hr

f& z-&II'gjC (Qfj &'& Before the ] Igg COPYRIGHT ROYALTY JUDGES Washington, D.C.

In the Matter of

Distribution of the Docket No. 14-CRB-0010-CD (2010-13)

2010, 2011, 2012, and 2013 Cable Royalty Funds

WRITTK&N DIRECT STATEMENT REGARDING ALLOCATION METHODOLOGIES OF PROGRAM SUPPLIERS

VoI,vME I or II WRITTEN TESTIMONY AND EXHIBITS

Gregory O. Olaniran D.C. Bar No. 455784 Lucy Holmes Plovnick D.C. Bar No. 488752 Alesha M. Dominique D.C. Bar No. 990311 Mitchell Silberberg 4 Knupp LLP 1818 N Street NW, 8th Floor Washington, DC 20036 (202) 355-7917 (Telephone) (202) 355-7887 (Facsimile) goo msk.corn lhp msk.corn [email protected]

Attorneysfor Program Suppliers

December 22, 2016 Before the COPYRIGHT ROYALTY JUDGES Washington, D.C.

In the Matter of

Distribution ofthe Docket No. 14-CRB-0010-CD (2010-13)

2010, 2011, 2012, and 2013 Cable Royalty Funds

WRITTEN DIRECT STATEMENT REGARDING ALLOCATION METHODOLOGIES OF PROGRAM SUPPLIERS

The Motion Picture Association ofAmerica, Inc. ("MPAA"), its member companies and other producers and/or syndicators of syndicated movies, series, specials, and non-team sports broadcast by television stations ("Program Suppliers"),'n accordance with the procedural schedule set forth in the July 21, 2016 Order Regarding

Discovery issued by the Copyright Royalty Judges ("Judges"), hereby submit their

Written Direct Statement Regarding Allocation Methodologies ("WDS-A") in the consolidated 2010-2013 Cable Royalty Distribution Proceeding. Program Suppliers are submitting this introductory memorandum in order to summarize the evidence they will present in the royalty allocation phase of this proceeding, and to state the royalty shares

' listing ofMPAA-represented Program Suppliers who submitted royalty claims for the 2010-13 cable royalty years was included as a part ofMPAA's January 21, 2015 and July 6, 2015 Petitions to Participate filed in connection with this consolidated proceeding. that Program Suppliers are seeking for the 2010, 2011, 2012, and 2013 cable royalty funds ("2010-13 Funds").

I. PROCEDURAL HISTORY

On September 9, 2015, the Judges issued an order formally consolidating royalty distribution proceedings regarding the 2010-13 Funds and sought written submissions from the parties in an effort to clarify the definitions of claims categories for purposes of this proceeding. Thereafter, on November 25, 2015, the Judges adopted a set of eight

Agreed Categories for which the Judges may approve an allocation ofroyalties for the

2010-13 Funds.'hose Agreed Categories are as follows:

"Canadian Claimants." All programs broadcast on Canadian television stations, except: (1) live telecasts of Major League Baseball, National Hockey League, and U.S. college team sports, and (2) programs owned by U.S. copyright owners. "Commercial Television Claimants." Programs produced by or for a U.S. commercial television station and broadcast only by that station during the calendar year in question, except those listed in subpart (3) ofthe Program Suppliers category. "Devotional Claimants." Syndicated programs of a primarily religious theme, but not limited to programs produced by or for religious institutions. "Joint Sports Claimants." Live telecasts ofprofessional and college team sports broadcast by U.S. and Canadian television stations, except programs in the Canadian Claimants category.

See Notice OfParticipants, Notice OfConsolidation, And Order For PreliminaryAction To Address Claims Categories at 2 (September 9, 2015).

See Notice OfParticipant Groups, Commencement OfVoluntaryNegotiation Period (A/location), AndScheduling Order at Exhibit A (November 25, 2015). "Music Claimants." Musical works performed during programs that are in the following categories: Program Suppliers, Joint Sports Claimants, Commercial Television Claimants, Public Television Claimants, Devotional Claimants, Canadian Claimants. "National Public Radio." All non-music programs that are broadcast on NPR Member Stations.

"Program Suppliers." Syndicated series, specials, and movies, except those included in the Devotional Claimants category. Syndicated series and specials are de6ned as including (1) programs licensed to and broadcast by at least one U.S. commercial television station during the calendar year in question, (2) programs produced by or for a broadcast station that are broadcast by two or more U.S. television stations during the calendar year in question, and (3) programs produced by or for a U.S. commercial television station that are comprised predominantly of syndicated elements, such as music videos, cartoons, "PM Magazine," and locally-hosted movies.

"Public Television Claimants." All programs broadcast on U.S. noncommercial educational television stations.

The Judges further recognized that these eight Agreed Categories are intended to be

"mutually exclusive," such that every program eligible to receive cable statutory license royalties in this proceeding falls within only one ofthe Agreed Categories.

On December 6, 2016, the parties to this proceeding notified the Judges that they had reached a settlement concerning the National Public Radio ("NPR") royalty share for each ofthe 2010-13 Funds. Then, on December 14, 2016, the parties informed the

Judges that they reached a settlement concerning the royalty share to be awarded to the

Music Claimants for each ofthe 2010-13 Funds. In light ofthese two allocation

See id. at Exhibit A.

See id settlements, the Judges need only to determine royalty shares for the remaining six

Agreed Categories in this proceeding. The six Agreed Categories for which allocation remains in controversy are the Canadian Claimants Group ("CCG"), Commercial

Television Claimants ("CTV"), Devotional Claimants ("Devotionals"), Joint Sports

Claimants ("JSC"), Program Suppliers, and Public Television Claimants ("PTV"). In this

WDS-A, MPAA will set forth its testimony and evidence in support of royalty allocation awards for the Program Suppliers category.

II. SUMMARY OF PROGRAM SUPPI.IKRS'ASK

In this proceeding, the Judges must allocate royalties that cable system operators

("CSOs") paid to the Copyright Office for the privilege of distantly retransmitting broadcast stations to distant subscribers pursuant to the statutory license set forth in

Section 111 of the Copyright Act for the years 2010, 2011, 2012, and 2013. The standard that the Judges and their predecessors have used for allocating Section 111 royalties in past proceedings is the "the relative marketplace value of the distant broadcast signal programming retransinitted by cable systems" during the royalty years at issue. Program

Suppliers agree that the Judges should allocate the 2010-13 Funds based on the relative market value standard.

Program Suppliers have long maintained that the relative market value of distantly retransmitted programming is most clearly expressed through measurements of distant viewing, because viewing is consumption. Thus, it is axiomatic that consumers subscribe to a cable system to watch the programming made available via their subscriptions. The

See 75 Fed. Reg. 57063, 57065 (September 17, 2010). more programming a subscriber watches, the happier the subscriber is, and the more likely he or she will continue to subscribe. Thus, a reasonable measure of the relative market value of a retransmitted program is the relative level of subscriber viewing of that program.

Because each of the Agreed Categories is simply an aggregation of distantly retransmitted individual programs, an aggregation of the distant viewing for all of the programs falling in each of the Agreed Categories provides a reasonable and reliable measure ofthat catego&y's relative market value. Accordingly, Program Suppliers are presenting evidence of aggregated distant viewing for each Agreed Category that remains in controversy, measured by Nielsen, and estimated by our economist witness, Dr. Jeffrey

S. Gray. Dr. Gray explains in his testimony why his distant viewing estimates provide reliable evidence of relative market value, and should be the basis for the Judges'oyalty allocation awards to Program Suppliers in this proceeding. Program Suppliers also present the testimony of other industry witnesses who confirm that Nielsen viewing information is the most reliable metric for determining the relative market value of programming on distant signals retransmitted by CSOs.

In past proceedings, the Judges and their predecessors relied on a survey of CSOs performed by Bortz Media K Sports Group, Inc. (the "Bortz Survey") in order to determine the relative market value of distant signal programming. As explained in more detail in Program Suppliers'DS-A, Program Suppliers disagree that a survey of CSOs

See Notice at Exhibit A (describing the mutually exclusive Agreed Categories as "non-exhaustive descriptions of the types ofprograms or other creative works that fall within each ofthe agreed categories of claimants (Agreed Categories) to which categories the Judges may approve an allocation of cable retransmission royalties."). Cinemax ("MAX") for over 29 years. Mr. Pasquale explains the business model for cable premium networks and discusses how, during his tenure at HBO, HBO subscribed to and relied on Nielsen ratings information to evaluate whether its programming was sufficiently attracting and retaining subscribers. While Mr. Pasquale designed and conducted many surveys for HBO, Mr. Pasquale will explain that HBO relied on Nielsen data for accurate information regarding what programming was valuable to subscribers, and found a correlation between subscribers not watching HBO/MAX and their decision to drop their subscription. Mr. Pasquale will also explain why premium cable networks are a reasonable proxy for distant broadcast stations in the context of this proceeding and why, in his view, Nielsen viewing data provides significant evidence of the market value of programming and cannot be replaced by surveys.

Paul Lindstrom, is a Senior Vice President with Nielsen, where he is responsible for research design and analysis as part of the Nielsen Media Analytics group. Mr.

Lindstrom will provide information about the Nielsen viewing data on which Program

Suppliers rely in this proceeding, including his role in designing the custom analyses of viewing to distant cable households that Program Suppliers commissioned from Nielsen for 2010, 2011, 2012, and 2013.

Jeffre S. Gra Ph.D., is the founder and President of Analytics Research Group,

LLC. Relying on certain basic economic principles, Dr. Gray employs Nielsen data, multiple other data sources, and regression analyses to estimate the level of distant viewing to a random sample of stations carrying compensable works for each ofthe stations to RSNs, national cable networks, and new distribution options that compete for the attention of cable subscribers. Mr. Mansell supports his analyses with an array of data charting migration of games from the late 1990s to 2013. These data clearly show that the number of live Major League Baseball, National Basketball Association,

National Hockey League and National Collegiate Athletic Association games shown on over-the-air broadcast networks and local TV stations dramatically declined, while the number of live game telecasts available to cable subscribers via other delivery methods, including RSNs, national and cable networks, and live streaming over the Internet, substantially increased.

Jonda K. Martin, is the President and Owner of Cable Data Corporation

("CDC"), which collects, digitizes, and electronically stores the data contained in the cable operator statements of account ("SOAs") on file with the Copyright Office. Ms.

Martin will provide an overview of CDC's operations and data collection methodologies.

She will also describe data reports that CDC generated and provided to Program

Suppliers and their witnesses in connection with this proceeding, and the county analyses for each ofthe 2010-13 cable royalty years that she performed and provided for Nielsen to use in connection with distant viewing analyses they provided for this proceeding.

IV. DESIGNATED PRIOR TESTIMONY

Pursuant to 37 C.F.R. $ 351.4(b)(2), Program Suppliers hereby designate for incorporation in their WDS-A the following witnesses'estimonies from the 2004-2005

Phase I Cable Distribution Proceeding before the Judges, Docket No. 2007-3 CRB CD

2004-2005, and the 1989 Phase I Cable Distribution Proceeding before the Copyright

12 Royalty Tribunal, Docket No. CRT 91-2-89CD (copies ofwhich are included in Volume

II ofProgram Suppliers'DS-A):

DOCKET NO. 2007-3 CRB CD 2004-2005

WITNESS RECORD DESIGNATION

Alex Paen, Written Direct Testimony (filed June 1, Founder and President, Telco Productions. 2009), admitted as PS Exhibit 1.

Alan M. Rubin, Ph.D., Written Direct Testimony (filed June 1, Professor Emeritus and Director Emeritus 2009, corrected September 29, 2009), of the School of Communications Studies, admitted as PS Exhibit 4. Kent State University.

DOCKET NO. CRT 91-2-89CD

WITNESS RECORD DESIGNATION

Alan M. Rubin, Ph.D., Written Direct Testimony (filed November Professor of Communications and Director, 1991). Communication Research Center, Kent State University. Oral Testimony, Hearing Transcript pp. 5252-5459 (December 16, 1991).

Program Suppliers reserve the right to designate additional porlions ofthe records in prior proceedings if, afler examining the record designations of other parties, it appears that such additional portions are necessary for a complete and accurate understanding of the import of designated evidence.

13 V. PROGRAM SUPPLIERS'LLOCATION CLAIMS

Based on the evidence submitted to the Judges in this proceeding as a part ofthis

WDS-A, Program Suppliers are seeking the following percentage shares ofthe 2010-13

Funds:

Rovaltv Year Basic Fund (%) 3.75% Fund (%) Svndex Fund (%)

2010 52.10% 71.61% 96.50%

2011 51.13% 72.12% 96.50%

2012 36.40% 62.36% 96.45%

2013 45.12% 67.82% 96.45%

Program Suppliers reserve the right to change their allocation claims in light ofthe evidence presented by other claimants in this proceeding.

14 Respectfully submitted,

Gregory O. Olaniran D.C. Bar No. 455784 Lucy Holmes Plovnick D.C. Bar No. 488752 Alesha M. Dominique D.C. Bar No. 990311 Mitchell Silberberg 0 Knupp LLP 1818 N Street NW, 8th Floor Washington, DC 20036 (202) 355-7917 (Telephone) (202) 355-7887 (Facsimile) goo'sk.corn [email protected] amd msk.corn

Attorneysfor Dated: December 22, 2016 Program Suppliers

15 DIRECT TESTIMONY OF JANE V. SAUNDERS

I. BIOGRAPHICAL INFORMATION

My name is Jane V. Saunders. I serve as Managing Director k, Senior Vice

President, Rights Management Policy and Relations ("RMPR"), at the Motion Picture

Association ofAmerica ('MPAA"). Prior to joining MPAA in 1995, I was an attorney engaged in private practice for seven years in Atlanta, Georgia and the District of

Columbia. Immediately prior to joining MPAA, I worked for three years as an associate attorney handling compulsory license matters, including participating in royalty distribution proceedings before the Copyright Royalty Tribunal. I received my baccalaureate degree &om Dartmouth College and my law degree from Emory

University. I am a member, with inactive status, ofthe Georgia State and District of

Columbia Bar Associations. I am fluent in Spanish and French and have a working knowledge of German and Italian.

I represent MPAA and its member companies on the boards ofvarious collective management organizations, including AGICOA,'here I sit on the Executive Board; the

German collection society, GWFP USA; Danish collection societies, CAB and FilmKopi; and the Copyright Collective of Canada. I head MPAA's RMPR department which is responsible for copyright royalty policy as it relates to collective and statutory rights management both domestically and worldwide on behalf ofMPAA's members and, in some cases, other producers and distributors ofcompensable works.

'GICOA, the Association For The International Collective Management OfAudiovisual Works, is headquartered in Geneva, Switzerland. Direct Testimony ofJane V. Saunders Page 3 regarding setting rates and negotiating with the other collection society groups (i.e., claimants) regarding allocation of collected royalties.

In the U.S., I supervise all ofMPAA's activities in connection with the administration ofU.S. cable and satellite retransmission royalties (Sections 111 and 119 ofthe Copyright Act, respectively), including supervision ofthe royalty administration and enforcement ofthe compulsory licenses; supervision ofMPAA employees; engagement of outside vendors; procurement ofdata; and supervision of outside counsel.

In that regard, I manage a team who works closely with information technology contractors and with financial, legal and statistical professionals to provide fair and efficient distribution ofroyalties among our represented claimants.

II. PURPOSE OF TESTIMONY

First, I will describe the nature and extent ofProgram Suppliers'laims in this proceeding, including the different types ofprograms that comprise Program Suppliers'laim.

Second, I will discuss my international experience regarding the role of viewership as a measure ofvalue for purposes ofthe distribution ofretransmission royalties.

III. OVERVIEW OF PROGRAM SUPPLIERS'LAIM

Beginning with the first royalty distribution proceeding addressing the allocation of 1978 cable royalties, MPAA has been the defacto category representative of all

Program Suppliers claimants. For purposes ofthis proceeding, the Copyright Royalty

Judges ("Judges") have defined Program Suppliers as the owners of syndicated series, specials and movies, except those included in the Devotional Claimants category. See

Notice ofParticipants Groups, Commencement ofVoluntary Negotiation Period Direct Testimony ofJane V. Saunders Page 4

(Allocation), and Scheduling Order, Docket No. 14-CRB-0010-CD (2010-13) at Exhibit

A (Nov. 25, 2015). The Judges have further defined syndicated series and specials as including (1) programs licensed to and broadcast by at least one U.S. commercial television station during the calendar year in question; (2) programs produced by or for a broadcast station that are broadcast by two or more U.S. television stations during the calendar year in question; and (3) programs produced by or for a U.S. commercial television station that are comprised predominately of syndicated elements, such as music videos, cartoons, "PM Magazine," and locally-hosted movies. See id.

Program Suppliers include not only the major studios, but also dozens of smaller producers and syndicators &om both the U.S. and many parts ofthe world — all ofwhom have filed claims seeking a share ofthe pool for some or all ofthe 2010-13 cable royalty years. By way ofexample, the owners ofnonnetwork series, movies, specials, and non- team sports which air on commercial television broadcast stations retransmitted by cable systems who have agreed to representation by MPAA ("MPAA-represented Program

Suppliers") asserting claims to Section 111 royalties in this proceeding are set forth in

MPAA's 2010-2012 cable Petitions to Participate filed on January 21, 2015, and 2013 cable Petition to Participate filed on July 6, 2015.

Although Program Suppliers programs fit generally under the umbrella of series, movies, specials, and non-team sports, the width and breadth ofthose programs are quite remarkable. The programs include game shows, sitcoms, news magazines, interview shows, sports shows and sporting events, awards shows, health and fitness shows, and animal shows, as well as similar Spanish works. The following are examples ofProgram

Suppliers'rograins: Direct Testimony ofJane V. Saunders Page 5

~ Animated series and sitcoms, such as: FRIENDS (Warner Bros. Entertainment,

Inc.), MODERN FAMILY (Fox Ente~ent Group, Inc.), and THE SIMPSONS

(Fox Entertaixment Group, Inc.).

~ Movies, such as: X-MEN: THE LAST STAND (Fox Entertainment Group, Inc.),

NO COUNTRY FOR OLD MEN (Miramax Film NY, LLC) and RUNAWAY

BRIDE (Paramount Pictures Corporation).

~ Game shows, such as: FAMILY FEUD (Fremantlemedia North America, Inc.),

WHEEL OF FORTIJNE (Califon Productions, Inc.) and JEOPARDY! ( Jeopardy

Productions, Inc.).

~ Sports shows and sports-related programs, such as: 2011 WIMBLEDON

CHAINPIONSHIPS (IMG, Inc.), PISTONS WEEKLY (National Basketball

Association), THIS WEEK IN BASEBALL (Major League Baseball Properties, Inc.),

NASCAR RACING (NASCAR Media Group), 2010 AMERICAN SKI CLASSIC

(Jalbert Productions, Inc.) and WWE FRIDAY NIGHT SMACKDOWN! (World

Wrestling Entertainment, Inc.).

~ Awards shows and pageants, such as: AMERICAN LATINO AWARDS

(Latinafion, LLC), 6STH ANNUAL GOLDEN GLOBE AWARDS (dick clark

productions, inc.), and THE 65TH ANNUAL TONY AWARDS (The Fremantle

Corporation).

~ News shows, such as: MCLAUGHLIN GROUP (Oliver Productions, Inc.) and

WALL STREET JOURNAL REPORT WITH MARIA BARTIROMO (NBC

Universal, Inc.). Direct Testimony of Jane V. Saunders Pa e6

~ Reality shows, such as: AMERICAN IDOL (Fremantlemedia North America, Inc.),

SURVIVOR: HEROES VS. VILLAINS (CBS Broadcasting, Inc.), and THE

BACHELORETTE (Warner Bros. Entertainment, Inc.).

~ Animal shows, such as: WILD ABOUT ANIMALS (Steve Rotfeld Productions,

Inc.), ANIMAL RESCUE (Telco Productions, Inc.), and INCREDIBLE DOG

CHALLENGE (Eclipse Television, LLC).

~ Interview and talk shows, such as: THE OPRAH ~REY SHOW (CBS Studios,

Inc.) and THE ELLEN DEGENERES SHOW (Warner Bros. Entertainment, Inc.).

IV. VIEWING AND MARKET VALUE

I understand that the standard for allocation of royalties in this proceeding is the relative market value ofprograms. In their recent Phase II distribution decision regarding the distribution ofthe 2000-2003 cable royalties, the Judges concluded:

[V]iewership can be a reasonable and directly measurable metric for calculating relative market value in cable distribution proceedings. Indeed, the Judges conclude that viewership is the initial and 'predominant heuristic that a hypothetical CSO would consider in determining whether to acquire a bundle ofprograms for distant retransmission, subject to marginal bundling adjustments needed to maximize subscribership.

Distribution ofthe 2000, 2001, 2002, and 2003 Cable Royalty Funds, 78 Fed. Reg.

64984, 64996 (Oct. 30, 2013). I think the Judges might find it useful that my international experience working with retransmission systems, including in Europe and in

Canada, confirms that viewing of programs is the predominant metric used to distribute retransmission royalties paid by cable and satellite operators to producers, such as the

MPAA members. Direct Testimony ofJane V. Saunders Page 7

For example, Canada has a compulsory license system quite similar to that in the

U.S. In that system, nine collective societies (the Canadian equivalent of claimant groups), each representing a category ofrights owners, claim royalties paid by retransmitters under the compulsory license for the distant retransmission of audiovisual works. Viewing is the metric relied upon by the Canadian Copyright Board ("CCB"), a body comparable in mission to the Judges, to allocate shares oftotal royalties paid by retransmitters among the Canadian collectives. To that end, one or more ofthe claiming collective societies, which represent the rights holders, submit individually orjointly

(depending on the royalty year period in question) viewing studies to the CCB. These

studies are based upon viewing data for broadcast events during a given royalty year,

obtained through a monitoring organization called BBM Canada, and other metrics to winnow out distant subscriber viewing ofthe various audiovisual works claimed by each

collective. The CCB has relied on these viewing studies to allocate the collected royalties.

Thank you for the opportunity to present the information in this testimony. I hope

it will be helpful in the Judges'eliberations.

See Retransmlssion ofDistant Radio and Television Signals, CCB Decision, File 1989-1, at p. 57 (October 2, 1990) ("The cable industry itself insists that its business is to offer variety and choice of programming. There is, however, no measure ofthe value of individual programs to a cable operator; it is not possible to determine, for example, how much a cable operator would pay for a program that has twice the number ofviewers as another. But a cable operator, wishing to attract and keep subscribers, would make choices consistent with the preferences ofthose subscribers. Accordingly, the Board finds that the next best method ofdetermining the value ofprograms to a cable operator is the extent to which subscribers watch the programs that are available."); id. at p. 58 ("Whatever its imperfections, the Board concludes that the viewing approach is the best means available for establishing the different values of different programs."); see also Retransmission ofDistant Radio and Television Signals, CCB Decision, File 1991-10, at p. 191 (January 14, 1993) ("The Board also remains ofthe opinion that, within the respective pools, viewing provides the best method ofallocating the royalties collected, so long as the data documenting viewing is adequate."). DECLARATION OF SANK V. SAUNDERS

I declare under penalty ofperjury that the foregoing testimony is true and correct, and ofmy personal knowledge.

Executed on December A 2016

Joe V. Saunders CO X

O

I I

I Before the COPYRIGHT ROYALTY JUDGES Washington, D.C.

In the Matter of

Distribution of the Docket No. 14-CRB-0010-CD (2010-13)

2010, 2011, 2012, and 2013 Cable Royalty Funds

DIRECT TESTIMONY OF SUE ANN R. HAMILTON

DECEMBER 22, 2016 DIRECT TESTIMONY OF SUE ANN R. HAMILTON

I. BIOGRAPHICAL INFORMATION

My name is Sue Ann R. Hamilton. I am the Founder and Principal of Hamilton

Media LLC ("Hamilton Media"), a consulting firm dedicated to advising and representing public and private media companies, including both content creators and multi-platform distributors, regarding the finer points of content distribution via multiple system operators ("MSOs") as well as over-the-top ("OTT") video distributors. My work with Hamilton Media currently includes or has included: (a) content licensing and distribution consulting for program suppliers, such as Mark Cuban Companies'nd

AEG's AXS TV and HDNet Movies, as well as two large national broadcast network groups; (b) negotiating content deals for distribution platforms such as the National Cable

Television Cooperative ("NCTC") (with more than 850 cable system member companies), DBSD North America, Inc., and several other relatively new OTT video distribution entrants; (c) providing strategic consulting services for media companies, professional sports teams and leagues, and collegiate sports teams and conferences, as well as for non-team amateur and professional events and sports; and (d) serving as arbitrator, mediator, or litigation advisor in numerous disputes related to television rights, distribution, and license fees.

I received a Bachelor ofArts from Carleton College and my Juris Doctor from

Stanford Law School. After completing law school, I worked as an associate attorney, and later as a partner, at the law firm of Kirkland X Ellis, LLP. In 1993, I left the practice of law for a business position in the content acquisition group, or "programming" department, of Tele-Communications, Inc. ("TCI"), which later through merger became

ATILT Broadband, LLC ("ATILT"). There, my initial position was Assistant Director, then Director, and finally Vice President, Programming and Senior Vice President,

Satellite Services. My primary role was negotiating agreements to permit ATILT to distribute cable and broadcast networks and stations (but not individual programs or categories of programs, which in my experience, are virtually never negotiated for, or acquired, on an individual level) to cable television subscribers. After ATILT, I served for a short time as the acting head of the programming department at Adelphia

Communications Corporation during its bankruptcy, and then moved to Charter

Communications, Inc. ("Charter"), where I worked from 2003 to 2007, as head of the content acquisition team, first as Senior Vice President, Programming, and later as

Executive Vice President, Programming, reporting to the company's CEO.

During my tenure, Charter was the third largest MSO in operation in the United

States, serving approximately six million cable subscribers. My responsibilities at

Charter were very broad. They included:

~ Strategic planning, distribution management, and selecting cable networks

and broadcast signals to carry to serve Charter's subscribers;

~ Managing a roughly $ 1.5 billion programming budget;

~ Negotiating the terms of programming deals between Charter and such

companies as Disney/ABC/ESPN, CBS/Showtime (including Viacom at

that time), Time Warner/HBO/Turner, NBC Universal, Twentieth Century Fox (including Fox Sports), Discovery Networks, AXE/History, and AMC

Networks (including at the time Madison Square Garden Network

("MSG")), as well as independent regional sports networks such as YES

Network and New England Sports Network ("NESN");

~ Leading negotiations with major sports leagues for potential carriage of

networks like NFL Network and NBA TV; and

~ Selecting broadcast stations for distant carriage by Charter.

While at Charter, I was recognized by both CableFAX magazine and Multichannel

News for my role as a leader in the cable and telecommunications industry.

As a part of my recent work with Hamilton Media, I have served as an expert witness on behalf of EchoStar Satellite LLC (a/k/a Dish Network) in federal court regarding a retransmission consent dispute with a broadcaster. I have also served as a non-publicly disclosed expert witness in some matters that have settled before trial or mediation. This is my first testimony in a royalty distribution proceeding before the

Copyright Royalty Judges ("Judges"). A copy of my curriculum vitae is attached to my testimony as Exhibit A.

II. PURPOSE OF TESTIMONY

Based on my review of materials related to this and prior cable royalty distribution proceedings,' understand the purpose of this proceeding to be the allocation of royalties

'o prepare for my testimony, I reviewed the following documents: (1) prior testimony of cable operator witnesses Judith Allen, Michael Egan, June Travis, and Judith Meyka submitted in connection with the 1998-99 and 2004- 2005 Cable Phase I Proceedings, and the 2004-2009 Cable and 1999-2009 Satellite Phase II Proceeding; (2) the testimony submitted by Dr. Robert W. Crandall and Howard B. Homonoff in the 2004-2005 Cable Phase I Proceeding; (3) the Bortz Media & Sports Group, Inc. Report on Cable Operator Valuation Of Distant Signal Non- paid by cable systems to the Copyright Office pursuant to Section 111 ofthe Copyright

Act. The parties entitled to those royalties, who are organized by claimant groups and corresponding program categories, are those copyright owners whose nonnetwork programs aired on broadcast television stations retransmitted out of market during the years, 2010, 2011, 2012, and 2013. In determining the allocation of the 2004 and 2005 cable royalties among the claimants, the Judges relied on the results of a survey of cable system executives known as the Bortz Survey. Specifically, the Judges relied on the respondents'llocation of a fixed budget amount of the different program categories as evidencing the relative market value ofthose program categories.

The purpose ofmy testimony is to discuss (a) how cable systems actually select programming; (b) the nature of compensable programming in these proceedings and the volume of programming appearing on distant broadcast signals retransmitted by cable operators; and (c) to offer my views on how cable operators are likely to value non- network programs as shown on distantly retransmitted broadcast signals, in a hypothetical market where the cable compulsory license does not exist. In these discussions, where appropriate, I juxtapose my actual experience in the cable industry with some of the assumptions underlying the Bortz Survey.

III. PROGRAMMING SELECTION DECISIONS AT CABLE SYSTEMS

Over the years, cable systems have become increasingly consolidated into large regional and national MSOs. This consolidation has had a significant impact on the way

Network Programming: 2004-05 ("Bortz Survey"); (4) the 2004-2005 Cable Final Distribution Order issued by the Judges, 75 Fed. Reg. 57063 (September 17, 2010) ("2004-05 Decision"); and (5) the Judges'otice OfParticipant Groups, Commencement OfVoluntary Negotiation Period (Allocation), And Scheduling Order (November 25, 2015) ("Notice"). cable operators make decisions about the selection ofthe cable networks and the broadcast stations they choose to carry. In my experience, during 2010-13, virtually all major MSOs had a centralized hierarchy in place requiring senior-level management to approve channel line ups for all cable systems within the MSO, regardless of geography.

Programming decisions were no longer made at the individual system level. For example, when I was in charge of programming at Charter, I was responsible for making programming decisions for literally hundreds of cable systems located in approximately

40 states.

During my time at Charter, my programming decisions were designed to select the cable networks and broadcast stations that I thought would best contribute to subscriber attraction and retention for my cable systems. In order to make that determination, the factors I considered were (1) actual and/or projected subscriber viewing behavior, (2) legacy carriage, (3) whether carriage of a particular network or station was necessary due to bundling of stations by content providers, and (4) cost to my cable system for acquiring the network or station in terms of my overall programming budget. I discuss each of these factors below.

A. THE IMPACT OF SUBSCRIBER VIEWING BEHAVIOR ON PROGRAMMING DECISIONS

In my experience, cable systems would never contemplate either adding or dropping a channel without analyzing actual and/or projected subscriber viewing behavior associated with the programming. I would usually request (and receive) Nielsen information regarding programming as a part of a pitch from a content provider seeking carriage on my system. I would also consider other factors evidencing the programming's actual or potential subscriber appeal, such as carriage by competitor

MSOs, and subscriber commentary via surveys and social media. This information was important to me in understanding what my subscribers wanted to watch—and, in turn, what sort of programming would either attract or retain them as Charter subscribers.

B. THE IMPACT OF LEGACY CARRIAGE ON PROGRAMMING DECISIONS

Another very important factor that influenced my programming decisions, especially in the case of distant signals, is legacy carriage. In my experience, cable operators would typically not make programming decisions regarding distant signals on an annual basis but rather over a term of years, typically coinciding with the three-year retransmission consent election cycle. When distant signals would come up for renewal, legacy carriage (and viewership and subscriber satisfaction or dissatisfaction associated with legacy carriage) was an important factor to me in deciding whether to continue the carriage or drop it. In my experience, it is very difficult for any cable system to drop any channel, especially any channel that it has been carrying for a long time. Every channel carried has its subscriber constituencies, even if viewing levels of individual content or the nature of that content vary over time. A cable system's decision to drop a channel is very often cited by cable subscribers as a reason for them to switch to another cable provider. Thus, I would consider and weight legacy carriage, and subscriber satisfaction or dissatisfaction associated with that carriage, in my programming decisions. C. THE IMPACT OF BUNDLING ON PROGRAMMING DECISIONS

Another important consideration in my programming decisions as a cable operator is the concept of bundling. In the past, cable operators sometimes chose to carry an individual station based on their dominant impression ofthe type of programming carried by that particular station. However, with the increased consolidation of both MSOs and content providers, MSOs are more likely to acquire multiple cable and broadcast stations as a part of a bundle of stations, rather than selecting each station individually. This was particularly true during the 2010-13 time frame in the case of the distant signal WGN-TV

("WGN") out of the Chicago DMA. During the period from 1994 through at least 2010, cable operators generally were required to carry WGN as a part of a bundle of programming provided by Tribune Media ("Tribune") if they chose to carry other in- market Tribune stations. In my experience, cable systems would carry WGN as a distant signal simply because it was required as a part of the Tribune bundle, without regard for the particular content appearing on WGN. The original decision to carry WGN was made to provide subscriber access to other Tribune-owned stations, particularly major in- market broadcast network affiliates, and not necessarily because of content retransmitted on WGN. The continuation of WGN carriage after it was unbundled from Tribune station retransmission consent was primarily due to the legacy carriage considerations discussed immediately above rather than the content itself. The broad carriage of WGN, in turn, perpetuated carriage of other distant market signals, because payment of the mandatory copyright fee made the carriage of other distant signals in these markets effectively "free." D. THE IMPACT OF COST ON PROGRAMMING DECISIONS

Another important factor that I considered in making programming decisions was the cost of the programming versus my anticipated gain or loss in subscribers associated with the carriage (or non-carriage) ofparticular stations. Distant signals make up a very small portion of a cable operator's overall programming budget. When I was working at

Charter, my annual programming spend was approximately $ 1.5 billion, of which the costs related to the carriage of distant signals were only a small fraction. During that time, the cost associated with carrying distant stations was immaterial when compared with the potential loss of subscribers that could result from dropping a distant station, especially if the distant station in question was being offered by my MSO competitors.

Thus, if I expected a potential loss of subscribership related to dropping a particular legacy distant station, I was far more likely to continue carrying the station rather than drop it.

IV. COMPENSABLE PROGRAMMING IN THESE PROCEEDINGS

As I have stated, the goal of this proceeding is to determine the relative market value of nonnetwork programming aired on distant broadcast signals that were simultaneously retransmitted by cable systems during the 2010-13 calendar years.

According to the Judges'rder, program category definitions adopted for these proceedings are as follows:

'or the purposes ofthis proceeding, network programs on broadcast signals and programs on cable networks are not compensable. For the same purposes, only programming by ABC, CBS, and NBC are considered network programs. Programs by Fox are considered nonnetwork programs. "Canadian Claimants." All programs broadcast on Canadian television stations, except: (1) live telecasts ofMajor League Baseball, National Hockey League, and U.S. college team sports, and (2) programs owned by U.S. copyright owners.

"Commercial Television Claimants." Programs produced by or for a U.S. commercial television station and broadcast only by that station during the calendar year in question, except those listed in subpart (3) ofthe Program Suppliers category.

"Devotional Claimants." Syndicated programs of a primarily religious theme, but not limited to programs produced by or for religious institutions.

"Joint Sports Claimants." Live telecasts ofprofessional and college team sports broadcast by U.S. and Canadian television stations, except programs in the Canadian Claimants category.

"Music Claimants." Musical works performed during programs that are in the following categories: Program Suppliers, Joint Sports Claimants, Commercial Television Claimants, Public Television Claimants, Devotional Claimants, Canadian Claimants.

"National Public Radio." All non-music programs that are broadcast on NPR Member Stations.

"Program Suppliers." Syndicated series, specials, and movies, except those included in the Devotional Claimants category. Syndicated series and specials are defined as including (1) programs licensed to and broadcast by at least one U.S. commercial television station during the calendar year in question, (2) programs produced by or for a broadcast station that are broadcast by two or more U.S. television stations during the calendar year in question, and (3) programs produced by or for a U.S. commercial television station that are comprised predominantly of syndicated elements, such as music videos, cartoons, "PM Magazine," and locally-hosted movies. "Public Television Claimants." All programs broadcast on U.S. noncommercial educational television stations.

The foregoing program definitions followed the claimants'ast practice ofusing these definitions to define the programming within their respective claimant groups. This past practice formed the basis ofthe program categories appearing in the Bortz Survey, except for the works falling within the Music Claimants and the National Public Radio program categories, which were not a part ofthe Bortz Survey.

A. THE PROGRAM CATEGORY DEFINITIONS USED ARE UNIQUE TO THESE PROCEEDINGS AND INCONSISTENT WlTH THE GENERAL CABLE INDUSTRY CLASSIFICATION OF PROGRAM GENRES.

It is important to recognize that the program categories that have been adopted for this proceeding and in the Bortz Survey are quite different from the industry understanding of what programming typically falls in a particular programming genre. In my experience, the genres ofprogramming in the industry are more broadly defined than the program category definitions that the Judges have adopted, especially with regard to sports programming. For example, in my experience, most people working in the cable industry would not immediately realize that the definition of sports programming established for the Joint Sports Claimants'rogram category covers only "live telecasts ofprofessional and college team sports."'hat definition excludes other broadcast programming commonly understood in the industry to fall under the umbrella of "sports,"

'ee Notice at Bxhibit A.

"See id at 1 and Bxhibit A; see also 2004-05 Decision at 57064, n.12.

The Bortz Survey questionnaire describes the same Joint Sports Claimants program category as "Live professional and college team sports" broadcast during the royalty years at issue by the U.S. commercial stations identified by the survey interviewer.

10 such as NASCAR and Formula One racing; PGA and LPGA golftournaments; professional tennis matches; individual and team performance "ninja" and "warrior" races; cycling, running, and swimming competitions; and even the Olympics themselves

— all quintessential "sports" in most cable operators'inds — and all of which fall in the

Program Suppliers'ategory if shown on broadcast signals for purposes ofthese proceedings.

Also, in my experience, most cable operators would not necessarily recognize that pre- and post-game shows, coaches'eekly interview programs, and highlight compilation telecasts (including those replaying professional and college team sports games segments) fall within the Program Suppliers'ategory, rather than the Joint Sports

Claimants'ategory. Nor would they recognize that station-produced programs such as high school games and even sports segments in local newscasts that cover professional and college teams (including game highlights) fall in the Commercial Television program category. Because ofthe incongruence ofthe definitions ofprogram categories in this proceeding versus industry practice, I would expect cable operators to have difficulty accurately assessing programming value (or accurately allocating shares oftheir fixed budget as the Bortz Survey requires) to the program categories established by the Judges in this proceeding.

Moreover, when assessing the value of sports carriage, cable operators do not routinely differentiate between network sports and nonnetwork sports as required to be done to assign value in this case. In fact, during the 2010-13 time period, cable operators spent an average of 33-35% oftheir overall cable television programming budget on

11 cable sports channels. The most significant share ofthat spending goes towards NFL,

NBA, NHL, and MLB games appearing on national cable sports networks (like ESPN) and regional sports networks ("RSNs"), and I understand such games are not compensable for purposes ofthis proceeding. Thus, when the term "sports" is used vaguely and without a specific definition or examples ofwhat events would be included or excluded, I would expect cable operators'pending on cable network sports to predispose their opinion ofthe cost (and therefore, in their minds, the corresponding market value, as determined by the value to their subscribers) of sports programming.

Meaning, a cable operator unfamiliar with the very specific definitions adopted for purposes ofthese proceedings (which I would submit would almost always be the case) would almost certainly confuse which programs fall within the operative categories in

this proceeding versus commonly understood industry genres, and in particular, would vastly overstate the amount of sports programming in the Joint Sports Claimants'rogram

category in response to an opinion survey such as the Bortz Survey.

