However, we would like to explore further whether metered viewing can be used in a valid way.

 Summing up, we have accepted the criticism of the diary method in terms of resolving doubts in favor of the adversely affected claimants. We do not accept that metered viewing should have been used in the 1989 proceeding, but would like to see what its impact would be in future proceedings. However, none of the criticisms invalidate the Nielsen study. We find that the Nielsen study is a reliable and important piece of evidence for the Tribunal's allocation decision.

 The Bortz survey--validity. As was the case with the Nielsen study, the Bortz survey has taken important steps to improve the validity and reliability of its results.

 The high standards of procedure that obtained in the 1983 survey were again followed in the 1989 survey. In addition, the Bortz survey made some key improvements. The Devotional and Canadian Claimants were added as categories. The survey was conducted 60% during 1989, the year in question, and 40% shortly afterwards. This is in contrast to the 1983 survey which was conducted in 1985. This especially improved the survey, because of the unreliability of long-term recall. Furthermore, the key question was modified from asking about the total value of distant signal programming to allocating a fixed budget. We agree that this made the survey more closely resemble tasks that the cable operator performs in his or her business.

 Certain questions concerning the reliability of the results remain. First, no category definitions or descriptions were given, despite our concerns expressed in the 1983 proceeding. The result, as Mr. Ducey described, was that the respondents gave their answers based on their "dominant impression" of each category. We have no way of knowing if these dominant impressions helped or hurt any particular category. It could have been a "wash" But it could have also altered the results.

 Second, although the recall problem was greatly improved, there was still the difficulty for each respondent to recall even in 1989 all the programming that was on each distant signal that year. *15301 Unlike the Nielsen diary-keeper who is keeping track each day, or at most weekly, the cable operator could have had at best a dominant impression of what was broadcast that year.

 Third, we express concern about the qualifications of the respondents. 11.1% of the respondents appeared to have been unqualified. Of the remaining 89.9%, no qualifying question was asked about budget preparation or knowledge.

 Fourth, we give credit to Dr. Rubin's criticism that the brief telephone interviews (5-10 minutes to answer 11 questions) may have affected the survey. Given that the key question asks about allocating a cable system's budget, the answers were given in a remarkably short time, whereas in the "real world," such decisions are considerd and revised by many cable personnel, taking into account a multitude of factors.

 These four criticisms concerned us the most and have affected our allocation decisions as discussed below. However, as with the Nielsen study, we consider the Bortz survey to be valid, and a key part of our determination.

 Before turning to the ultimate issue, the relevance of each survey, we need to address the question of whether Program Suppliers complied with the Tribunal's orders in this proceeding and to what degree that should affect our decision.

 Nielsen study--compliance with Tribunal's orders. MPAA commissions the Nielsen Company each year to perform a special viewing study based on the diaries that Nielsen has produced in the ordinary course of its business. MPAA acted promptly to inform Nielsen of the Tribunal's orders on two issues: the confidence intervals and access to the diaries. Nielsen was late in producing the confidence intervals and for confidentiality reasons, Nielsen restricted access to the diaries, rendering it impractical and burdensome for the other parties to inspect and copy the diaries.

 Concerning the confidence intervals, Section 301.49(f)(1) requires that each sample survey shall include with it "the validity and confidence limits on major estimates." MPAA has not provided this information with its direct case in past years, but neither has the Tribunal nor any claimant requested it. However, we agree with the parties that this is required information, and in future proceedings, we direct MPAA to submit the confidence intervals for each Phase I category for each cycle study with the direct case. Whether the confidence intervals are necessary for each measured broadcast station is an open question and we invite pleadings on this issue at the time the Tribunal ascertains the next controversy.

 In this proceeding, however, we credit the Nielsen's efforts to comply with the Tribunal's order were in good faith. We have also accepted Mr. Rubens' criticism that the confidence intervals do not tell us that Nielson's figures can be projected to the universe. This can be seen from the numerous adjustments we have already made to the Nielsen data. Therefore, the delay in submitting the confidence intervals did not adversely affect Program Suppliers' allocation.

 Concerning access to the diaries, once again, MPAA acted promptly, but Nielsen, acting on its own view of business confidentiality, restricted access. We agree with Sports that Nielsen's confidentiality can be protected and, at the same time, reasonable access can be made possible. We also agree that Sports was unduly burdened in this proceeding, having given full access to its survey, while being denied access to Nielsen's study. However, Ms. Stein has shown the Tribunal the errors that result from the diary method and the Tribunal has given credit to nonnormative viewing. We believe that Ms. Stein's testimony was an adequate substitute, and therefore, we have striven to decide this determination on the merits. However, before the next proceeding, a resolution of the access issue must be reached, and we invite pleadings.