B. SPORTS MIGRATION HAS IMPACTED THE VOLUME OF COMPENSABLE SPORTS PROGRAMMING AVAILABLE ON DISTANT SIGNALS.

Cable operators typically attach a high value to sports programming carried by their systems. However, the vast majority of any cable operator's programming budget devoted to so-called "sports" programming is not related to the value of nonnetwork live professional and college team sports programming carried on distantly retransmitted broadcast signals. Rather, that budget has focused increasingly on sports programming on cable networks. This was especially true over the 2010-13 time period. Over the last

12 decade there has been a significant migration of live team sporting events from broadcast television and syndication to other distribution outlets, such as ESPN, RSNs, cable sports channels owned and operated by different sports teams, leagues, and conferences, and general interest cable networks such TNT and TBS. This migration of sports from broadcast to cable has increased since 2005, and based on my industry experience and work with different professional and college sports teams, leagues, and conferences, I am confident that this trend will not be reversed. Due to such migration, the volume of non- network live professional and college team sports available on distant broadcast stations is now very small, and is dwarfed by the volume of nonnetwork syndicated series, movies, specials, and non-team sports that are available on broadcast signals. In my experience, a cable operator asked to value sports programming for purposes of this proceeding would not be in a position to accurately identify that particularly small subset of sports programming that falls within the Joint Sports Claimants'efinition for this proceeding absent additional information about the nature and volume of programming that falls within the category definition established by the Judges.

V. VALVATION OF DISTANT SIGNAL PROGRAMMING

Cable system operators have a goal of attracting and retaining cable subscribers to generate revenue in the form of monthly subscription fees and maximize potential viewership for increased advertising revenue. To do this, cable operators select a complement of broadcast stations and cable networks, and create a channel line-up of programming to make available to their subscribers. Generally, households subscribe to cable because they want to watch the variety of programming that cable operators choose

13 to carry, and retain their subscriptions as long as they continue watching the programming being carried. Households that are not interested in watching the programming offered by the cable system would be unlikely to subscribe, or to retain their subscription. Accordingly, subscriber viewing information is very important to cable operators in assigning value to distant signal programming, along with the other factors I described earlier in my testimony,

Current regulations prevent cable operators that carry distant broadcast signals from altering those signals, thus, preventing them from inserting advertising directed at the distant market. However, if the regulatory prohibition on altering distant signals were not in place, I would expect that cable operators would negotiate the right to alter distant signals so they could insert advertising directed to their distant markets, just as they do on cable networks today. In such a hypothetical market where inserting advertising on distant signals was permitted, actual and projected subscriber viewing information would have increased value to cable operators.

In a hypothetical, unstructured market, I would expect that cable operators would still want to negotiate the right to carry and retransmit entire distant signals (not individual programs), and would engage in such negotiations with broadcasters, likely as a part of retransmission consent negotiations. This would be much more efficient than engaging in separate negotiations with individual rightsholders for specific content over distant retransmission. I expect that actual and projected subscriber viewing information would be critical to such negotiations with broadcasters in such a hypothetical, unstructured market. Based on my experience working in the cable industry, such

14 subscriber viewing information would be the most reasonable metric for determining the relative market value of distant signal programming in such a hypothetical, unregulated market.

VI. CONCLUSION

Thank you for the opportunity to present this information in this proceeding. I hope that it will assist you in your deliberations.

15 DECLARATION OF SUE ANN R. HAJMH.TOW

I declare under penalty of perjury that the foregoing testimony is true and correct, and ofmy personal knowledge. Executed on December 20, 20M ~ZP Sue AGn R. Hamilton EXHIBIT A SUE ANN R. HAMILTON 155 Madison Street, Suite 200 Denver, Colorado 80206 (o) 303.777-2878 (m) 303.887-2079 sue@ha mlltonmedla.net

Professional Exoerlence 2007-present Hamilton Media LLC Denver, CO Founder and Principal

Advise and represent public and private media companies, including cable and broadcast networks, as well as multi-platform distributors and major consulting firms, regarding various aspects of content distribution such as business forecasting and modeling, market assessment, and overall strategy, including retransmission consent for broadcasters. Develop license fee structures, distribution strategies, and other economic and non-economic deal points; negotiate agreements incorporating such terms. Act as an arbitrator, mediator, and expert witness and litigation advisor in various matters, both as a pre-trial/pre-arbitration advisor and expert in non-disclosed representations, and as an expert witness in the case of OREGON 7V LLC v. DISH

Network LLC (D. Ore. 2014), where I testified in the United States District Court for the District of Oregon on behalf of DISH in a retransmission consent dispute with a broadcaster. A detailed list and description of client representations is available upon request.

2003-2007 Charter Communications, inc. St. Louis, MO; Greenwood Village, CO Executive Vice President, Programming, 2005-2007; Senior Vice President, Programming, 2003-2005

Strategically planned and managed distribution of content for nation's then-third largest cable operator with approximately 6,000,000 video subscribers. Maintained PRL responsibility for annual programming budget of approximately $1.5 billion. Acquired rights across multiple platforms for cable and advanced applications, including cable, broadband, HD, and VOD. Negotiated financial and operational terms of agreements with senior executives of all major programming providers, including Disney/ABC/ESPN, NBCUniversal, Twentieth Century Fox, Viacom/CBS/Showtime, Time Warner/HBO/Turner Networks, and Discovery, ARE, and AMC Networks, as well as motion picture studios. Oversaw content-related field activities of five regional divisions with systems in 40 states and large corporate staff as well as supervising work with outside law firms, accounting firms, and consultants. Actively coordinated with operations, marketing, advertising sales, engineering, and accounting groups to achieve company objectives. Represented company on quarterly earnings calls. Worked closely with largest shareholder Vulcan Ventures (run by controlling shareholder Paul Allen) and members of Board of Directors, and regularly participated in Board meetings and monthly President's Council meetings with Vulcan. Reported to President/CEO.

2003 Adelphia Communications Corporation Coudersport, PA; Englewood, CO Consultant to President/COO; acting head of programming department.

1993 - 2002 AT&T Broadband, LLC; Tele-Communications, lnc. Englewood, CO Vice President, Programming and Senior Vice President, Satellite Services, Inc., 1997-2002; Director, 1994-1997; Assistant Director, 1993-1994 Negotiated affiliation agreements and managed relationships with content providers for the then-largest cable multiple system operator. Strategized rollouts of digital cable, VOD, and interactive applications. Negotiated company's first VOD hardware and software agreements. Voluntarily left following sale to Comcast Corporation and relocation of position to Philadelphia.

1986 - 1993 Kirkland & Ellis Chicago, IL; Denver, CO Partner, 1992-1993; Associate, 1986-1992

Represented domestic and international clients in complex commercial transactions and securities matters, including mergers, acquisitions, divestitures, leveraged buyouts, financing arrangements, public offerings, private placements, loan workouts, foreclosures, bankruptcies, and receiverships.

Education

JD, 1986 Stanford Law School Stanford, CA Associate Managing Editor, Stanford Law Review, Volume 38; Member, Volume 37. Vice President, Federalist Society. Editor, Stanford Journal of International Law.

BA, 1982 Carleton College Northfield, MN Sociology and Anthropology, Magna Cum Laude. Phi Beta Kappa and Mortar Board honor society.

Northwestern University Evanston, IL Completed freshman and sophomore years before transferring.

Honors Named among the "CableFAX 100 Most Influential People in Cable" in 2004 and 2005. Designated a "Top 50 Most Powerful Women in Cable" for 2004, 2005, and 2006 by CableFAX Magazine. Recognized as 2006 "Wonder Woman" by Multichannel News and the New York chapter of Women in Cable and Telecommunications. I

I

I Before the COPYRIGHT ROYALTY JUDGES Washington, D.C.

In the Matter of

Distribution of the Docket No. 14-CRB-0010-CD (2010-13)

2010, 2011, 2012, and 2013 Cable Royalty Funds

Direct Testimony of Jan Pasquale

December 22, 2016 Testimonv of Jan Pasauale

I. Personal Background And Experience

My name is Jan Pasquale. I am an independent media consultant with over 34 years of experience in television, including with premium cable networks Home Box Office

("HBO")/Cinemax ("MAX"); with the broadcast station representative, Petry Television; with the advertising agencies, McCann-Erikson and BBD&O; and with media research supplier,

Nielsen. Prior to my work as an independent media consultant, I served as Senior Vice President . ofAudience, Marketing, and Web Analytics Research for both HBO and MAX. I worked for

HBO for over 29 years, before retiring in August 2012.

As head ofresearch at HBO, my responsibilities included providing scheduling recommendations for HBO's original and theatrical programming, taking into consideration audience estimates and appeal as well as content sensitivity; and conducting HBO/MAX monthly viewing diary studies to analyze HBO/MAX audience viewing and satisfaction. I also collaborated with Nielsen to develop studies and analyze metered data on audience viewing patterns and the correlation between viewing levels and HBO subscriber retention rates.

I received a Bachelor ofArts degree in Psychology, curn laude, from Boston College. A more detailed summary ofmy professional background is attached hereto as Exhibit 1.

II. Purpose Of Testimony

The purpose ofmy testimony is to (1) describe the business model for premium cable networks; (2) discuss how HBO, a premium cable network, relied on Nielsen ratings to evaluate its programming; (3) explain why premium cable networks are a reasonable proxy for distant broadcast stations in the context ofthis proceeding; and (4) explain why, in my experience,

Nielsen viewing data provides significant evidence ofthe market value ofprogramming and cannot be replaced by surveys. III. Description Of Business Model For Premium Cable Networks

I understand that this proceeding concerns the allocation of certain royalties among

copyright owners, or their representatives, whose programs aired on broadcast stations

retransmitted out ofmarket by cable system operators ("CSO(s)") during 2010, 2011, 2012 and

2013. The retransmission royalties were paid by CSOs for the aforementioned years and are to be allocated to the copyright owners who, for the purpose ofthis proceeding, have been grouped

into broad program categories. Also, I understand that the basis for allocation ofthese royalties

is relative market value ofthe programming airing on the stations as one encounters it in a hypothetical market where the compulsory license regulation does not exist.'remium

cable networks, such as HBO/MAX, Showtime, and The Movie Channel, are subscription-based television services that, when carried by CSOs, do not run commercial advertisements or generate advertising-based revenue. Rather, premium cable networks generate the bulk oftheir revenue &om a portion ofthe monthly subscriber fees paid to CSOs. Typically, premium cable networks are paid at wholesale rates for each subscriber obtained by the CSO.

CSOs carry premium cable networks in order to provide a wide variety ofprogramming to attract new subscribers to the CSOs'elevision services and retain their current subscribers.

CSOs pay an agreed upon licensing fee for each premium channel they carry, typically on a cents-per-subscriber basis, with the fees paid relative to the actual or prospective subscriber interest in the channel. Thus, offerings ofthe national broadcast networks (ABC, CBS, NBC,

FOX) and their affiliates, distant broadcast stations, premium cable networks (e.g., HBO/MAX,

SHO/TMC, STARZ/ENCORE, EPIX), premium cable sports networks, on demand channels, on demand origination channels offered by CSOs, and pay-per-view are all essential in maintaining a healthy business for the CSO.

'ee Distribution ofthe 2004 and 2005 Cable Royalty Funds, 75 Fed. Reg. 57063, 57065 (September 17, 2010); Distribution ofthe 1998 and 1999 Cable Royalty Funds, 69 Fed. Reg. 3606, 3608 (January 26, 2004). IV. HBO Relies On Nielsen Ratings To Evaluate Its Programming

HBO has been a Nielsen subscriber since the early 1980s, and has found Nielsen viewing data to be very valuable to both HBO's decision making process and CSOs'usiness decisions about carriage and positioning ofHBO. HBO has utilized Nielsen ratings in its programming decisions regarding both what programming to choose initially and what programming to keep on the air. HBO's many uses ofNielsen ratings include:

~ To study whether its subscribers are watching its programs. HBO has found a

correlation between subscribers not watching HBO/MAX and the decision to drop

their subscription.

~ To provide viewing metrics to CSOs and satellite distributors, the consumers and

industry press, and the financial community.

~ To determine its programs'udience size in comparison to other premium

channel programs, cable network programs, and broadcast programs.

To inform program acquisition costs for its programming, especially in the

context ofrenewals.

In general, higher ratings mean higher program acquisition costs. Ratings are instrumental in determining the value of programming for HBO. Further, investors watch the ratings levels on HBO to determine the value ofthe HBO brand. While HBO has employed other, additional methods of analysis to identify whether subscribers are watching HBO/MAX, including survey data, custom diary data, and focus groups, Nielsen ratings have been considered the key metric for accurately measuring viewership and value ofHBO/MAX content.

Understanding whether its subscribers are watching its programming is integral for HBO to manage its business, which is why HBO continues to subscribe to Nielsen at great cost. V. Premium Cable Networks Like HBO Are A Reasonable Proxy For Distantly Retransmitted Broadcast Stations

In general, the individual value ofa broadcast station program depends on anticipated or estimated actual viewing and the amount ofmoney advertisers are willing to pay to advertise during the airing ofthat program. Although premium networks do not have in-program advertisements and thus do not generate advertising revenues, the value ofa program on a premium network is similarly tied to viewing because such networks must ensure that their programming remains sufficiently attractive to their subscribers who are watching their programming. Ifa premium cable network's programming remains attractive, CSOs, in turn, will continue to offer such networks to their subscribers.

I understand that in the compulsory license world, when broadcast stations are retransmitted out ofmarket, their content may not be altered in any form. I assume that with this limitation, the retransmission does not generate any new advertising revenues for the broadcast stations in the distant market. That absence ofadvertising revenues makes the broadcast stations similar to premium cable networks, which also do not generate advertising revenues.

Notwithstanding, I would expect CSOs to find Nielsen ratings similarly valuable in deciding what broadcast stations to retransmit. Broadcast stations are supported by advertising revenues, which are driven by ratings levels oftheir programming. Like HBO and other premium cable networks, broadcast stations seek to maximize their audience by offering content that will attract and retain that audience. Broadcast stations also benefit &om higher ratings by generating higher advertising revenues. The motivation to ofFer programming that will attract a bigger audience does not change for the broadcaster regardless ofwhether the station is retransmitted only locally or out ofmarket. Based on my experience working at HBO, the bigger the audience for the television channel, the more attractive it is to a CSO as it will both keep existing and generate additional subscribers. Thus, for the same reasons as HBO, CSOs would rely on Nielsen ratings to accurately measure viewership ofdistant broadcast programming and would ultimately use such data to make business decisions about what distant broadcast signals to import in order to attract and retain cable subscribers.

Currently distant broadcast stations'rograms are unable to generate advertising revenue; however, distant broadcast stations must continue to offer programming that attracts and retains the CSO's viewing subscribers to justify the CSO's carriage ofthe signal, just like the HBO model. Accordingly, useful comparisons can be drawn between how the premium cable network

HBO's programming is analyzed to determine its ability to attract and retain its subscribers and how, in a market absent the compulsory license regulation, programming available on distant signals would be analyzed. In a scenario where there is no regulation, I would not expect to find the current limitation on alteration ofdistant signals. Instead, in an unregulated market CSOs would have the ability to, and would likely attempt to, sell advertising on distant signals targeting the distant market, utilizing viewing. Thus, viewing would be important to programming value on distant signals.

VI. Nielsen Data Provide Significant Evidence Of Market Value And Cannot Be Replaced By Surveys

HBO has utilized both Nielsen data and subscriber surveys, and over the course ofmy time at HBO, I personally designed hundreds ofsuch surveys. Surveys can be useful but are fraught with issues, including respondent error (e.g., whether respondents understand the questions being asked and/or are unable or unwilling to provide accurate responses); response bias (e.g., whether respondents are answering a question for a specific agenda, such as saving a particular show from cancellation); improper value scales (e.g., not providing a neutral choice for respondents); and non-randomization of questions (e.g., questions are asked in the same order or improperly placed within general survey themes). In addition, surveys merely scratch the surface ofan issue and tend to be easily subject to biases, such as answers being driven by particular things happening in the life ofthe respondent on that given day or reflecting what the respondent thinks the interviewer wants to hear. Focus groups are similarly prone to problems because they only address "why" a person is thinking a certain way, and therefore are not considered to be a quantitatively reliable methodological tool.

Based on my experience in utilizing different methodologies for understanding consumer behavior, one methodology still stands out as critical in gauging audience interest — Nielsen meters and other sources that physically capture viewing, as they rely less upon personal observation/opinion and more on viewing behavior as recorded by machines. In fact, during my time at HBO, we never considered discontinuing HBO's Nielsen subscription just because we were also conducting surveys and focus groups. Nielsen information has been and remains very important for HBO and its programming decisions. Nielsen data is less subjective than surveys and reflects actual behavior. %hen Nielsen data is available, it would be improper in my view not to consider it.

VII. Conclusion

Premium cable networks rely on ratings to determine what programming to offer their viewers in order to maximize their audiences and attract carriage by CSOs even though these networks are supported by subscription, and not advertising revenues. In determining what channels to carry, I would expect CSOs to similarly consider ratings ofthe content on a distant broadcast station as a metric for determining the station's potential to attract and retain subscribers—in effect, its value. As HBO has found, there is often a correlation between program viewing and value. Nielsen ratings would enable CSOs to accurately capture viewing in an unbiased manner, and accordingly, they are the best methodology for CSOs to employ in deciding what distant broadcast stations to carry. Declaration ofJan Pasauale

I declare under penalty ofperjury that the foregoing testimony is true and correct, and of my personal knowledge.

Executed on December ~ 2016.

Jan IPasguaie EXHIBIT 1 Curriculum Vitae of Jan Pas uale

Mr. Pasquale is an independent media consultant with over 34 years of experience in television. His experience includes analyzing trends in audience research viewing and audience/subscriber attitudes and behavior as well as examining new technology adoption, media research methodologies, web/social media metrics and site usability, and program valuation based on consumer and cost perspectives, original program elements, and qualitative research (e.g., focus groups and internet site usability one-on-ones). His experience also includes modeling program audience estimates. Mr. Pasquale is well versed in research tools, such as Nielsen audience metrics, survey construction, modeling, web based metrics, and other secondary vendor research. He received his Bachelor ofArts degree in Psychology, curn laude, from Boston College in 1978.

Prior to his retirement in 2012, Mr. Pasquale served as Senior Vice President ofAudience, Marketing, and Web Analytics Research for both Home Box Office ("HBO") and Cinemax ("MAX"). His career at HBO began in May of 1983 and spanned over 29 years. As head of research at HBO, Mr. Pasquale was responsible for Audience, Marketing, and Web/Social Media. He managed a staff of 19 and had oversight of a $9.4 million budget. His internal HBO clients included all areas of HBO (i.e., Scheduling, Original Programming, Consumer Marketing, Promotion, Corporate Affairs, Media Relations, Finance, Affiliate Marketing, HBO Digital Platforms, IT, HBO International, HBO Enterprises).

Mr. Pasquale's responsibilities at HBO included providing recommendations to HBO Scheduling as to how to schedule HBO's original and theatrical programming, such as time slots, number of plays, number oftimes to bring back on HBO, and on-air promotion considerations, including audience appeal and content sensitivity. Mr. Pasquale also provided audience estimates and post analysis to HBO Scheduling and Consumer Marketing to develop scheduling strategies and advertising considerations for specific content.

Mr. Pasquale conducted HBO/MAX monthly viewing diary studies from the early 1980s to the early 1990s with the purpose of capturing viewing ofHBO and MAX programs, subscriber satisfaction, and subscriber sentiment. This research was used to provide early indicators for "viewing" versus "satisfaction" versus "attitudinal" metrics. He also represented HBO in the first Time Warner divisional negotiations with Nielsen, constructing a long-term corporate contract and initiating a project to determine the correlation between HBO viewing levels and disconnect rates. Mr. Pasquale also worked with Nielsen to develop new research methodologies to enable the reporting ofHBO content on HBO Multiplex Channels, HBO On Demand, and HBO Go.

Mr. Pasquale provided HBO Original Programming with concept analysis, audience estimates, and subscriber and non-subscriber attitudes toward cast and storylines, among other things. He also provided HBO Corporate Affairs with varied analyses concerning audience, new technology, content, and competitive network trends.

In addition to his role as Senior Vice President ofAudience, Marketing, and Web Analytics Research, Mr. Pasquale held positions in the following research industry groups during his tenure at HBO: the Media Ratings Council ("MRC") General Committee, MRC Executive Television Committee, and served as Co-Chair ofthe Cable & Telecommunications Association for Marketing. He also served on several panels of industry experts with discussions focused on television and new technology issues, and as a senior member ofthe Time Warner Executive Research Council. As member ofthe Executive Research Council, he played a role in creating a state-of-the-art Time Warner Media Lab. The lab has been hailed as a valuable tool for the television industry in gaining deep insights into consumer behavior and evolving media habits through usability labs, interactive focus groups, biometric devices, and eye tracking.

Prior to joining HBO, Mr. Pasquale served as a Director ofResearch at Petry Television from 1982 — 1983. Petry Television represents broadcast station groups in attracting advertising revenue as well as consults on station program line-ups and the acquisition ofprograms for the purchase ofboth national spot and local advertising. Mr. Pasquale's responsibilities included working with station groups, such as Hubbard Broadcasting and Post-Newsweek, to evaluate local market reports for sales stories/pieces, and reviewing local market diaries at Nielsen's Florida facility for comments and viewing anomalies (e.g., right station channel, but wrong call letters).

From 1981 — 1982, Mr. Pasquale served as a Director of Research at the advertising agency McCann-Erickson where, among other things, he created national viewing estimates for all broadcast national programs to be used for upfront advertising purchases and the scatter market and evaluated national broadcast network viewing trends, program type performance, magazine buys, and cross media reach/frequency levels. Mr. Pasquale worked at the advertising agency Batten, Barton, Durstine, and Osborn from 1980 — 1981 where he similarly created national viewing estimates for all broadcast national programs to be used for upfront advertising purchases and the scatter market. He also conducted tracks of broadcast network programs to determine viewing trends and demographic appeal and assisted media planners in advertising purchases.

Prior to his work at advertising agencies, Mr. Pasquale served as a Senior Research Analyst at Nielsen from 1979 — 1980 where he worked within Nielsen Station Index — the local side ofthe business. Among other things, he responded to audit questions from the Electronic Media Ratings Council (now known as the MRC- Media Ratings Council) and conducted local station out ofmarket viewing studies to assist station clients attract national and local market advertisers. I

I

0 fO C

I

CL I nItelsen

Paul B. Lindstrom Sr. Vice President Media Analytics

TESTIMONY OF PAUL B. LINDSTROM

My name is Paul Lindstrom. I am a Senior Vice President with Nielsen. I am responsible for research design and analysis as a part of the Nielsen Media Analytics group. Nielsen is a global leader in information services for the media and entertainment industries. Nielsen serves the information and marketing needs of television and radio broadcasters, cable networks, advertisers, agencies, media planners, music companies, publishers, motion-picture studios, distributors and exhibitors, and the Internet industry.

The Nielsen name is synonymous with television ratings. Ratings are the percent of the universe of U.S. households tuned to a TV program during the average minute or quarter-hour. Nielsen ratings provide an estimate of the U.S. television audience size and are a barometer for viewing choices and preferences. Viewing information is important to broadcast networks, local and national syndicated programs, local cable system operators, multi-system cable operators (MSOs), and satellite carriers.

As more local cable ad sellers sell local advertising time on cable channels, they need an agreed "currency" in order to maximize the value of their advertising time. Nielsen ratings offer that currency. Nielsen's charter as an independent measurement service is to provide both the buyer and seller of time with unbiased estimates of viewing behavior. I. BACKGROUND AND EXPERIENCE

I have worked for Nielsen for thirty-eight years and have spent the majority of that time designing custom research with a particular focus on new television viewing sources and audience measurement of new services that might compete with television. These have included cable television, pay-TV, satellite services, over-the- air subscription television, VCRs, PCs, on-line services, the Internet, DVDs, cinema, and most recently, place-based and location-based digital networks. I am currently responsible for national custom research and custom research for local cable. I work with clients to determine the best methodologies to answer their audience research questions. In the television area, these methods can involve either the analysis of existing databases of previously collected meter data, local television diary samples, or the development of new databases through the use of new single-client sponsored data collections.

Through the years I have worked on projects as varied as the pre-launch concept tests for ESPN, The Weather Channel and DirecTV, the design of Nielsen's

Syndicated Pay Cable, VCR Usage, Syndicated Satellite and Home Technology

Reports,'he CommerceNet Study of Internet Usage, the Nielsen Cinema Audience

Report, and Nielsen On Location Media. I have been involved in all of the studies that the Motion Picture Association of America ("MPAA") and MPAA-represented

Program Suppliers ("Program Suppliers") have directed Nielsen to conduct for proceedings before the Copyright Royalty Tribunal, the Copyright Arbitration Royalty

'ielsen's Syndicated Satellite Report was the first study to utilize diary data to examine satellite viewing. The Home Technology Report is a trending study which has now produced estimates of the growth of new technologies over the last 20 years. Elllc]lHcEl

Panel, and the Copyright Royalty Judges since 1980. Also, I have testified before those bodies.

II. PURPOSE OF TESTIMONY

The purpose of my testimony is to provide an overview of Nielsen's sampling process and method for generating television ratings, and to explain the type of

Nielsen data on which Program Suppliers are relying in this proceeding.

III. SAMPLING AND TV RATINGS

The Nielsen rating you may see reported in newspapers or magazines is simply a statistical estimate of the number of homes tuned to a program. For example, a rating of 5 for a network television program means that 5% of the estimated U.S. television homes at the time of measurement are estimated to have been tuned in to that program at any point in time.

Above, I describe a rating as a "statistical estimate." Ratings are based, not on a count of all television households, but on the count within a sample of television households selected from all television households. The sample results are then projected to national totals. We also sometimes use the phrase "share" to quantify audience viewing levels. "Share" is an estimate of the households tuned to a particular channel or program. In other words, a rating measures what percentage of the universe of television households tuned in to a program, while a share measures what percentage of the number of television households in use are tuned in to any particular channel or program. IV. NIELSEN DATA USED IN THIS PROCEEDING

During 2010-2013, Nielsen utilized meters as its basic data collection instrument. A meter is an electronic device attached to a television set in a particular household that detects the channel to which the television is being tuned. The data from these meters are then converted into household ratings. Household meter data was collected year-round in Nielsen's metered markets during 2010-2013.

Program Suppliers contacted Nielsen and sought to obtain custom analyses of national household metered viewing data to cable distant households for the 2010-

2013 royalty years. Accordingly, Nielsen designed for Program Suppliers custom analyses of national household metered viewing data for each of the 2010-2013 years.

I understand that Dr. Jeffrey Gray makes use of Nielsen's custom analyses in his economic analysis.

Our team of professionals designed custom analyses of Nielsen national household metered viewing data for 2010-2013 which estimate actual distant viewing by cable households. The methodology for our custom analyses, in brief, is as follows:

1) Dr. Gray supplied Nielsen with a list of his sample stations for the 2010-

2013 cable royalty years. I understand that Dr. Gray relied on data from Cable Data

Corporation ("CDC") in order to select the stations in his samples for each year.

2) Based on county analyses it performed, CDC provided Nielsen with the identity of the counties considered local to each station in Dr. Gray's samples.

3) For the 2010-2013 cable custom analyses, Nielsen eliminated all non-cable viewing of programs for Dr. Gray's sample stations. Further, it separated all viewing nielsen to each station that occurred within the station's local area (as determined by CDC's county analyses) from viewing that occurred outside the station's local area. Nielsen then provided a report to Dr. Gray separately identifying both local viewing and distant viewing among cable households for the stations in Dr. Gray's 2010-2013 samples. This was reported in the form of quarter hours of viewing by households.

V. "ZERO VIEWING" INSTANCES

One concern raised in past Phase II proceedings, and which may be raised in the allocation phase of this proceeding, is the so-called "zero viewing" instances that appear in Nielsen's custom analysis of national household metered viewing data. The appearance of these "zero viewing" instances is consistent with what I would expect to find in a custom analysis of viewing to distant signals by cable subscribers, for at least two reasons. First, it is important to recognize that Nielsen's custom analyses excluded all distant viewing to programs that are not compensable in this proceeding.

This included distant viewing to ABC, CBS, and NBC network programs and programs that were not simultaneously broadcast on both WGN's local feed and

WGN's satellite feed (known as WGN-A). Where noncompensable programs aired,

Nielsen's custom analyses properly reported a zero viewing value. Second, the amount of actual viewing minutes to certain distant signals is very small. Where the viewing minutes to particular distant signal programs were so small as to be statistically insignificant, Nielsen's custom analysis would assign a zero viewing value.

Thank you for the opportunity to testify in this proceeding. pie/SeP

DECLARATION OF PAUL B. LINDSTROM

I declare under penalty of perjury that the foregoing testimony is true and correct, and ofmy personal knowledge.

Executed on December 2016

mdstrom I

I I

I I

C

'CQl 6) 2 C 0 U I Before the COPYRIGHT ROYALTY JUDGES Washington, D.C. In re CONSOLIDATED PROCEEDING DISTRIBUTION OF CABLE NO. 14-CRB-0010-CD ROYALTY FUNDS (2010-13)

TESTIMONY OF JEFFREY S. GRAY PH.D.

December 22, 2016 Table of Contents

I. Qualifications .

II. Background And Overview of Royalty Allocation Process . III. Economic Value ofProgramming: Relative Market Value Depends on Viewership.. A. Application ofthe Relative Market Value Standard. B. Measuring Relative Market Value: Volume and Viewership

1. Volume. 2. Viewership C. Data Relied Upon to Measure Relative Market Value of Programming ... 1. Nielsen Data. 12 2. Gracenote Data. 13 3. CDC Data... 14 4. CRTC Program Logs 14 D. Economic Analysis: Estimating and Imputing Distant Viewing 15 E. Relative Market Value of All Programming 15

1. Nmnber of Compensable Programs, Program Retransmissions, and Volume Statistics 15 2. Program Viewing Statistics. IV. Conclusion: Royalty Share Allocations 20 Appendix A: Curriculum Vitae ...... ,...... 21 Appendix B: Stations Sampled for Cable Analysis.. 24 Appendix C: Cable Regression Models - Specifications A Results 27 I. QUALIFICATIONS

1. I, Jeffrey Gray, am an economist and President of Analytics Research Group, LLC

("ARG"). ARG provides expert analysis concerning economic, statistical and data issues.

2. I received training in economics and statistics at the University of Pennsylvania, where I earned a Ph.D. in economics. In 1991, I was appointed to a one-year position on the staFofthe President's Council ofEconomic Advisers, where I concentrated on the economic impact of government policies and regulation. Prom 1993 to 1997, I served on the faculty ofthe University of Illinois, where I taught graduate and undergraduate courses covering survey techniques, demand analysis, labor economics, and statistics.

My research has been published in some of the top peer-reviewed journals in the economics profession, including The American Economic Review. I have received grants to pursue my research from the U.S. Department ofLabor, the U.S. Department of

Agriculture, and the Research Board ofthe University of Illinois. I have presented my research Qndings before a variety of seminars at universities, meetings ofprofessional societies and conferences on specialized topics in the United States and abroad.

Throughout my professional career, I have been asked to serve as a referee for leading economics journals, such as The American Economic Review and the Review of

Economics and Statistics, concerning the appropriate application of economics and statistics.

3. I have served as a consultant for companies, law firms, and government agencies on a variety of economic and statistical issues related to antitrust, copyright and patent infringement, and complex commercial disputes. My consulting work has included analyzing economic markets as well as valuing copyrighted material and assessing efficient price and advertising levels. I have been engaged by cable system operators

("CSOs") to analyze the content and viewership of certain channels and by music performance rights owners to determine the economic value ofthe right to perform copyrighted music. I have provided expert testimony before the Copyright Royalty

Judges ("Judges"), as well as in state, federal and international courts, and have presented my research methodology and analytical findings before the Securities and Exchange

Commission, the Texas Commissioner of Insurance, and the New York and

Massachusetts State Offices of the Attorney General.

4. My curriculum vitae, which includes a list of my publications in the last ten years, and a list of cases in which I have testified in the last four years, is attached as Appendix

A. This report is based upon information currently available to me; I reserve the right to supplement this report should additional information be made available.

II. BACKGROUND AND OVKRVIKW OF ROYALTY ALLOCATION PROCESS

5. I understand that the purpose of this proceeding is to allocate the 2010, 2011,

2012, and 2013 cable royalty funds ("2010-2013 Cable Royalties") paid by CSOs under statutory (compulsory) licenses established by Section 111 of the Copyright Act

("Section 111").

6. Compulsory licenses allow CSOs to retransmit broadcast television signals out-of- market (i.e., on a distant basis) without the need to negotiate individual private license agreements with the multitude of copyright owners whose programs air on those signals.

Economists refer to the mme and expense associated with negotiating such private license agreements as transaction costs. Section 111 effectively eliminates the transaction costs that would occur in a market without the compulsory license and sets the rates for the compulsory license fees paid by the CSOs. These statutorily-set fees are subject to periodic adjustments.'SOs pay the licensing fees based primarily on the number and type of distant stations each CSO chooses to carry. After collecting the royalty payments, the Copyright Office distributes them among eligible copyright owners of compensable programs aired on the distant signals (or their representatives),'ither by agreement among the claimants, or pursuant to the determination in a caMe royalty distribution proceeding held before the Judges.

7. The CRT decided in 1980 to conduct its distribution proceeding in two phases. In

Phase I, the Judges allocated royalties among broad categories of self-organized claimants." In Phase II, royalties have been divided among individual claimants or their

'he periodic adjustments to the royalty fee rates were initially made by the Copyright Royalty Tribunal ("CRT"). Following abolition ofthe CRT, the adjustments were overseen by Copyright Arbitration Royalty Panels ("CARPs") appointed by the Librarian of Congress. The CARPs were subsequently replaced by the current system ofJudges.

The compulsory license fee for CSOs was based upon the number of "distant signal equivalents" ("DSEs") that a cable system imported, valuing a distant independent station as one and a network- affiliated station or educational station as 1/4. In general, the number of DSEs carried by a CSO is multiplied by a DSE rate to establish the percentage oftheir gross revenues charged for importing distant television signals.

'ligible compensable programs are non-network broadcast programs aired on simultaneously retransmitted distant signals during the 2010-2013 cable royalty years.

Historically, for cable Phase I Proceedings there have been eight broad categories ofprogramming: (1) Program Suppliers; (2) Joint Sports Claimants ("JSC"); (3) Commercial Television Claimants ("Commercial Television'"'); (4) Public Television Claimants ("Public Television"); (5) Devotional representatives within each ofthe broad program categories. However, as explained by the Judges in this proceeding, nothing in the Copyright Act or the Judges'egulations requires that the proceeding be split into multiple phases. Indeed, the de6nitions ofthe broad categories in the historical Phase I Proceedings, and in the current proceeding, are based on an agreement between the participating claimants and not, to my understanding, based on any market-accepted de6nition ofprogramming.

8. While the broad category de6nitions were agreed to by the parties, each allocation parlicipant's programming claims are simply an aggregation ofthe distantly retransmitted individual programs that fall within each agreed claimant category. Some ofthe agreed upon claimant category definitions may appear counter-intuitive to the market. For example, the Program Suppliers category is comprised ofproducers and/or distributors of syndicated series, movies, specials, and non-team sports, excluding devotional programs.

Thus, certain sports programming that commonly airs on distant broadcast signals such as

NASCAR racing, professional bowling, golf, and the Olympics fall into the Program

Claimants ("Devotionals"); (6) Canadian Claimants Group ("Canadian Claimants"); (7) Music Claimants; and (8) National Public Radio ("NPR"). The Judges adopted these eight categories ofprogramming for this proceeding as well. See Notice of Participant Groups, Commencement ofVoluntary Negotiation Period (Allocation), and Scheduling at Exhibit A (November 25, 2015) ("Notice").

Chapter III oftitle 37 ofthe Code of Federal Regulations.

See Notice at 3; see also Order Regarding Discovery at 4, n.7 (July 21, 2016).

See Notice at Exhibit A (describing the mutually exclusive Agreed Categories as "non-exhaustive descriptions of the types ofprograms or other creative works that fall within each ofthe agreed categories of claimants (Agreed Categories) to which categories the Judges may approve an allocation of cable retransmission royalties."). Suppliers category and not the Joint Sports Claimants ("JSC") category, which consists of only live telecasts of professional and college team sports.

9. In addition to sports programming falling into more than one claimant category, certain retransmitted programming that is broadcasted on public television stations,8 as well as on Canadian television stations, could belong to the Program Suppliers category.

III. ECONOMIC VALUE OF PROGRAMMING: RELATIVE MARKET VALUE DEPENDS

ON VIEWERSHIP

10. The criterion for dividing the annual royalty pools among claimants is the "relative market value" of the copyrighted programs.

A. Application ofthe Relative Market Value Standard

11. Relative market value corresponds to the price at which the right to retransmit a

program carried on a distant broadcast signal would change hands between a willing buyer (a CSO) and a willing seller (a copyright owner), neither being under any compulsion to buy or to sell and both having reasonable knowledge ofrelevant facts.'he

"willing buyer" in this hypothetical negotiation is the CSO because it chooses which

"See 66 Fed. Reg. 55653, 55655 n.3 {October 28, 1996) {"An example of a program which would not be in the Public Broadcasting Service category, because it fell within another category, would be the movie 'Platoon'hat was broadcast by a PBS station. That program would properly fall within the Program Suppliers category.").

'ee 75 Fed. Reg. 57063, 57065 {September 17, 2010).

" This definition is consistent with the definition offair market value written by the U.S. Supreme Court: "The fair market value is the price at which the property would change hands between a willing buyer and a willing seller, neither being under any compulsion to buy or to sell and both having reasonable knowledge of relevant facts." United States v. Cartwright, 411 U. S. 546, 93 S. Ct. 1713, 1716-17 {1973). 13. It is axiomatic that consumers subscribe to a CSO to watch the programming made available via their subscriptions. The more programming a subscriber watches, the happier the subscriber is, and the more likely she will continue to subscribe, all else equal.'herefore, a measure of the happiness, or "utility," an individual subscriber gets from a specific program is the number of minutes that subscriber spent viewing the program offered to him or her by the CSO. A measure of the utility all subscribers get, in total, from a specific program is the total level of subscriber viewing ofthe program.

Thus, even though CSOs are the buyers ofthe programming bundles, a reasonable

measure ofthe relative market value of a retransmitted program is the relative level of

subscriber viewing of that program. The higher the subscriber viewing, all else equal, the

greater the subscribers'tility and the greater the CSOs'etention of subscribers.

14. Thus, the relative market value of an individual program ultimately depends upon the consumption of the programming as measured by its level of viewing by subscribers.

As explained by actual copyright owners, audience size — as measured by viewership — is

central when making licensing deals with broadcast stations and cable networks in the

world outside the compulsory licensing scheme." Moreover, in an attempt to attract and

retain customers, CSOs want to carry stations with high viewership programming such as

off-network syndicated television series that originally attracted a loyal following in their

'conomists refer to this axiom underlying consumer preferences as the axiom of "nonsatiation." See, Economics and consumer behavior, Angus Deaton and John Muellbauer, 1980, p.28.

" See MPAA Written Direct Statement Vol. II, Prior Designated Testimony, at Tab A, Docket No. 2007- 3 CRB CD 2004-2005, Written Direct Testimony ofAlex Paen, pp. 11-12 (filed June 1, 2009). network runs and continue to do so in syndication.'hey also desire to carry stations with high viewership programming, such as fIrst-run syndicated programs that they believe will garner satisfactory audience levels."

15. Since this proceeding involves allocating Gxed royalty pools as part ofa compulsory licensing scheme, it is appropriate to consider pertinent information concerning the relative economic value ofprogramming, namely, program consumption as measured by actual program viewing. Purposefully ignoring estimates of actual viewing or ratings could lead to individual copyright owners, as well as categories of copyright owner representatives, ofvaluable programming receiving disproportionately small royalty awards compared to the relative market value oftheir copyrighted programming.

8. Measuring Relative Market Value: Volume and VI'eldership 16. Subscriber preferences are revealed by which distant stations and programs they choose to watch. Below, I discuss two measures ofrelative economic value of programming: prograrroning volume and programming viewership.

1. Volume

17. Holding costs constant, CSOs.will choose to carry distant signals with programming that will attract and retain as many subscribers as possible. In theory, the rational (i.e., economic-optimizing ) CSO will choose to carry distant signals with the

'ee Written Direct Testimony ofAlex Paen, p. 12.

'ee id atpp. 5-6,9-10. most preferred programming airing at the most preferred times. The total volume of minutes ofprogramming retransmitted by CSOs effectively represents the volume of programming purchased by the CSOs, albeit in a regulated setting. Therefore, total program volume represents rational CSOs'hoices and provides a measure of the relative economic value ofthe programming to the CSOs.

18. While total program volume provides useful information concerning the relative value ofprogramming to CSOs, that measure alone is not suQicient. In general, the value ofprograms to the CSO and their subscribers may differ depending on the time slot during which the programs are aired. A 30-minute program aired during primetime might be more valuable to a CSO and their subscribers than an hour-long program shown in the middle ofthe night. Moreover, programs of identical duration shown at the same time of day may have very different values to CSOs and their subscribers. That is, programming volume alone does not convey a complete picture ofthe relative value of the programs. The reason volume alone is an insufhcient measure ofvalue is because it ignores audience size.

2. Viewership

19. Audience size, which is determined through program viewership, is a primary interest ofprogrammers and therefore the most direct measure ofa program's relative value.'rom the CSO's perspective, the more a program attracts subscribers to watch and to keep watching, the more valuable the program is to the CSO's net-revenue

'edia Promammina: Strategies and Practices„8 ed., S.T. Eastman and DA. Ferguson, 2009, p. 40. maximizing goal of retaining and growing subscriber count. From the subscriber's perspective, relatively low viewership of a given program signifies relatively low valuation of that program. Absent the bundling of programs, economic theory implies that a program with no viewership will most likely not continue to be carried. 17

20. Program viewership as a measure of relative market value is consistent with the economic theory of revealed preference.'n the present context, it means a CSO's willingness to pay for a particular type ofprogram, or the station that airs these programs, is a function of the programming's contribution to the cable system's ability to attract and retain subscribers and thereby maximize net revenue.

21. The two most recent Phase I rulings relied heavily on surveys of CSOs attempting to quantify the relative value of Phase I program categories." However, even cable networks routinely analyze viewing levels because they understand that this measure is the best available indicator of what attracts and retains subscribers. Moreover, surveys are fundamentally flawed ifthe programming categories are not correctly understood by survey respondents.

22. Consistent with the economic arguments described above, I analyze programming voluine and viewing to determine relative market value of six broad claimant categories of individual programming: (1) Program Suppliers; (2) JSC; (3) Commercial Television;

"See geneva/ly, Economics and consumer behavior, Angus Deaton and John Muellbauer, 1980.

"IrI., p. 51-53.

'ee 75 Fed. Reg. at 57065; 69 Fed. Reg. 3606, 3609 (January 26, 2004).

10 (4) Public Television; (5) Devotionals; and, (6) Canadian Claimants. I apply a two-step approach:

~ First, I calculate measures of the relative volume of programming by claimant

category. This provides good, but imperfect, indicators ofthe relative value of

the sets of programs at issue in this proceeding.

~ Second, I calculate the relative viewership ofprogramming by claimant

category. This is done on a program-by-program basis, then aggregated up to

the agreed upon broad claimant categories. As described above, given the

available data, this is the most direct measure of relative value: if costs are

deemed constant, higher subscriber viewership suggests higher relative market

value of the programming.

C. Data Relied Upon to Measure Relative Market Value ofProgramming 23. I rely upon Nielsen viewing data, Gracenote, Inc. ("Gracenote") 'rogramming data, and Canadian Radio-television and Telecommunications Commission ("CRTC") program logs, to study the volume and viewing information of compensable programs from 2010 through 2013. I also rely upon Cable Data Corporation ("CDC") carriage data that includes information on the number of CSO subscribers of each distantly retransmitted signal analyzed. Due to cost considerations in obtaining the Nielsen and

Gracenote data for all stations distantly retransmitted by CSOs in every royalty year, I

My testimony does not address an allocation to either Music Claimants or NPR.

'ribune Media Services merged with Gracenote, Inc. on June 12, 2014, as a division of Tribune Media Company. The division now operates under the Gracenote name.

11 3. CDC Data

2S. The CDC data are information catalogued by CDC from statements of accounts

("SOAs") that CSOs filed with the Licensing Division ofthe Copyright Office semi- annually. These data include information regarding the distant signals carried, the number of subscribers receiving each distant signal, and the estimate of fees generated by each signal during years covered by this proceeding. Based on the CDC data, over

1,000 stations were distantly retransmitted by CSOs each year from 2010 to 2013.

4. CRTC Program Logs

29. The CRTC requires stations broadcasting in Canada to submit monthly program logs ("CRTC Program Logs"). These CRTC Program Logs include information such as station call signs, the program title and actual start time and end time of each program transmitted by each Canadian station, and an indicator for the country of origin of each program. I understand that programming airing on Canadian stations that originated outside the United States and which was secondarily transmitted into the United States is compensable in the Canadian Claimants category. Also, syndicated programming and movies that aired on Canadian stations which originated from the United States are compensable as Program Suppliers programs. Similarly, live telecasts of Major League

Baseball, National Hockey League, and U.S. college team sports airing on Canadian

" See, Written Direct Testimony of Jonda Martm.

'ee CRTC website for more information, htto://www.crtc.ac.ca/. stations are compensable in the JSC category. ' used these CRTC Program Logs to determine the country of origin of the programming that aired on the Canadian stations and assigned such programs to either the Program Suppliers, Canadian Claimants, or JSC categories as appropriate.

D. Economic Analysis: Estimating andImputing Distant Viewing

30. To determine the relative market value of all compensable programs that aired on stations that were distantly retransmitted by CSOs, I calculate the relative distant viewing ofthose programs. I provide a reasonable estimate of relative distant viewing levels by relying upon the data sources described in the previous section. As I explain later in this testimony, I establish a mathematical relationship between local and distant viewing levels using various program characteristics during those years. I then extrapolate that mathematical relationship using a regression analysis to estimate distant viewing for all compensable programs each year for the 2010-2013 Cable Royalty years.

E. Relative Market Value ofAll Programming 31. Considering the various datasets described above, my analysis demonstrates the breadth of programming and the extent to which it is retransmitted in distant markets by

CSOs.

2. Number ofCompensable Programs, Program Retransmissions, and Volume Statistics

32. Table 1 below present summary statistics comparing compensable programs in the six-broad claimant categories in the (1) number of retransmissions of compensable

"See Notice at Exhibit A.

15 programs; and (2) the volume (in minutes) ofprogramming that aired on stations

distantly retransmitted by CSOs. From 2010 to 2013, between 12.3 million and 14.6

million compensable programs aired on the stations retransmitted by CSOs. This

programming represented between approximately 501 million and 613 million

retransmitted minutes of compensable programming.

Table 1: Levels and Shares ofRetransmissions and Volume by Royalty Year Share ofAll Minutes of Share ofAll Year Claimant Category Retransmissions Retransmissions Retransmissions Volume Canadian Claimants 57,225 0.46% 2,122,047 0.42% Commercial Television 1,462,553 11.84% 65,052,837 12.96% Devotionals 875,464 7.09% 36,297,516 7.23% 2010 Program Suppliers 6,964,783 56.38% 273,216,974 54.44% Public Television 2,969,458 24.04% 121,429,914 24 19% JSC 22,917 0.19% 3,766,093 0.75% Total 12,352,400 100% 501,885,381 100%

Canadian Claimants 205,451 1.41% 10,062,905 1.73% Commercial Television 1,594,802 10.94% 72,537,208 12.50% Devotionals 1,540,798 10.57% 53,386,932 9.20% 2011 Program Suppliers 7,984,259 54.78% 311,730,950 53.72% Public Television 3,221,460 22.10% 128,137,417 22.08% JSC 28,227 0 19% 4,407,898 0.76% Total 14,574,997 100% 580,263,310 100%

Canadian Clannants 190,680 1.44% 7,186,266 1.27% Commercial Television 2,006,702 15.15% 110,254,196 19.42% Devotionals 604,350 4.56% 23,086,212 4.07% 2012 Program Suppliers 5,108,608 38.57% 204,883,036 36.08% Public Television 5,316,379 40.14% 219,327,673 38.62% JSC 18,551 0.14% 3,106,191 0.55% Total 13,245,270 100% 567,843,574 100%

Canadian Claimants 113,649 0.80% 4,520,712 0.74% Commercial Television 2,018,588 14.23% 84,254,857 13.74% Devotionals 905,633 6.39% 35,912,773 5 86% 2013 Program Suppliers 7,257,384 51.18% 321,859,971 52.50% Public Television 3,819,006 26.93% 158,274,057 25.82% JSC 66,768 0.47% 8,229,864 1.34% Total 14,181,028 100% 613,052,234 100%

16 33. Across the 2010-2013 Cable Royalty years, with the exception of 2012, each claimant category's share of the total number retransmissions and the volume of retransmissions is relatively steady. In 2012, there were signi6cantly more public television stations retransmissions in the sample as well as more sub-channels. Many of these sub-channels were in turn retransmitted by CSOs, contributing to an increase in

Public Television's share of both retransmissions and volume.

34. As described earlier, the relative minutes, or volume, ofprogramming retransmitted provides an imperfect metric ofthe relative value ofthe two sets of programs. The volume measure does not take into account what time of day the retransmission took place, the number of cable subscribers who had access to the distantly retransmitted broadcast, or the number ofhouseholds who watched the show.

Thus, the share of viewing minutes provides a superior measure of relative value.

2. Program Viewing Statistics 35. Due to the low frequency of distant viewing and the size ofthe sample Nielsen uses to measure total U.S. household viewing, there are many instances ofno recorded distant viewing of compensable retransmitted programs in the Nielsen Household Meter

Data. However, it is possible to obtain reliable estimates of distant viewing for all retransmitted programs by also relying on Nielsen measures ofhousehold viewing in each retransmitted station's local market.

" Sub-channels are where the main channel broadcasted several separate signals on the same channel.

17 C-5 presents the 95% confidence intervals associated with each viewership share calculation. 'hese viewership shares correspond to reasonable cable royalty shares.

Table 2: Distant Viewing Levels amd Shares by Royalty Year

Yeur Claimant Category Distunt Viewing Share ofDistunt Viewing Canadian Claimants 18,270 1.59% Commercial Television 181,739 15.81% Devotionals 6,788 0.59% 2010 Program Suppliers 598,824 52.10% Public Television 313,137 27.24% JSC 30,696 2 67% Total 1,149,454 100%

Canadian Claimants 34,203 3.40% Commercial Television 122,130 12.15% Devotionals 10,068 1.00% 2011 Program Suppliers 513,746 51.13% Public Television 292,267 29.09% JSC 32,390 3.22% Total 1,004,804 100%

Canadian Claimants 31,949 3.09% Commercial Television 163,892 15.87% Devotionals 5,054 0.49% 2012 Program Suppliers 375,983 36.40% Public Television 430,093 41.64% JSC 26,008 2.52% Total 1,032,979 100%

Canadian Claimants 32.271 4.42% Commercial Television 76,110 10.42% Devotionals 4,764 0.65% 2013 Program Suppliers 329,715 45.12% Public Television 244,627 33.48% JSC 43,240 5.92% Total 730,727 100%

" The confidence intervals are calculated applying the bootstrap methodology. See E&on, B.; Tibshirani, R. (1986). "Bootstrap Methods for Standard Errors, Confidence Intervals, and Other Measures of Statistical Accuracy." Statistical Science 1(1), 54-77.

19 39. As reported in the final column in Table 2, Program Suppliers'able viewership shares are 52.10% in 2010, 51.13% in 2011, 36.40% in 2012, and 45.12% in 2013.

Program Suppliers'ecrease in distant viewing shares in 2012 and 2013 correspond with an increase in Public Television distant viewing in those years.

CONCLUSION: ROYALTY SHARK ALLOCATIONS

40. Based upon the economic theory of consmner behavior, my analysis indicated that relative program viewership provides a reasonable and reliable measure of the relative economic value of distantly retransmitted programing. All else equal, the higher the viewing of distantly retransmitted programming, the higher the value ofthat programming to consumers and logically to CSOs. Therefore, following this theory, to determine what I believe to be reasonable and reliable relative market values of the 2010-

2013 claimant categories, I analyzed data concerning program volume and program viewing patterns of a randomly selected set of stations each year from 2010 to 2013.

Table 1 above reports each claimant category's level and share of volume of retransmissions from 2010-2013. Table 2 above reports each claimant category's distant viewing share and therefore its share of the total 2010-2013 Cable Royalties for each royalty year.

20 APPENDIX A: CURRICULUM VITAE

Jeffrey S. Gray, Ph.D. President Analytics Research Group LLC 912 F Street NW Washington, DC 20004

Education S, Background Summary Ph.D., Economics, University of Pennsylvania B.A., Economics (with honors) University of California Santa Cruz

Dr. Gray has over 20 years of experience in economic and statistical consulting, survey design, sampling methodologies, and complex database analytics. He is an authority on economic markets, statistical methods, and economic damages. His research has been published in some of the top peer-reviewed journals in the economics profession including The American Economic Review and the Journal of Human Resources. Dr. Gray has presented his findings before a variety of seminars at universities, meetings of professional societies and conferences on specialized topics in the United States and abroad. Dr. Gray has received recognition and financial support to pursue his research from the U.S. Department of Labor, the U.S. Department of Agriculture, and the Research Board of the University of Illinois. Throughout his career Dr. Gray has served as referee for professional journals assessing the appropriate application of economics and statistics.

Dr. Gray has conducted studies for corporations, government agencies and law firms on a variety of economic and statistical issues. Dr. Gray has served as a testifying expert on behalf of both plaintiffs and defendants addressing class certification, liability and/or damages issues. He has provided written or oral expert testimony in state, federal, and international courts and presented analytical findings before the Securities and Exchange Commission, the Texas Commissioner of Insurance, the Government of Singapore, and the New York and Massachusetts State Offices of Attorney General.

In addition to leading the economic and statistical consulting practices at Huron Consulting Group and Deloitte Financial Advisory Services LLP, Dr. Gray has served on the staff of the President's Council of Economic Advisers and on the faculty of the University of Illinois where he taught graduate and undergraduate courses covering consumer demand analysis, labor economics, and statistics. He earned a Ph.D. in economics from the University of Pennsylvania.

21 Professional Experience

~ Analytics Research Group LLC, Washington, DC o President, Washington DC, 2013 — Present

~ Deloitte Financial Advisory Services LLP, Washington, DC o Principal and Leader of Economics Practice, Washington DC, 2010- 2013

~ Huron Consulting Group, Boston, MA o Managing Director & National Leader, Economics, 2006 — 2009 ~ Deloitte Financial Advisory Services LLP/Deloitte & Touche LLP: FAS, Boston, MA o Principal-ln-Charge, Boston, MA, 2004 — 2006 o Economist & Principal, Economic Consulting, 2002 — 2006

~ Arthur Andersen LLP, Boston, MA & Chicago, IL o Director, Economic Consulting, 2001 — 2002 o Economist, 1999 — 2002

~ Welch Consulting, College Station, TX o Senior Economist, 1996 — 1999

~ University of Illinois, Urbana, IL o Assistant Professor, 1993 — 1997

~ President's Council of Economic Advisors, Washington, DC o Staff Economist, 1991 — 1992

~ University of Pennsylvania, Philadelphia, PA o Research, Teaching Assistant and Instructor, 1989 — 1991

Professional Affiliations ~ American Economic Association ~ American Finance Association ~ American Statistical Association

Referee Responsibilities ~ American Economic Review, Demography, Economic Inquiry, International Economic Review, Eastern Economic Journal, Journal of Human Resources, Journal of Labor Economics, Review of Economics and Statistics, Social Science Quarterly, Sociological Forum.

22 Publications and Presentations (Prior 10 Years) ~ Jeffrey S. Gray. Class Action Litigation: Working with Economics and Statistics Experts, invited presentation, Washington, DC, September 2013.

~ Jeffrey S. Gray. Patent Infringement Damages: Approaches and Trends, Moderated Panel on Intellectual Property in the Life Sciences, May 2010.

~ Jeffrey S. Gray. Institutional Investors: Protecting Your Assets- Prudent Investing, Moderated Panel on Fiduciary Litigation issues, February 2009.

~ Jeffrey S. Gray. Subprime Fallout: Prudent Investing & Economic Damages. Professional Liability Underwriting Society Conference, Boston, MA. October 2008.

~ Jeffrey S. Gray with Carl Tannenbaum and Laurence Kotlikoff, Was the Credit Crisis Foreseeable? Moderated Panel, April 2008.

~ Eugene Canjels, Jeffrey S. Gray and Michel J. Vanderhart. Does Everyone Overstate the Number of Hours They Work? An Examination ofSurvey Response Bias Among Salaried and Hourly Workers, White Paper, April 2005.

Expert Testimony & Affidavits (Prior 4 Years) ~ In the Matter of Distribution of the 2004, 2005, 2006, 2007, 2008 and 2009 Cable Royalty Funds, before the Copyright Royalty Judges, Washington D.C., Doc No. 2012-6 CRB CD 2004-2009 (Phase II), and in the Matter of Distribution of the 1999, 2000, 2001, 2002, 2003, 2004, 2005, 2006, 2007, 2008 and 2009 Satellite Royalty Funds, before the Copyright Royalty Judges, Washington D.C., Doc No. 2012-7 CRB CD 1999-2009 (Phase II), expert affidavits and trial testimony (2014-2015).

~ In the Matter of Distribution of the 2000, 2001, 2002, and 2003 Cable Royalty Funds, before the Copyright Royalty Judges, Washington D.C., Doc No. 2008-2 CRB CD 2000- 2003 (Phase ll), expert affidavits and trial testimony (2013).

~ Michael Brown, Brian Singer et al v. Canadian Imperia! Bank of Commerce, proceeding under the Class Proceedings Act, 1992, Court File No. 08-CV-00365119CP, Ontario Superior Court of Justice, Canada; expert affidavit and oral cross-examination (2011).

~ Wayne B. Gould et al v. Western Coal Corporation, et al., proceeding under the Class Proceedings Act, 1992, Court File No. CV-09-391701-00CP, Ontario Superior Court of Justice, Canada; two expert affidavits (2011).

23 APPENDIX B: STATIONS SAMPLED FOR CABLE ANALYSIS

2010 2011 2012 2013 Distant Distant Distant Distant Station Subscribers Station Subscribers Station Subscribers Station Subscribers WGN-DT 41,361,722 WGN-DT 43,106,794 WGN-DT 42,459,172 WGN-DT 42,522,609 WLIW-DT 743,494 CBUT 966,581 CBUT 868,203 CBUT 893,666 WNET-DT 661,353 WLIW-DT 680,208 WLIW-DT 613,759 WLIW-DT 644,340 WPIX-DT 605,742 WNET-DT 643,737 WPIX-DT 590,292 WPIX-DT 571,383 CBUT-DT 519,880 WPIX-DT 611,976 WBRE-DT 589,716 WNET-DT 516,323 WUAB-DT 502,043 WNBC-DT 443,643 WNET-DT 515,030 CKSH 367,635 CBUT 495,028 WLRN-DT 440,554 WNBC-DT 432,652 KZSW-LP 356,147 WNBC-DT 457,453 CKSH 355,378 WTWO-DT 397,445 WPHL-DT 345,778 WNYJ-DT 443,386 KCET-DT 345,004 CKSH 376,637 WWOR-DT 304,576 KTNC-DT 362,773 WPHL-DT 341,002 WPHL-DT 354,663 WNBC-DT 296,559 WPHL-DT 361,462 WWOR-DT 330,783 KZSW-LP 324,535 WZDC-LP 295,427 KCET-DT 355,090 WRNN-DT 296,865 WWOR-DT 314„518 WSBE-DT 277,067 WWOR-DT 330,262 WSEE-DT 285,877 WTXF-DT 308,018 KCET-DT 265,628 WRNN-DT 309,964 KZSW-LP 279,651 WRNN-DT 293,264 WNJN-DT 264,498 WSEE-DT 302,564 CBMT 271,354 CBMT 274,453 CBMT 260,888 WSBE-DT 242,579 WSBE-DT 251,487 WSBE-DT 274,148 WYIN-DT 243,832 WSBK-DT 240,355 WUFT-DT 237,247 WSEE-DT 274,126 WLIW-DT2 239,948 CKSH-DT 236,355 WSBK-DT 220,690 KCET-DT 272,285 WLIW-DT3 239,948 KZSW-LP 222,205 WGBX-DT 214,346 WGBX-DT 233,897 WLIW-DT4 239,763 KCTS-DT 219,278 CFTO 213,637 WPRU-LP 232,171 ~-DT 238,553 WTXF-DT 217,279 KTNC-DT 211,995 WSJX-LP 232,171 WLVT-DT 223,568 WNJT-DT 207,593 WTXF-DT 208,845 WUFT-DT 222„840 WTTW-DT 207,447 KCTS-DT 202,059 WNJN-DT 226,262 WUFT-DT 220,375 CELT 201,175 WYIN-DT 201,027 CFTO 225,240 WLVT-DT2 215,031 WGBX-DT 196,230 WTTW-DT 197,967 WYIN-DT 220,030 WLVT-DT3 215,031 WGBH-DT 191,634 WXIX-DT 192,049 CBET 212,586 CFTO 210,241 KCAL-DT 188,472 CELT 191,437 WLIW-DT2 209,720 WGBH-DT 204,149 WJZ-DT 184,684 KQED-DT 189,265 WLIW-DT3 209„720 WSBK-DT 199,676 CBMT-DT 184,474 WGBH-DT 189,072 WLIW-DT4 207,296 KCTS-DT 196,645 KICU-DT 161,348 WNJN-DT 184„922 WSJP-LD 206,904 CELT 188,028 KERA-DT 154„646 KCAL-DT 181,998 WGBH-DT 205,789 WNJT-DT 187,323 WPRU-LP 137,494 WLUT-DT 178,912 WLVT-DT 203,523 KQED-DT 181,115 WSJX-LP 137,494 WENH-DT 175,878 WSBK-DT 202,406 WENH-DT 181,077 KYW-DT 136„254 WLUT-DT3 174,221 CELT 201,644 WJZ-DT 173,911 CFTO-DT 126,564 WPTO-DT 160,795 WJZ-DT 196,034 WNYJ-DT 168,803 WSBE 116,830 KERA-DT 153,072 WLVT-DT2 193,931 KCAL-DT 166,764 WFUT-DT2 115,485 WNJT-DT 146,290 WLVT-DT3 193,931 WHYY-DT 166,246 WBAL-DT 113,901 WNJT-DT2 142,397 KCTS-DT 192,067 WGBX-DT 164,515 KTHV-DT 109,220 WHYY-DT 136,737 KQED-DT 183,750 WXIX-DT 159,311 WSB-DT 108,701 WPRU-LP 133,599 WZDC-LP 180,293 CBET 157,644 WZDC-LP 108,300 WSJP-LD 133,599 WXIX-DT 179,438 KERA-DT 154,289 WCVB-DT 107,534 WSJX-LP 133,599 WTTW-DT 177,361 KBTC-DT 152,766 WEIU-DT 107,128 KUHT-DT 126,726 WENH-DT 175,613 WETA-DT 145,010 WOTM-LP 105,778 WPVI-DT 123,447 WHYY-DT 173,644 WTAM-LP 143,157 KAET-DT 105,173 WBNS-DT 117,523 WWME-CA 173,604 WCMU-DT 142,603 WNET-DT2 104,678 KICU-DT 116,588 WNJT-DT 168,757 WTTW-DT 142,600 WNET-DT3 104,678 WDIV-DT 115,424 KERA-DT 155,565 WAUG-LD 139,738 WNEP-DT 102,560 WPSG-DT 114,434 KBTC-DT 153,098 WSEE-DT 136,966 KNBC-DT 100,612 WTAM-LP 114,296 WROC-DT 152,860 KUHT-DT 111,236 WHA-DT 100,517 WUNC-DT 111,526 WCAU-DT 151,701 WBAL-DT 109,024 WPTO-DT 99,448 WNYJ-DT 106,681 WCMU-DT 147,625 WUNC-DT 108,110 WBZ-DT 95,788 WEIU-DT 105,993 WETA-DT 138,421 KOCE-DT 101,855

24 WTBY-DT 95,510 WBAL-DT 105,602 WTAN-LP 129@29 CKWS 99,186 WTAM-LP 94,619 KATV-DT 105,440 WSB-DT 111,897 WSB-DT 98,337 KOPB-DT 89,812 KDKA-DT 103,604 WDIV-DT 111,546 KAET-DT 91,579 WBPX-DT 88,819 WOTM-LP 99,007 WUNC-DT 108,913 WTBY-DT 90,729 CHLT 83,385 WLIW-DT4 97,797 KATV-DT 99,117 WNYW-DT 90,161 KABC-DT 83,371 WUAB-DT 96,059 WTBY-DT 93,808 WCAU-DT 88,498 WNDU-DT 83,187 KTCA-DT 95,888 KAET-DT 93,681 CBFT 85,637 KRMA-DT 82,703 WTBY-DT 95,474 KNBC-DT 92,777 WMBC-DT 84,662 WSYX-DT 79,721 WMBC-DT 92,504 WNYW-DT 91,556 WSYX-DT 82,968 WPCW-DT 79,329 WRMD-CD 92,367 WYIN-DT3 91,474 WQED-DT 82,650 WNJN 77,965 WSFL-DT 86,820 KDKA-DT 90,166 WFME-DT4 80,033 WRMD-CD 77,909 WCVB-DT 83,785 WMBC-DT 89,837 KABC-DT 74,622 WFLD-DT 77,301 WSIU-DT 81,666 WRIC-DT 85,027 KCOP-DT 74,622 KTCA-DT 76,016 KOPB-DT 81,269 CBFT 84,355 KCWC-DT 74,250 WMCN-DT 74,555 WWSI-DT 81,102 WCVB-DT 83,723 WNVT-DT 73,295 WWSI-DT 74,202 WFPA-CA 80,217 KRON-DT 83,661 KRWG-DT 72,293 CKWS-DT 73,288 WPCW-DT 79,227 WBZ-DT 80,457 WPCW-DT '0,553 WCMU-DT 72,967 WNPB-DT 78,400 WHDH-DT 80,167 IBSEN-DT 70,072 KUSM-DT 72,521 WNED-DT 77,945 WSYX-DT 78,651 XHAB 63,362 KLRN-DT 72,005 KCRA-DT 77,363 KCOP-DT 78,416 XHOR 63,362 KOFY-DT 71,940 KABC-DT 76,932 WHA-DT 76,686 WSBT-DT 62,507 KRON-DT 71,940 KCOP-DT 76,932 WIS-DT 75,601 WPSU-DT 60,934 WSFL-DT 69,587 KOFY-DT 75,859 KYW-DT 75,176 WHA-DT 59,103 WMUR-DT 65,106 KLRN-DT 71,699 WMK-DT 71,757 WYES-DT 58,742 WVIZ-DT 64@52 WKYT-DT 70,031 WJAN-LP 68,586 WZME-DT 53,235 WDIV-DT 63,936 WMUR-DT 67@02 WEFS-DT 67,981 WBEC-DT 52,267 KEYT-DT 61,536 WSBT-DT 63,781 KDIN-DT 66,614 WKYT-DT 51,204 WAUG-LP 59,850 WWBT-DT 62,417 XHAB 62,720 WFYI-DT 48,994 WBIQ-DT 58,897 WNDU-DT 62,323 WNDU-DT 59,830 WHVL-LD 48,727 WUSA-DT 56,517 KRMA-DT 60,811 KCWC-DT 59,346 KTNC-DT 48,059 WPXI-DT 51,921 WVXF-DT 60,765 WVXF-DT 57,633 WUSF-DT 47,196 WBGT-CD 46,786 WVIZ-DT3 60,635 WDBJ-DT 57,466 WBRA-DT 46,765 KYMB-LD 46,235 XHAB 60,550 WSAH-DT 55,220 WVXF-DT 46,684 WFXT-DT 44,154 KLCS-DT 59,651 WNED-DT 54,402 WPXI-DT 46,261 KWTV-DT 44,062 WSAH-DT 57,940 WMAQ-DT 52,693 WTAJ-DT 45,846 WWMT-DT 42,702 WVIZ-DT 57,376 WKMJ-DT 52@91 KYW-DT 45,678 WLNY-DT 42,530 WJAN-LP 57,334 KLCS-DT 52,350 WBIN-DT 45/14 WBOC-DT 42,186 WUSA-DT 55/57 KTHV-DT 51,238 KETA-DT 42,635 WOSU-DT 37,576 WFME-DT4 54,583 WUSA-DT 49,938 KWGN-DT 42,050 KCSO-LD 36,014 KFOR-DT 48,553 KICU-DT 49,937 WAIQ-DT 41,486 KTWO-DT 32,547 KVIE-DT 47,717 KRMA-DT 49,551 WATC-DT 38,317 WIAT-DT 31,580 WBGT-CD 47,270 KARK-DT 49,167 WDRB-DT 32,820 WSEC-DT 31,519 WPGH-DT 47,126 KEYT-DT 48,245 WAFF-DT 30,951 WSAH-DT 29,949 WYES-DT 46,817 WCMV-DT 46,287 WPGH-DT 28,149 WXIA-DT 29,420 WPBY-DT 43,916 WXSP-LP 45,181 WIAT-DT 27,365 WPBS-DT 26,596 WYTV-DT 42,051 KUSA-DT 43,677 WNEP-DT 26,949 WWPX-DT 26,583 WWL-DT 41,440 WFYI-DT 41,151 WCAX-DT 26,523 KPLC-DT 25,955 WHUT-DT 40,870 WBIQ-DT 39,847 WEBA-DT 26,078 CBWT-DT 25,917 WThU-DT 40,003 WPBS-DT 38,790 WITN-DT 25,148 WLAE-DT 25,868 WALA-DT 38,667 WNPT-DT 38,705 KOAT-DT 24,791 WCAX-DT 25,691 ~-DT2 30,630 KTRK-DT 35,440 WNYF-LP 23,454 KSMQ-DT 25,036 XEW 29,593 WBIN-DT 34,565 KTMJ-CD 21,232 WCHS-DT 24,164 WCAX-DT 26,076 WGTV-DT 34,153 KETV-DT 20,412 WJEB-DT 24,137 WMHT-DT 25,802 WMHT-DT 33,718 KTEJ-DT 19,122 WKMG-DT 24,062 WLIO-DT 25,595 WNEP-DT 33,057 WTVQ-DT 18,988 WNEO-DT 23,808 WEWS-DT 25,369 WTMJ-DT 32,404 WIPB-DT 18,676 WOLF-DT 23,475 KMGH-DT 24@60 WVLA-DT 30,634 WRTV-DT 17,999 WBNG-DT 21,554 WUHF-DT 24,193 WDCQ-DT 28,380 KOCE-DT2 17,711 WTVR-DT 20,945 WLS-DT 24,187 KHNE-DT 23,762 WUNP-DT 17,477

25 WICZ-DT 19,953 WSAV-DT 22,848 WLS-DT 22,931 WAFB-DT 16,776 WCCO-DT 19,866 WBQD-LP 19,418 KBSV-DT 22,799 KTSC-DT2 15,885 WNEM-DT 19,718 WPSD-DT 18,335 CBUT-DT 22,672 WTVR-DT 15,186 WTGL-DT 17,013 KCRG-DT 16,382 WSFA-DT 21,225 KIIN-DT 14,416 WPBA-DT 16,876 WHEC-DT 13,685 KSTP-DT 18,696 WOKZ-CA 11,512 KNXT-DT 16,844 WHAM-DT2 11,977 WMYO-DT 14,860 KPTV-DT 11,470 WHBQ-DT 15,595 WMPT-DT3 11,544 WRAL-DT 13,905 KSAW-LD 9,182 WKBD-DT 13,767 WTTV-DT 10,762 WMPT-DT3 12,823 WYCC-DT2 9,022 WOUB-DT 10,748 WVIT-DT 10,471 WGVU-DT 12,459 WEIR-DT 8,851 WCJB-DT 10,660 WHAM-DT 10,443 KBFX-CA 12,381 WPXA-DT 8,744 WJET-DT 9,932 KBFX-CA 10,162 WLBZ-DT2 12,122 KGWC-DT 8,731 KUON-DT 9,932 KTFT-LP 9,426 WMNS-LP 11,880 WYBE-DT 8,542 KHBS-DT 9,811 KEVN-DT 9,422 WNSC-DT 11,591 WKOP-DT 8,455 KTFT-LP 9,658 CIMT 8,253 KSPR-DT 10,561 WGNT-DT 8,393 WEEK-DT 8,744 WTOM-DT 8,055 WGME- 8,370 WAFF-DT2 8,195 KVIA-DT 8,582 KHOU-DT 8,026 WDFM-LP 8,310 WMBD-DT 8,188 KACV-DT2 8,003 KACV-DT2 7,881 WKOP-DT 8,153 WLPB-DT 6,578 WBGU-DT3 7,844 KMOV-DT 7,792 WVIR-DT 6,958 WFVX-LD 6,489 WSTM-DT3 7,765 KMCI-DT 6,955 WLEF-DT 6,921 KBHE-DT 5,573 WFRZ-LP 7,448 KNPB-DT 6,834 KTEN-DT 6,576 KCCI-DT 4,852 WLLZ-LP 6,859 KTEN-DT 6,827 KJBO-LP 6,070 WJET-DT 4,603 KSTC-DT 5,864 KNLJ-DT 6,736 WHTX-LP 5,939 WKYC-DT 3,478 KHON-DT 5,813 WCTX-DT 6,030 KETV-DT2 4,914 WBGH-CA 3,347 KSNF-DT 5,030 WUWT-LP 4,762 WMHT- 3,660 WAXN-DT 2,132 KATU-DT 4,958 KQEG-CA 3,981 WKYC-DT 3,627 WFMZ-DT 1,620 KCCW-DT 3,761 KOZL-DT 3,770 KCCW-DT 3,274 WREG-DT 1,469 WNMN-DT 3,433 WHTV-DT 2,586 WILL-DT2 2,457 WTLW-DT2 1,451 WTO5 2,572 WPXN-DT 2,519 WTTE-DT 2,319 WLYH-DT 1,284 WMAZ-DT 2,359 WUPA-DT 2,185 WTVE-DT 2,183 KYTX-DT3 1,230 WNIN-DT 2,043 WITV-DT 2,112 WTWV-DT 2,176 WHMB-DT 1,191 WCIU 1,068 WGCB-DT 1,709 KMOS-DT 1,918 WJYS-DT2 1,070 WBMM-DT 895 WPBM-LP 1,554 WNIN-DT 1,301 KWCM 1,070 WWJ-DT 782 KLTJ-DT 1,391 WXPX-DT 1,207 WHWC-DT3 1,011 WTHR-DT 700 WUNL-DT 1,190 KLFY-DT 871 WENY-DT3 736 WPXV-DT 567 KWSD 1,188 KPBN-LP 714 KSWO-DT2 530 WTCE-DT 558 WWJ-DT 1,156 KSIN-DT2 675 KUNW-LP 454 KRPV-DT 465 KOCE-DT2 1,107 WMAZ-DT 522 WGB0-DT 341 WFXS-DT 409 WHTN-DT 975 WTNZ-DT 417 WBFF-DT3 307 317 KJTV 418 KRBC-DT 226 KTPX-DT 29 WMYD 142 KAZT-DT 290 KAMR-DT 186 WQAD-DT3 WPXD 142 KOMU-HD 70 WHBF-DT 176 WDKA-DT 99 WMDT-DT 27

26 APPENDIX C: CABLE REGRESSION MODELS - SPECIFICATIONS dk RESULTS

Table C-1: Poisson Regression Results, 2010

Robust Coefficient Standard Distant Viewers Estimate Error Z-score 95% Confidence interval Log of Local Ratings 1,050.15 13.3501 78.66 1,023.99 1,076.32 Log of Market Size 0.4848 0.0004 1,252.68 0.4840 0.4856

Time of Day (Quarter Hour) -0.1873 0.0081 -23.12 -0.2032 -0.1714 -0.2555 0.0081 -31.54 -0.2714 -0.2396 -0.3575 0.0083 -42.96 -0.3738 -0.3412 -0.3406 0.0083 -40.99 -0.3569 -0.3243 -0.4848 0.0086 -56.22 -0.5017 -0.4679 -0.5726 0.0087 -65.61 -0.5897 -0.5555 -0.6902 0.0090 -76.85 -0.7078 -0.6726 -0.6343 0.0090 -70.71 -0.6519 -0.6167 10 -0.7771 0.0094 -82.60 -0.7955 -0.7586 -0.8570 0.0097 -88.45 -0.8760 -0.8380 12 -0.9624 0.0101 -95.26 -0.9822 -0.9426 -0.9547 0.0102 -93.86 -0.9746 -0.9347 14 -1.0497 0.0105 -99.56 -1.0704 -1.0290 15 -1.1413 0.0108 -105.93 -1.1625 -1.1202 16 -1.1474 0.0109 -105.34 -1.1687 -1.1260 17 -1.0971 0.0108 -102.05 -1.1181 -1.0760 18 -1.0846 0.0107 -101.70 -1.1055 -1.0637 19 -1.0524 0.0105 -100.25 -1.0730 -1.0318 20 -1.0802 0.0106 -102.10 -1.1010 -1.0595 21 -0.9411 0.0098 -96.21 -0.9603 -0.9219 22 -0.8831 0.0097 -90.72 -0.9022 -0.8640 23 -0.8352 0.0095 -87.63 -0.8538 -0.8165 24 -0.8290 0.0094 -87.94 -0.8475 -0.8105 25 -0.7048 0.0092 -76. 27 -0.7229 -0.6867 26 -0.4898 0.0087 -56.08 -0.5069 -0.4727 27 -0.3414 0.0082 -41.56 -0.3575 -0.3253 28 -0.1826 0.0079 -23.02 -0.1981 -0.1670 29 -0.2573 0.0092 -27.98 -0.2754 -0.2393 30 -0.0961 0.0088 -10.91 -0.1134 -0.0789 -0.0054 0.0085 -0.64 -0.0221 0.0112 0.0955 0.0082 11.68 0.0795 0.1116 0.1525 0.0079 19.37 0.1371 0.1679 34 0.2289 0.0077 29.64 0.2138 0. 2440 35 0.2064 0.0078 26.51 0.1911 0.2216 36 0.2894 0.0076 38.09 0.2745 0. 3043 37 0.1546 0.0075 20. 72 0.1400 0.1693 38 0. 1056 0.0076 13.94 0.0908 0.1205 39 0. 1055 0.0076 13.90 0.0906 0.1204 40 0.1308 0.0075 17.34 0. 1160 0. 1456 41 0.0861 0.0076 11.33 0.0712 0. 1010 42 0.0037 0.0077 0.48 -0.0114 0.0188 43 0.0347 0.0077 4.52 0.0196 0.0498 0.0700 0.0076 9.18 0.0551 0.0850 45 0.0588 0.0079 7.49 0.0434 0.0742 46 -0.0007 0.0080 -0.09 -0.0165 0.0150