 Relevance. We now turn to a discussion of the relevance of each survey. In the early distribution proceedings, it was suggested that a seemingly fair and objective way to divide the royalties was to take the fees generated by a cable system and to allocate them to each program based on the time each program was on the air. This was called the time-plus-fee generation approach.

 But this approach turned out be the unfairest of all because it assigned equal value to a value to a program aired at 2 a.m. and a program aired during prime time.

 Some objective distinction between programs was sought and the Nielsen study provided it, because it added the necessary ingredient to weight the value of each program--reliable estimates of actual viewing by distant cable subscribers.

 This improved the analysis greatly and gave the Tribunal what it called its "starting point." Why was it only a starting point and not the final answer? Because we recognized that viewing per se did not necessarily correspond to marketplace value. Even in the broadcast industry which relies heavily on viewing data, ratings do not precisely predict value, because of the viewers' age, income level and other demographics.

 However, in the cable industry, viewing is even a lesser predicter of value. As discussed earlier, cable's goal is to attract and retain subscribers, and will offer "niche" services, often unrelated to the volume of viewing, to induce segments of the population to subscribe. Consequently, the Tribunal has often made allocation decisions far removed from the raw Nielsen data. For example, in the 1983 decision, while Sports had a 9.5% share of the Nielsen study (see chart above), it was awarded a 16.35% share of the basic fund, 72% higher than its Nielsen share.

 The Tribunal has made these allocations upon a showing by a claimant group of the intensity or avidity of its viewers that drive cable systems to respond to them out of proportion to their actual viewing.

 Yet, the non-program suppliers claimants argue for an even greater departure from the moorings of viewing. They belive that cable operators are the ones to know how much each program type means to them, and the sum of their answers should give a much truer picture of the would-be distant-signal marketplace.

 However, Mr. Bortz conceded that all his survey provides is how much each operator would be likely to spend. It does not take into account the seller's side. Dr. Besen made a similar point, that without taking into account supply, the results are questionable. Further, Dr. Reid conceded that practicial factors such as availability and convenience can not be controlled for in a constant sum question.

 Dr. Besen contended that the Bortz survey gives relative total value, not relative marketplace value, so that the results may not be applicable to the Tribunal's task at all.

 Taking all this into account, and realizing what we said above, that the cable operator is giving a quick answer using his or her dominant impression of what was aired in 1989 and what the category boundaries are, the Tribunal is not prepared to fully embrace the results of the Bortz survey.

 What we are prepared to do, however, is to test the survey against the other testimony in the record to see if there exists any corroboration. Where corroborating evidence existed to sustain that a program category had an intense viewership or commanded valuable license fees, then the results of the Bortz survey were given substantial weight. However, even where there was no corroborating evidence, the Bortz *15302 survey, by itself, did tend to buoy a program category that might have otherwise decline due to the Nielsen data.

 How this has been done is described below in our discussion of each Phase I group. We note, however, prior to our discussion, that in the 1983 decision, Sports, PBS and the Devotional claimants had allocations higher than their Nielsen share because of a recognized avid viewership. Program Suppliers and NAB had allocations lower than their Nielsen share because syndicated series and station-production programs had not been previously shown to be a special factor to attract or retain subscribers.

 Program Suppliers

  In this proceeding, Program Suppliers had a 74.166% Nielsen share and a 45.53% result in the Bortz survey. The discrepancy between the two surveys is wide, and is only slightly ameliorated by our agreement with Dr. Besen that the Bortz survey, by splitting Program Suppliers into two categories, movies and syndicated series, adversely affected Program Suppliers' total Bortz result.

 For clues as to why this might be, we note first that the Bortz survey gave almost twice as much value to movies--31.2% (pre-adjustment)--than to syndicated series--16.9% (pre-adjustment). This indicates that although syndicated series had high totals in viewing, cable operators do not expect to have to pay that much for them. This was corrobarated by Mr. Wussler, a former president of WTBS, who testified that the broadcast license fees for such shows as "The Beverly Hillbillies" and "Petticoat Junction" were relatively cheap.

 Program Suppliers sucessfully showed high prices for movies. Movies do well in the Nielsen and Bortz surveys and in individual testimony. But we believe that syndicated series has been rated higher than it should have been by the Tribunal. Given the new weight we accord the Bortz survey and the fact that it has been corroborated by other testimony, the Tribunal makes a downward adjustment of Program Suppliers' allocation from 67.1% to 60%.