27 47 -0.0106 0.0081 -1.32 -0.0264 0.0052 0.1165 0.0078 14.96 0.1012 0.1318 49 0.1911 0.0074 25.87 0.1766 0. 2056 50 0.1984 0.0075 26.41 0.1837 0.2132 51 -0.0221 0.0078 -2.82 -0.0374 -0.0067 52 -0.0616 0.0079 -7.81 -0.0771 -0.0462 53 0.0939 0.0077 12. 20 0.0788 0.1089 0.0847 0.0077 10.97 0.0696 0.0999 55 0.0700 0.0077 9.06 0.0549 0.0852 56 0.1095 0.0076 14. 34 0.0945 0.1244 57 0.1519 0.0076 20.11 0.1371 0. 1667 58 0.1285 0.0076 16.93 0.1137 0.1434 59 0.1408 0.0076 18.45 0.1259 0.1558 60 0.1438 0.0076 18.96 0.1289 0.1587 61 0. 1829 0.0074 24.66 0.1684 0.1975 62 0.1627 0.0074 21.96 0.1482 0.1772 0.1824 0.0074 24.56 0.1678 0.1969 0.2073 0.0074 28.06 0.1928 0.2218 65 0.3954 0.0070 56.16 0.3816 0.4092 66 0.3575 0.0071 50.44 0.3436 0.3714 67 0. 3833 0.0070 54.67 0.3696 0.3971 68 0.4429 0.0069 64.01 0.4294 0.4565 69 0.5249 0.0067 77.78 0.5117 0.5381 70 0.5348 0.0067 79.33 0.5216 0.5480 71 0.4951 0.0069 72.07 0.4817 0.5086 72 0.5570 0.0068 81.75 0.5436 0.5703 73 0.6996 0.0067 104.26 0.6864 0.7127 74 0.6858 0.0068 101.20 0.6725 0.6990 75 0.7515 0.0066 113.71 0.7386 0.7645 76 0.8646 0.0065 133.20 0.8519 0.8773 77 0.9164 0.0065 141.75 0.9037 0.9290 78 0.9446 0.0065 145.60 0.9319 0.9573 79 0.9030 0.0064 141.42 0.8905 0.9156 80 1.0651 0.0064 167.41 1.0526 1.0775 81 1.2130 0.0064 190.44 1.2005 1.2255 82 1.1078 0.0064 171.83 1.0951 1.1204 1.1264 0.0065 174.50 1.1138 1.1391 84 1.1310 0.0065 174.86 1.1184 1.1437 85 1.0433 0.0067 155.98 1.0302 1.0564 86 0.8987 0.0068 132.29 0.8854 0.9120 87 0.8808 0.0068 128.64 0.8674 0.8943 88 0.8245 0.0069 119.90 0.8110 0.8380 89 0.9788 0.0065 150.97 0.9661 0.9915 90 0.7855 0.0066 118.19 0.7725 0. 7985 91 0.6556 0.0068 96.38 0.6422 0.6689 92 0.5901 0.0069 85.81 0.5766 0.6036 93 0.6259 0.0066 94.96 0.6130 0.6389 94 0.4277 0.0068 62.95 0.4144 0.4410 95 0.2558 0.0072 35.75 0.2418 0.2699 96 0.0929 0.0075 12.36 0.0781 0.1076

Program Type -0.7754 0.0042 -183.32 -0.7837 -0.7671 -0.9129 0.0047 -192.87 -0.9222 -0.9036 -0.4101 0.0197 -20.86 -0.4487 -0.3716 -1.9916 0.2374 -8.39 -2.4568 -1.5264 0.4353 0.0048 90.23 0.4258 0.4447 -1.5119 0.0083 -181.45 -1.5283 -1.4956 -0.7757 0.0139 -55.74 -0.8030 -0.7484 -0.5317 0.0050 -105.77 -0.5416 -0.5219 10 -2.2343 0.0161 -138.71 -2.2659 -2.2027 -0.9464 0.0161 -58.94 -0.9779 -0.9150 12 -0.5380 0.0054 -99.56 -0.5486 -0.5274 13 -0.4908 0.0066 -74.75 -0.5036 -0.4779 14 -0.5351 0.0048 -110.83 -0.5446 -0.5256 15 -0.5424 0.0053 -101.82 -0.5529 -0.5320 -0.6221 0.0050 -124.22 -0.6319 -0.6123 17 -0.4245 0.0041 -104.74 -0.4325 -0.4166 18 -0.6880 0.0041 -168.22 -0.6961 -0.6800 19 -1.5147 0.0045 -338.67 -1.5234 -1.5059 20 0.1313 0.0051 25.95 0.1214 0.1412 21 0.4614 0.0067 68.85 0.4483 0.4746 22 0.1380 0.0081 17.07 0.1221 0.1538 23 -1.0783 0.0065 -167.01 -1.0910 -1.0657 24 -2.9598 0.0096 -309.50 -2.9785 -2.9410 25 -0.8181 0.0046 -178.66 -0.8270 -0.8091 26 0.0401 0.0071 5.67 0.0262 0.0539 27 -1.5299 0.0361 -42A1 -1.6006 -1.4592 -0.7426 0.0083 -89.74 -0.7588 -0.7264 -0.6162 0.0039 -158.15 -0.6238 -0.6085 30 -0.2113 0.0095 -0.2299 -0.8061 0.0042 -193.74 -0.8143 -0.7980 -0A492 0.0070 -64.07 -0.4629 -0.4354

Constant 0.0081 -904.29 -73109 -7.2793

29 Table C-2: Poisson Regression Results, 2011

Robust Coefficient Standard Distant Viewers Estimate Error 2-score 95% Confidence Interval Log of Local Ratings 6,702.91 77.9067 86.04 6,550.21 6,855. 60 Log of Marl&et Size 0.4918 0.0004 1,179.47 0.4910 0.4927

Time of Day (Quarter Hour) -0.1692 0.0086 -19.65 -0.1861 -0.1523 -0.1936 0.0086 -22.39 -0.2106 -0.1767 -0.2795 0.0089 -31.47 -0.2969 -0.2621 -0.3250 0.0089 -36.47 -0.3425 -0.3076 -0.4480 0.0091 -49. 23 -0.4658 -0.4301 -0.4863 0.0092 -52.92 -0.5043 -0.4683 -0.6106 0.0095 -64.09 -0.6293 -0.5919 -0.6905 0.0099 -69.61 -0.7099 -0.6711 10 -0.8001 0.0103 -78.01 -0.8202 -0.7800 -0.8345 0.0105 -79.79 -0.8550 -0,8140 -0,9126 0.0107 -0.8916 -0.9737 0.0109 -89.00 -0.9951 -0.9522 -1.0608 0.0113 -93.97 -1.0829 -1.0387 -1.1333 0.0116 -97A2 -1.1561 -1.1105 -1.2101 0.0118 -102.29 -1.2333 -1.1869 -1.2103 0.0119 -101,70 -1.2336 -1.1869 -1.,1488 0.01'17 -1.1717 -1.0914 0.0112 -97A5 -1.0694 20 -1.2242 0.0117 -104.61 -1.2013 -1.0921 0.0113 -96.77 -1.1142 -1.0699 22 -0.9889 0.0110 -90.29 -1.0104 -0.9674 -0.9945 0.0111 -89.85 -1.0162 -0.9728 -0.9036 0.0108 -83.79 -0.9248 -0.8825 -0.5886 0.0099 -59.56 -0.6079 26 -0.3968 0.0093 -42.68 -0A150 -0.3785 27 -0.2483 0.0089 -27.89 -0.2657 -0.2308 28 -0.0198 0.0084 -235 -0.0364 -0.0033 0.1351 0.0091 14.90 0.1173 0.1529 0.2222 0.0092 24.03 0.2040 0.2403 0.1604 0.0089 17.95 0.1429 0.1779 0.2448 0.0085 28.68 0.2281 0.2615 0.1501 0.0084 17.91 0.1336 0.1665 0.2494 0.0082 30.25 0.2333 0. 2656 35 0.2500 0.0082 30.59 0.2340 0.2660 36 0.2817 0.0082 34.50 0.2657 0.2977 37 0.2586 0.0079 32.85 0.2432 0.2740 0. 2244 0.0079 28.46 0.2089 0.2398 39 0. 2039 0.0079 25.74 0.1883 0.2194 40 0. 2100 0.0079 26.57 0.1946 0.2255 0.2701 0.0079 34.29 0.2547 0. 2855 42 0.2276 0.0080 28.55 0.2120 0.2433 0.1784 0.0080 22.26 0.1627 0.1941 44 0.1957 0.0079 24.63 0.1801 0.2113 45 0.1750 0.0081 21.73 0.1592 0.1908 46 0.1379 0.0081 16.96 0.1220 0.1538 47 0.1601 0.0081 19.74 0. 1442 0.1760 48 0.1738 0.0081 21.52 0.1580 0.1896 0.2959 0.0076 38.92 0.2810 0.3108

30 50 0.2593 0.0077 33.81 0. 2442 0.2743 51 0.1655 0.0079 20. 89 0.1499 0.1810 52 0.2112 0.0080 26.46 0.1955 0.2268 53 0.3627 0.0077 47.08 0. 3476 0.3778 54 0.3403 0.0078 43.69 0.3250 0.3556 55 0.3531 0.0078 44.99 0.3378 0.3685 56 0.3846 0.0078 49.12 0.3693 0.3999 57 0.3483 0.0078 44.70 0.3330 0.3635 0.3029 0.0079 38.53 0.2875 0.3183 59 0.3270 0.0079 41.50 0.3116 0.3425 60 0.3492 0.0078 44.92 0.3340 0.3645 61 0.4319 0.0076 56.79 0.4170 0.4468 62 0.3469 0.0078 44.57 0.3317 0.3622 63 0.3458 0.0078 44.38 0.3305 0.3611 64 0.4074 0.0077 52.94 0.3923 0.4224 65 0.4601 0.0076 60.77 0.4453 OA750 66 0.3929 0.0076 51.38 0.3779 0.4078 67 0.4255 0.0076 55.73 0.4105 OA405 68 0.4335 0.0076 57.33 0.4187 OA483 69 0.5688 0.0074 76.46 0.5543 0.5834 70 0.5495 0.0074 0.5349 0.5640 0.5682 0.0076 75.24 0.5830 72 0.6499 0.0074 87.61 0.6353 0.6644 0.7664 0.0072 106,19 0.7523 0.7805 0.7930 '0.0072 110.63 0.7789 0.8070 0.6931 0.0075 0.7078 0.8136 0,0084 96.98 0.7972 0.8300 0.9612 0,0077 125.16 0.9461 0.9762 0.0070 1.0114 1.0389 79 0.0070 142.47 0.9794 1.0067 80 1.1217 0.0071 159.06 1.1079 0.0069 187.74 1.3105 1.1453 0.0070 1.1316 1.1591 0.0070 165.63 1.1508 1,1783 1.1637 0.0070 165.96 1.1500 1.1775 85 1.0678 0.0072 148.37 1.0537 1.D819 86 0.9044 0.0073 123.11 0.8900 0.9188 87 0.8533 0.0074 115.86 0.8389 0.8677 0.8261 0.0074 111.29 0.8116 0.8407 1.0494 O.OD72 144.87 1.0352 1.0636 90 0.9005 0.0073 122.67 0.9149 91 0.8250 0.0073 113.28 0.8107 0.8393 92 0.6944 0.0074 93.51 0.6798 0.7090 93 0.6605 0.0072 92.04 0.6464 0.6746 94 0.5059 0.0073 69.20 0.4916 0.5202 95 0.1985 0.0079 25.09 0.1830 0. 2140 96 0.0075 0.0083 0.90 -0.0089 0.0238

Program Type -0.5914 0.0042 -141.63 -0.5996 -0.5833 -0.7826 O.OD47 -166.66 -0.7918 -0.7734 -0.5079 0.0247 -20.53 -0.5564 -0.4594 0.0000 (omittedj -0.0804 0.0057 -14.03 -0.0916 -0.0691 -0.8678 0.0074 -117.30 -0.8823 -0.8533 -0.6568 0.0502 -13.07 -0.7553 -0.5584 -0.4216 0.0053 -79.00 -0.4321 -0.4112 10 -1.0608 0.0132 -80.44 -1.0866 -1.0349 -0.2775 0.0136 -20.46 -0.3041 -0. 2509 12 -0.4136 0.0053 -77.47 -0.4241 -0.4032 -0.1738 0.0072 -24.20 -0.1879 -0.1597 14 -0.2342 0.0048 -48.59 -0.2437 -0.2248 15 -0.4162 0.0052 -79.74 -0.4264 -0.4059 -0.6169 0.0051 -122.12 -0.6268 -0.6070 17 -0.3808 0.0040 -94.71 -0.3887 -0.3729 18 -0.7901 0.0042 -186.76 -0.7984 -0.7818 19 -1.1051 0.0044 -249.53 -1.1138 -1.0965 20 0.4576 0.0051 89.84 0.4476 0.4676 21 0.9870 0.0084 118.00 0.9706 1.0034 22 -24.3465 0.0212 -1,148.09 -24.3881 -24.3049 -0.4159 0.0060 -69.29 -0.4277 -0.4041 24 -1.9789 0.0090 -220.33 -1.9965 -1.9613 25 -0.4793 0.0044 -108.55 -0.4880 -0.4707 0.3033 0.0081 37.50 0.2874 0.3191 27 -1.7441 0.0542 -32. 19 -1.8503 -1.6379 28 -0.4326 0.0090 -47.84 -0.4503 -0.4148 29 -0.5178 0.0039 -133.97 -0.5253 -0.5102 30 0.0068 0.0140 0.49 -0.0206 0.0342 -0.5459 0.0041 -132.41 -0.5540 -0.5378 -0.4052 0.0077 -52.34 -0.4203 -0.3900

Constant -7.8395 0.0086 -912.35 -7.8563 -7.8227

32 Table C-3: Poisson Regression Results, 2012

Robust Coefficient Standard Distant Viewers Estimate Error 2-score 95% Confidence Interval Log of Local Ratings 25,365.42 58.8317 431.15 25,250.11 25,480.73 Log of Market Size 0.4303 0.0004 1,102.68 0.4295 0.4311

Time of Day (Quarter Hour) -0.1444 0.0078 -18.50 -0.1597 -0.1291 -0.2588 0.0080 -32.55 -0.2744 -0.2432 -0.4446 0.0083 -53.79 -0.4608 -0.4284 -0.4097 0.0084 -49.00 -0.4261 -0.3933 -0.5570 0.0086 -64.65 -0.5739 -0.5401 -0.6191 0.0087 -71.22 -0.6361 -0.6021 -0.6782 0.0087 -77.94 -0.6952 -0.6611 -0.8049 0.0090 -89.51 -0.8226 -0.7873 10 -0.9339 0.0095 -97.92 -0.9526 -0.9152 -0.9609 0.0097 -99.32 -0.9799 -0.9420 -1.0184 0.0099 -102.50 -1.0379 -0.9990 -0.9987 0.0099 -101.29 -1,0180 -0.9793 -1.0311 0.0101 -101.84 -1.0509 -1.0112 -1,1544 0.0104 -111.24 -1.1747 -1.1340 -1.2319 0.0107 -115.16 -1.2529 -1.2110 -1.1256 0.0103 -109.01 -1.1459 -1.1054 -1,1935 0.0106 -112.25 -12144 -1.1727 -1.1899 0.0102 -116.53 -1.2100 -1.1699 20 -1.2052 0.0104 -115.88 -1.2256 -1.1848 -1.0324 0.0100 -103.18 -1.0520 -1.0128 22 -1.0633 0.0100 -106.36 -1.0829 -1.0438 -1.0856 0.0101 -107.63 -1.1054 -1.0659 24 -0.9438 0.0096 -98.68 -0.9626 -0.9251 -0.6979 0.0091 -76.43 -0.7158 -0.6800 -0.5580 0.0087 -63.95 -0.5751 -0.5409 -0.5895 0.0089 -66.57 -0.6069 -0.5721 -0.3974 0.0084 -47.14 -0.4140 -0.3809 29 -0.3661 0.0090 -40.75 -0.3837 -0.3485 30 -0.2267 0.0087 -26.21 -0.2437 -0.2098 -0.1423 0.0085 -16.73 -0.1590 -0.1256 -0.0320 0.0083 -3.88 -0.0482 -0.0158 33 -0.0996 0.0084 -11.88 -0.1160 -0.0832 -0.0231 0.0082 -2.82 -0.0391 -0.0070 0.0596 0.0079 7.51 0.0440 0.0751 36 0.1152 0.0079 14.66 0.0998 0.1306 37 0.1712 0.0073 23.52 0.1569 0.1855 0.1544 0.0073 21.24 0.1401 0.1686 39 0.1574 0.0073 21.71 0.1432 0.1716 40 0.1333 0.0073 18.19 0.1189 0.1476 0.0338 0.0077 4.40 0.0188 0.0489 42 -0.0535 0.0078 -6.82 -0.0689 -0.0382 43 -0.1329 0.0081 -16.45 -0.1487 -0.1170 -0.1273 0.0093 -13.72 -0.1455 -0.1091 45 -0.1645 0.0080 -20.43 -0.1802 -0.1487 -0.2233 0.0082 -27.09 -0.2394 -0.2071 47 -0.2006 0.0080 -24.92 -0.2164 -0.1848 48 -0.2295 0.0081 -28.30 -0.2454 -0.2136 49 -0.0723 0.0075 -9.67 -0.0870 -0.0577

33 50 -0.1150 0.0076 -15.22 -0.1298 -0.1002 51 -0.1676 0.0078 -21.53 -0.1828 -0.1523 52 -0.1481 0.0078 -19. 10 -0.1633 -0.1329 0.0412 0.0076 5.39 0.0262 0.0562 54 0.0513 0.0077 0.0362 0.0664 55 0.0718 0.0077 9.38 0.0568 0.0868 56 0.1382 0.0076 18.22 0.1234 0.1531 57 0.1253 0.0075 16.74 0.1106 0.1400 58 0.0526 0.0076 6.89 0.0376 0. 0676 59 0.0629 0.0076 8.22 0.0479 0.0778 60 0.1157 0.0075 15.37 0.1010 0.1305 0. 2566 0.0071 36.37 0.2428 0. 2704 62 0.2221 0.0071 31.18 0.2081 0.2361 63 0.2577 0.0071 36.22 0.2438 0.2717 0.3252 0.0070 46.48 0.3115 0.3389 65 0.3441 0.0069 49.72 0.3306 0.3577 0.2768 0.0070 39.41 0.2631 0.2906 67 0.2633 0.0070 37.54 0. 2495 0. 2770 68 0.3239 0.0069 47.10 0.3104 0.3374 0.3485 0.0070 49.95 0.3348 0.3622 70 0.3153 0.0069 45.38 0.3017 0.3289 71 0.3225 0.0071 45.65 0.3087 0.3364 72 0.3867 0.0069 55.65 0.3731 0.4003 0.4757 0.0067 70.57 0.4625 0.4889 74 0.4907 0.0067 73.69 0.4777 0.5038 75 0.4513 0.0068 66.35 0.4380 0.4646 76 0.5599 0.0066 84.34 0.5469 0.5729 77 0.7370 0.0063 116.16 0.7246 0.7494 78 0.6993 0.0064 109.18 0.6867 0.7118 79 0.6936 0.0063 109.67 0.6812 0.7060 80 0.7481 0.0063 118.31 0.7357 0.7604 81 1.0707 0.0063 169.56 1.0583 1.0830 82 0.9636 0.0064 151.27 0.9512 0.9761 83 0.9581 0.0064 149.31 0.9455 0. 9707 0.9978 0.0064 156.74 0.9854 1.0103 85 0.9827 0.0063 155.36 0.9703 0.9951 86 0.8456 0.0064 131.29 0.8330 0.8583 87 0.8233 0.0065 127.48 0.8107 0.8360 0.7895 0.0065 121.84 0.7768 0.8022 89 0.7955 0.0066 120.66 0.7826 0.8084 90 0.6337 0.0067 94.24 0.6205 0.6469 91 0.6245 0.0066 93.99 0.6114 0.6375 92 0.5321 0.0067 79.00 0.5189 0.5453 0.4769 0.0066 72.44 0.4640 0.4898 94 0.4157 0.0067 62.43 0.4026 0.4287 95 0. 2459 0.0070 34.92 0.2321 0.2597 96 0.0333 0.0075 4.42 0.0185 0.0480

Program Type -0.6288 0.0040 -156.93 -0.6367 -0.6210 -0.6774 0.0045 -150.50 -0.6863 -0.6686 -0.5263 0.0157 -33.43 -0.5571 -0.4954 -0.3190 0.0065 -48.79 -0.3318 -0.3061 -0.7585 0.0065 -116.21 -0.7713 -0.7457 -16.2495 0.0595 -272.90 -16.3662 -16.1328 -0.3583 0.0054 -66.78 -0.3688 -0.3477 10 -1.3444 0.0119 -113.01 -1.3677 -1.3211 -0.6958 0.0161 -43. 29 -0.7273 -0.6643 12 -0.3481 0.0047 -74. 32 -0.3573 -0.3389

34 13 -0.1080 0.0074 -14.68 -0.1224 -0.0936 14 0.0016 0.0048 0.34 -0.0077 0.0110 15 -0.3598 0.0051 -70.01 -0.3699 -0.3498 16 -0.3198 0.0045 -70.79 -0.3287 -0.3110 17 -0.1054 0.0037 -28.10 -0.1127 -0.0980 -0.6031 0.0039 -155.61 -0.6107 -0.5955 19 -1.2210 0.0044 -276.35 -1.2296 -1.2123 20 0.5766 0.0068 84.77 0.5633 0.5899 21 0.9240 0.0082 112.23 0.9079 0.9401 22 -0.9008 0.1290 -6.99 -1.1536 -0.6481 -0.4854 0.0052 -92.50 -0.4957 -0.4752 24 -2.2066 0.0107 -205.97 -2.2276 -2.1856 25 -0.4632 0.0041 -111.67 -0.4713 -0.4550 26 0.6093 0.0081 74.79 0.5933 0.6252 27 -0.9432 0.0452 -20.85 -1.0319 -0.8546 28 -0.2578 0.0087 -29.56 -0.2749 -0.2408 29 -0.0653 0.0036 -18. 21 -0.0724 -0.0583 30 0.0926 0.0122 7.61 0.0687 0.1164 31 -0.4662 0.0039 -118.31 -0.4739 -0.4584 32 0.4022 0.0068 58.72 0.3888 0.4157

Constant -7.1103 0.0078 -909.78 -7.1256 -7.0949 Table C-4I Poisson Regression Results, 2013

Robust Coefficient Standard Distant Viewers Estimate Error 2-score 95% Confidence Interval Log of Local Ratings 57,723.42 282.7888 204.12 57,169.17 58,277.68 Log of Market Size 0.5378 0.0006 956.52 0.5367 0.5389

Time of Day (Quarter Hour) -0.1135 0.0100 -11.35 -0.1331 -0.0939 -0.0699 0.0100 -6.99 -0.0895 -0.0503 -0.1760 0.0103 -17.02 -0.1963 -0.1558 -0.1816 0.0103 -17.61 -0.2018 -0.1614 -0.3123 0.0106 -29.46 -0.3331 -0.2916 -0.2983 0.0107 -27.90 -0.3193 -0.2774 -0.3745 0.0109 -34.48 -0.3958 -0.3532 -0.4376 0.0111 -39.32 -0A594 -0.4158 10 -0A639 0.0113 -0.4861 -0.4417 -0.5223 0.0115 -45.55 -0.5447 -0A998 -0.6001 0.0117 -51,39 -0.6230 -0.5772 -0.5556 0.0114 -48.86 -0,5778 -0.6025 0.0116 -51.86 -0.6252 -0.5797 -0.6966 0.0119 -0.7200 -0.6732 -0.7032 0,0120 -58.69 -0.7267 -0.6798 -0.7202 0.0123 -58A2 -0.7443 -0,6960 -0.7614 0,0124 -61.31 -0.7858 -0.7371 -0.8025 0.0124 -0.8269 -0.7781 20 -0.7564 0.0122 -62. 11 -0.7803 -0.7326 -0.6438 0.0123 -52.40 -0.6678 -0.6197 22 -0.6564 0.0122 -53.80 -0.6803 -0.6324 -0,6782 0.0122 -55.62 -0.7021 -0.6543 24 -0.4913 0.0119 -41.15 -0.5147 -0.4679 -0.1210 0.0105 -11.55 -0.1415 -0.1004 -0.1018 0.0104 -9.79 -0.1222 -0.0815 27 -0.2919 0.0110 -26.55 -0.3135 -0.2704 -0.2148 0.01.07 -20.00 -0.2359 -0.1938 0.0865 0.0104 8.34 0.0662 0.1069 30 0.2791 0.0099 28.17 0.2597 0.2985 0.2508 0.0099 0.2314 0.2703 0.3166 0.0097 32.47 0.2975 0.3357 0.1875 0.0099 18.91 0.1681 0.2069 34 0.2625 0.0098 26.85 0.2433 0.2816 35 0. 2688 0.0098 27.44 0.2496 0.2880 36 0.3236 0.0097 33.46 0.3047 0.3426 0.4528 0.0089 50.70 0.4353 0.4703 0.4761 0.0089 53.56 0.4587 0.4935 0.4451 0.0090 49. 67 0.4276 0.4627 40 0.4550 0.0089 51.15 0.4375 0.4724 0.3295 0.0093 35.62 0.3114 0.3477 42 0.1944 0.0095 20.38 0.1757 0.2131 43 0.0818 0.0098 8.36 0.0626 0.1010 44 0.0195 0.0100 1.96 0.0000 0.0391 45 -0.0422 0.0101 -4.16 -0.0621 -0.0223 46 -0.0869 0.0103 -8.43 -0.1072 -0.0667 47 0.0055 0.0100 0.55 -0.0141 0.0252 48 0.0027 0.0100 0.27 -0.0169 0.0223 49 0.1800 0.0095 18.96 0.1614 0.1986

36 50 0. 2066 0.0094 22.02 0.1882 0.2250 51 0.0098 9.71 0.0758 0.1141 0.0950'.1015 52 0.0097 10.46 0.0825 0.1205 0. 3667 0.0093 39.39 0.3485 0.3849 54 0.3349 0.0095 35. 28 0.3163 0.3535 55 0.3932 0.0095 41.60 0.3747 0.4117 56 0.4160 0.0095 43.83 0.3974 0.4346 57 0.3518 0.0094 37.34 0.3334 0.3703 58 0.3414 0.0096 35.60 0.3226 0.3602 59 0.3241 0.0097 33.36 0.3051 0.3431 60 0.3685 0.0096 38.51 0.3497 0.3872 0.5467 0.0091 59.96 0.5289 0.5646 62 0.5373 0.0092 58.51 0.5193 0.5553 0.5786 0.0091 63. 69 0.5608 0.5964 64 0.6537 0.0089 73.09 0.6361 0.6712 65 0.7193 0.0088 81.63 0.7020 0.7365 66 0.7041 0.0089 79.54 0.6868 0.7215 67 0.7334 0.0088 83.50 0.7162 0.7506 68 0.7508 0.0086 86.82 0. 7339 0.7678 0.7979 0.0086 92.44 0.7810 0.8149 70 0.7152 0.0088 81.73 0.6981 0.7324 71 0.7475 0.0088 85.34 0.7304 0.7647 72 0.8722 0.0085 102.24 0.8555 0.8889 0.8849 0.0092 96.22 0.8668 0.9029 74 0.8467 0.0085 100.05 0.8301 0.8633 75 0.9360 0.0085 110.15 0.9194 0.9527 76 1.0939 0.0083 132.47 1.0777 1.1100 77 1.1617 0.0081 144.07 1.1459 1. 1775 78 1. 1743 0.0081 144.95 1.1584 1. 1901 79 1.2590 0.0079 158.40 1.2434 1.2746 80 1.3431 0.0079 170.16 1.3276 1. 3585 81 1.6043 0.0078 205. 21 1.5890 1. 6196 82 1.4823 0.0079 187.05 1.4667 1.4978 83 1.5232 0.0079 193.01 1.5077 1.5386 1.5298 0.0079 194.02 1.5143 1.5452 85 1.4888 0.0080 185.83 1.4731 1.5045 86 1.3839 0.0081 171.11 1.3680 1.3997 87 1.3363 0.0081 164.39 1.3204 1.3523 88 1. 2834 0.0082 157.31 1.2674 1.2994 89 1.4162 0.0081 175.52 1.4003 1.4320 90 1.2395 0.0082 150.91 1.2234 1.2556 91 1.1278 0.0083 135.59 1.1115 1.1441 92 0.9815 0.0085 115.94 0.9649 0.9980 93 0.7871 0.0084 93.34 0.7705 0.8036 94 0.6153 0.0086 71.34 0.5984 0.6322 95 0.4878 0.0089 54.71 0.4703 0.5053 96 0.3266 0.0094 34.56 0.3081 0.3451

Program Type -0.5828 0.0042 -139.26 -0.5910 -0.5746 -0.5415 0.0048 -112.02 -0.5509 -0.5320 -0.4009 0.0153 -26.17 -0.4309 -0.3708 0.4202 0.0055 76.47 0.4094 0.4309 -0.7442 0.0067 -111.47 -0.7573 -0.7312 -20.6398 0.0384 -536.84 -20.7151 -20.5644 -0.3959 0.0069 -57.54 -0.4094 -0.3824 10 -1.1484 0.0116 -99.40 -1.1710 -1.1257 -1.0185 0.0223 -45.62 -1.0623 -0.9748 12 -0.2655 0.0050 -52.77 -0.2753 -0.2556

37 -0.3612 0.0120 -30.08 -0.3847 -0.3377 14 -0.3051 0.0053 -57.85 -0.3155 -0.2948 -0.2647 0.0051 -51.86 -0.2747 -0.2547 16 -0.3507 0.0046 -76. 39 -0.3597 -0.3417 17 -0.1551 0.0038 -40. 50 -0.1626 -0.1476 18 -0.6871 0.0042 -164.45 -0.6953 -0.6789 19 -0.8979 0.0046 -194.61 -0.9070 -0.8889 20 1.0321 0.0054 191.88 1.0215 1.0426 21 0.7147 0.0092 78.02 0.6967 0. 7326 22 -20.0405 0.0141 -1,422.33 -20.0681 -20.0128 23 -0.4598 0.0054 -85.18 -0.4703 -0.4492 24 -1.4496 0.0084 -172.95 -1.4661 -1.4332 25 -0.4519 0.0044 -102.95 -0.4605 -0.4433 26 0.5246 0.0097 54.34 0.5056 0.5435 27 -0.5430 0.0489 -11.10 -0.6389 -0.4471 28 -0.3756 0.0099 -38.06 -0.3950 -0.3563 29 -0.2568 0.0037 -69.28 -0.2640 -0.2495 30 -0.6591 0.0238 -27.74 -0,7057 -0.6125 -0.2378 0.0040 -59.45 -0.2456 -0.2299 0.3677 0.0072 51.28 0.3536 0.3817

Constant -9.0163 0.0105 -855.44 -9.0369 -8.9956 Table C-5: 95% Confidence Intervals Corresponding to Distant Viewing Estimates, 2010-2013

2010 2011 2012 2013

Canadian Claimants 1.57% - 1.61% 3.37% - 3.43% 3.07% - 3.12% 4.37% - 4.47%

Commercial Television 15.68% - 15.92% 12.03% - 12.26% 15.74% — 16.02% 10.30% - 10.51%

Devotionals 0.56% - 0.63% 0.93% — 1.06% 0.45% - 0.52% 0.62% — 0.69%

Program Suppliers 5 1. 95% — 5 2.28% 50.97% - 51.32% 36.23% — 36.56% 44.84% - 45.35%

Public Television 27.11% - 27.36% 28.95% — 29.22% 41.50% - 41.80% 33.29% — 33.69%

JSC 2.60% — 2.74% 3.11% - 3.37% 2.43% - 2.61% 5.55% - 6.30%

39 DKCLARATIOX OF JEFFREY S. GRAY PH.D.

I declare under penalty ofperjury that the foregoing testimony is true and correct, and of my personal knowledge.

Executed on Decemberd+, 2016

Jeffrey S. Gray, Ph.D. R ll B 0 O

I

ll

lI

I Before the COPYRIGHT ROYALTY JUDGES Washington, D.C.

In the Matter of

Distribution of the Docket No. 14-CRB-0010-CD (2010-13)

2010, 2011, 2012, and 2013 Cable Royalty Funds

DIRECT TESTIMONY OF MARTIN R. FRANKKL, PH.D.

DECEMBER 22, 2016 TESTIMONY OF MARTIN R. FRANKKL, PH.D.

A. BACKGROUND

My name is Martin R. Frankel. I am a Professor of Statistics and Computer Information

Systems at Baruch College, City University ofNew York. I have held this position at various levels (Assistant, Associate and Full Professor) for more than 30 years. I also hold a professorial appointment on the Graduate Faculty ofthe City University ofNew York. I have taught courses in basic statistics as well as advanced courses in statistical areas such as Statistical Sampling and

Multivariate Statistical Methods. I am the author or co-author of 3 books, 10 book chapters and more than 60 articles and papers on various applications of statistics and computers. I have authored and co-authored publications in statistical journals on statistical methods and theory. I have co-authored publications in non-statisticaljournals involving the application of statistical methods to applied areas such as Medicine, Public Policy, Marketing, Physics and Media

Research.

My education includes a Bachelor ofArts (Mathematics) from the University ofNorth

Carolina, and an Master ofArts (Mathematical Statistics) and Ph.D. (Mathematical Sociology) from the University of Michigan.

I have served as an officer in a number ofprofessional organizations. These posts include: President of the Market Research Council, Chair of the American Statistical Association

Section on Survey Research, Chair ofthe Standards Committee for the American Association of

Public Opinion Research, and Chair ofthe Quality and Methods Council ofthe Advertising

Research Foundation. I ain an elected member ofthe International Statistical Institute and an elected Fellow ofthe American Statistical Association. In 2007, I was inducted into the Market

Research Council Hall ofFame. I have served as a consultant to more than 50 business, government and industry organizations since 1971. Over the course ofmy career, I have applied statistical methods to a number of diverse areas such as Criminal Justice, Anti-Trust, Rail Tariffs, Magazine Readership,

TV Viewing and Radio Listening, Voter Discrimination, Nuclear Licensing, Cable TV

Forecasting and Network Election Coverage. Since 1996, I have served as a member ofthe NBC

Election Decision Team and have represented NBC on the Election Pool Statistical Committee.

A number ofthe papers and articles that I have authored and co-authored are cited in a number oftests and manuals. One ofmy papers has been included in the publication "50

Landmark Papers in Survey Statistics." A copy ofmy curriculum vitae is attached hereto as

AppendixA.

B. NATURE OF MY ASSIGNMXNT

I selected samples of cable systems for the 2010, 2011, 2012, and 2013 royalty years that were used for surveys seeking information about how cable system operators ("CSOs") value programming on broadcast stations they retransmitted on a distant basis during 2010, 2011,

2012, and 2013. The programming on these retransmitted stations was organized into the following eight (8) categories:

a. News and Community Events b. Syndicated Series c. Movies d. Live, play-by-play coverage ofprofessional and college team sports e. Other sports programming f. Devotional programs g. Programs broadcast only on PBS stations h. Canadian Station programs

I selected the samples of cable systems for each royalty year and provided them to

Horowitz Research, Inc., which conducted surveys asking CSO executives to allocate a percentage of a fixed dollar amount to each ofthe eight programming categories based on the value of each type ofprogramming to their respective cable systems.

C. SELECTION OF CABLE SYSTEM SAMPLES

Each sample selection and survey occurred the calendar year following the royalty year.

After the end of each royalty year to be surveyed, Cable Data Corporation ("CDC") provided me with a list (i.e., the universe) ofForm 3 cable systems submitting remittances to the Copyright

Office for the first semi-annual period in each ofthe 2010, 2011, 2012, and 2013 cable royalty years. The 2010 sampling &arne consisted of 1,003 Form 3 systems with total royalties of

$76,700,063. The 2011 sampling &arne consisted of 913 Form 3 systems with total royalties of

$87,961,106. The 2012 sampling &arne consisted of 808 Form 3 systems with total royalties of

$96,654,632. The 2013 sampling &arne consisted of 798 Form 3 systems with total royalties of

$ 101,253,909. These Form 3 reports were the most currently available reports for each royalty year to be studied. For each royalty year to be surveyed, I selected a sample of 300 cable systems. The measure of size chosen for the sample selection was the amount ofremittance (i.e., royalty fee payments) for each system.

I selected each year's sample using random disproportionate stratified sampling, which is

designed to minimize the margin ofrandom sampling estimation error by sampling cable

systems that remit higher cable royalty payments with higher probabilities when compared to

cable systems with smaller cable royalty payments. I drew each year's sample &om four (4)

strata of cable royalty classes (based on cable royalty payments) in order to optimize the

estimation of cable system operator dollar allocations to each ofthe above listed programming

categories. Optimization of estimation means that the sample design is constructed to minimize the margin ofrandom sampling estimation error associated'with the dollar amounts and proportions being estimated. This type of sample allocation is sometimes known as Optimal

Neyman Allocation'nd the stratum formation is based on methods often known as the cume- square root rule for stratum formation.

Since each royalty year's sample size was 300, all systems which accounted for more than 1/300'" (i.e., 0.3%) ofthe total cable royalty payments were placed in a self-representing stratum (shown as stratum 4 in the tables below), which assured that all systems representing more than 0.3% ofthe total cable royalty payments were selected with certainty (probability equal to 1.0). Systems which accounted for less than 0.3% ofthe total cable royalty payments were subdivided into the remaining three (3) sampling strata. The "cume-square root rule" was used to form the stratum boundaries for each of these three (3) strata. The Optimal Neyman

Allocation was used to determine the sample size to be selected within each ofthe three (3) strata.

Tables 1 - 4 below show the stratum definitions for the 2010, 2011, 2012 and 2013 sample selections, respectively. Each table shows, by stratum, the range of cable royalty amounts, the number of systems in the Form 3 universe (N), the mean royalty amount for the systems, the standard deviation and the number of systems in the sample (n).

'. Edwards Deming, Sample Design in Business Research, Wiley, 1960. Wayne A. Fuller, SaInpling Statistics, Wiley, September 2011 TABLE 1 - STRATUM DEFINITIONS BY ROYALTY AMOUNT FOR 2010

Stratum Low High Universe of Mean Royalty Sample Size Systems (N) Royalty Standard (n) Deviation

$0 $26,499 460 $ 14,220 $5,715 25

$26,500 $64,499 271 $41,345 $ 10,489

$64,500 $173,603 174 $101,435 $29,114 130

$175,527 $1,779,077 98 $421,477 $302,048 98

TABLE 2 - STRATUM DEFINITIONS BY ROYALTY AMOUNT FOR 2011

Stratum Low High Universe of Mean Royalty Sample Size Systems (N) Royalty Standard (n) Deviation

$0 $25,499 396 $14,029 $5,466

$25,500 $62,499 240 $40,243 $10,326

$62,500 $ 162,013 161 $101,435 $29,114 87

$168,075 $2,348,097 116 $494,011 $405,872 116

TABLE 3 - STRATUM DEFINITIONS BY ROYALTY AMOUNT FOR 2012

Stratum Low High Universe of Mean Royalty Sample Systems (N) Royalty Standard Size (n) Deviation

$0 $23,000 331 $13,259 $4,970 40

$23,001 $57,000 205 $36,608 $9,578 48

$57,001 $140,000 136 $89,879 $23,019 76

$147,001 $2,681,527 136 $533,368 $446,286 136 TABLE 4 - STRATUM DEFINITIONS BY ROYALTY AMOUNT FOR 2013

Stratum Low High Universe Mean Royalty Sample Size of Royalty Standard Systems Deviation (n) (N)

$0 $23,000 336 $ 12,792 $4,987 41

$23,001 $57,500 194 $36,802 $9,612 45

$57,501 $ 144,500 132 $92,447 $24,610

$ 144,501 $3,283,512 136 $570,689 $339,777 136

UALITY ASSURANCE

For the sample selection process, I repeated the computations fiom scratch at least twice and then cross-checked the results. Further, I compared the allocation and sample selection with prior years to assure that the changes seemed reasonable.

E. SAMPLE ESTIMATION

After selecting each year's sample in the manner set forth above, I provided the sample to

Horowitz Research to be used for the year's survey. Once Mr. Horowitz completed his survey of cable system operators, I used the survey responses to provide a weighted estimate ofthe percent dollar allocation that all cable systems would assign to the eight programming categories.

Tables 5 — 8 below show the resulting weighted estimate of percentages and the associated standard errors assigned to the eight programming categories for royalty years 2010—

2013.

'he formulas used for calculating the weighted estimates and standard errors are shown in Appendix B attached hereto. TABLE 5 — 2010 PERCENTAGE DISTRIBUTION OF ROYALTY DOLLARS BY PROGRAMMING TYPE

STANDARD PROGRAM TYPE PERCENT ERROR News 2 Communi 12.38 1.49 S ndicated series 20.28 1.70 Movies 17.15 1.17 Live Pro Col Team S orts 31.94 2.17 Other S orts 6.77 0.65 Devotional 3.78 0.78 PBS 7.69 1.69 Canadian 0.01 0.01

TABLE 6 — 2011 PERCENTAGE DISTRIBUTION OF ROYALTY DOLLARS BY PROGRAMMING TYPE STANDARD PROGRAM TYPE PERCENT ERROR News & Communi 12.85 1.02 S ndicated series 17.56 1.07 Movies 11.43 0.79 Live Pro Col Team S orts 27.13 1.53 Other S orts 10.80 0.79 Devotional 5.92 0.67 PBS 13.31 1.66 Canadian 1.00 0.86

TABLE 7 — 2012 PERCENTAGE DISTRIBUTION OF ROYALTY DOLLARS BY PROGRAMMING TYPE

STANDARD PROGRAM TYPE PERCENT ERROR News k, Communi 15.72 0.87 S ndicated series 15.97 1.03 Movies 12.14 0.72 Live Pro Col Team S orts 25.50 1.50 Other S orts 9.02 0.65 Devotional 5.74 0.40 PBS 15.05 1.82 Canadian 0.87 0.36 TABLE S — 2013 PERCENTAGE DISTRIBUTION OF ROYALTY DOLLARS BY PROGRAMMING TYPE

STANDARD PROGRAM TYPE PERCENT ERROR News & Community 9.54 1.00 Syndicated series 16.26 1.60 Movies 12.39 1.26 Live Pro Col Team Sports 35.28 4.85 Other Sports 7.40 0.75 Devotional 3.48 0.44 PBS 15.39 3.35 Canadian 0.35 0.15 Declaration of Martin R. Frankel

I declare under penalty of perjury that the foregoing testimony is true and correct, and of my personal knowledge.

Executed on December /, 2016.

Martin R. rankel, Ph.D. APPENDIX A

CURtuCULUM VITAE

MARTIN RICHARD FRANKEL

OFFICE: Department of Statistics

Baruch College, CUNY 17 Lexington Avenue New York, New York 10010 (646) 312-3350 [email protected]

MAILING: 1000 N US Highway 1, Unit A301 Jupiter, FL 33477 K203) 912-6611 [email protected]

EDUCATION: A. B. (Mathematics) University of North Carolina, 1965 M.A. (Mathematical Statistics) The University of Michigan, 1967 Ph. D. (Mathematical Sociology) The University of Michigan, 1971

PROFESSIONAL EXPERIENCE:

PRESENT: Professor of Statistics and Computer Information Systems, Baruch College, City University of New York (1980-present). Deputy Chair (1994-2001).

1975-79: Associate Professor of Statistics, Baruch College, City University of New York

1973-74: Assistant Professor of Statistics, Graduate School of Business, University of Chicago. 1971-72: Assistant Professor of Statistics, Baruch College, City University of New York.

1965-70: Research Associate, Survey Research Center, Institute for Social Research, the University of Michigan. {Research Assistant 65-68).

1973-1996: Senior Statistical Scientist, MORC, The University of Chicago

1996-2012: Senior Statistical Scientist, ABT Associates, Cambridge, MA.

PUBLICATIONS:

BOOKS:

~ Frankel, M.R, Inference from Survey Samples: An Empirical Investigation. Ann Arbor: Institute for Social Research, the University of Michigan. 1971. 196 pages.

~ Kish, L, Fran/el, M.R. and Van Eck, N., SEPP: Sampling Error Program Package. Ann Arbor: Institute for Social Research, the University of Michigan. 1972, 50 pages

~ Andersen, R., Kasper, J. and Frankel, M.R., Total Survey Error: Applications to Improve Health Surveys. San Francisco: Jossey-Bass. 1979, 296 pages

CHAPTERS IN BOOKS:

~ Martin R. Frankel and Lester R. Frankel. "Probability Sampling," in Robert Ferber, Ed. Handbook of Marketing Research. New York: McGraw Hill, 1974. p. 230- 246.

~ Martin R. Frankel and Lester R. Frankel, "Some Recent Developments in Survey Sample Design," in Jain, A.K. et. al., Eds. Marketing Research: Applications and Problems. New York: John Wiley & Sons, 1982.

~ Martin R. Frankel, "Probability Sampling," in Rossi and Andersen, Eds. Handbook of Survey Research. New York: Academic Press, 1983. p. 21-67.

~ Martha J. Banks, Ronald Andersen, Martin R. Frankel, "Total Survey Error," in Incomplete Data In Sample Surveys Vol. 1, Pt II. New York: Academic Press, 1983.

~ Martin R. Frankel, "Master Samples" in Kotz and Johnson, Encyclopedia of Statistical Sciences, John Wiley and Sons, 1985, 285. ~ "Going to the Oracle for Strategic Planning: The Delphi Process," in Glass, H. E. ed. Handbook of Business Strategy (1988/1989 Yearbook) Boston: Warren, Gorham & Lamont, 1989. (with D. Kudon)

~ "The Effect of Interviewer Characteristics and Expectations on Response," Survey Research Methods, Singer, E and Presser, S. eds. 1989 (with E. Singer and M Glassman).

~ "Statistical Design and Estimation: Discussion," in Kasprzyk et. al. eds. Panel Surveys, New York: John Wiley k Sons, 1989.

~ "Forecasts in Strategic Planning," in Glass, H., ed. Handbook of Business Strate~r, Second Edition, Boston: Warren, Gorham 5 Lamont, 1991. (with D. Kudon)

~ Martin R. Frankel, "Resampling Procedures for Sample Surveys," in Hall, John, ed. Encyclopedia of Biostatistics, London: John Wiley and Sons, 1998.

~ Martin R. Frankel, "Probability Sampling", Handbook of Survey Research, J. Wright, Peter Marsden - Emerald Group Pub Ltd (2010) - Hardback - 886 pages- ISBN 1848552246

ARTICLES:

~ Leslie Kish and Martin Frankel, "Balanced Repeated Replications for Analytical Statistics," Proceedings of the Social Statistics Section, American Statistical Association. 1968.

~ Leslie Kish and Martin Frankel, "Balanced Repeated Replication for Standard Errors," Journal ofthe American Statistical Association 1970, 65, 1071-1094.

~ Martin Frankel, "Inference from Cluster Samples," Proceedings of the Social Statistics Section, American Statistical Association, 1971.

~ Leslie Kish and Martin Richard Frankel, "Inference from Complex Samples (with discussion)," Journal of the Royal Statistical Society (Series B), Volume 36, No. 1, 1974. 1-37.

~ James R. Murray, Michael Minor, Norman Bradburn, Robert Cotterman, Martin Frankel and Alan Pisarski, "Evolution of Public Response to the Energy Crisis," Science, Vol. 184, No. 4134, 1974. 257-263. ~ Martin R. Frankel, "Development of Broadcast Rating Standards for Standard Error Estimates," Proceedings of the Social Statistics Section, American Statistical Association, 1975.

~ Martin R. Frankel, "Software for Surveys — Are Existing Packages Adequate for Valid Statistical Inference?" Proceedings of the Ninth Interface Symposium on Computer Science and Statistics. New York: 2 ed., Prindle, Weber-Schmidt, Inc. 1976.

~ Ronald Anderson, Judith Kasper and Martin Frankel, "The Effect of Measurement Error on Differences in Hospital Expenditures," Medical Care 14: 1976, 932-949.

~ Martin R. Frankel and Lester R. Frankel, "Some Recent Developments in Sample Survey Design," Journal of Marketing Research, Vol. XIV, No. 3, 1977. p 280- 293.

~ Eleanor Singer and Martin R. Frankel, "Informed Consent Procedures in Telephone Interviews," American Sociological Review, Vol. 47, No. 3, 1982. 416-426.

~ Martin R. Frankel, "Ascription in Magazine Audience Research," in Henry, H., Ed., Readership Research: Theory and Practice. London: Sigmatext, 1982.

~ Eleanor Singer, Martin R. Frankel and Mare B. Glassman, "The Effect of Interviewer Characteristics and Expectations on Response," Public Opinion Quarterly, Vol. 47, No. 1, 1983, 68-83.

~ Martin R. Frankel and Michael G. Occhiogrosso, "Radio Audience Accumulation over Multiple Weeks," Journal of Advertising Research„Vol. 25, No. 3, 1985. p 23-30.

~ Donald Sadowsky, Carol Kunzel and Martin Frankel, "Predictors of Dentist' Level of Knowledge Regarding the Recommended Prophylactic Regimen for Patients with Rheumatic Heart Disease," Soc. Sci. Med., Vol. 21, No 8, 1985, 899-907.

~ Martin R. Frankel, "A Probability Sample of the Homeless Population of Chicago," Proceedings of the Section on Survey Research Methods, American Statistical Association, 1986, 176-177.

~ Shulamith T. Gross and Martin R. Frankel, "New Algorithms for Multidimensional Sample Allocation," Proceedings of the Section on Survey Research Methods, American Statistical Association, 1986. ~ Martin R. Frankel and Lester R. Frankel, "Fifty Years of Survey Sampling in the United States," Public Opinion Quarterly, Vol. 51:S127-S138, 1988

~ Martin R. Frankel, "Current research practices: general population sampling including geodemographics," Journal of the Market Research Society, 31,4, 1989.

~ Shulamith T. Gross and Martin R. Frankel, "Confidence Limits for Small Proportions in Complex Samples," Communications in Statistics, 20(3), 951-975, 1991.

~ Norman M. Bradburn, Martin R. Frankel, Reginald Baker, "A Comparison of Computer Assisted Personal Interviews (CAPI) with Personal Interviews in the National Longitudinal Survey of Labor Market Behavior - Youth Cohort," in Proceedings of the 1991 Annual Research Conference. Washington, DC: U. S. Bureau of the Census, 389-397, 1991.

~ H. Taylor and M. R. Frankel, "Suicide Highest in Wide-Open Spaces," American Demographics, 14(4) April, 1992 p. 9.

Robert J. Blendon, Karen Donelan, Carole VanDeusen Lucas, Kenneth E. Thorpe, Martin Frankel, Ronald Bass, Humphrey Taylor, "The Umnsured and the Debate Over the Repeal of the Massachusetts Universal Health Care Law," Journal of the American Medical Association, 267(8), 1992 1113-1117.

~ J. R. Friedman, X.Q. Guo, M. S. Lubell and M. R. Frankel, "Reexamination of tests of the Wannier threshold law for two-electron escape," Physical Review A, 46(1), July 1, 1992, 652-655.

~ Richard F. Mollica, Karen Donelan, Svang Tor, James Levelle, Christopher Elias, Martin Frankel and Robert J. Blendon, "The Effect of Trauma and Confinement on Functional Health and Mental Health Status of Cambodians Living in Thailand- Cambodia Border Camps" Journal of the American Medical Association, August 4, 1993, Vol. 270. Pp. 581-586

~ Leslie Kish, Martin R. Frankel, Vijay Verma and Niko Kaciroti, "Design Effects for Correlated (Pi - Pj)" Survey Methodology,, Vol. 21, No. 2 December 1995. 117- 124.

~ Karen Donelan, Robert J. Blandon, Craig Hill, Catherine Hoffman, Diane Rowland, Martin Frankel and Drew Altman, "Whatever Happened to the Health Insurance Crisis in the United States?" Journal of the American Medical Association, October 23/30 1996, Vol. 276. Pp. 1346-1350. ~ Martin Frankel and Benjamin King, "A Conversation with Leslie Kish," Statistical Science, 1996, Vol. 11, No. 1, 65-87.

~ Joseph J. Marbach, Gerald T. Ballard, Martin R. Frankel and Karen G. Raphael, "Patterns of TMJ Surgery: Evidence of Sex Differences," Journal of the American Dental Association, Vol. 128, May 1997, 609-614.

~ Carroll Seron, Martin Frankel, Douglas Muzzio, et. al., "A Report of the Perceptions and Experiences of Lawyers, Judges, and Court Employees Concerning Gender, Racial and Ethnic Fairness in the Federal Courts of the Second Circuit of the US," Annual Survey of American Law, New York University School of Law, 1997 Vol, Issues 1 and 2, 415-527.

~ Yael Caspi, Charles Poole, Richard F. Mollica and Martin Frankel, "Relationship of Child Loss to Psychiatric and Functional Impairment in Resettled Cambodian Refugees," The Journal of Nervous and Mental Disease, Vol. 186, No. 8., August 1998, 484-491.

~ Samuel A. Bozzette, Sandra H. Berry, Naihua Duan, Martin R. Frankel, et.al., "The Care of HIV-Infected Adults in the United States,'* New England Journal of Medicine, Volume 339, No. 26, Dec. 24, 1998, 1897-1904.

~ Gifford DG, Holloway RG, Frankel M, Albright CL, Meyerson R, Griggs R, Vickrey BG "A randomized trial to implement practice recommendations. Design and Methods of the Dementia Care Study," Controlled Clinical Trials, Vol 20, 1999, 369- 385.

~ David R. Gifford, Robert G. Holloway, Martin R. Frankel et.al, "Improving Adherence to Dementia Guidelines through Educational and Opinion Leaders, A Randomized, Controlled Trial," Annals of Internal Medicine, Vol. 131, No. 4, Aug. 17, 1999, 237-246.

~ Martin R. Frankel, Martin F. Shapiro, Naihua Duan, et.al., "National Probability Samples in Studies of Low-Prevalence Diseases. Part II: Designing and Implementing the HIV Cost and Services Utilization Study Sample," Health Services Research Vol 34, No 5, Part I (December, 1999), p. 969-992. One of Three Papers Selected by Association for Health Services Research Article of the Year Award: 2000.

~ Mare L. Berk, Claudia L. Schur, Leo R. Chaves and Martin Frankel, "Health Care Use Among Undocumented Latino Immigrants," Health Affairs, Vol 19, No. 4 July/August 2000, 51-64. ~ Trena M. Ezzati-Rice, Martin R. Frankel, David C. Hoaglin, John D. Loft, Victor G. Coronado, and Robert A. Wright, "An Alternative Measure of Response Rate in Random-Digit-Dialing Surveys That Screen for Eligible Subpopulations," Journal of Economic and Social Measurement 26, 2000, pp. 99-109.

~ Carroll Seron, Gregg Van Ryzin, Martin Frankel and Jean Kovath, "The Impact of Legal Counsel on Outcomes for Poor Tenants in New York City's Housing Court: Results of a Randomized Experiment," Law 8z Society Review, 35,2, 2001, pp. 419- 434.

~ Susan E. Cohn, Mare L. Berk, Sandra H. Berry, Naihua Duan, Martin R. Frankel, et. al. "The Care of HIV-Infected Adults in Rural Areas ofthe United States," Journal of Acquired Immune Deficiency Syndromes, 28, 2001, pp. 385-392.

~ Martin Frankel, "Election Night Estimation — 11th Morris Hansen Lecture, Discussion," Journal of Official Statistics, Vol 18, No. 2, 2002, pp 181-184

~ Martin R. Frankel, K.P. Srinath, Michael P. Battaglia, David Hoglan, Philip J. Smith, Robert A. Wright, and Meena Khare "Adjustments for Nontelephone Bias in Random-Digit-Dialing Surveys," Statistics in Medicine, 2003, Vol 22, pp 1611-26.

~ Martin R. Frankel, "RDD Surveys: Past and Future," Eighth Conference on Health Survey Research Methods, Cohen SB, Lepkowski JM, eds. Hyattsville, MD: National Center for Health Statistics. 2004.

~ Michael W. Link, Michael P. Battaglia, Martin R. Frankel, Larry Osborn, and Ali H. Mokdad, "Address-based versus Random-Digit-Dial Surveys: Comparison of Key Health and Risk Indicators", American Journal of Epidemiology, V164, 8, September 12, 2006

Michael W. Link, Michael P. Battaglia, Martin R. Frankel, Larry Osborn, and Ali H. Mokdad, "Reaching The U.S. Cell Phone Generation: Comparison of Cell Phone Survey Results with an Ongoing Landline Telephone Survey", Public Opinion Quarterly, Vol 71 No. 5 2007, p 814-839.

~ Michael P Battaglia, Martin R. Frankel, and Michael W. Link, "Improving Standard Poststratification Techniques for Random-Digit-Dialing Telephone Surveys," Survey Research Methods (2008) Vol. 2, No 1, pp. 11-19.

~ Michael W. Link, Michael P. Battaglia, Martin R. Frankel, Larry Osborn, and Ali H. Mokdad, "A Comparison of Address-Based Sampling (ABS) Versus Random-Digit- Dialing (RDD) for General Population Surveys", Public Opinion Quarterly, Vol. 72 No. 1 2008, p 6-27. ~ Michael P Battaglia, Martin R. Frankel, and Michael W. Link, "Improving Standard Poststratification Techniques for Random-Digit-Dialing Telephone Surveys," Survey Research Methods (2008) Vol. 2, No 1, pp. 11-19

~ Battaglia, M.P., Link, M.W., Frankel, M.R., Osborn, L., and Mokdad, A.H. 2008. An Evaluation of Respondent Selection Methods for Household Mail Surveys. Public Opinion Quarterly, 72: pp. 459-469.

~ Michael P. Battaglia, Martin R. Frankel, Ali H. Mokdad. 2008. Statistical Challenges Facing Cell Phone Survey. 2008 Proceedings of the Section on Survey Research Methods of the American Statistical Association.

~ K.P. Srinath, Martin R. Frankel, David C. Hoaglin, and Michael P. Battaglia. 2009. Compensating for Noncoverage of Nontelephone Households in Random-Digit- Dialing Surveys: A Comparison of Adjustments Based on Propensity Scores and Interruptions in Telephone Service. journal of Official Statistics, Vol. 25, No. 1, 2009.

~ Patrick S. Sullivan, Marta Juhasz, A.D. McNaghten, Martin Frankel, Sam Bozzette 5 Martin Shapiro, "Time to first annual HIV care visit and associated factors for patients in care for HIV infection in 10 US cities,*'IDS Care, 2011, p. 1-7.

~ S. Hu, Lina Balluz, Michael Battaglia and Martin Frankel, "The impact of cell phones on public heath surveillance," Bulletin of the World Health Organization, 2010, 88:799.

~ S. Hu, Lina Balluz, Michael Battaglia and Martin Frankel, "Improving Public Health Surveillance Using a Dual Frame Survey of Landline and Cell Phone Numbers, American Journal of Epidemiology (2011) 173 (6): 703-711

~ Martin R. Frankel, A.D. McNaghten, Martin F. Shapiro, Patrick S. Sullivan, et. Al. "A Probability Sample for Monitoring the HIV-infected Population in Car in the U.S. and in Selected States," The Open AIDS Journal 2012,6 (Suppl I:M2) 67-76

~ Martin R. Frankel, Michael P Battaglia, Lina Balluz, Tara Strine, "When data are not missing at random: implications for measuring health conditions in the Behavioral Risk Factor Surveillance System," BMJ Open 2012:2 1-9. OTHER REPORTS:

~ "The Health Studies Survey of Workers'ompensation Recipients," Research Report of the Interdepartmental Worker's Compensation Task Force, Vol. 7, U. S. Department of Labor, 1979.

~ High School and Beyond - Sample-Design-Report. Chicago: National Opinion Research Center, 1981.

~ The Profile of American Youth: Technical Sampling Report. Chicago: National Opinion Research Center, 1981.

~ Arbitron Replication II: A Study of the Reliability of Radio Ratings, Fishergate Publishing: Annapolis, Md., 1982.

~ "A Comparison of Reach and Frequency Estimates: Single Versus Dual Interview Approaches," in Henry, H., Ed., Readership Research: Montreal 1983. Amsterdam: Elsevier Science Publishers B.V. (with Adam Richard)

~ "The State of Audience Measurement in the United States," in La Mosaique Televisuelle: Du petit ecran au centre de controle, Congres ARCQ, Montreal, 1985.

~ "The Attribution of Missing Data: Damned If You Do and Dammed If You Don',", Fifth Annual ARF Research Quality Workshop: Transcript Proceedings. New York, Advertising Research Foundation, 1987.

~ A Household Survey of the Health Insurance Status of Massachusetts Residents: Final Report, Department of Health Policy and Management, Harvard School of Public Health, 1990

~ 1988 Physician's Practice Costs and Income Survey: Final Report, NORC, University of Chicago, 1991 (co-author)

~ "A Comparison of Computer-Assisted Personal Interviews (CAPI) With Personal Interviews in the National Longitudinal Survey of Labor Behavior-Youth Cohort," Proceedings 1991 Annual Research Conference, Bureau of the Census, U. S Bureau of the Census, 1991 leo-author)

~ "Repatriation and Disability: A Community Survey of Health, Mental Health, and Social Functioning of the Khmer Residents of Site Two. Volume 1 Khmer Adults," Working Document Harvard Program in Refugee Trauma, Harvard School of Public Health and World Federation of Mental Health, 1992 (co-author). ~ Arrowhead 12 - The Effect of Interview Attempts on Survey Results, New York: Advertising Research Foundation, 1992 (co-author).

~ National Education Longitudinal Study of 1988 - First Follow-Up: Student Component Data File User's Manual Vol. 1, National Center For Education Statistics, 1992 (co-author)

~ "Sample Representativeness in Electronic Data Bases." Media Research on the Leading Edge, New York: Advertising Research Foundation, 1992. pp. 75-78.

~ "Two Papers on the Use of Computer-Assisted Personal Interviews in the National Longitudinal Survey of Youth, U.S. Department of Labor, Report NLS 92-2. pp. 1-8 and pp. 1-14. (co-author)

~ "Estimation of Turnover in Readership Surveys," Readership Research Symposium 6: Session Papers, 1993: London: RSL. PP 525-530.

~ "Dropping Out of School: 1982 and 1992," U.S. Department of Education, Office of Educational Research and Improvement, Issue Brief, August 1993. PP 1-2. (co-author)

~ "Integrating Results of Physician Practice Cost Surveys, Final Report," Health Policy Research Consortium, Heller Graduate School, Brandeis University, October 1992. pp. 1-73. (co-author)

~ "Benchmarking Readership Levels in the New Study of Media and Markets to the Survey of American Readership," Worldwide Readership Research Symposium VII (Berlin), 1995. pp. 329-332. (lead author) Winner of Best Technical Paper Prize.

~ "The Interviewer Effect on Readership Levels," Worldwide Readership Research Symposium VII (Berlin), 1995. pp. 329-332.

~ "Enhanced Ascription in Magazine Audience Research," Worldwide Readership Symposium VIII (Vancouver), October, 1997 pp. 117-130.

~ National Education Longitudinal Study of 1988, Users Manual, National Center for Education Statistics, September 1994, Co-Author.

~ Bloomberg SJ, Osborn L, Luke JV, Olsen L, and Frankel MR. "Estimating the prevalence of uninsured children: An evaluation of data from the National Survey of Children with Special Health Care Needs, 2001." National Center for Health Statistics, Vital Heath Stat 2(136), 2004

10 PROFESSIONAL HONORS."

~ Harry O'eill Outstanding Achievement Award, NYAPPOR, 2013

~ Warren J. Mitofsky Innovators Award, American Association for Public Opinion Research, 2011

~ Elected to Hall of Fame, Market Research Council, 2007

~ Elected to Membership, International Statistical Institute, 1980

~ Fellow, American Statistical Association, 1978

SELECTED PROFESSIONAL ACTIVITIES:

~ Chair, Committee on Standards, American Association for Public Opinion Research, (1988-89).

~ Chair, Research Quality Council, Advertising Research Foundation, 1988- present.

~ Chairman, Section on Survey Research Methods, American Statistical Association (1975-1976).

~ Associate Editor, Journal of the American Statistical Association. (1973-1981).

~ Member of Editorial Board, Public Opinion Quarterly (1977-1983, 1986- 1990).

Member of Editorial Board, Sociological Methods and Research (1979-1983).

~ Chairman, American Statistical Association Advisory Committee to the U. S. Census (1981), Member (1975-1981).

~ Member of Editorial Board, Encyclopedia of Statistical Sciences, John Wiley &.

Sons, t 1980-1985).

~ Member, Panel on Occupational Safety and Health Statistics, Committee on National Statistics, National Research Council, National Academy of Sciences. (1985-1987)

11 ~ Member, Panel on the Functionality and Usability of Data from the American Community Survey, Committee on National Statistics, National Research Council, National Academy of Sciences. (2004-2005)

~ Member, External Advisory Committee, Program in Survey Methodology, The University of Michigan (2002-2004).

Fellow, American Statistical Association Fellow, (1978)

Fellow, International Statistical Institute (elected 1980)

President, Market Research Council, 1995

Biographee: Who's Who in America

Biographee". American Men and Women of Science

Hall of Fame, 2007„Market Research Council

SELECTED APPEARANCES INVOLVING THE PRESENTATION AND INTERPRETATION OF STATISTICAL EVIDENCE FOR LEGAL AND ADMINISTRATIVE PROCEEDINGS

~ SUPREME COURT OF THE STATE OF NEW YORK: Introduced the use of statistical sampling for the determination of the weight of narcotics. Accepted by the Court and Jury. (1997)

~ FEDERAL COURT (IOWA): Expert witness survey methods for U.S. Department of Justice, Anti-Trust Division (1995)

~ FEDERAL COURT (DETROIT): Expert witness survey sampling and methods for A.C. Nielsen Company (1993)

~ FEDERAL COURT (NEW YORK) Expert witness sampling and estimation. J Lefcourt (1985)

~ NUCLEAR REGULATORY COMMISSION: Expert witness statistical sampling and quality for Commonwealth Edison Company ILicense hearing for Byrom and Braidwood Nuclear Power Stations. (1980-1984)

~ INTERSTATE COMMERCE COMMISSION. Expert witness statistical sampling and estimation on behalf of Union Pacific Railroad merger with Western and Missouri Pacific Railroads. (1975-6)

12 ~ FEDERAL COURT (ALABAMA): Expert witness in statistics and demography on behalf of Southern Poverty Law Center for redistricting of Alabama after 1970 Census. (1973-4)

~ COPYRIGHT ARBITRATION ROYALTY PANEL: Expert witness on behalf of the Motion Picture Association of American, (1989, 2003, 2007)

13 Estimation ofProportion ofRoyalty Dollars among Program Types for Sample Survey and Standard Errors

Let h=stratum index h=l,...,H (H=4}.

Let ih= index ofreporting sample sysh:m i within stratum h.

Let jih=index ofprogram type j for sample system i stratum h.

Let Nh = number of systems in the population for stratum h.

Let nh = number of sample reporting systems in stratum h.

Let P j;h=proportion of royalty dollars allocated to program type j by sample system i in stratum h.

Let X h = royalty dollars remitted by system i in stratum h.

Yj'h = P j'h X ih royalty dollars allocated to program type j by sample system i in stratum h.

X„=g Xa = total royalty dollars remitted by all systems in stratum b. iral

Xh Wh = „" = proportion ofroyalty dollars remitted by all systems in stratum h. QX„ h=l x„=/X„= total royalty dollars remitted by all completed sample systems in stratum b. y,.h=gYjih = total royalty dollars allocated to program type j by sample systems in stratum h.

h=l

14 +X'„-(x,',/n„) 1=1 SX1 11„-1

1 var(r.„)= (I-(n„/N„))[(n„xsy.„)+(n„xsx„xr. xr.„)-(2xr.„xnaxsxy „)j

h=l

15 K

2. K 8 Q' Before the COPYRIGHT ROYALTY JUDGES Washington, D.C.

In the Matter of

Distribution of the Docket No. 14-CRB-0010-CD (2010-13)

2010, 2011, 2012, and 2013 Cable Royalty Funds

DIRECT TESTIMONY OF HOWARD HOROWITZ

DECEMBER 22, 2016 TABLE OF CONTENTS Page

I. Background and Experience.

II. Purpose ofTestimony.

Development ofthe Horowitz Survey .,4 A. Questionnaire Design. .. 5 B. Sample Design. 7 C. Survey Methodology. .,8 1. Screening and Qualification ofRespondents: .'8 2. Field House: ', D Survey Instrument. E. Sample Disposition 12, F. All Systems Disposition. 14

Survey Results I. I, 1,4 1. Assigned Value to Non-Network Programming Carried by Distant Signal Stations in 2010-2013 Royalty Years ...... l 114 2. Weighted Value to Non-Network Programming Carried by Distant Signal Stations in 2010-2013 Royalty Years ...... I, 116 3. What Most Influences Operators'ecisions Regarding What Distant Broadcast Stations to Carry: 16 4. Program Categories on Distant Stations that Were Featured in System's Advertising and Promotions: ll7 5. Respondents'erceived Importance to Subscribers ofProgram Categories on Distant Stations: 18

V. Conclusion. 21

APPENDIX A 23 APPENDIX B 41 APPENDIX C 55 APPENDIX D 62 I. Background and Experience

I am the Founder and ChiefExecutive Officer ofHorowitz Research, Inc. ("Horowitz Research"), a firm based in New York specializing in market research since its inception in 1985. Prior to forming Horowitz Research, I served as the Vice President of Cable Video Research Center, Opinion Research Corporation (division ofArthur D. Little) ("CVRC"). There, I engaged in consumer research design for the cable industry, including survey questionnaire development, data analysis and reporting, channel carriage, content and channel development. Also, at CVRC, I designed and implemented surveys on the viewership and value ofnew and emerging cable networks among both consumers and local cable system operators. Before that, I worked as an analyst at Dresner, Morris and Tortorello, where I focused on questionnaire development for political polling surveys and market research for new Pay TV services, program guides, and emerging cable networks. I received a BA. in Psychology and a M.A. in Political Science from New York University, graduating with honors and as a member ofPhi Beta Kappa. For over thirty years, I have advised many major media companies, including Multichannel Video Programming Distributors ("MVPDs"), cable systems, television networks, developers of interactive applications, Internet companies and other content providers to assess the impact on consumers ofnew technological developments and increased competition in the marketplace. Recently, my work has also focused on trade and consumer research in the areas of digital media, including broadband and mobile content, services and technologies. My research work in the industry has helped to support the evolution of programmed cable television in the United States. Among the highlights ofmy early work was the NCTA Segmentation Study, a first of its kind conducted in 1983. The project examined the market for cable television and the value ofnew cable channels in urban and suburban markets, which at the time was uncharted territory. Through this study, I helped label key audience segments that became standard in the industry, and helped define the target audience for such programming ventures as MTV, ESPN, The Discovery Channel, AN.: Networks and Nickelodeon, among other culture- entertainment-information networks. Horowitz Research provides research and consulting services to the television, cable, telecommunication, and broadband industries,!including studies for multiple system operators ("MSOs") and cable systems on customer satisfaction, customer acquisition and retention, charnel carriage, and ad'option'fdigital. technology — both at the residential and business levels., Our jMVPD clients hkvd included Bresnan Communications, Buckeye Cable System, Cablevision Systems, Comcast, ( ox Communications, Grande Communications., In,sight Communications, Medi.acorn, RCN, Time Warner, and Verizon. We also conduct vieveinj'~ n&arket potential, and program development research for numerous cable networks and broadcasters, Moreover, Horowitz Research is well-recognized for its research on urban, m.ult:icultural consumers, and Latino consumers in particular. In addition, Horowitz Research h.as developed a series of industry studies that are widely recognized on the market for multichannel~, broadband, and mobile services, and in particular, on adoption and use ofthese services in multicultural communities. These studies include State of(.able and Digital Media published 'since 1993; State ofCable and Digital Media Multicultural Edition, published since 1999 (and originally titled State ofBroadband Urban Markets):, Focus: Latino; Focus: APican America; Viewing the 'iewer: an In-Home Ethnography ofTelevision View~ers;~Multiplatform Content and Services; and Multiplatform Content and Services Multicultural Edition.

II. Purpose of Testimony

As I understand it, this proceeding concerns the allocation ofroyalties deposited with the Copyright Office by cable system operators ("CSOs"), for the royalty years, 2010, 2011, 2012, and 2013, for the privilege ofretransmitting broadcast television signals out oftheir local market areas. Also, I understand that the standard for allocating the royalties is the relative market value ofthe different types ofthe nonnetwork programming aired on the CSO-retransmitted signals in terms of its power to attract and retain subscribers.

In prior proceedings, the decision makers have relied on the results of surveys presented by Bortz Media and Sports Group, Inc. on behalf of the Joint Sports Claimants ("Bortz Survey"). That survey asked respondents to allocate a fixed budget amount among program categories specified in the survey questionnaire. The decision makers then relied on the results of the survey to approximate the allocation of royalties among the defined group of claimants that represented the program categories in the proceeding.

Our cable operator survey ("Horowitz Survey") was designed to carefully replicate the methods and procedures of the Bortz Survey that was done for the 2005 royalty year and presented during the 2004-2005 Cable Phase I Proceeding.'ur task was to improve upon the Bortz Survey by solving some of its information and category weaknesses that were noted by the Copyright Royalty Judges ("Judges") in the 2004-2005 Cable Phase I Proceeding.

Accordingly, my testimony will (1) describe the development ofthe Horowitz Survey; (2) provide the results ofthe Horowitz Survey; and (3) discuss the differences between the Horowitz Survey and the Bortz Survey and why those differences make the Horowitz Survey a better survey. Notwithstanding these improvements, it is my opinion that the Horowitz Survey is not a substitute for behavioral data such as viewing.

In sum, the Horowitz Survey sets itself apart from the Bortz Survey by providing greater clarity in the following respects:

~ Creating the separate "Other Sports" program category to distinguish between sports programming that should be categorized as Program Suppliers'such as NASCAR auto races, professional wrestling, and figure skating) versus sports programming that should be categorized as Joint Sports Claimants';

'he 2005 Bortz Survey was substantially the same cable operator survey as had been presented by the Joint Sports Claimants in prior proceedings. ~ Providing "warmup"'uestions that would elicit well-reasoned, non-reflexive responses;

~ Enhancing program category definitjions by providing representative examples for each program category;

~ Customizing the,survey questI.onnaire to focus only on content on distantly retransmitted signals actually carried by respondent's system (including taIloring questionnaires for CSOs that carry WGN-only, PBS-only, and Canadian-only stations as distant signals;

~ Repeatedly reminding the respondents ofthe distant signals that the CSO actually carried throughout the interview; and

~ Reminding respondent not to assi.gn any value to programs that are substituted for WGN's blacked out programming.

III. Development of the Horowitx Survey

The Horowitz Survey was designed to mirror the Bortz Survey, which purported to assess the relative market value of the non-network programming carried on distant signals from the point ofview of CSOs — the major premise being that (1) CSOs players a certain value on the programming carried on theke distant broadcast signals when making their decisions regarding whether or not to carry~them and (2) those CSOs can assess a proportion ofthat value to each ofthe relevant categories ofprogramming actually carried on each respective signal. While I played a principal role in the design, development and executIion ofthe Horowitz Survey, development ofthe Horovvitz Survey was a collaborative effort that also meshed the knowledge and experience of several ~ additional individuals, including Dr. Alan Rubin, Dr. Martin Frankel, Dr. Ellie Hakak, and MPAA staff, Marsha Kess.ler.

Dr. Hakak, an econometric consultant, has over 25 years of experience in economic analysis, quantitative market research, strategic planning, and competitive analysis in the private sector, government and. One major step we took in development ofthe Horowitz Survey was to address a significant concern with the field implementation ofthe Bortz Survey. Over the past 10 to 15 years, management of cable system operations has become much more centralized, moving the locus of decision-making about what signals to carry further away f'rom local systems and markets to regional and even national headquarters. Notwithstanding these changes, we set about a process by which we located the nexus of decision making and identified - by title and ultimately by self-qualification — the executive with the decision- making authority over the carnage ofdistant broadcast signals for each CSO in our sample. As you see in our specification ofthe sample in this report, some individual executives had decision making authority to select carriage of signals for multiple cable systems in our sample.

A. Questionnaire Design

Our approach to the questionnaire design was to bring greater clarity to the questions which we believed would result in more accurate responses from the CSOs responding to the Horowitz Survey. In this vein, we corrected what we believed were several flaws in the Bortz Survey. One such flaw was the Bortz Survey's omission of a major program category, Other Sports. Other Sports refers to non-team sports, such as NASCAR racing, golf, figure skating, and horse racing that fall within the Program Suppliers category that I understand the Judges established for this proceeding. Other Sports programming is different than, and should be distinguished from, the program category Live Coverage ofProfessional and College Team Sports, which is claimed by Joint Sports Claimants. This correction was necessary to allow respondents to distinguish and value these two key sports categories separately. It is clear in the Horowitz Survey results that correcting for this crucial difference affected the allocations.

academia. Dr. Frankel is an expert statistician with extensive professional experience statistics, and especially with sample design. His background and experience are detailed in his testimony in this proceeding. Dr. Rubin is a retired Professor Emeritus and Director Emeritus ofthe School of Communications at Kent University. Ms. Kessler was MPAA's Vice-President, Retransmission Royalty Distribution, for 28 years prior to her retirement in 2010. %bile at MPAA, her responsibilities included overseeing the distribution of cable and satellite statutory license royalties to MPAA-represented claimants. Both Dr. Rubin and Ms. Kessler have testified in prior royalty distribution proceedings before the Judges. See, e.g., Docket No. 2007-3 CRB CD 2004-2005, Written Direct Statement Of Program Suppliers, Direct Testimony ofAlan M. Rubin, Ph.D. and Direct Testimony ofMarsha E. Kessler (filed June 1, 2009). Equally important, we provided the respondent with. a frame ofreference with which to answer the allocation question. The warm-up questions in the early part ofthe i Horowitz Survey questionnaire set up a well-reasoned, nion~reflexive allocation ofvalue by first asking the respondent to assess the importance of each program category to their subscribers. The questionnaire then asks about the advertising/promotional usage ofeach category, and asks the respondent to assess the importanCe 5f consideration of subscribers to decisions about channels. Only after asking the warm-up questions did we ask the respondents to allocate a relative dollar value to the program categories.