 This award includes a continuing credit for harm, based on our previous record, but quantifiable evidence is still lacking. Last, we were unable to credit Mr. Cooper's analysis of viewing-to-time ratio. NAB demonstrated on cross-examination that the ratio may be more a function of the program available on a superstation than for any supposed viewer intensity.

 The Joint Sports Claimants

  There is a wide discrepancy between the Nielsen data (11.045%) and the Bortz survey (32.37%) when it comes to Sports. The Tribunal has never tied Sports down to its Nielsen figures, but the testimony in this proceeding more than ever supports an upward adjustment toward the Bortz survey result.

 First, it was demonstrated in several ways that the period from 1983 to 1989 saw an impressive increase in the marketplace value of sports. Sports rose in the Nielsen data by 16%; Dr. Lemieux showed that sports flagship stations were carried more than ever by distant cable systems; and Commissioner Stern testified to the dramatic rise in NBA telecast rights.

 It is true that Dr. Woodbury showed that some of the rise in sports rights on cable networks was attributable to the ability of cable systems to insert their own advertising, an option not available when systems retransmit a broadcast station, and it is also true that the cable network package represents the best of the national games. Further, Dr. Woodbury showed that other sports telecast rights did not increase so dramatically as the NBA rights. Yet, Dr. Woodbury's own analysis did not include a number of sports contracts in existence from 1983 to 1989, or negotiated in 1989, and the contracts he did analyze were based on media reports which afforded no opportunity to the parties to carefully examine. As a result, Sports' testimony on the rise in license fees from 1983 to 1989 was credited.

 We also gave some credit to the contracts negotiated in 1989 because they represented current marketplace valuations, but we did not give full credit because they also represented the future value of sports in 1990-94.

 Intensity of viewership was shown by Program Suppliers' viewing-to-time ratio which was 10.5 to 1 in the four-cycle data. While we stated earlier that NAB had impeached the value of the viewing-to-time ratio analysis, Sports' result in this study was so impressive, it went beyond any difference between the reach of superstations and the reach of regional stations.

 Given these changed circumstances, and the intensity corroborative of the Bortz survey, the task for the Tribunal was how to quantify Sports' proper allocation. Mr. Wussler and Mr. Werner testified credibly about the disparity between viewing percentages and licensing fees/revenues in analogous markets. Sports on WTBS accounted for 30% to 35% of advertising revenues, yet only 13.4% of the six-cycle Nielsen viewing. Baseball on ESPN received more than 30% of ESPN's program budget, yet generated less than 15% of ESPN's viewing. In each instance, a more than two to one ratio appeared between value and viewing. In our view it would be a fair simulation of the marketplace to allocate Sports somewhat higher than twice its Nielsen share. Therefore, we have made an upward adjustment of Sports from 16.35% to 23.8%.

 The award to Sports includes a credit for harm. We accepted Commissioner Vincent's testimony that the payments made by superstations to baseball only somewhat abated the harm Sports incurs by distant signal retransmission.

 The Commercial Television Claimants (NAB)

  NAB has traditionally received less than its Nielsen viewer share for lack of harm, and for less benefit to the cable system than other program types. 51 FR 12811. In this proceeding, NAB offered a source of harm in its proposed findings--the belief that many station-produced programs have no opportunity to be syndicated because they are already carried by cable systems in the regional areas where syndication would most likely be successful. However, no testimony was offered on this point, and therefore the argument could not be weighed.

 Concerning benefit, the Tribunal has in the past credited station-produced programming for having regional appeal, but not beyond the close-in regions. Therefore, NAB provided the Tribunal in this proceeding with data to show that more of the viewing public now reside in those close-in areas of regional appeal, and that more of the distant signal importation has been within 150 of the broadcast station or in the same state.

 However, there was a countervailing trend that also occurred from 1983 to 1989. Network-affiliates, which have more station-produced programs, declined proportionally in both instances of carriage (27.4% to 22.8%) and fees generated (7.6% to 4.9%). So while the total number of signals were "closer" distant signals, the mix of those signals definitely tended to more independent stations and fewer network stations, meaning fewer station-produced programs were retransmitted. The effect of this different mix is revealed in the trend of the Nielsen data which also declined for NAB, 6.879% down to 6.347%.