In addition, we clarified the definition ofthe content ofeach program category in the Horowitz Survey by providing examples ofrepresentative programs for the about'he categories. This is particularly important to insure that respondents reasoned well distinction among various categories instead ofreflexively assigning values to the wrong categories. For example, a respondent could confuse non-network sports content with network sports content (which is not compensable in these proceedings). Another example would be a respondent confusing Other Sports content, which falls in the Program Suppliers category, with Live Coverage ofProfessional and College Team 'ports, which falls in the Joint Sports Claimants category. We think the Horowitz Survey's use ofproper categorization, relevant and probingiintroductory questioning, and incorporation ofprogramming examples makes for a major improvement over the Bortz Survey.

Further, we customized and streamlined each Horowitz Survey questionnaire to ensure that the questionnaire focused only on content carried on the complement of signals carried by the respondent's system. To dol so) wh divided the cable systems into five carriage groups based on the types of distant broadcast:signals carried by the systems:

~ Canadian Only (Group A): systems that carry'only C'anadian stations as distant signals; ~ PBS Only (Group B): systems that carry only ~PBS stations as distant signals; ~ WGN Only (Group C): systems that carry only WGN stations as distant signals; ~ Non-Network Affiliated Stations (Group D): systems that carry only independent (i.e., non-network affiliated) stations as distant signals; ~ Network Affiliated and Other Stations (Group E): systems that carry network- affiliated stations, only or in combination with other type of stations, as distant signals.

We customized and automated questionnaires for each ofthe above carriage groups such that each questionnaire focused only on programming offered on the stations within the carriage group ofthe responding system. To commence the interview, the interviewer would call up, by computer, information about the particular system to be interviewed. The computer responded with a pre-programmed survey questionnaire for that particular cable system with the questions tailored specifically for the programming offered by the stations carried by that system. For example, a respondent on behalf of a system carrying PBS-only signals, would not be asked questions about categories of programs other than PBS because the questionnaire would focus only on programming offered by PBS stations. This system helped to avoid potential interviewer error in selecting the correct version ofthe questionnaire. 8. Sample Design

Cable Data Corporation ("CDC") provided detailed information about the universe ofthe Form 3 cable systems that retransmitted at least one distant signal during each royalty year in question. That detail included cable system name, location ofthe cable system, counties corresponding to the geographical footprint ofthe cable system, relevant distant signals carried by that cable system, city and state of origin ofthose distant signals, and the affiliation status of each distant signal ("Cable System Detail"). Dr. Frankel selected a random sample of 300 cable systems for each ofthe 2010-2013 royalty years Rom the universe ofForm 3 cable systems, and provided those samples to Horowitz Research. I understand that the details on the parameters and procedures of sample selection are laid out in Dr. Frankel's testimony which is also being presented in this proceeding. In an effort to increase the number of successfully completed interviews, Horowitz Research worked simultaneously with the random samples selected by Dr. Frankel and the broader uni.verse of all systems that carried distant signals in the specific royalty year (798 systems in 2013, 815 systems in 2012, 913 systems in 2011, and 1003 systems in 2010).

C. Survey Methodology

1. Screeni~n and Qgalification of Res~onclents:

Horowitz.Research properly screened and qualified respondents. To qualify, the CSO executives reached had to affirm that they were "the person at their cable system most responsible for programming decisions made by their system" in the particular 'oyalty year in questi.on.

Horowitz.Research employed the services of Sue Panzer, a cable industry executive, to qualify executives as truly being the person responsible for programmiitig decisions at the speci.fic system level.'s. Panzer reached the CSO executives, screened them and scheduled calls for the field house. Respondents who were pre-screened were not made aware of th.e topic of'the survey prior to being called by the field house. In each ofthe four surveys, we found that for most MSOs, channel carriage deciIsions were made at the national level, not at the local level.

As noted, some respondents had authority to select carriage of signals for multiple cable systems in our sample. For those respondents, he/she was only asked to respond to one survey for all the systems with the same channels. Otherwise, we conducted separate surveys.

2. Field House:

Horowitz.Research engaged the services of Global Marketing Research Services, Inc., ("GMRS"), a field house with over 20 years of experience, to conduct the interviews. Two trained executive interviewers conducted the interviews. Each interview was approximately 15-minutes long, and the interviewer offered,$ 125

's. Panzer's curriculum vitae attached hereto as Appendix D. honorarium to the respondent for participating in the survey. GMRS was not informed as to the identity ofthe client or about the purpose of this research. The surveys were fielded over a period of about four to five months, respectively, for each royalty year, as follows:

Field Period

June 1, 2011 - October 31, 2011 2010

May 1, 2012 - September 21, 2012 2011 June 3, 2013 - September 30, 2013 2012 May 19, 2014 - September 24, 2014

Horowitz Research directed GMRS to try each number up to 30 times to reach a respondent. However, once reached, executives were not re-contacted more than 3 times before being counted as a refusal. D. Survey Instrument

The survey questionnaire is substantially the same for all survey/royalty years. It begins with screening questions to qualify respondents so that the executive interviewer can verify that the respondent is the person at the cable company "most responsible for programming decisions" made by or for the cable system in the particular royalty year in question.

Once the respondent is qualified, the questionnaire next hones in on the specific types of programming that is of interest to the interviewer — non-network programming on the distant broadcast stations carried by the cable system during the relevant royalty year. The questionnaire informs the respondent that "distant broadcast stations," mean free, over-the-air, V.S. commercial and non-commercial TV stations imported from an outside market or markets by the respondent's system. The questionnaire identifies the actual distant signal(s) — call letter and city — carried by the respondent's system and repeats the signal call letters throughout the questionnaire.

" The complement of distant signals carried by the system and each signal's programming information varied from year to year. The questionnaire repeats the list of program categories and the distant stations carried by the respondent's system multiple times throughout as follows:

~ News and Community Events broadcast on (STATION(S)).

~ Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality i shows, game shows, and other series broadcast on (STATION(S)). Examples. include programs such as Everybody Loves Raymond, Seinfeld, American Idol, Jeopardy, and The Dr. Oz Show. [30 Rock, Adelante Chicago, Everybody Loves Raymond, and People To People for WGN Only isystems]: ~ Movies such as feature films, Movies oftble %'eek, And:specials broadcast on~ (STATION(S)). Examples include movies 'such ah Pirates ofthe Caribbean: The Curse ofthe Black Pearl, Crash, Ghostbusters, The Matrix, X-.Men: The Last

Stand, Signs, and Girl, Interrupted. [Gladiator, The Lord OfThe Rings: The i Return ofthe King, and Home Alone 2: Lost ~ln New York for WGN Only systems] ~ Live, play-by-play coverage ofprofessional and college team sports broadcast on (STATION(S)). Examples include Major League.Baseball, NBA professional

basketball, NFL professional football, NHL professi'onal hockey, NCAA college I

football and basketball, and Major League l Sobcei. [Major League Baseball, and National Basketball Association for WGNiOnly systems.]: ~ Other sports programming broadcast on (STATION(S)). Examples include NASCA'R auto races, professional wrestling, and figure skating broadcasts. [Horse Racing for WGN Only systems.] ~ Devotional programs such as shows with religious themes or religious taIk shows broadcast on (STATION(S)). Examples include Kenneth Copeland Ministries, Creflo Dollar, 700 Club, and Joel Osteen Ministry. [Tomorrow's World and: Creflo Dollar for WGN Only systems.] ~ Programs broadcast only on PBS station (STATICN(S)). Examples include Masterpiece Classic: Downton Abbey Season III, Masterpiece Mystery!, PBS NewsHour, and Sesame Street.

10 ~ Programs broadcast only on Canadian stations (STATION(S)). Examples include Steven & Chris, The Social, Coronation Street, Busytown Mysteries, and CBC News.

The questionnaire text and examples above pertain to the 2013 royalty year, a complete a copy ofwhich is attached as Appendix A. We updated programming examples every year, and no other changes beyond programming examples were made to the questionnaire year-to-year. As I stated earlier, the questionnaire included instructions on what questions, items and language should be used for each ofthe system categories. The field house used these instructions to program the survey for each ofthe five categories/groups. Horowitz Research staff checked to ensure the accuracy ofthe programming.

Additionally, the Horowitz Survey features introductory questions to focus respondents on the specific types ofprogramming and stations and on the value these might have to their system. The introductory questions are as follows:

~ "In considering what distant broadcast stations to carry, what would you say most influences your decision: programming that is important to your cable system, programming that is popular and important to your current or potential subscribers, or some other matter or consideration I have not mentioned?"

~ "In your advertising and promotional efforts in [royalty year], did your cable system feature any programs actually broadcast by the distant stations we have been talking about?"; "Which ofthe following program categories actually broadcast by the distant television stations we have been talking about did you feature in your advertising and promotions in [royalty year]?"

~ "We have been discussing the several types ofprogramming on your distant

television station(s). Using a 1 to 5 scale, where "1" means "not important at all," and "5" means "very important," please tell me how important you think each of the following types ofprogramming on these distant television stations is to your subscribers."

11

In the last and core question ofthe Horowitz Survey (the "payoff" question), the respondent is asked to assume she has a fixed dollar amount to allocate for the non- network programming broadcast during the royalty year in question on the distant signal'tations carried and, considering the value of each program category to her cable system, to tell us what percentage, if any, ofthe fixed dollar amount: she would allocate for each i program category.

For operators of systems that carry network-affiliated distant signals, the Horowitz Survey emphasizes that by "non-network programs," it means programs on these stations that are not distributed by the ABC~ NBC~ os CBS television: networks. In addition, the survey directs respondents responding on behalf of.CSOs that carry WGN distant signals not to assign any value to programs that are substituted for WGN's blacked out programming.

K. Sample Disposition

Table 1.1 below shows the total number of responding cable systems for each royalty year and the total number ofexecutives responding On behalf ofthose systems.

SCABS NcSI4 ea5em 1'EPll@VCC6'013 gill) Qgfil&i(Q '200 38

2012 , 228, 42 .

2011 , 174, 4S

2010 ,136, 34 . AVERAGE 184.5 40.75

For each of the 2010-2013 royalty years, the:dispositions of our attempts to contact the persons responsible for programming at cable systems. in the samples are presented in the following series oftables:

12 Completed surveys 200 Refusals Not able to reach (voice mails, etc.) 45 No/inaccurate contact information 50 Total 4 of systems in the sample 300

I ~

Completed surveys 228 Refusals 14

Not able to reach (voice mails, etc.) 28 No/inaccurate contact information 30 Total 0 of systems in the sample 300

I ~

Completed surveys 174 Refusals

Not able to reach (voice mails, etc.) 83 No/inaccurate contact information 35 Total 0 of systems in the sample 300 Completed surveys 136 Refusals 18 Not able to reach (voicemails, etc.) No/inaccurate contact information Total 0 of systems in the sample 300

Tables 10.1 — 10.4 in Appendix C show the dispositions of our attempts to contact the persons responsible for programming at all cable systems for each ofthe 2010-2013 royalty years. F. All Systems Disposition

In Appendix B, Tables 7.1 — 7.4, for all systems surveyed, I show the number of respondents and the corresponding number of systems for which they provided responses for each survey year. Also, in Appendix B, Tables S&1 & 8.4, for all systems surveyed, I show the titles ofthe respondents surveyed and the corresponding number of cable systems whose executives bore the titles. Also, we compared how all cable systems surveyed were distributed across the 5 carriage groups detailed in:section III.A., and how the cable systems in the Form 3 universe were distributed across the same 5 carriage groups. The results show that both sets ofcable systems were very similarly distributed across all 5 carriage groups. See Appendix B Tables '9.1' 9.4.:

IV. Survey Results

1. Assigned Value to Non-Network Proerammine Carried bv Distant Sienal Stations in 2010-2013 Rovaltv Years

The payoff question in the Horowitz Survey asks each cable operator respondent to assume he or she had a fixed dollar amount to allocate for the non-network programming actually broadcast during the 2010-2013 royalty:years, respectively, on the distant signal stations carried and, to consider the value of each type ofprogramming

(i.e., identified program categories) to their cable system. The Horowitz Survey further I

14 asks the respondent to show what percentage allocations she would make, if any, ofthe fixed dollar amount for each type ofprogramming. The wording ofthis question replicates the question in the Bortz Survey, with two differences: (1) we included a programming category not included in the Bortz Survey: "other sports programming" (other than live professional and college team sports), such as NASCAR races, professional wrestling, and figure skating broadcasts, and (2) we included examples for each programming category for further clarification (please see section III.D.). The results of the value question are as follows:

I ~ - ~ ~ ~ ~ . s ~ ~ ~ ~ ~ - . ~ ~ ~ - ~ ~ -o ~ ~ II - ~ ~ ~

¹ of Completed Surveys 200 228 174 136 News and community events 12% 15% 12% 15% Syndicated series 19% 18% 18% 17% Movies 13% 12% 12% 'l5% and Live coverage of professional college 27% 28% 31% teams sports Other sports programming 9% 9% 12% 8% Devotional programming 4% 6% 4% 4% on PBS Programs broadcast only 12% 11% 12% 9% stations broadcast only on Canadian Programs 1% 1% 1% 0% stations PROGRAM SUPPLIERS TOTAL 41% 39% 40% 40%

The results from the 300 system sample are consistent with the results Rom the all systems surveyed. See Appendix B, Table 11.

15 2. Weighted Value to Non-Network Proerammine Carried bv Distant Sienal Stations in 2010-2013 Rovaltv Years

Dr. Martin Frankel used the Horowitz Survey responses to provide a weighted

estimate ofthe percent dollar allocation that all cable systems would assign to the

programming categories. Dr. Frankel's weighted results are provided below, and are also

reported in Dr. Frankel's separate written direct testhmoniy m this proceeding.

IRRt MR% a ~ News and community events 9.54% 15.72% 12.85% 12.38% Syndicated series 16.26% 15.97% 17.56% 20.28% Movies 12.39% 12.14% 11.43% 17.15% Live of professional and coverage college 35.28% '25.'50% 27.13% 31.94% teams sports

Other sports programming 7.40% 9.02% 10.80% 6.77% Devotional programming 3 48% 5 74% 5.92% 3.78% Programs broadcast only on PBS station 15.39% '15'.05% 13.31% 7.69%

broadcast on Canadian ' Programs only 0.35% 0.87% 1.00% 0.01% stations PROGRAM SUPPLIERS TOTAL 36 05' '37.'3% 3g. 79% 44.2Q%

3. What Most Influences Operators'ecisions Reeardine What Distant Broadcast Stations to Carrv:

The Horowitz Survey also asked respondents about the factors that most influence

decisions regarding what distant broadcast systems to cavy. As shown in the table

below, programming that is popular and importanti to cumenit and potential subscribers

overwhelmingly influences the systems'ecisions. to carry distant signals. ~ I ~ - -e . ~ -e ~ ~ II

0 of Systems Surveyed 228 /74 736 to Programming Popular and Important 84% 80% 79% 85% Current and Potential Subscribers

to the Cable Programming Important 10% 11% 35% 4% System

Other 10% 13% 12% 14%

4. Pro ram Cate pries on Distant Stations that Were Featured in S stem's Advertisin and Promotions:

As shown in the table below, only three (3%) and four percent (4%) ofthe systems surveyed featured programs carried by the distant stations in their advertising and promotional efforts during 2013 and 2012, respectively. The data &om the 2011 and

2010 royalty years axe higher. ln general, across all years, most CSOs did not include program categories on distant stations in their advertising and promotional efforts. The following table portrays the data on the program categories featured in advertising and promotion:

17 ~ .l. M I 0 of Systems Surveyed '200 '28:: 174 136 Any (Net) 3% 4% 29% 9%

News and community events &1% 7% 1%

Syndicated series (1% 2% 2% 1% Movies &1% 5% 1% Live coverage of professional and college teams ,'ports '1% 2% 25% 7%

Other sports programming 2% 2%

Devotional programming : 1% Programs broadcast only on PBS station 2% Programs broadcast only on Canadian stations N/A No programs broadcast by distant stations featured '7% 93% 64% 88%

5. Respondents'erceived Importance to Subscribers ofi Pro@ram Cateeories on Distant Stations:

Ratings ofhow important the respondents thought each program category was to i their subscribers is provided in the following series oftables:

18 ~ ~ 0 ~ ~ ~ 0 ~ ~ ~ ~ 0

~ ~ ~ ~ ~ ~ I ~ o ~ ~

~ ~ ~ 0 ~ ~ ~ ~ 0 ~ ~ ~

News and community events 31% 10% 59% Syndicated series 28% 39% 33% Movies 26% 42% 32% Live coverage of professional and college teams 64% 33% 3% sports Other sports programming 22% 29% 49% Devotional programming 5% 9% 86% Programs broadcast only on PBS station 34% 40% 26% Programs broadcast only on Canadian stations 38% 25% 38%

~ ~ ~ ~ 4 ~ I ~ 0 ~ ~ 0 ~

~ ~ ~ I ~ a I .. ~ I ~ ~

0 ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~

News and community events 29% 21% 50% Syndicated series 37% 30% 33% Movies 21% 29% 50% Live coverage of professional and college teams 73% 12% 15% sports Other sports programming 31% 33% 37% Devotional programming 8% 15% 77% Programs broadcast only on PBS station 35% 3?% 26% Programs broadcast only on Canadian stations 58% 29% 14%

19 ~ ~ 4 0 0 0 ~ . e. ~ o o II .. 0 I II e 0 0 0 0 ~ ,'

~ ~ ~ 0 0 4 l ~ ~

~ ~ ~ 0

News and community events 28'/o 25% 4?% Syndicated series 33o/o 40% 2'?% Movies 19o/o 34% 4?o/. Live coverage of professional ancl college teams 84o/o 8% 90/ sports Other sports programming 32'/o 26% 42% Devotional programming 6% 9% 85% Programs broadcast only on PBS statioln 49% 18% 33% Programs broadcast only on Canadian stations 50'/o 25% 25%

"~ ~ 0 0 0 0 0 0 ~ 0 +,pg e I .. 0 I I o e 0 0 ~ '

~ ~ 0 I' ~ 4 ~ 4 0 ~

News and community events 31% 18% 51I o/o Syndicated series 30% 26% 44% Movies 47% 36% Live coverage of professional and college teams 85% 6% sports 9'%fo/o Other sports proclrarnming 19% 54'/o Devotional programming 6% 4% 9;1% Programs broadcast only on PBS station 28% 38% 26% Programs broadcast only on Canadian stations 100%

20 V. Conclusion

The premise ofthe Bortz Survey is that how local or regional CSO executives allocate their programming budget reflects their proportional assessment of the relative market value of the categories of programming actually carried by the distant signals on their respective systems. In designing the survey, our task was to improve upon the Bortz Survey by solving some ofthe information and category weaknesses as pointed out by the Judges in the 2004-2005 Cable Phase I Proceeding. The improvements we made included, (a) creating the separate "Other Sports" category to allow respondents to distinguish and value this category separately f'rom Live Coverage of Professional and College Team Sports, (b) providing "warmup" questions that would elicit well-reasoned, non-reflexive responses; and (c) enhancing program category definitions with examples and customizing the survey questionnaire to focus only on content on distantly retransmitted signals actually carried by respondent's system, among others. We subjected both the implementation and the content ofthe Bortz Survey to constructive criticism, corrected its flaws and greatly improved upon it in several important ways.

The evidence strongly indicates that the Bortz Survey results would be different and in line with the Horowitz Survey findings ifthe Bortz Survey included a separate "other sports" category and offered programming examples. The different results would have important implications for the proceedings and findings ofthe Judges.

It became apparent from the implementation ofthis research that, for most MSOs, channel carriage decisions are not made at the local level, which impacts a major underlying premise of the Bortz Survey as a direct measure ofvalue. Therefore, in my view, the reasoned judgment of the cable system executives surveyed about the allocation oftheir budget among the program categories carried to subscribers living in distant markets is not a substitute for behavioral data about subscribers. While past decision makers have relied on results of cable operator surveys as proxies for relative market value ofthe programming categories for attracting and retaining subscribers, the surveys remain attempts to ascertain market value indirectly through the perceptions ofthe buyer

21 (Le., the cable system) with his or her subscribers in mind, instead ofthrough direct observation ofthe actual subscribers. The buyer or decision-maker in these operator surveys is an intermediary whose purpose is to bring subscribers the value that they actually pay for when subscribing to cable service. Done well, such a survey may illuminate the criterion by which to allocate royaltIiesl

Still, whatever the reasoned judgment of executives 'about the value ofthe program categories carried on distant signals to their subscribers, any cable operator survey should not be considered a substitute for behavioral data on viewing. Viewmg is both a reflection of and, I believe, the real, direct measure ofvalue. Viewing by subscribers is what the operator pays for when they carry a network or broadcast station& and viewing is what subscribers/customers pay for when buying TV services/channels. l Thus, in my opinion, viewing should determine the allocation ofroyalties among claimants.

22 Horowitz Research

APPENDIX A — Questionnaire

2013 Cable Operator Questionnaire

NAME: (06-08) (09-11) (12-15) TIME STARTED: CITY AND STATE: (16-17) TELEPHONEN: / / (Area Exchange LENGTH (MINUTES): Code Number) RECORD SEX: (18) INTERVIEW'ER: MALE 1 FEMALE 2 DATE: TIME ENDED: SYSTEM

DISTANT STATIONS CARRIED (FROMSAMPLE): STATION: CITY.

INTRODUCTION

(IF PERSON NAMED ON SAMPLE IS NOT AVAILABLE, ASK TO SPEAK WITH (IN THIS ORDER): GENERAL MANAGER, PROGRAMMING DIRECTOR, OR THE MARINATING DIRECTOR. IF NONE IS AVA1LABLB, SCHEDULE A CALL BACK WHEN GENERAL MANAGER IS AVAILABLB). (WHEN PERSON NAMED IN SAMPLE, OR GENERAL MANAGER, PROGRAIMI'.vIING DIRECTOR, OR MARKETING DIRECTOR IS REACHED, RECORD NAME AND TITLE ABOVE, THEN).

Good morning/afternoon. My name is and I'm calling &om Horowitz Research, a well- known survey research company in the cable industry. We are conducting a national survey among randomly selected cable systems regarding the programming on distant signals they carry. This survey will take approximately 15 minutes and we are offering an honorarium of $ 125 to participants. (ONLY IF PERSON VOLUNTEERS THAT HE/SHE CANNOT OR DOES NOT WANT TO ACCEPT HONORARIUM, OFFER TO MAKE DONATION TO A CHARITY OF HIS/HER CHOICE IN HIS/HBR NAME).

All responses will be held in strict confidence; only aggregate data are reported. May we talk now? IF NOT, SCHEDULE A TIME TO CALL BACK.

23

1. First, are you the person at your cable system most responsible for programming decisions made by your system in 2013?

1. Yes 2. No 3. DON'T KNOW

IF YES, PROCEED TO Q.2. IF NO/DON'T KNOW, Could you please direct me to the person at your cable system most responsible for programming decisions in 2013? REPEAT INTRODUCTION AND Q.l. NON-NETWORK CARRYING SYSTEMS: 'e CANADIAN ONLY SPREADSHEET (GROUP A) PBS ONLY SPREADSHEET (GROUP B) WGN ONLY SPREADSHEET (GROUP C) OTHER NON-NETWORK SPREADSHEET (GROUP D)

NETWORK CARRYING SYSTEMS: NETWORK SYSTEMS SPREADSHEET (GROUP E)

2. We are interested specifically in [NETWORK CARRYING; SYSTEMS (E): non-nehvork] programming on the distant broadcast stations you carried during 2013. [NETWORK CARRYING SYSTEMS (E): By "non-nehvorkprogramming," wd mkan Progmnming other than national network programming coming from ABC, CBS, and NBC networks.:] By "distant broadcast 'stations," mean free, over-the-air, U.S. commercial and non-eordmeicial TV stations imported from ani outside market or markets by your system.

Industry data show that during 2013 your system carried the following distant broadcast statilonsl: PNSERT STATION(S) 4 CITIES FROM LIST]

CALL LETTERS CITY (COLUMN J ON SAMPLE) (COLUMN L ON SAMPLE)

I will be asking you about the [NETWORK CARRYING SYSTEMS (E): non-network] programs on those distant stations.

In considering what distant broadcast stations to carry,'hat would you say most influences your

decision: programming that is important to your cable system, programming that is popular and I important to your current or potential subscribers, or some other matter or consideration I have not '. mentioned? (MULTIPLE RESPONSES ACCEPTED)

1. Programming that is important to your cable system, 2. Programmmg that is popular and important to your current or potential subscribers 3. Or some other matter I have not mentioned (Please specify ) 4. DON'T KNOW

24 3. Now I want to ask you some questions specifically about the [NETWORK CARRYING SYSTEMS (E): non-network] programming actually carried on PNSERT STATION(S) 4 CITIES FROM SAMPLE — COLUMNS J AND I; WGN ONLY SYSTEMS (C): ONLY REFER TO STATION, NO CITY REFERENCE NEEDED]. In your advertising and promotional efforts in 2013, did your cable system feature any programs actually broadcast by the distant stations we have been talking about? [NETWORK CARRYING SYSTEMS (E): Again, by "non-networkprograms," I mean programs that were not broadcast by the ABC, CBS, or NBC television networks.]

1. Yes - PROCEED TO Q.4 2. No - SKIP TO Q.S DON'T KNOW/REFUSED - SKIP TO Q.5

4. (IF YES IN Q.3) Which ofthe following program categories actually broadcast by the distant television stations we have been talking about did you feature in your advertising and promotions in 2013? [NETWORK CARRYING SYSTEMS (E): Again, do not consider programs on these'Y'OR stations coming &om the ABC, NBC, or CBS television networks.] (READ EACH PROGRAMMING CATEGORY FROM THE LIST BELOW). INTERVIEWER: RECORD YES, 'N'OR NO, OR 'DK'OR DON'T KNOW (READ ENTIRE LIST; MULTIPLE RESPONSES ACCEPTED)

(ROTATE 'a'HROUGH 'h'N RANDOM ORDER — SEE NOTE FOR ITEMS'd'ND 'e')

LIST FOR GROUP D (NON-NETWORK CARRYING SYSTEMS, EXCLUDING WGN ONLY SYSTEMS): GROUP A — ONLY ASK ITEM H GROUP B — ONLY ASK ITEM G a) News and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as Everybody Loves Raymond, Seinfeld, American Idol, Jeopardy, and The Dr. Oz Show. c) Movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Pirates of the Caribbean: The Curse of the Black Pearl, Crash, Ghostbusters, The Matrix, X-Men: The Last Stand, Signs, and Girl, Interrupted.

25 ROTATE 'd'ND 'e'OGETHER AND ALWAYS RBAD 'e'FTER 'd'.

d) Live, play-by-play coverage ofprofessional and college team sports broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]).. Examples include Major League Baseball, NBA professional:basketball, NFL professional football, NHL professional hockey, NCAA college football and basketball, and Major League Soccer. e) Other sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]).

Examples include NASCAR auto races, professional wrestling, and Ggure skating broadcasts. I

Devotional programs such as shows with religious themes or religious talk shows broadcast. on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Kenneth Copeland Mimstries, Creflo Dollar, 700 Club, and Joel Osteen Ministry.

g) ONLY ASK TO SYSTEMS THA.T CARRY PBS STATION — GROUP B, OR GROUPS '

AND E WITH 1 FOR COLUMN R: ProIIrarhs broadcast only on PBS station (INSERT PBS STATIONS (IF ANY) FROM THE LIST. [PBS = 1 IN COLUMN R]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP iIF THERE ARE NO PBS STATIONS ON LIST).

Examples include Masterpiece Classic: Downton Abbey Season III, Masterpiece Mystery!, I PBS NewsHour, and Sesame Street. h) ONLY ASK TO SYSTEMS THAT CARRY A iCANADIAN STATION — GROUP A, OR GROUPS D AND E WITH 1 FOR COLUMN S: Programs broadcast only on Canadian stations (INSERT CANADIAN STATIONS (IF ANY) FROM LIST. [PBS ='1 IN

COLUMN S]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARB'NO CANADIAN STATIONS ON LIST). Examples include Steven Ec Chris, The Social, Coronation Street, Busytown Mysteries, and CBC News.

LIST FOR GROUP C (WGN ONLY SYSTEMS): a) News and Community Events broadcast on (INSERT. STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS = [CANADIAN 1 IN COLUMN S]). 'nd b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as 30 Rock, Adelante Chicago, Everybody Loves Raymond, People To People. c) Movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Exmnples include movies such as Gladiator, The Lord Of The Rings: The Return ofthe King, and Home Alone 2: Lost In New York. ROTATE 'd'ND 'e'OGETHER AND ALWAYS READ 'e'FTER 'd'.

d) Live, play-by-play coverage of professional and college team sports broadcast on (INSERT = STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Major League Baseball, and National Basketball Association.

e) Other sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Horse Racing.

Devotional programs such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN = COL~ R] AND CANADIAN STATIONS [CANADIAN 1 IN COLM/K S]), Examples include Tomorrow's World and Creflo Dollar.

g) OMITTED h) OMITTED

LIST FOR GROUP K (NETWORK CARRYING SYSTEMS)." a) Non-network News and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]).

b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as Everybody Loves Raymond, Seinfeld, American Idol, Jeopardy, and The Dr. Oz Show. c) Non-network movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Pirates of the Caribbean: The Curse of the Black Pearl, Crash, Ghostbusters, The Matrix, X-Men: The Last Stand, Signs, and Girl, Interrupted.

27 ROTATE 'd'ND 'e'OGETHER AND ALWAYS READ 'e" AFTER 'd'.

d) Non-network live, play-by-play coverage of professional and college team sports broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN

COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Major League Baseball, NBA professional basketball, NFL professional football, NHL professional hockey, NCAA college football and basketball, and Major League Soccer. Li'ST,'UT e) Other Non-network sports programming broadcast on (INSERT STATIONS:FROM EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]).

Examples include NASCAR auto races, professional wrestling, and figure skating broadcasts. i f) Non-network Devotional programs such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST,~BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Kenneth Copeland Ministries, Creflo Dollar, 700 Club, and Joel Osteen Ministry. g) ONLY ASK TO SYSTEMS THAT CARRY PSS STATION — GROUP 3, OR GROUPS D AND E WITH 1 FOR COLUMN R: Programs broadcast only on PBS station (INSERT PBS STATIONS (IF ANY) FROM THE LIST. [PBS = 1 IN COLUMN R]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP .IF THERE ARE NO PBS STATIONS ON LIST). Examples include Masterpiece Classic: Downton Abbey Season III, Masteipiece Mystery!, PBS NewsHour, and Sesame Street. h) ONLY ASK TO SYSTEMS THAT CARRY A CANADIAN STATION — GROUP A, OR GROUPS D AND E WITH 1 FOR COLUMN S: Programs broadcast only on Canadian stations (INSERT CANADIAN STATIONS (IF ANY) FROM LIST. [PBS =l 1 IN COLUMN S]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO CANADIAN STATIONS ON LIST). Examples include Steven 8'c Chris, The Social,'oronatian Street, Busytown Mysteries, and CBC News.

28 5. (ASK EVERYONE) We have been discussing the several types of programming, on your distant "1" "5" television station(s). Using a 1 to 5 scale, where means "not important at all," and means "very important," please tell me how important you think each ofthe following types of programming on these distant television stations is to your subscribers. [NETWORK CARRYING SYSTEMS (E): In your answers, please include only non-network programs on these distant broadcast stations. By "non-networkprograms," I mean programs on these stations that are not coming &om the ABC, NBC, or CBS television networks.] How important would you say (READ ITEM) is to your subscribers? 1,2,3,4 or 5? DON'T KNOW/ NOT IMPORTANT AT ALL VERY IMPORTANT REFUSED 1 2 3 4 5

(ROTATE 'a'HROUGH 'h'N RANDOM ORDER — SEE NOTE FOR ITEMS'd'ND 'e')

LIST FOR GROUP D (NON-NETWORK CARRYING SYSTEMS, EXCLUDING WGN ONLY SYSTEMS): GROUP A — ONLY ASK ITEM H GROUP B — ONLY ASK ITEM G a) News and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). b) Syndicated series such as sitcoms, drama, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as Everybody Loves Raymond, Seinfeld, American Idol, Jeopardy, and The Dr. Oz Show. c) Movies such as feature 61ms, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Pirates of the Caribbean: The Curse of the Black Pearl, Crash, Ghostbusters, The Matrix, X-Men: The Last Stand, Signs, and Girl, Interrupted. ROTATE 'd'ND 'e'OGETHER AND ALWAYS READ 'e'FTER 'd'.

d) Live, play-by-play coverage ofprofessional and college team sports broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Major League Baseball, NBA professional basketball, NFL professional football, NHL professional hockey, NCAA college football and basketball, and Major League Soccer.

29 e) Other sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN CGLUIMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include NASCAR auto races, professional wrestling, and figure: skating broadcasts..

Devotional programs such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANAl3IAN = 1 IN COLUMN S]). Examples include Kenneth Copeland Ministries, Creflo Dollar, 700 Club, and Joel Osteen Ministry.

g) ONLY ASK TO SYSTEMS THAT CARRY PBS STATION — GROUP B, OR GROUPS D AND E WITH 1 FOR COLUMN R: Programs broadcast only on PBS station (INSBRV

PBS STATIONS (IF ANY) FROM THE LIST,. [PBS = 1 IN COLUMN R]. DO NOT I INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO PBS STATIONS ON LIST). Examples include Masterpiece Classic: Downton Abbey Season III, Masterpiece Mystery!, PBS NewsHour, and Sesame Street. h) ONLY ASK TO SYSTEMS THAT CARRY A'CA'NA'DIAN STATION — GROUP A, OR GROUPS D AND E WITH 1 FOR COLUMN S: Programs broadcast only on

Canadian stations (INSERT CANADIAN STATIONS gF ANY) FROM LIST. [PBS =~ 1 IN ~ COLUMN S]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO CANADIAN STATIONS ON LIST). Examples include Steven & Chris, The Social,~ Coronation Street, Busytown Mysteri~es, and~ CBC News.

LIST FOR GROUP C (WGN ONLY SYSTEMS): a) News and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN:STATIONS [CANADIAN = 1 IN COLUMN S]). b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as 30 Rock, Adelante Chicago, Everybody Loves Raymond, and People To People. c) Movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Gladiator, The I.ord Of The Rings: The Return ofthe King, and Home Alone 2: Lost In New York

30 ROTATE 'd'ND 'e'OGETHER AND ALWAYS READ 'e'FTER 'd'.

d) Live, play-by-play coverage ofprofessional and college team sports broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Major League Baseball, and National Basketball Association.

e) Other sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Horse Racing.

f) Devotional programs such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Tomorrow's World and Creflo Dollar.

g) OMITTED h) OMITTED

LIST FOR GROUP E (NETWORK CARRYING SYSTEMS): a) Non-network news and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 1N COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as Everybody Loves Raymond, Seinfeld, American Idol, Jeopardy, and The Dr. Oz Show. c) Non-network movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Pirates of the Caribbean: The Curse of the Black Pearl, Crash, Ghostbusters, The Matrix, X-Men: The Last Stand, Signs, and Girl, Interrupted.

31 ROTATE 'd'ND 'e" TOGE":TEIER AND ALWAYS READ 'e'FTER 'd'.

d) Non-network live, play-by-play coverage of professional and college team sports broadcast on (INSERT STATIONS FROM LIST,, BUT EXCLUDE PBS STATIONS [E»B S = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUIvtN S]) Examples include Major League Baseball, NBA professional basketball, NFL professional football, NHL professional hockey, NCAA college football and basketballl, and Major League Soccer. e) Other Non-network sports prograrruning broadcast on (INSERT STATIONS FROM LIST„ BUT EXCI.UDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include NASCAR auto races, professional wrestling, anti figure skating broadcasts.

f) Non-network Devotional programs, such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE'. PBS STATIONS = [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN 1 IN COLUMN Sl). Examples include Kenneth Copeland Ministries, Creflo Dollar, 700 Club, and. Joel Osteen Ministry.

' g) ONLY ASK TO SYSTEMS TIIAT CARRY PBS STATION — GROUP 8, OR CR0UPS AND K WITH 1 FOR COLUMN R: Programs broadcast only on PBS station (INSERT PBS STATIONS (IF .ANY) FROM THE LIST. [PBS = 1 IN COLUMN R]. DO NOT INCLUDE ANY OTEIER S'1"ATIONS. SKIP IF THERE ARE NO PBS STATIONS ON LIST). Examples i.nclude Masterpiece Classic: Downton Abbey Season III, Masterpiece Mystery!, PBS NewsHour, and Sesame Street. h) ONLY ASK TO SYSTEMS T!HAT CARRY A CANADIAN STATION — GROUP A, OR GROUPS D AND K WITII 1 FOR COLUMN S: Programs broadcast only oh Canadian stations (INSERT CANADIAN STATIONS (IF ANY) FROM LIST. [PBS = 1 IN COLUMN S]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE N0 CANADIAN STATIONS ON LIST). Examples include Steven 2 Chris, The Social, Coronation Street, Busytown Mysteries, and CBC News.

6. Now, considering everything we have been discussing, I would like you to estimate the relative valr~e to your cable system of each type of [NETWORK CARRYING SYSTKMS (K): non-netzvorl~] programming actually broadcast during .'2013 by [INSERT STATION(S) FROM LIST — COLUMN J]. We would like you to be very precise about this; can I ask you first to write down the types of [NETWORK CARRYING SYSTEMS (K): non-nerzvork] programming on these distant stations? Please write them down in. the order I read them. Here they are; (READ LIST) (ROTATE 'a'HROUGH 'h'N RANDOM ORDER — SEE NOTE FOR ITEMS 'd'ND 'e'). READ SLOWLY AND ALLOW TIME FOR RESPONDENT TO WRITE CATEGORIES DOWN.

LIST FOR GROUP D (NON-NETWORK CARRYING SYSTEMS, EXCLUDING WGN ONLY SYSTEMS): GROUP A — ONLY ASK ITEM H GROUP B — ONLY ASK ITEM G a) News and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as Everybody Loves Raymond, Seinfeld, American Idol, Jeopardy, and The Dr. Oz Show. c) Movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Pirates of the Caribbean: The Curse of the Black Pearl, Crash, Ghostbusters, The Matrix, X-Men: The Last Stand, Signs, and Girl, Interrupted. ROTATE 'd'ND 'e'OGETHER AND ALWAYS READ 'e'FTER 'd'.

d) Live, play-by-play coverage ofprofessional and college team sports broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Major League Baseball, NBA professional basketball, NFL professional football, NHL professional hockey, NCAA college football and basketball, and Major League Soccer. e) Other sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include NASCAR auto races, professional wrestling, and figure skating broadcasts.

f) Devotional programs such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Kenneth Copeland Ministries, Creflo Dollar, 700 Club, and Joel Osteen Ministry.

g) ONLY ASK TO SYSTEMS THAT CARRY PBS STATION — GROUP B, OR GROUPS D AND E WITH 1 FOR COLUMN R: Programs broadcast only on PBS station (INSERT PBS STATIONS (IF ANY) FROM THE LIST. [PBS = 1 IN COLUMN R]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO PBS STATIONS ON LIST). Examples include Masterpiece Classic: Downton Abbey Season III, Masterpiece Mystery!, PBS NewsHour, and Sesame Street.

33 h) ONLY ASK TO SYSTEMS THAT CARRY A CANABIAN STATION — GROUP A, OR GROUPS D AND E WITH 1 FOR COL'UMN 8: Programs broadcast only on Canadian stations (INSERT CANADIAN STATIONS (IF AN~ Y) FROM LIST. [PBS =.1 IN COLUMN S]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO CANADIAN STATIONS ON LIST). Examples include Steven A Chris, The Social, Coronation Street, Busytown Mysteries, andi CBC News.

LIST FOR GROUP C (WGN ONLY SYSTEMS): a) News and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). b) Syndicated series such as sitcoms, dramas, childretn's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as 30 Rock, Ade1ante Chicago, Everybody Loves Raymond, and People To People. c) Movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN K] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Gladiator, The Lord ofthe Rings: The Rein~ ofthe King, and Home Alone 2: Lost In New York ROTATE'd'ND 'e'OGETHER AND ALWAYS READ 'e'FTER 'd'.

d.) Live, play-by-play coverage ofprofessional and college team sports broadcast on. (INSERT. STATIONS FROM LIST, BUT EXCLUDE PBS ST%.TIONS [PBS = 1 IN COLUMN K] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Major League Baseball, and National Basketball Association.

e) Other sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R]: AND CANADIAN STATIONS

[CANADIAN = 1 IN COLUMN S]).

Examples include Horse Racing. Devotional programs such as shows with religious themes or religious talk shows'broadcast'n (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS.= 1 IN COLUMN R] AND CANADIAN STATIONS. [CANADIAN = 1 IN COLUMN S]).. Examples include Tomorrow's World and Creflo Dollar.

g) OMITTED h) OMITTED

34 LIST FOR GROUP E (NETWORK CARRYING SYSTEMS): a) Non-network News and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as Everybody Loves Raymond, Seinfeld, American Idol, Jeopardy, and The Dr. Oz Show. c) Non-network movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Pirates of the Caribbean: The Curse of the Black Pearl, Crash, Ghostbusters, The Matrix, X-Men: The Last Stand, Signs, and Girl, Interrupted. ROTATE'd'ND 'e'OGETHER AND ALWAYS READ 'e'FTER 'd'.

Non-network live, play-by-play coverage ofprofessional and college team sports broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Major League Baseball, NBA professional basketball, NFL professional football, NHL professional hockey, NCAA college football and basketball, and Major League Soccer. e) Other Non-network sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include NASCAR auto races, professional wrestling, and figure skating broadcasts.

Non-network Devotional programs such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]) Examples include Kenneth Copeland Ministries, Creflo Dollar, 700 Club, and Joel Osteen Ministry.

g) ONLY ASK TO SYSTEMS THAT CARRY PBS STATION — GROUP B, OR GROUPS D AND E WITH 1 FOR COLUMN R: Programs broadcast only on PBS station (INSERT PBS STATIONS (IF ANY) FROM THE LIST. [PBS = 1 IN COLUMN R]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO PBS STATIONS ON LIST). Examples include Masterpiece Classic: Downton Abbey Season III, Masterpiece Mystery!, PBS NewsHour, and Sesame Street. h) ONLY ASK TO SYSTEMS THAT CARRY A CANADIAN STATION — GROUP A, OR GROUPS D AND E WITH 1 FOR COLUMN S: Programs broadcast only on Canadian stations (INSERT CANADIAN STATIONS (IF ANY) FROM LIST. [PBS = 1 IN COLUMN S]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO CANADIAN STATIONS ON LIST). 35

Examples include Steven & Chris, The Social,, Coronation Street, Busytown Mysteries, and CBC News.

,'ean Assume you had a fixed dollar amount to allocate for the [NETWORK CARRYING SYSTEMS (E): 'f

non-neteorkj progranuning actually broadcast during 20.13 on PNSERT STATION(S) FROM LIST— COLUMN J]. [NETWORK CARRYING SYSTEMS 'E): Again, by "non-nebvork progrumv,", I programs on these stations that are not coming from the 'ABC, NBC, or CBS television networks.] DO NOT READ TO PBS ONLY AND CANADIAN: ONLY STATIONS (DO NOT READ TO 'ROUPS

A OR B): Considering the value of each type of programming to your cable system,l what I percentage, if any, of the 6xed dollar amount would you allocate for each type of prograuiming? Please write down your estimates, and make sure they add to 100 percent.

PBS ONLY STATIONS (GROUP B): Considering the'value iof the programs broadcast only on PBS station (INSERT PBS STATIONS) to your cable system, what pe'rcentage, if any, of the fixed dollar amount would you allocate for this type ofprogramming?

CANADIAN ONLY STATIONS (GROUP A): Considering the value of the programs broadcast only on Canadian station (INSERT CANADIAN STATIONS).to your cable system, what percentage, if any, the fixed dollar amount would you allocate for this type iofprogramming?

In formulating your percentage, please think about all the factors we have been discussing, including

using this programming in your advertising and promotions in 2013 to attract and retain customers, the i importance ofthis prograrruning to you and your subscribers, and:any other considerations you niay have..

Remember, you are only estimating the relative value.of .each type of [NETWORK CARRYING SYSTEMS (E): non-network] programming actuauy broadcast in 2013 on: [INSERT STATION(S) FROM LIST — COLUMN J].

P'OR WGN SYSTEMS (GROUP C, OR GROUPS D AND K WITH 1 IN COLUMN T): Please do not assign any value to programs that are substituted for WiGN's blacked'out programming.]

Once you are done, we will review your allocations together. Let me know when you are done.

Across all the distant stations you cariy, and considering the value to your cable system, what percentage, if any, ofthe fixed dollar amount would you allocate to (READ EACH CATEGORY. SAME RANDOM SEQUENCE AS ABOVE).

Percent

LIST FOR GROUP D (NON-NETWORK CARRYING SYSTEMS, EXCLUDING WGN ONLY . SYSTEMS): GROUP A — ONLY ASK ITEM H GROUP B — ONLY ASK ITEM G a) News and Community Events broadcast on (INSERT. STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS P'BS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as Everybody Loves Raymond, Seinfeld, American Idol, Jeopardy, and The Dr. Oz Show. c) Movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Pirates of the Caribbean: The Curse of the Black Pearl, Crash, Ghostbusters, The Matrix, X-Men: The Last Stand, Signs, and Girl, Interrupted. I ROTATE''ND 'e'OGETHER AND ALWAYS READ 'e'FTER 'd'.

Live, play-by-play coverage ofprofessional and college team sports broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Major League Baseball, NBA professional basketball, NFL professional football, NHL professional hockey, NCAA college football and basketball, and Major League Soccer. e) Other sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 1N COLUMN S]). Examples include NASCAR auto races, professional wrestling, and figure skating broadcasts.

Devotional programs such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Kenneth Copeland Ministries, Creflo Dollar, 700 Club, and Joel Osteen Ministry.

g) ONLY ASK TO SYSTEMS THAT CARRY PBS STATION — GROUP B, OR GROUPS D AND E WITH 1 FOR COLUMN R: Programs broadcast only on PBS station (INSERT PBS STATIONS (IF ANY) FROM THE LIST. [PBS = 1 1N COLUMN R]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO PBS STATIONS ON LIST). Bxainples include Masterpiece Classic: Downton Abbey Season III, Masterpiece Mystery!, PBS NewsHour, and Sesame Street.

ONLY ASK TO SYSTEMS THAT CARRY A CANADIAN STATION — GROUP A, OR GROUPS D AND E WITH 1 FOR COLUMN S: Programs broadcast only on Canadian stations (INSERT CANADIAN STATIONS (IF ANY) FROM LIST. [PBS = 1 IN COLUMN S]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO CANADIAN STATIONS ON LIST). Examples include Steven & Chris, The Social, Coronation Street, Busytown Mysteries, and CBC News.

LIST FOR GROUP C (WGN ONLY SYSTEMS): a) News and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 1N COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]).

37 b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows, game shows, and other series broadcast on (INSERT STA ltIONS FROM LIST, BUT EXCLUDE

PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]).

Examples include programs such as 30 Rock, Adelante chicago, Everybody Loves Raymond, and People To People. c) Movies such as feature films, Movies ofthe Week, and specials broadcast on (INSERT 'TATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include movies such as Gladiator, The Lord ofthe Rings: The Return of the King, and Home Alone 2: Lost In New York ROTATE 'd'ND 'e'OGETHER AND ALWAYS'EAD 'e'FTER 'd'.

d) Live, play-by-play coverage ofprofessional and college team sports broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = il IN COLUMN S]). Examples include Major League Baseball, and National Basketball Association.

e) Other sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN CCJLUIMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Horse Racing.

Devotional programs such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Tomorrow's Worl.d and CreQo Dollar.

g) OMITTED h) OMITTED

LIST FOR GROUP E (NETWORK CARRYING SYSTEMS): a) Non-network News and Community Events broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS =11 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). 'hows, b) Syndicated series such as sitcoms, dramas, children's shows, talk shows, reality shows,'ame and other series broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND C~AWADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include programs such as Everybody Loves Raymond, Seinfeld, American Idol,

Jeopardy, and The Dr. Oz Show.

c) Non-network movies such as feature Qlms, Movies ofthe Week, and specials broadcast ori. (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS:STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]).'xamples include movies such as Pirates of ithei Caribbean: The Curse of the Black Pearl,

Crash, Ghostbusters, The Matrix, X-Men: The Last Stand, Signs, and Girl, Interiwpted..

38 ROTATE 'd'ND 'e'OGETHER AND ALWAYS READ 'e'FTER 'd'.

d.) Non-network live, play-by-play coverage ofprofessional and college team sports broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include Major League Baseball, NBA professional basketball, NFL professional football, NHL professional hockey, NCAA college football and basketball, and Major League Soccer. e) Other Non-network sports programming broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN = 1 IN COLUMN S]). Examples include NASCAR auto races, professional wrestling, and figure skating broadcasts.

Non-network Devotional programs such as shows with religious themes or religious talk shows broadcast on (INSERT STATIONS FROM LIST, BUT EXCLUDE PBS STATIONS = [PBS = 1 IN COLUMN R] AND CANADIAN STATIONS [CANADIAN 1 IN COLUMN S]) Examples include Kenneth Copeland Ministries, Creflo Dollar, 700 Club, and Joel Osteen Ministry.

g) ONLY ASK TO SYSTEMS THAT CARRY PBS STATION — GROUP B, OR GROUPS D AND E WITH 1 FOR COLUMN R: Programs broadcast only on PBS station (INSERT PBS STATIONS (IF ANY) FROM THE LIST. [PBS = 1 IN COLUMN R]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO PBS STATIONS ON LIST). Exainples include Masterpiece Classic: Downton Abbey Season III, Masterpiece Mystery!, PBS NewsHour, and Sesame Street.

— 11) ONLY ASK TO SYSTEMS THAT CARRY A CANADIAN STA.TION GROUP A, OR GROUPS D AND E WITH 1 FOR COLUMN S: Programs broadcast only on Canadian stations (INSERT CANADIAN STATIONS (IF ANY) FROM LIST. [PBS = 1 IN COLUMN S]. DO NOT INCLUDE ANY OTHER STATIONS. SKIP IF THERE ARE NO CANADIAN STATIONS ON LIST). Examples include Steven 8c Chris, The Social, Coronation Street, Busytown Mysteries, and CBC News.

TOTAL (a through h)

39 LVTERVIE@'ER PLEASE NOTE:

DOES NOT APPLY FOR PBS ONLY AND FOR CANADIAN ONLY SYSTEMS (GROUPS A, B): TOTAL MUST ADD TO 100 PERCENT; PROMPT RESPONDENTS IF IT DOES NOT.

Now I'm going to read back the categories and your estimates. Remember, you are only estimating the, relative value of the [NETWORK CARRYING SYSTEMS (E): non-nehvork] programming actually '. broadcast on the following distant television stations in 2013: (INSERT COMMERCIAL STATION[S] '. AND CITY [IES] FROM LIST — COLUMNS J AND I)! (RETREAD THB CATEGORIES,AND, RESPONSES IN RANDOM SEQUENCE, IN ORDER TO ALLOW THE RESPONDENT TO REVIEW

THE ESTIMATES. [NOTE: STATIONS AND CITIES~ WILLl NOT:BE READ, EXCEPT POR PBS I AND CANADIAN STATIONS. DO NOT READ EXAMPLES.])

Are there any changes you would like to make? (RECORD ANY. CHANGES BY CROSSING OUT ORIGINAL RESPONSE AND WRITING IN REVISED RESPONSE NEXT TO IT. DOES NOT APPLY FOR PBS ONLY AND FOR CANADIAN ONLY SYSTEMS (GROUPS A, B): TOTAL MUST STILL ADD TO 100 PERCENT; PROMPT RESPONDENTS IF IT DOES NOT.)

This concludes our interview. Thank you very much for your time today. I have noted on my call record that I am speaking to a [MAX/WOMAN, BY OBSERVATION]. Is that correct?

CONFIRM GENDER: 1. MALE 2. FEMALE

ARRANGE FOR PAYMENT OF HONORARIUM — BITtHEIR 'ItO RESPONDENT OR, IF HE/SHB SO CHOOSES, CHARITY. RECORD THE NAME OF THE CHARITY.

40 APPENDIX B — Additional Survey Data The tables that follow show the distribution of surveys by MSO for each, respective, royalty year 2010- 2013.

I HeFII ~jllLTT lit [ I o s i~ 7p'aa s isa

',3 0',1 II i& ~~ 0. II 0 tK s II C I I1L~ 501la (i I 4 Cl 0 ~ ~ &e i, oe 4 ~ i'e. I ~~ion. eeill 8 IL&i 4.'0 I II 'i+J Ea&J..C I ~ E) L~l + 0.

li~HXI.4 I IIVL r4 i I Armstrong Communications/Utilities Atlantic Broadband/Grande Comm. ATILT 60

BCI Allegiance 2 1 Berkeley Cable Television 1 1 Brandenburg Telecom 1 1 Buckeye Cablevision INC 1 1 Cablevision 27 2

Catawba Services 1 1 Centurytel Broadband Services 7 1 Cequel Communications/Suddenlink 54 3 Charter Communications" 103" 5 Comcast 48 9 Community Antenna Service 1 1 Cox 40 5 Cumberland Cellular INC 1 1 Etex Communications 1 1 Fayetteville Public Utilities 1 1 General Communications 4 1 Innovative 2 1 Loretel/Tekstar 2 1 Massillon Cable TV 2 1 Mediacom 69 6 Metrocast/Gans Communications 9 1 NEP Datavision 1 1 Northland Cable Television 3 1 North State Communications 1 1 Premiere Communications 1 Puerto Rico Cable 1 1

41 Resort Television RCN Telecom Services Service Electric South Central Rural Telephone Starvision Sweetwater Cable Thames Valley Communications The U. TV Cable United Verizon 14 Vicksburg Wave Western W Communications White WOW TOTAL F06

~Includes Falcon, Helicon, HP 1, Marcus, Robin, R'iM&, Ritter, and Renaissance systems.

42 ~ ~ ~ ~

I ~

Allegiance Antietam Cable TV Armstrong Communications/Utilities Atlantic Broadband/Grande Comm. 10 AT8T 61 Buckeye Cablevision INC Cablevision 25 Cass Cable Cequel Communications/Suddenlink Charter Communications" Comcast 109'8 Cox Cumberland Cellular INC Frontier Communications General Communications Groton Utilities Innovative Knology Massillon Cable TV Mediacom Metrocast/Gans Communications NEP Datavision Northland Cable Television Northwest Communications Co-op RCN Telecom Services Service Electric South Central Rural Telephone Time Warner Cable 61 Verizon 15 TOTAL 525 55

~ Includes Falcon, Helicon, HP I, Marcus, Robin, Rifkin, Ritter, and Renaissance systems.

43

Ee 'FIIllxiimo am~ I 2 4'llegiance

Antietam Cable TV 1 Armstrong Communications/Utilities 5 1

Atlantic Broadband/Grande Comm. 3 1

ATBcT 58 1

Bristol Tennessee Essential 1

Cable One 15 1 Cablevision 30 3 Cequel Communications/Suddenlink 5 1 Charter Communications" 54 5

City of Burlington 1 1

City of Columbia 1 1 Comcast 118 5 Cox 32 3

Fayetterville Public Utilities Il 1

Frontier Communications 1

General Communications 1 Grande Communications 6 1

Insight Communications 5 1 James Cable 1 1 Knology 13 1 Massillon Cable TV 1 1 Mediacom 70 2 Metrocast/Gans Communications 9 1

Millington CATV 1 1

NEP Datavision 1 1 Northland Cable Television 1 1

Premiere Communications 1 RCN Telecom Services 2

Service Electric 1

South Central Rural Telephone 1 1

Time Warner NY Cable 38 . 3 Verizon 13 1 TOTAL 502 49 o Includes Falcon, Helicon, HP1, Marcus, Robin, Rifkin, Ritter, and Renaissance systems.

' 'e' SY rll lIJ i ~ i&Nli 4 WI4 li 1.%

" O' ' o. ii i i i ~ il.: le L. E1 i 0 ~ i! Hoi I a(i;=ro . s s, -. (& I I i ~ 4 ~

t . ~ i'o I 4i!]ieeiil ~ . I i:0 I s Ish t '.ia|Mel iMJ 4. o o Il

L&i iiEK'Si I iii&46 I 41,OLI Allegiance Armstrong ATILT Cablevision Caribbean Communications Catawba Charter" 26 Comcast 105 Cox GCI Hiawatha Broadband Insight Midwest 5 Mediacom 39 Metrocast/Gans NPG Cable RCN Telecom Services Ritter Cable Service Electric South Central Rural Telephone Southern New England Tel. Co. Time Warner NY Cable US Cable of Coastal- Texas Wide Open West TOTAL 297 40 o Includes Falcon, Helicon, HP1, Marms, Robin, Rifkin, Ritter, and Renaissance systems.

45 The tables that follow show titles of the executives interviewed for each, respective, royalty year 2010- 2013.

:i)Ill ..jhow oj si II kljil j~k.j% a&e&~p jij jjcelj j= III i&|- .t$iIREiwI%lgv'e,, , Illj rr ' &gy s — 'IISS R ~ $84 j +jjj~gjgf)gl Ip Q~IIpi[ I ID ' Q ~ . I cJ Director of Marketing/Sales Senior Vice PresidentNice President of .10 Content/Programming Vice President/General Manager Vice President of Product Management Senior Vice President/ Vice President of Marketing/Sales/Marketing Operations(Branding Regional Sales Vice President of Customer Service Manager Director of Content Acquisitions Product Specialist Senior/Marketing Manager Director of Programming Superintendent

Programming Products Coordinator : 2:

Director of Engineering : 2: President 1

Chief Financial Officer 1 Chief Product Officer 1

Chief Executive Officer . 1 Chief Programming Officer 8 I V&AII I II IL&I 'Ill' kU; A&&@II :ii&a uv rwimr= Director of Marketing/Sales Senior Vice President/ Vice President of Marketing/Sales/Marketing Operations/Branding Vice President/General Manager Senior Vice President/Vice President of Programming Senior/Marketing Manager Director of Programming Director of Operations Director of Product Marketing/Development President Manager Operations Manager Product Manager Vice President of Business Development Vice President of Acquisitions Director of Sales Assessment Chief Financial Officer Vice President of Product Management Network Manager Programming Consultant Product Specialist Regional Vice President Programming/Contracts Coordinator Chief Programming Officer

47

I C ~ it't A 6'ilail M Il I1@I~ Senior Director/Director of Marketing/Sales ,'ice President of Marketing/Salles/Marketing Operations/Branding

Regional Vice President

Programming/Contracts Coordinator Corporate Programming Manager Senior Director/Director of Programming 'enior Vice President/Vice President of Programming Vice President/General Manager Video Product Manager Division Vice President Content Analysts Director of Product Management Executive Director of Programming C.E.O. Marketing Product Manager Network Manager Regional Director of Marketing Programming Consultant Product Specialist President

48 '~ ~ iAii 1 'Oi i t3:C. I II Eg .. i lilt 4'l CL4ii Ni iisiil HF K&iiiiti Vice President of Programming/of Programming and Marketing/of Marketing Vice President/Regional Vice President Director of Marketing Director of Product Management/Product Manager Director of Programming Sales and Marketing Director Chief Operating Officer President/CEO Executive Vice President of Cable TV Vice President of Content Strategy and Operations Vice President of Content and Product Management Vice President of Operations Director of Area Operations Director of Engineering Director of Sales and Marketing Director of Video Content Operations Content Analyst Marketing Communications Marketing Operations General Manager Product Marketing Manager Senior Marketing Manager Video Product Manager

49 The following tables show, for each royalty year, h.ow the cable sy,stems are distributed into the five carriage groups compared how all of systems surveyed (both within andI outside of the random sample) are d:istr.ibuted into the same groups

0 I e

Canadian Only (Group A) &'1% &1%

PBS Only {Group IB) 5% 40/

WGN Only (Group C) 3tl% 33%

Non-Network (Group D) 25% 28%

Network (Gl oup E) 8% 35%

'$&)p~t I o "~

Canadian Only (Group A) &1% &1%

PBS Only (Group B) 4'/0 4%

WGN Only (Group C) 30% 39%

Non-Network (Group D) 27% 24'%1% Network (Group E) 38%

50 Canadian Only (Group A) 1% 1%

PBS Only (Group B) 6% 6%

WGN Only (Group C) 32% 34%

Non-Network (Group D) 22% 21%

Network (Group E) 39% 39%

51 Canadian Only (Group A) 1% 3%

PBS Only (Group B) 6% 5%

WGN Only (Group C) '35% 34% '1% Non-Network (Group D) 20%

Network (Group E) '37% 38%

52 Tables 10.1 — 10.4 in Appendix C shows the disposition of our attempts to contact the persons responsible for programming at cable systems for each of the 2010-2013 royalty years.

I ~

Completed surveys 505 Refusals 24 Not able to reach (voice mails, etc.) 161 Not accurate contact information 108 Total 0 of systems 798

I ~

Completed surveys Refusals

Not able to reach (voice mails, etc.) 182 Not accurate contact information Total 4 of systems 815

I ~

Completed surveys 502 Refusals 36 Not able to reach (voice mails, etc.) 71 Not accurate contact information 303 Total 0 of systems 912

53 e,.

Completed surveys 297

Refusals 11 79 Not able to reach (voicemails, etc.) 288 No/inaccurate contact information 239

Total re of systems 1,003

As shown in the table be1low, th.e results &om the 300 system sample are consistent with the results from the all systems surveyed.

- ~ r) e e . e e e e e 0 ~ -: ~ e ~ e ~ - ~ ~ I a e

0 of Systems Surveyed 505 525 502 297

News and community events 14"/ 'I 6% 13% 14% '15% '8% Syndicated series 17% 'I 80/o 1 8% 17%

Movies 12% 'I 10/0 13% 16'/0 i22% i Live coverage of profes. ional and 28% 29~/o 30% 3'1% '37%

college teams sports '/A Other sports programming 9% 9'/ 9% Devotional programming 60/i) 6% 5% 4% 5()/ Programs broadcast only on PBS 9% 9% 9% 6% 3()/ station broadcast on Programs only '1% 3% 0% Canadian stations 1% 1%

/1//PAA TOTAL 38% 38% 42% 40% i40% i

54 Horowitx Research Company Profile

HOROWITZ RESEA.RCH,

INC.

MARKET RESEARCH A

CONSULTING www.horowitzResearch.corn BA.CD~ROUND

Horowitz is a leading provider of market and consumer research for cable operators, television networks

and content providers, and technology companies. Our strength lies in an in-depth understanding'of'eople and their relationships to entertainment and technology in today's rapidly evolving and competitive media environment.

Our mission is to provide clients with the crucial consumer marketing information they need to make

strategic business decisions. At its inception in the 1980s, Horowitz Associates focused on market potential research for cable operators and progrmnmers. Principals of the f3rm developed the NCTA Segmentation Study, a first of its kind, conducted in 1983 in urban/suburban markets that were either newly built or on the drawing board at the time. The premise of the study was that new satellite technology, and the programming proposed for it, would open up these markets to cable services

heretofore uncharted territory in the history of television technology. We answered the question of the 'esults day: Who will pay for television when television is freeY The development of such programming, ventures as The Discovery Channel„AXE Networks, and Nickelodeon was a direct outgrow)th 'of the

ofthat seminal research.

We have been studyirrg consumers of media products ever since, aud have continued to be on the forefront ofresearch on new TV and broadband technologies. As a leader in new and irmovative research 'ncluding, technologies, our research—including our videographies—has been featured at various industry venues among others, The Cable Show (NCTA), The ARF (Advertising Research Foundation)'s Re:Think and Audience Measurement conferences, the ANA (Association of National Advertisers) annual event, the MRA (Marketing Research Association) conference, various AHAA (Association of Hispanic Advertising Agencies) conferences, at CTAM Summits and CTAM Research Confe'rences, at various Hispanic Television Surrunits, annually at our Cultural Insights Forum,, and at many regional cable association events.

Another important perspective we bring to the table is our understanding of the shifting demographics of America and the implications of this on the media industry. This perspective, that "Multicultural America" is America, Iis the framework within wl;.ich we contextualize our understanding of consumers and the evolving media landscape, particularly when it involves adoption and usage of new technologies among Millennials and all futme generations. As such, in addition to our general market expertise, we are a leading provider of Hispanic and multicultura]. research and have in-language research capabilities. RESEARCH MKTEIODOLOGY AND SERVICE

Horowitz Research is a boutique, full-service firm offering a full range of quantitative and qualitative services, as well as a series of annual syndicated studies. Custom Research Services

Quantitative Survey Researc)h For Service Providers: Marketing, Positioning and Branding Surveys + Packaging, Pricing and Bundling ofTelevision, Internet and Phone Services + Surveys for the Development, Distribution and Marketing ofNew Technologies Customer Satisfaction Tracking Surveys MSO and Cable System Programming and Channel Carriage Surveys

For Content I'roviders: Network. Programming Surveys Surveys to Support Affiliate Marketing and Sales Research for Web Site Development and Evaluation Advertising Effectiveness and Tracking Surveys

Qualitative Research

Focus Groups, Dyads and Triads One-on-One Interviews Online Communities Ethnography Executive Interviewing

Secondary Data Analysis Syndicated Studies

Horowitz Research, Inc. produces a series of major industry studies including:

State of Cable & Digital Media State of Cable & Digital Media, Multicultural Edition Multiplatform Content &, Services Multiplatform Content & Services, Multicultural Edition State of OTT State of OTT, Multicultural Edition FOCUS Latino FOCUS African America Client List

Our client list for custom and/or syndicated research includes (but is not limited to) ABC, AETN (A&E, History, Lifetime), Alterna TV, Arbitron, Baja Broadband, BBC, Buckeye Cable, Cable Bahamas, Cablevision, Charter Communications, Cine Sony, CNN, Comcast, The Corporation for Public Broadcasting, Cox Communications, Discovery Networks, Disney, ESPN, Food Network, Gallery Player, Grande Communications, HBO, Mediacom, Music Choice, the MPAA, MyDTV, NBCUniversal (NBC, Telemundo), NESN, Onelink Puerto Rico, RCN, QVC, Scripps, Sesame Workshop, Showtime Networks, Sirius XM, SkiTAM/Adaptive Spirit, Sony Music, Tempo, Time Warner Cable, TV Guide, TVSMONDE, Univision, USA Networks, Viacom Media Networks, Verizon FiOS, The Weather Channel, Tribune Media Services, and Worldgate, among others.

57 i I FORUMS AN:D CONFERENCES

Horowitz Research'nnual Iindustry Cultural Insights Forum brings together key industry figures to meet, network, plan, and learn about the consumer market for new media services and technologies. The 16'" annual Forum took place on March 9, 2016. The topic of this year' Forum was Content Matters: Race, )Diversity and Media.

Three hundred executives from the cable, adlveitising and marketi.ng industries attended t]he event. Speakers and paneli,sts includ.ed: Peter E. Blacker,i EVP Digital Meclia and Emerging

Businesses, NBCUniversal Telernundo Enterprises;, James .Brown, E'.VP,:Head of Content Distribution, Revolt; Mi.chelle Webb, E'.xecutive Direc'tor,'ontent Strategy and Acquisition, Verizon FiOS; Michael Tribolet, Chairman, Co-Founder, and 'CEO, YipTV; D'Angela Proctor, SVP, Original Programming, TV One,; Joseph I,awson, ( ontent StTategy and Acquisition, Go90, Verizon FiOS; Thomas Graym.an, VP, Brand andi. Consumer Research, Spike TV; Apoorva N. 'irector, Gandhi, VP, Multicultural Affairs, Marriott International; Antonio Briceno, Deputy Managing U.S. and Canada, BeIN Sports; and Patricia HAdden, SVP and Head of Mhrkbtirjg, Digital Enterprises, NBCUniversal.

Other recent conference appearances ancl presentations by principals of.Horowitz Research, Inc. include:

+ News Technology Summ;it (September 2016) Speed to Air: Winning the Social, Digital and TVNews Wars, Panelist NMSC (September 2016) Binge on! Streaming 's Implications for the Future of TV, Panelist + Mid-America Cable Show (September 2016) The Three M's: Mobile, Millenm'al, and Multi-Cultural: Can "t Have One J/ithout the Other, Moderator Mid-America Cable Show (September 2016) Future Trends and Technologies ofthe Industry, Panelist + The Independent Show (July 2016) Navigating Today's Turbulent Video and Broadband Business, Moderator + CANTO 32"" Annual Conference and Tradie Exhibition Show (August 20~16)~ Attracting and Retaining Millennials i'. the Age ofOTT, Presenter The TV of Tomorrow Show (June 2016) TV Used to Be Easy. What Did You Guys DO to It?, PanelI.st + Multicultural Breakfast at INTX I,'May 2016) Providing Programmingfor Multicultural Audiences, Co-Moderator + NAB Show (April 2',016) Research Presentati'on — The Value ofBroadcast! Content in Today 's Multiplatform World, Presenter NAB Show (April 2',016) Getting to anom Y'ou: 2VIore Research + More Data = More viewers?, Panelist Media Insights & I&.ngagement 2016 (February 2016) Think Tank: Hispanics, Media and the.Bottom Line, Panelist + Cablefax TV Innovation Su:mmit (October 20I5) 58 We Do (Content) Windows: What's Working, What's Not and What's the Future~, Moderator NAMIC (September 2015) BUSINESS SOLUTIONS—Research: Big Data 2.0, Panelist New Video Frontiers (September 2015) OTT: Friend or Foe?, Presenter Emma Bowen Foundation Summer Conference (June 2015) The Business Casefor Diversity and Inclusion, Presenter

RECENT ARTICLES AND PRESS MENTIONS (PARTIAL LIST)

To Win with Hispanics, There's No One-Size-Fits-All Approach, Horowitz Study Finds, Press Release (July 26, 2016) Report: Millennials Stream at Least HalfofTheir TV and Love Their Netflix, CED Magazine (May 4, 2016) Broadcast is a top 'ask'or OTT streaming packages, Rapid TV News (April 28, 2016) Why are so many great black TV shows missing from streaming services?, The Washington Post (April 27, 2016) BBC appoints new drama chief, More broadcast opportunities and Sky spending double on sort coverage: DTG Daily News April 21st, DTG (April 21, 2016) NAB 2016: Access to TV Everywhere, OTT Crucial for Consumers, Horowitz Says, Broadcasting Cable (April 20, 2016) Opportunity for Over-the-Top Providers in New Video Ecosystem; Broadcast Content is Key, Press Release (April 19, 2016) Urban Millennials Not Dropping Cable Like It's Hot, Multichannel News (March 9, 2016) Apple TV: Is connected TV Arriving Right on Time?, The Huffington Post (September 8, 2015) Univision's blurry picture, Economist (September 5th, 2015) [itvt] Column: The iTV Doctor Is In! Who Are the New TV Consumers?, ITVT (September 1, 2015) Jumping Into the Mobile Marketing Waters, Advertising Age (July 9, 2015) Mobile: Fueling the Marketing Feeding Frenzy, Advertising Age (July 6, 2015) Multicultural TV Users Favor OTT Viewing, Mediapost (June 24, 2015) Multicultural Talent Is Surging on TV and Winning Mainstream Audiences, Adweek (May 18, 2015) How to Be Relevant in a Multicultural TV World, MediaBizBloggers (April 17, 2015) Multicultural Content Goes Multiplatform, Multichannel News (April 6, 2015) Streaming Back to TV, Multichannel News (March 30, 2015) OTT Claims Bigger Viewing Share, Multichannel News (March 30, 2015) The Cord-Cutting, Multicultural Millennial Way, MediaBizBloggers (March 29, 2015) Curious Behavior, Multichannel News (March 16, 2015) Youth Will Be Served, But How? Multichannel News, (March 9, 2015)

59 + Review: Sling TV offers streaming alternative to'cable,'hicago Tribune (February 25, 2015)

Why Multichannel TV Subscribers Won't Go: OTT-Only, Emarketer (February l9, i 2015) + Waiting for the Fat Lady to Sing, Multichannel News (February 19, 2015) + Survey: You'e Missing Out Without Cable, Multichannel News. (February 9, 20.15) Survey: Multiplatform Viewers Cite Importance ofBroadcast: Networks, Broadcasting 4 Cable (February 2, 2015) + Shifting Focus In a Multiplatform Age, Multichannel News (January 5, 2015)

BIOGRAPHIES OF PRINCIPALS AND KEY PERSONNEL

Howard Horowitz, President. Horowitz Research,'n'c.

Howard Horowitz is a renowned specialist in the marketing of television, computer and. Internet . services to consumers. A political scientist by training and market researcher by profession, Mr.: Horowitz has served as advisor to many of the major media accompanies. Over the past thirty, years, Mr. Horowitz's work has helped shape and advance the standard for success in marketing i

cable television, home entertainment services, and broadband and mobile content and services..

Horowitz Research has been called upon by many clients, including MVPDs, cable systems, television networks, developers of interactive applications, Internet companies and other content .

providers to assess the impact on consumers of new technological developments and increased competition in the marketplace. Howard's most recent work focuses on trade and consumer

research in the areas of digital media including broadband and mobile content, service5 phd l

technologies. Horowitz Research is also well-recognized for its research on urban, multicultural . consumers, and Latino consumers in particular. Howard Horowitz is a Cable Pioneer (Class of 2012) who began his career in the industry thinly i'nited years ago. His innovative research helped usher in the era o'f pr'ogrammed cable television in the

States, highlighted by his work on the NCTA Segmentation Study, a first of 'its.kind, . conducted in 1983. Howard continues to be an innovator, introducing new and powerful i research methodologies. Howard and his company were the first among cable research

companies to introduce ethnography as a research methodology to understand the complexity.of . how consumers interact with technology in their homes,'heir places of business, and out of the

home. Mr. Horowitz and his staff at Horowitz Research continue to put into practice new and . cutting-edge research methodologies to fully capture the nuances of the rapidly changing cable; and telecommunications industry.

Mr. Horowitz and his staff at Horowitz Research have developed a series of industry studies that,

are widely recognized as essential data on the market for multichannel, broadband, and mobile .

services, and in particular, on adoption and use of these services among America's Hispanic, . Black, Asian, and other multicultural communities. These annual studies include titles such,as, State of Cable and Digital Media, Multiplatform Contenti and Services, FOCUS: Latino, and i FOCUS: A&ican America.

60 Mr. Horowitz has a B.A. in Psychology (1973) as well as a M.A. in Political Science (1977) &om New York University. He graduated with honors and is Phi Beta Kappa. curia Riera Senior Vice President o Research and Strate

As Horowitz's Senior Vice President of Research and Strategy, Nurin oversees the entire research team, as well as outside vendors. Nuria is responsible for ensuring quality control and working with her team on all types of methodologies and projects to bring our clients the insights they are looking for. During her 18-year tenure at Horowitz, Nurin has worked on virtually all types of qualitative and quantitative projects ranging &om brand and network evaluation and loyalty, customer service tracking, pricing and packaging, advertising and marketing communications effectiveness, brand/corporate image, and strategic reviews.

Originally from Barcelona, Spain, Nuria brings a unique cultural perspective as well as a wealth of knowledge from her international experience with research. Always looking to expand Horowitz's capabilities, Nuria is constantly on the lookout for new and improved research methodologies that will best serve Horowitz's clients.

Adriana Waterston Senior Vice President o Insi hts and Strate

Adriann Waterston is Senior Vice President ofInsights and Strategy at Horowitz Research.

Adriana is n research junkie with a flair for finding the story behind the statistics, teasing the meaning out of the measurement, and revealing the faces behind the facts. A seasoned quantitative and qualitative market researcher with expertise on consumers and their attitudes towards their media tools, Adrinna has moderated hundreds of groups and conducted in-home ethnographic research on a bevy of topics ranging from advertising and marketing effectiveness; media and entertainment consumption; unmet needs in technology and media; pricing and packaging; travel; customer service satisfaction; lifestyles and family dynamics; and brand/corporate inmage.

Adrinna oversees all of Horowitz's millennial, multicultural nnd Latino research endeavors. Well-known for her leadership in this space, Adriana is often quoted in the press and is a frequent speaker at industry events and conferences. She is also the lead organizer and co-host of the annual Cultural Insights Forum, a New York City event that draws a crowd of over 350 media and advertising executives.

Adriana has been named one of the industry's "Most Influential Minorities in Cable" by Cablefax Magazine and has received a CTAM TAMI award for her work in multicultural marketing. She co-authored The Practical Guide to Multicultural Marketing, which won the Bronze Global Ebook Awards in the Multiculhu.al Non-Fiction category in 2013.

Born in San Juan, Puerto Rico, Adriana is a graduate ofBrandeis University in Waltham, MA.

61 APPENDIX D

SUE PANZER President, Sue Panzer Consulting LLC

Sue Panzer is a sales and marketing executive with a successful track record leading and motivating national sales organizations to deliver exceptional top and bottorri-line financial results. She is specifically skilled in solidifying long-term client relationships, negotiating challenging agreements and developingi strategic affiliate marketing initiatives. Since 2009, she has served as the Presi'dent of Sue Panzer Consulting LLC, advising domestic and international companies in the telecommunications and human resource industries on business development, sales strategy, marketing and communications.

Sue was the VP, Affiliate Ad Sales 8 DistributionlMhrkhtikg for the top cable programmer, Lifetime Entertainment Services, where she led a team to develop strategic local ad sales and affiliate marketing promotions. Prior, she was the VP, National Distribution responsible for leading the sales team to increase revenue and distribution of Lifetime's three channels.

Previously, Sue was the VP, Area General Manager at the cable programmer, Showtime Networks. During her 11 year tenure, she increased revenue and distribution of Showtime and The Movie Channel, ran the regional operations, managed sales trainings and developed incentive programs.

Sue graduated from the University of Richmond 4itlii aiB.A. in Education. After teaching fifth grade in Richmond, VA, she returned to graduate school full-time to earn her MBA in Marketing from The College of William and Mary in Williamsburg, VA. In additioni to i serving on the Board of Directors of the MBA Alumni Association, Sue served asi a i National Board member for "Women in Cable 8 Telhcdmrhunications" and as President of the NY chapter where she won the "Chapter of the Year Award". From 2003 through 2006, she also served as President/Co-Founder of the "Fire Island Golden Wagon Film Festival" in Ocean Beach, NY. DECLARATION OF HOWARD HOROWITZ

I declare under penalty of perjury that the foregoing testimony is true and correct, and of my personal knowledge.

Executed on December, 2016

oward orowitz I

C O CD M PV A CD U p Before the COPYRIGHT ROYALTY JUDGES Washington, D.C.

In the Matter of

Distribution of the Docket No. 14-CRB-0010-CD (2010-13)

2010, 2011, 2012, and 2013 Cable Royalty Funds

DIRECT TESTIMONY OF JOEL STKCKKL, PH.D.