 In 1983, we declined to credit the Bortz survey as much for NAB as we did for Sports, because "the Tribunal had independent evidence corroborating *15303 Sports' argument" and we did not have such for NAB. This remained so in the 1989 proceeding. The testimony of Messrs. Viener and Davies showed instances where NAB programs had appeal, but in relative terms, we still find that the viewer intensity is higher for the other Phase I claimants than for any that exist for NAB.

 However, based on a higher credit we give for the Bortz survey and for the probability that NAB's viewing may not have received full credit from diary-keepers in the Nielsen study, NAB's award is adjusted upward from 5% to 5.7%.

 The Noncommercial Television Claimants (PBS)

  By every measure, PBS has declined from 1983 to 1989 in its proportional share of the distant-signal marketplace. Instances of carriage, which we agree should have been considered at 8.06% in 1983, still declined from 7.6% to 7.4% in 1989. Fees generated declined from 2.4% to 2%. The Nielsen viewing data declined from 4.380% to 3.074%. The Bortz survey result declined from 3.40% to 2.3%.

 PBS has offered various special factors to counteract this decline. PBS offered that it has more content per program hour than commercial television; it is videotaped more; it is not geared for high ratings, so the cume rating is more relevant; and PBS stations are careful to avoid head-to-head duplicative schedules.

 These factors all relate to the special appeal of PBS, that it offers the cable system diversity, and that it has an intense viewership. We have in the past agreed with PBS and have awarded it an allocation somewhat higher than its Nielsen share to take this into account.

 Therefore, these were already incorporated into PBS' valuation, and the downward change in circumstances necessitates a downward adjustment of PBS' share from 5.2% to 4%, a share still higher than the Nielsen or Bortz results. We credit that PBS may have been adversely affected by flaws in diary-keeping, and we also credit PBS with the upward end of the confidence limits in the Bortz survey due to its small sample size.

 Concerning harm, PBS' argument that creation of PBS stations have been precluded by distant-signal importation was not credited in this proceeding. More development of that argument is needed.

 Finally, each PBS witness spoke of the quality of PBS' programs, and Messrs. Katzman and Carey offered what they believed was objective evidence of quality in terms of awards and critical recommendations.

 Quality has been a secondary criterion in the Tribunal's allocation decisions since the first proceeding. Evidence on quality has been received, but ultimately no distribution decision has been made on quality. The reason should be clear. It is a subjective evaluation with serious First Amendment implications.

 Freedom of speech means not only freedom from government censorship. It also means that the government cannot financially burden any speech because of the content of that speech. While Congress may affirmatively fund PBS, the Tribunal is in a different position. For the Tribunal to give PBS royalties on the basis that its speech has "quality"--that it is in some sense worthier than other programs--is to deny royalties to other artistic expressions because their speech is less worthy. That would be, in essence, penalizing speech for its content.

 PBS' witnesses have all made attempts to define quality. Those definitions were all different. PBS has even attempted objective standards, but these objective standards fail becacuse they are only the sum of the subjective opinions of individuals.

 Significantly, PBS, in its proposed findings, did not ask the Tribunal to credit PBS with quality per se, but only to credit PBS with the degree to which the perception of quality is valued by cable operators and induces them to import PBS stations. We agree with PBS. In this proceeding and in future proceedings, quality will no longer be a criterion in the Tribunal's distribution because of its conflict with the First Amendment.

 The Devotional Claimants

  The period from 1983 to 1989 witnessed a decline in the objective indicators for the Devotional Claimants. Their Nielsen share dropped from 0.618% to 0.243%, a 60% decline. Similarly, their share in their own time-plus-fee generation study declined from 4.36-4.63% to 1.857%, also a 60% drop.

 Dr. Clark explained the decline at least in part by the change in the schedule of WWOR, a major superstation. In 1983, 10% of WWOR's schedule was devoted to religious programs. In 1989, that had dropped to less than 0.25%.

 Ordinarily, this would result in a lower award for the Devotional Claimants. However, in 1989, the Bortz survey included the Devotional Claimants for the first time, and cable operators said they would allocate more than 4% of their budget to religious programs. Yet the 4% represented only buyers' side. If cable operators went into the marketplace, would they find the price of devotional programs much cheaper, or even zero?

 Mr. Robertson explained that looking at the broadcast industry is not the appropraite analogy, because religious programmers do not want their programs interrupted by commercials. To induce stations to give up their only source of revenue, religious programmers sometimes must pay to have their programs aired.

 But the cable marketplace is different, and the proof that religious programming has some marketplace value is The Family Channel which can command a five cents per subscriber fee. Consequently, despite the decline in carriage, the Tribunal assesses that devotional programming is the sort of "niche" service that a cable operator would want to offer subscribers, despite the low viewership. We adjust the award for Devotional Claimants upward from 1.1% to 1.25%. This is still far below the Bortz survey result, because we believe that the price of such programs is much less than what the cable operator is willing to spend. This award includes a credit for the faults of diary-keeping.

 Concerning harm, Mr. Robertson's testimony on the siphoning effect of distant-signal viewing causing local religious programmers to pay more to have their programs aired needs further development and was not credited in this proceeding.

 The 3.75% Fund

  In the 1983 proceeding, we concluded that: PBS was not entitled to share in the 3.75% fund because PBS stations are not paid for at the 3.75% rate; the awards to the Devotional Claimants and the Canadian Claimants should be reduced because religious and foreign-language specialty stations are not paid for at the 3.75% rate; and that Program Suppliers and Sports should accrue the benefit of the lower awards made to PBS, the Devotional Claimants and the Canadian Claimants.

 No evidence was introduced in this proceeding to challenge these earlier conclusions. Consequently, compared with the 1989 basic fund, in the 3.75% fund we have raised Program Suppliers and Sports, lowered the Devotional Claimants and given no award to PBS. The allocations are as follows: Program Suppliers--62.6%; Sports--26%; NAB--5.7%, Music--4.5%, Devotional Claimants-- 0.95%, Canadian Claimants--0.25%.

 *15304 The Syndex Fund

  In the 1983 proceeding, we stated that PBS was not entitled to share in the syndicated exclusivity surcharge fund because the FCC syndex rules prior to 1981 did not apply to noncommercial stations.

 PBS has challenged that conclusion by noting that cable systems paid over $400,000 in 1989 in syndex surcharges for the importation of distant PBS stations, and that the Copyright Office has interpreted these payments as being required.

 PBS is correct in its factual assertions, but that does not alter our conclusion. The former FCC rules stated that a commercial station which contracted for the exclusive license to broadcast a program had the right to demand blackout of any imported signal that had that program on it. Therefore, the Copyright Office was correct in requiring a surcharge on imported PBS stations because a commercial station could have demanded blackout of a program on a PBS signal.

 But the royalties that accrue from that payment are meant to be distributed to the program supplies of the commercial station that lost the blackout right in 1981, not to the station that is no longer the subject of blackout. It was the syndicator's loss of blackout protection that the surcharge was intended to compensate. Since noncommercial stations could never demand blackout, the repeal of the FCC blackout rules had no effect on suppliers of programs to noncommercial stations, and there was no loss to compensate.

 Concerning PBS program producers and the former pre-clearance rights of blackout, when a producer syndicates to both PBS and commercial stations, these owners are part of the Program Suppliers group as well, and when they receive compensation in that group, they receive a share of the syndex fund, so no further compensation to them in the PBS category is warranted.

 As a result, we again conclude that the only Phase I categories entitled to share in the syndex fund are Program Suppliers and Music. The claimants in these categories have reported a settlement based on the 1983 allocation decision. Therefore, the Tribunal will award Program Suppliers 95.5% and Music 4.5% of the syndex fund.

 Allocations

  After subtracting the stipulated award to National Public Radio of 0.18%, the Tribunal has adopted the following allocation to categories of claimants in Phase I of the 1989 cable copyright royalty fees available for distribution:

 
 ------------------------------------------------------------------------------
         Category Basic (percent) 3.75 (percent) Syndex
                                                                   (percent)
 ------------------------------------------------------------------------------
Program suppliers ..................... 60.00 ........... 62.60 ......... 95.50
Joint sports claimants ................ 23.80 ........... 26.00 ............. 0
Commercial television .................. 5.70 ............ 5.70 ............. 0
Music .................................. 4.50 ............ 4.50 .......... 4.50
Noncommercial television ............... 4.00 ............... 0 ............. 0
Devotional claimants ................... 1.25 ............ 0.95 ............. 0
Canadian claimants ..................... 0.75 ............ 0.25 ............. 0
Commercial radio .......................... 0 ............... 0 ............. 0
 ------------------------------------------------------------------------------
 

 Dated: April 20, 1992.

 Cindy Daub,

 Chairman.

 (FR Doc. 92-9684 Filed 4-24-92; 8:45 am)


Please refer to the IP Mall Copyright Royalty Tribunal homepage for details on this valuable resource.