DECEMBER 22, 2016 TABLE OF CONTENTS

Page

I. Qualifications . II. My Understanding of Compulsory License Procleedlingh.

A My Assignment. J i B S»~~~~ ofConclusions . III. Fundamental Principles of Survey Design. A Survey Research Requires Clear Objectives ..i. B Questions Should Be Clear and Understandable. C. The Choice of Survey Medium. IV. The Studies at Issue: The Bortz and Horowitz Siuvhys I. J A The Bortz Survey .14 B. The Horowitz Survey .18 V. The Borax and Horowitz Surveys Cannot Satis@ thh Rhsedrch Objectives In the Current, Matter 21

A The Bortz-Horowitz Survey Objectives Do Not Match the Statutory Requirements ~... 21 B The Bortz-Horowitz Survey Format Asks Respondents to Focus on an Arti6cial Construct and Presume an Activity That They Do Not Engage In In and Does Not Exist In Their Daily Business Activities. 23 The Constant Sum Resource Allocation in the Money Question is Ambiguous. Furthermore, that Ambiguity Introduces Inconsistencies In the Unit ofAnalysis aud

Biases Results in Favor of Smaller Cable Syslteni O)editors...... l... 25 I D The Constant Sum Resource Allocation in the Money Question Does.Not Elicit Data ~

that Correspond to the Relative Market Value Question~. 26 .

The Constant Sum Money Question is Too Complex to.Produce Valid Data...... 28 . F. Constant Sum Questions in General Do Not Capture Real World Behavior...... 34 G The Bortz/Horowitz Survey Should Not Be Administered By Telephone ...... 36

H The Lesser ofTwo Evils L.. i L J & '...... 37

Surveys ofCable Programming Executives are Poor Ways to Measure Relative Market . Value in this Matter J I J i ...... 39 A. Objective Market Results are Preferable to Subjective Judgments, Especially When the Judgments Are Flawed as Described Above..' ' The Relevant Opinions for Projecting Marketplace Results are Not Those of Cable Executives; They are Those of Cable Customers.'40,'9

APPENDIX A l I L J .43 APPENDIX B 57 DIRECT TESTIMONY OF JOEL STECKEL Ph.D.

I. Qualifications

I am a Professor ofMarketing and the Vice Dean for Doctoral Education at the Leonard

N. Stern School ofBusiness, New York University, where I have taught since January 1989. I was the Chairperson ofthe Marketing Department for six years, Gom July 1998 to June 2004.

Prior to being promoted to Vice Dean, I was the faculty director ofthe Stern School Doctoral

Program for 6ve years, &om May 2007 to July 2012. I also currently serve as the Acting

Chanperson ofthe Accounting Department at NYU Stern.

I have also held either permanent or visiting faculty appointments at the Graduate School ofBusiness, Columbia University; the Anderson Graduate School ofManagement, U.C.L.A.; the

School ofManagement, Yale University; and the Wharton School, University ofPennsylvania. I received my B.A. summa curn laude Rom Columbia University in 1977, and M.B.A., M,A., and

Ph.D. degrees from the ~on School, University ofPennsylvania in 1979, 1980, and 1982, respectively. I was elected to Phi Beta Kappa at Columbia University and Beta Gamma Sigma at the Wharton School. These are the national honor societies for the respective disciplines I studied at these institutions.

I was the Founding President ofthe INFORMS institute for Operations Research and

Management Science) Society for Marketing Science, the foremost professional group for the development and application ofmanagement science theory and tools in marketing. In addition,

I am a member ofthe American Marketing Association, the American Statistical Association, the

Association for Consumer Research, the American Psychological Association, the American

Association for Public Opinion Research, the International Trademark Association INTA), and the Society for Consumer Psychology. My fields of specialization withiii marketing include marketing research methodology, i marketing and branding strategies, electronic commerce, and managerial decision making, I am an author of three book. and o ver 40 articles. In the course ofmy scholarly research, teaching, and consulting work, I have studied issues ofmarketing research, branding, and their roles in consumer choice and marketing strategy,.

During the course ofmy professional career, I!have designed, conducted, supervised, and/or evaluated hundreds of c:onsumer surveys. In that work. I have formulated and evaluated sampling strategies, supervised and rendered. opinions on sample selections, designed questionnaires, analyzed data, and interpreted results. I have also evaluated similarly purposed survey work performed by others. This work includes branding, false advertising, and asset valuation related studies in the cable, entertainment, satellite, and telecommunications induktri6s.

One ofmy books is a textbook entitled Marketing Research. This book has been adopted

at several ofthe country's major business schools. During one ofmy sabbaticals I served as an

in-house consultant at the market research firm., Directions for Decisions (DFD), headquartered in Jersey City, New Jersey. DFD's growth allowed it to be acquired by Kl i R.esearch, another'esearch

firm, headquartered in Norwalk, Comiecticut! My professional qualifications are described further in my cuiriculurn vitae, which is attached as Appendix A..

I have served as an expert witness on marketing research, marketing strategy., branding„ trademark, and issues related to consumer decision making in a variety of litigation matters!, In the past five years, I testified as an expert witness in the matters listed in Appendix B.

My experience most relevant to this ass:igmnent, however, is my editorial work, I have! sat on the editorial boards ofmany of our majoi journals over the years. Currently, I serve as a

co-Editor-in-Chief of the journal Marketing Letters. In that capacity I evaluate over 200 research ~ studies each year. I am a gatekeeper. I decide what gets published and what does not. As such, my evaluations of the scientific reliability and validity of each study are subject to the scrutiny of the academic community. The community considers any study that does not conform to the scientific standards ofmy profession that appears in the journal as a black mark on my record. I consider the fact that the journal's publisher, the international Grm, Springer-Verlag, has kept me on past the expiration of my term (July 2014) as validation of my performance in evaluating research.

I am receiving compensation for my testimony in this proceeding, However, my compensation is not contingent upon my opinions or ofthe outcome of the instant case.

II. My Understanding of Compulsory License Proceedings

In 1976, Congress amended the Copyright Act to estabhsh a compulsory hcense for cable systems which retransmit broadcast television stations to distant — that is, out-of-market or non- local - subscribers. To maintain their licensee status, cable systems must, among other things, pay royalties to the Copyright Office semi-annually for the benefit of the copyright owners whose content is aired on the distantly retransmitted stations (i.e., distant signals). In creating the cable compulsory license, Congress focused on the distinction between local and distant signals carried by cable operators. Only copyright owners ofprogramming on stations retransmitted on a distant basis are eligible to receive compulsory license royalties paid by cable systems. Congress limited the compulsory license to programming distantly retransmitted by cable operators based on their conclusion that a cable system's retransmission of local programming within the local market did not harm the copyright owner, as the signal was already available to the public free of charge in the station's local market through over-the-air broadcasting. Thus, cable compulsory license fees pertain only to programming on distant . broadcast signals.

Periodically, proceedings are held to determine how the cable compulsory royalties should be allocated to the copyright owners whose programs 8reicarried on the distantly retransmitted stations. Congress did not set a specific statutory standard for cable royalty allocations, but as the proceedings have evolved over the years, the standard has become the

"relative marketplace value ofthe distant broadcast signal programming retransmitted by cable systems."'n past proceedings, such marketplace value has been determined in terms ofthe attraction and retention ofsubscribers. The question, as I,understand it, becomes how speci6c distant signal broadcast programming impacts the attraction and xetention ofsubscribers. In other words, the allocation mechanism must recognize how speci6c programing impacts consumers'ecisions to contract with and remain with cable systems.

For the purposes ofthis proceeding, the copyright.owners who are entitled to the royalties at issue are organized into eight groups based on their pro~mming types ("Agreed Categories") as follows:

"Canadian Claimants." All programslbroadcasti on Canadian: television stations, except: (1) live telecasts ofMajor League Baseball, National Hockey League, and U.'S. college team sports, and (2) programs owned by U.S. copyright owners. "Commercial Television Claimants." Pro~s produced by or for a U.S. commercial television station and broadcast only by that station during the calendar year in question, except those listed in subpart (3) ofthe Program Suppliers category.

"Devotional Claimants." Syndicated progmms of a:primarily religious theme, but not limited to programs produced by or for religious institutions.

'istribution ofthe 2004 and 2005 Cable Royalty Funds, 75 FecL Reg. 57063, 57065 (September 17, 2010) "Joint Sports Claimants." Live telecasts ofprofessional and college team sports broadcast by U.S. and Canadian television stations, except progmns in the Canadian Claimants category.

"Music Claimants." Musical works performed during programs that are in the following categories: Program Suppliers, Joint Sports Claimants, Commercial Television Claimants, Public Television Claimants, Devotional Claimants, Canadian Claimants.

"National Public Radio." All non-music programs that are broadcast on NPR Member Stations.

"Program Suppliers." Syndicated series, specials, and movies, except those included in the Devotional Claimants category. Syndicated series and specials are de6ned as iucluding (1) programs licensed to and broadcast by at least one U.S. commercial television station during the calendar year in question, (2) programs produced by or for a broadcast station that are broadcast by two or more U.S. television stations during the calendar year in question, and (3) programs produced by or for a U.S. commercial television station that are comprised predominantly of syndicated elements, such as music videos, cartoons, "PM Magazine," and locally-hosted movies.

"Public Television Claimants." All programs broadcast on U.S. noncommercial educational television stations.

Prior to December 1993, royalty distribution proceedings were conducted by the

Copyright Royalty Tribunal ("CRT"). In December 1993, Congress abolished the CRT and transferred its responsibilities to the Librarian ofCongress and the Copyright Once who established ad hoc Copyright Arbitration Royalty Panels ("CARPs"). More recently, the proceedings have been atimi~istered through the Library ofCongress by the Copyright Royalty

Board ("CRB"). The Copyright Royalty Judges of the CRB ("Judges") determine how the royalties at issue are allocated among the eight groups ofclaimants.

See Notice ofParticipant Groups, Commencement of Voluntmy Negotiation Period (Allocation), and Scheduling Order at Exhibit A (November 25, 2015) ("Notice"). For most of the distribution proceedings after the enactment ofthe cable compulsory license, both the CRT and the CARP relied significantly bn evidence ofviewing ofprograms as the bases for allocating royalti.es among the claimant groups! For the proceedings to distribute the 1998-1999 and the .2004-2005 royalties, despite several critical attacks, a CARP and the

Judges, respectively, found that surveys sponsored by the Jo!int Sports Claimants ("JSC'"'),

representatives ofthe copyright holders oflive professional and college team sports programming, and conducted by the Bortz Media Ec Sports Group, Inc. of Denver ("Bortz")

"provided...the best starting point for evaluating an award."4 The Bortz studiies have been

supported (with or without adjustments) by the Commercial Television Claimants, the

Devotional Claimants, and. the Public Television Claimants.

In response to the Bor!Iz studies, the Motion Picture Association of America ("MPAA"), which represents Program Suppliers claiimants, commissioned Horowitz Research, Inc, to design a survey ("Horowitz Survey") that carefully replicates the methods ancl procedures ofparallel 'ade surveys conducted by Bortz ("Bortz Survey"), but attempts to improve upon the previous efforts

by Bortz by solving some ofthe information and category weaknesses as pointed out in the

2004-2004 Cable Phase I Proceeding. The: Bortz Survey and the Horowitz Survey xnodels fojm, the basis for my testimony.

'ee Program Suppliers v..Librarian ofCongress, 409 F.3d 395, 401-02 (D.C. Cir. 2005). They also relied on evidence ofviewing to distribute royalties among claimants within the Progratn Suppliers categoty, See Independent Producers Group v. Copyright Royalty Board, 792 F.3d 132, 142 (D.C. Cir,. 2015).

Distribution ofthe I9!&8 and 1999 Cable Royalty Funds,75 Fed. Reg. 57063, 57068 (September 17, 2010).

See Program Suppliers'ritten Direct Statement, Docket No. 14-CRB-0010-CD (2010-13), Direct Testimony of Howard Horowitz at 3 (December 22, 2016) ("Horowitz: Testimony"). I have been asked by Mitchell, Silberberg k, Knupp, LLP, counsel for the MPAA and its represented Program Suppliers in the 2010-2013 Cable Royalty Funds allocation proceedings

("2010-13 Proceedings") to provide a professional opinion on the validity ofthe Bortz and

Horowitz Surveys as the bases for determining the allocation ofroyalties in this proceecHng, In particular, I have been asked to opine on the following questions:

1. Which ofthe Bortz and Horowitz Surveys is better suited for the issue at hand?

2. Can surveys of cable operators be informative in general about the issue at hand?

3. Are other approaches more informative than cable operator surveys about the issue at

hand?

This testimony summarizes the opinions that I have formulated to date. In case further review and analysis of information provided to me subsequent to the filing of this report leads to new or revised opinions, I reserve the right to amend or supplement the opinions contained in this report.

B. Summa o Conclusions

My work in this matter has led me to the following conclusions:

1. Neither the Bortz Survey nor the Horowitz Survey is sufficiently capable of

assisting the Judges in determining the relative market value ofthe programming

at issue in this proceeding;

2, While neither the Bortz Survey nor the Horowitz Survey provides a sufficient

basis for measuring marketplace value, the Horowitz Survey does overcome some

of the flaws in the Bortz model, thus making it preferable to the Bortz Survey;

3. Surveys of cable operators are inadequate in general for measuring marketplace

value or return; and 4. At least two other research approaches would.provide data more useful for i

assessing marketplace return: (l) analysis!ofmarket: data, and {2) surveys of

cable customers.

I begin my analysis of the issues in this case by outlining some basic principles of survey research that are essential for surveys to be reliable and valid. I then describe the Bortz and

Horowitz Surveys and apply those principles of survey research ito those surveys as described. I conclude by proposing alternative approaches to stu.dying the issues at hand. Should further information become available to me throughout this proceeding,: I reserve the right to supplement or amend these conclusions.

HI. Fundamental Principles of Survey Design

Courts have long accepted consumer surveys in lieu ofthe impractical presentation of voluminous data and testimony fi'om an entire population ofcustomers. These surveys are

conducted with the objective of generalizing measured opinions, attitudes, and. actions fmm a i sample to a much larger population. The Federal Judicial Centei's Manualfor Complex

Litigation cites seven technical factors that are among. those that a survey must conform to ~

(hereafter "MCL Criteria"). These are:

i) The population was clearly chosen and defined;

ii) The sample chosen was representative ofthat yopulation;

iii) The data gathered were accurately reported;

iv) The data were analyzed in accordance with accepted statistical principles;

v) The questions asked were clear and not leading; .

vi) The survey was conducted by qualified persons following proper interview I procedures; and

Federal Judicial Center, Manualfor Complex Litigation, 4, Section 11A93. vii) The process was conducted so as to ensure objectivity.

I show that, when measured against the principal objective of this proceeding, both the Bortz

Survey and the Horowitz Survey fail to comport with many of these criteria.

A. Surve Research Re uires Clear Ob'ectives

As my own textbook says, "research objectives should be both specific and limited. One of the greatest causes of dissatisfaction with research is vaguely worded or overly optimistic objectives, which are rarely achieved." Specific objectives guide the research, give it focus, and help ensure objectivity. In studies introduced as evidence in false advertising cases, for example, objectives typically test whether an allegedly misleading statement in a particular ad causes consumers to take away some specific impression from that ad. 8 Without precise well-defined objectives, it is impossible to draw precise well-defmed conclusions.

B. uestions Should Be Clear and Understandable

Survey questions need to be clear, precise, and unbiased. Even questions that appear clear can convey ambiguities to some respondents.9 Reliable survey research requires that each respondent understand the questions and the understanding is uniform across respondents. It is well established in the survey literature that ambiguous questions "written with ... words with

»10 various different meanings [are] ... more difficult for respondents to interpret." In The Art of

Lelunann, Donald R., Sunil Gupta, and Joel H. Steckel (1998), Marketing Research, Reading, MA: Addison- Wesley, p. 63.

For example, if a Marlboro cigarette ad hypothetically stated the following: "Nine out of ten doctors in New York city smoked Marlboro." The objective of a false advertising study could be to test whether such a statement caused consumers to conclude that Marlboro canies less of a health risk than other cigarettes.

Diamond, Shari S., "Reference Guide on Survey Research," Reference klanual on Scientific Evidence, Third Edition, Federal Judicial Center, 2011, p. 387.

'Qosnick J,, "Response Strategies for Coping with the Cognitive Demands for Attitude Measures in Survey," Applied Cognitive Psychology, 1991, p. 221. Asking Questions, Payne states, "[s]urvey questions ideally should be geared to embrace all levels ofunderstanding so that they have the same meaning for everyone."." He adds:

"The most critical need for attention to [surveyj wording is to make sure that the particular issue which the questioner has in mind is the particular issue on which the respondent gives his answers.

When faced with survey questions that are ambiguous or hard to interpret, prior literature .

has documented evidence that respondents tend to take shortcuts to reduce the cognitive effort i

used for interpreting the questions and formulating their ttns+ers. Speci6cally, respondents may i

reduce the cognitive effort necessary for generating answers by employing "cues in the question i itself to identify a response that seems easily defensible." This behavior is often referred to as satis6cing, and according to Krosnick and Presser:

Rather than expend the effort necessary to.provide optimal answers, respondents may take subtle or ~tic shortcuts. In the former case, respondents may simply be less tihoreugh in comprehension, retrieval, judgment, an'd response selection. Thev mav be less thoughtful about a question's meaning. search their memories less comprehensivelv: intemate retrieved information

less carefullv: or select a response choice lessipreciselv (emphasis: added).'n

other words, respondents who have difEculty understanding a question are unlikely to expend i the effort required to understand it.

In a more dramatic form of satisficing, often referredlto as strong satisficing, when faced . with an ambiguous question, respondents could. skip searching their memory completely and

" Payne, Stanley L., The Art ofAsking Questions, Princeton University Press, 1951, p.115.

'ayne, Stanley L., The Art ofAsking Quesiions, Princeton University Press, 1951, p.9.

Krosnick, J., A. Narayan, S. Smith, W. "Satisf|cing in Surveys: Initial Evidence" M.T. Braveman and J.K Sister (eds.) Advances in Survey Research. Jossey-Bass, 1996, p. 3l..

'rosnick J., Stanley P., "Question and Questionnaire Design,",,Marsden P., James W. (ed.), Handbook ofSurvey Research (Second Edition). Emerald Group Publishing Limited, 2010, p265:(emphasis added).

10 arbitrarily select an answer. According to Krosnick and Presser:

[R]espondents may interpret each question superficially and select what they believe will appear to be a reasonable answer. The answer is selected without reference to any internal psychological cues specifically relevant to the attitude, belief, or event ofinterest. Instead, the respondent may look to the wording ofthe question for

a cue, pointing to a response that can be easily selected and easily

defended ifnecessary. Ifno such cue is present, the respondent may select an answer completely arbitrarily.'uch

is likely to occur when the respondent is called upon to provide an answer to a question that presupposes the respondent has information related to or the ability to retrieve, but does not.

C. The Choice ofSurvev Medium

The past 30 years or so have seen survey researchers virtually invent a variety ofnew ways to collect survey data. Early in the history of survey research, data collection was typically done either face-to-face or via the mail. The late 1960s saw telephone surveys become more common. Today, computers and the Internet provide common vehicles to collect data (i.e. online or web surveys).

The four major modes of survey data collection, face-to-face, telephone, online, and mail each have their strengths and weaknesses. The proper choice depends on the survey task at hand and the inherent tradeoffs that the methods bring to that task. These tradeoffs include various sources of survey measurement error and cost. Sometimes data collection method choices are obvious. For example, a survey on literacy would have to be conducted either face-to-face or by telephone. Online and mail surveys would require respondents to read and those low in literacy would likely decline to participate or drop out after starting. Similarly, a survey on the usage

'rosnick J., Stanley P., "Question and Questionnaire Design," Marsden P., James W. (ed.), Handbook ofSurvey Research (Second Edition). Emerald Group Publishing Limited, 2010, p.265.

11 rates ofvarious websites would be most appropriately icondueted online since the only people i who use any websites are online.

Survey measurement error comes in many different forms.'hose that can be impacted by data collection mode include coverage error, nonresponse error, and measurement quality error. These must all be weighed against cost in deciding on a particular mode of data collection.

Coverage error refers to the inability ofa mode of~data collection to sample segments of the target population. Nonresponse error refers to the impact that failure to obtain data &om sampled individuals can have on desired measures; that is, when a segment ofpeople fail to cooperate with the survey request (either in whole or in part)~ survey measures can be biased.

Perhaps most important with respect to my analysis ofthe surveys at issue is measurement quality error. Data collected fmm individuals can be inaccurate for three reasons: a) respondents do not understand the question, b) respondents do not follow instructions, or c)

respondents are not willing to tell the truth.' have previously discussed what happens wh'en i

respondents do not understand the question. The same principle. applies to respondents followiing i instructions. Finally, respondents often are reluctant to reveal certain information about themselves or their attitudes for fear ofembarrassment. Measurement errors stemming from the

last ofthese are generally identi6ed as resulting from a 'social desirability bias.'or example,i in i a political poll conducted by one ofthe major parties, the xespondent might be more likely to express a preference for that party's candidate.

These three sources of error, coverage, nonresponse and measurement quality, impact the . appropriate mode of survey data collection. For example, ifcoverage error is an issue, either,

'f. Groves, Robert M., Floyd J. Fowler, Jr., Mick P. Couper, James M. Lepkowski, Eleanor Singer, and Roger Tourangeau (2009), Hoboken, NJ: Wiley, Chapter 2.

'~ Groves, Robert M., Floyd J. Fowler, Jr., Mick P. Couper, James M, Lepkowski, Eleanor Singer, and Roger Tourangeau (2009), Hoboken, NJ: Wiley, Chapter 5.

12 telephone, mail, or both combined merit consideration. Telephone coverage is about 97.5 percent ofAmerican households. Most households, but not all, can be accessed through voter registration lists. Web or Internet surveys have seen a rapid rise recently, but their coverage is still lower than other modes of data collection. Also, nonresponse error can be minimized through face-to-face interviews. It is harder to refuse an individual face-to-face than it is to discard a piece ofmail, hang up a phone, or ignore an email invitation to an online survey.

Survey mode also plays a role in the quality of data collected. In particular, survey mode can impact how well a complex question is understood. The underlying principle is that complex questions or ones with several response categories can tax the respondent's processing resources.

An example, based on a question in the Health Interview Survey, appears in the literature:

During the past 12 months, since January 1, 1987, how many times have you seen or talked to a doctor or assistant about your health"?

Do not count any time you might have seen a doctor while you were a patient in a hospital, but count all other times you actually saw or talked to a medical doctor of any kind.'his

question is comphcated to be sure. It covers face-to-face and telephone conversations, with doctors "of any kind" and "assistants." Respondents are also instructed to exclude such consultations that occurred in the hospital or ifthey did not concern the respondent's health.

Finally, there is a restricted time frame. Processing this question is likely to impose a burden on the respondent's working memory. Internal representations may omit part ofthe question's intended meaning. Even in the best of circumstances, satisficing strategies are likely to ensue.

For questions as complicated as the doctor count above, the choice of survey mode influences how well a respondent can actually process the question at hand. In particular,

'owler, Floyd J. (1992), "How Unclear Terms Affect Survey Data?", Public Opinion Quarterly, 56, 218-31; Torangeau, Roger, Lance J. Rips, and kenneth Rasinski (2000), The Psychology ofSurvey Response, Cambridge: Cambridge University Press, p. 38-40.

13

surveys that contain complex questions and or large numbers ofresponse categories bene6t &am .'r written presentation to respondents.'ritten presentations enable respondents to focus and concentrate on what the question actually means and keep the response categories in mind while '. processing the question. For these reasons, telephone surveys are inferior for complex questions 'o

questions with significant numbers ofresponse categories. Ifa respondent has the doctor count question in written form in front ofher, she can study it and more carefully account for what is to be included and what is to be excluded.

Mi~i~iring the most serious sources ofmeasurement error must be balanced against the costs ofthe different data collection modes. While each study is unique, my experience leads me

believe that face-to-face interviews are the most expensive, with mail and telephone being comparable. In general, Internet surveys are the least exyctisivei of all. Where the scales balancing cost with measurement error come out is a case-by-case proposition. However, the

one thing that cannot be compromised is data quality. Cutting costs at the expense of quality . data is never acceptable. As the old saying among social.scientists goes, "garbage in, garbage out."

IV. The Studies at Issue: The Bortz and Horowitz SurVeys

A. The BorN Survev

I Bortz has presented its surveys ofcable operators.on behalf ofJSC in many distribution proceedings with the goal ofdetermim~g how the compulsory licensing royalties cable systtemh pay to retransmit broadcast stations should be allocated. Bortz presented these surveys with minor differences implemented to account for ongoing cxjticj.smS. Previous decision makers

'illman, Don A., Jolene D. Smyth, and Leah Melani Guistian (2009),.Internet, Mail, andMixedMode Surveys:

The Tailored Design Method, 3 ed., New York: Wiley, p. 321. i

14 have used these surveys as a basis for distributing royalty payments to the Agreed Categories in past royalty distribution proceedings, I will focus on the study presented by Bortz to support the

allocation ofroyalties for the years 2004 and 2005 because it is the most recent and publicly

available,

Speci6cally the study had the following objective:

[T]o determine how cable operators valued, on a relative basis, the different categories ofnon-network distant signal television programing that they carried in those years.20

The Bortz Report justifies its approach by quoting an earlier report written by a CARP:

The critical significance of the Bortz surveys is the essential

question it poses to cable system operators, that is: What is the relative value of the type ofprogrammmg actually broadcast in terms of attracting and retaining subscribers? That is largely the

question the panel poses when it constructs a simulated market. Further, the question asks the cable system operator to consider the same categories we are presented here in the form of claimant — groups that is sports,'he movies, and the others. That is also what the panel must do.

2004-05 Bortz surveys estimated the relative values of the different categories ofnon-

network distant signal television programing and concluded that cable operators "would have

allocated the largest percentage of a distant signal programming budget (33.5 percent in 2004

and 36.9 percent in 2005) to live professional and collegiate sports

programming....approximately twice that ofthe next most highly valued program category."

Interestingly, the Bortz Report equates value to cable operators with how they would

have allocated their distant signal programming budget. I shall return to this point later as I

Cable Operator Valuation of Distant Signal Non-Network Programming: 2004-05, June 1, 2009 ('"Bortz Report").

'ortz Report, p. 2.

" Bortz Report, p. 3.

Bortz Report, p. 2.

15

believe this is a significant problem in the use of the Bortz (and Horowitz) Surveys. As a matter of econoniics, resources allocated do not directly translate to marketplace'undamental value one-to-one.

The sample Iin the Bortz Survey, was comprised fi'om "Form 3" systems, which the Bortz

Report claims account for over 95 percent of cable royalty payments. Bortz assembl.ed a stratified sample with four strata beirig dete~zrmned by copyright royalty pa.yments ($0-$20,628;

$20, 629-$59,628; $59,629-$2',07,129; $207,130 or more in 2004 $0-$2,3,844; $23,845-$65„344;

$65,345-$239,844; $239,8'45 or more in 2005). Cable systems were randomly drawn. &om'each stratum to comprise the final sample, Systems that either broadcast only PBS signals, ortiy

Canadian signals, or only PBS and Canadian signals were eliminated &om the analysis. In each ofthe 2004 and 2005 studies, ten such systems were discarded.

Potential respondents were contacted and survey data. were collected by telephone. The survey began by screening potential respondents to find the individual at the c:able system posit responsible for prograniming decisions. Ifthat was not the person picking up the phone, that person was asked to find the appropriate individual. The qualified respondent group consisted of general managers, marketing directors/managers, and programming directors/managers. The respondents were read a list of the distant signals their systems carried according to filings they made at the Copyright Offjice. They were instructed to consider only the non-network programming on those distant signals.

The telephone inteiview then proceedecl with several preparatory filler questions. These questions were inset&ed in the questiormaire at this poIint "to focus the responclent on the, value of

'ortz Report, p. 45, Table A-,, footnote. The report justifies this by expressing a desire to focus on Phase I program categories (Bo~&z Report, p. 32).

Cf. Bortz Report, Appendix B.

16 various programming types." More accurately, the questions appear to be designed to elicit a subset of the following pre-specified categories ofprogramrmng featured in 2004 (or 2005) advertising and promotion to attract and retain customers:.27

~ Movies

~ Live Professional and college team sports

~ Syndicated shows, series and specials

~ News and public affairs programs

~ PBS and all other programming broadcast by noncommercial stations

~ Devotional and religious programnung

~ All programming broadcast by Canadian station

~ Other (to be specified by the respondent).

Finally, respondents were then asked what I generally refer to as the "money question"; meaning, the one that directly supplies the data most relevant to the purposes of the survey. The money question in the Bortz survey asks respondents to allocate a distant signal non-network

29 programming budget among different program categories, Specifically, the question asks:

Now, I would like you to estimate the relative value to your cable system of each type ofprogramming actually broadcast by the stations I mentioned during 2004 (or 2005), other than any national programmmg from ABC, CBS, and NBC. That is, how much do you think each such type ofprogramming was worth if anything, on a comparative basis, in

Bortz Report, p. 12.

'ortz Report, Appendix B.

28 These 2004-05 program categories map into the program categories represented by the claimants in this proceeding as enumerated on page 6 as follows: Movies, and Syndicated shows, series and specials map into Program Suppliers; Live professional college and team sports map into JSC; News and public affairs programs map into Commercial Television Claimants; PBS and all other programming broadcast by noncommercial stations map into Public Televisions Claimants; Devotional and religious programs map into Devotional Claimants; and All programming broadcast by Canadian station map into Canadian Claimants, with certain exceptions.

Bortz Report, Appendix B.

17 terms ofattracting and retaining subscribers. We are only interested in

U.S. commercial station(s) , U.S. non-commercial station(s), and i Canadian station(s)

I'l read all the program types that were broadcast by these:stations to give you a chance to think about them; please write the categories down as I am reading them. (READ PROGRAM TYiPBS IN ORDER OF RANDOM SEQUENCE NUlMBBR) Assume you had a fixed dollar amount to spend in order to acquire all the pro~~~ing actually broadcast during 2004 (2005) by the stations I listed. What percentage, ifany, of the fixed dollar amount would you spend for each type ofyrol~~mi~g? Please write down your estimates, and make sure they add to 100 percent.

What percentage, if any, ofthe 6xed dollar amount would you spend on (READ FIRST PROGRAM TYPE)? And what percentage, ifany, would you spend on (READ NEXT PROGRAM.TYPE)? (COMPLETE LIST IN THIS MANNBR)

~ Movies broadcast during 2004 (2005) by the U.S. commercial stations I listed.

Live Professional and college team suorts broadcast during 2004 (2005) by the U.S. commercial stations I listed.

~ Svndicated shows. series and specials distributed to more than one television

station and broadcast during 2004 (2005) by the U.S. commercial stations I list&A l

~ News and public affairs promams produced by or for any ofthe U.S. commercial stations I listed, for broadcast during 2004 (2605) only by that station.

~ PBS and all other proexammine broadcast during 2004 {2005):by noncommercial station

~ Devotional and relimous promam~i~d broadcast during 2004 (2005) by the U.S commercial stations I listed.

~ All pro~~~i~a broadcast during 2004 (2005') bv Canadian station.

Respondents were asked to review and reconcile the percentages so that they added up to 100 percent. Finally, they were thanked and dismissed.

B. The Horowie Survev

In response to perceived flaws in the Bortz Survey, the MPAA commissioned Horowitz

Reseaxch, Inc. to design and conduct a survey that carefully replicates the methods and procedures ofparallel surveys conducted by Bortz on behalf of JSC. The Horowitz Survey was

designed as part of the effort to improve upon the previous efforts made by Bortz by solving

some of the information and category weaknesses as pointed out by the Judges in the 2004-2005

Cable Phase I Proceeding. 'imilar to the Bortz Survey, the Horowitz Survey had the objective

of assessing the relative value of the non-network programming carried on distant signals f'rom

the point ofview of local cable system executives, Also, similar to the Bortz Survey, Horowitz

Research conducted a series of studies over the years.

The Bortz Report describes the sampling procedure in great detail. For sample selection,

the Horowitz Report states that it relied on the work ofProfessor Martin Frankel, an expert

statistician, who has testified many times in these proceedings and who enjoys an excellent

reputation, I reviewed Professor Frankel's sampling procedure in connection with the Horowitz

Survey, and his procedure was similar to Bortz's.

As implied earlier, the Horowitz Surveys were adapted from the Bortz Survey.

According to the Horowitz Report, the survey instruments between the Bortz and Horowitz

surveys differed in several important ways:.34

1) Unlike Bortz which has only one category for sports, the Horowitz Survey distinguishes between live professional and college team sports broadcasts and other sports programmmg (such as NASCAR auto races, professional wrestling, and figure skating);

Horowitz Testimony at 3.

"Id. at 4.

"Id. at 4.

Horowitz Testimony, p. 3-4.

JSC are only entitled to royalties from live professional and team-college sports broadcasts. Notice at Exhibit A.

19 2) Horowitz enhances program category descriptions by providing examples;

3) The Bortz Survey asks for resource allocation &om a complete predeflned list (which does not separate live team and non-team sports). The Horowitz Survey customize8 its ilist ofprogram categories to be queried about accoidixig tb wbich ofthe following categories a cable system belongs tb: a. Non-network carrying systems; i. Canadian stations only ii. PBS stations only iii. WGN only iv. Other independent stations b. Network caxrymg systems;

4) The Bortz Survey repeatedly asks questions about types ofprograms

during 2004 (or 2005) across all stations "other than any national network programming from ABC, CBS and NBC.'n contrast, the Horowitz Survey continually remmds respondents about the specific broadcast stations at issue;

5) The Bortz Survey does not include Isystems that carry only PBS stations or systems that carry only Caxxadian stations as distant signals. The Horowitz Survey does;

6) The Horowitz Survey provides "warmup"i questions intended to enhance the likelihood ofwell-reasoned, non-reflexive responses later in the survey. The Bortz Survey does not do this; and

7) Unlike the Bortz Survey, the Horowitz Survey reminds respondents not to assign any value to programs that are substituted for WGN's blacked out programming. The Bortz Survey simply lists the program categories it is asking about. The Horowitz Survey gives explanatio'ns 'nd examples. For example, Bortz asks about "Syndicated shows, series and specials." In contrast, Horowitz asks about "Syndicated series, such as sitcoms, dramas, children's shows, talk shows, reality shows, and other series broadcast on INSERT STATIONS... Examples include program's such as Ev'erybody'oves Raymond, Seinfeld, i American Idol, Jeopardy, and the Dr. Oz Show." Similar clarifications are given for movies, devotional programs, and two categories of sports programming. These clarifications lengthen the survey; however, they do not complicate it. To the contrary, they resolve ambiguities that may exist in a respondent's interpretation ofthe question. To be sure, length does not necessarily complicate.

Bortz Report, Appendix B.

20 V. The Bortz and Horowitz Surveys Cannot Satisfy the Research Objectives In the Current Matter.

The Bortz and Horowitz Surveys are both incapable ofproviding valid data with respect to both the stated research objectives (i.e., to determine how cable operators valued, on a relative basis, the different categories ofnon-network distant signal television programing that they carried in those years) and those that I understand are required in this matter (i.e., to determine the relative marketplace value of the distant broadcast signal prograrrnning retransmitted by cable systems during 2010-2013). In particular, I note that the stated and required research objectives differ. The stated research objectives refer to how cable operators value the programming while the statutory objective is to determine the marketplace value ofprograms.

These are not necessarily the same.

A. The Bortz-Horowitz Surve 0~ ectives Do Not Matc& the Statttto

Marketplace value reflects the total financial return that an investment or asset will command in the relevant marketplace. In the context ofthis proceeding, it reflects the profits that a cable operator would accrue by investing in the rights to retransmit bundles of certain programming aired on distant broadcast signals. For the Bortz-Horowitz objectives (i.e., to measure cable operators'aluations) to correspond to the statutory requirements ofmarketplace value, cable operators would have to be able to perfectly assess the total market returns of each oftheir programming investments and assets. Ifthis were true, it would be a unique circumstance in the spectrum ofbusiness decision-making.

Bortz Report at 10.

Distribution ofthe 2004 and 2005 Cable Royalty Funds, 75 Fed. Reg. 57063, 57065 (September 17, 2010).

21

Ifmanagerial or executive judgment could predict marketplace returns, there would be no bad investments, all new products would succeed, and the market research: industry would 'anish.

However, like all h»r»n»s, managers, including cable operator respondents, suffer from limited information processing capacity. To cope, managers routinely resort to.intuition- and

heuristics-based decision-making processes. 'n day-to-day activities, judgmental heurhtics .

generally produce satisfactory outcomes. Unfortunately, heuristics also make decision-makers to a variety ofcognitive biases that oAen degrladeljudgment quality in more complex 'usceptible

situations. The literature documents dozens of such biasds." I I shall return to:these biases shortly when I discuss how the complexity ofthe money question distorts the data it produces.

In sum, the CARP's assertion about the constant sum question quoted in the Bortz Report

''that (it) is largely the question the panel poses when it constructs a simulated market's not correct. The panel is being asked to do something entirely different. As such, neither the Bortz

nor the Horowitz Survey processes provide an objective measure ofthe required construct i thereby violating the seventh MCL criterion described earlier.

Kabneman, Daniel (2003), "Maps ofbounded rationality: Psychology for behavioral economics."American Economic Review 93 (5): 1449-1475; Simon, Herbert A. (1957) Models ofman: Social and rational. New York: i Wiley.

'azerman, Max H. and Don A. Moore (2012) Judgment in managerial decision making. Eighth Edition. Hoboken, NJ: John Wiley and Sons, Inc.; Kahneman, Daniel and Amos Tversky (1979) "Prospect theory".Am analysis ofdecision under risk." Econometrica 47 (2): 263-291.:

Gigerenzer, Gerd and Reinhard Selten (2001), Bounded rationality: The adaptive toolbox. Cambridge, MA: MIT Press.

Bazerman, Max H. and Don A. Moore (2012) Judgment in managerial decision making. Eighth Edition. Hoboken, NJ: John Wiley and Sons, Inc.; Sutherland, Stuart. 2007. Irrationality. Pinter and Martin Ltd., UK.

Bortz Report, p. 2.

22 B, The Bortz-Horowitz Surve Format Asks Res ondents to Focus on an Arti tcial Construct and Presume an Activi That The Do Not En a e In In and Does Not Exist In Their Dail Business Activities

The Bortz and Horowitz Surveys each ask respondents to a) "estimate the relative value to your cable system of each type ofprogramming actually broadcast; ...(i.e.) how much do you think each such type ofprogrannmng was worth if anything, on a comparative basis, in terms of attracting and retaining subscribers," and b) to allocate "a 6xed dollar amount to spend in order to acquire all the programming actually broadcast..."

These requests presume that respondents have something in their minds akin to relative value of different types ofprogramming. The Horowitz Report points this out. It says,

The premise of the Bortz Survey is that how local or regional CSD executives allocate their programmmg budget reflects their proportional assessment of the relative market value ofthe categories ofprogramming actually carried by the distant signals on their respective systems,

This premise cannot possibly be true. Cable executives do not make decisions about individual programs or the various categories ofprogramming employed in this proceeding.

They make decisions about television stations and cable networks. A survey investigating the relative value of stations and networks might be able to tap into executives'ecision making regarding the value of entire channels. Dne investigating relative value ofprogramming categories that cut across stations cannot. Social scientists would refer to this as an (externally) invalid construct; that is, one with no manifestation in the real world.

Such a construct would be difficult for a respondent to evoke, because she will not have any experiences in memory upon which to base a judgment. To illustrate, imagine you are asked how tall someone you are standing next to is, then imagine you are asked how long his arm is.

'orowitz Testimony, p, 21.

23 People make judgments like the first all ithe time. Those are easy. They harcHy if ever make

judgments like the second. Those caimot be relied upon. ~

To further illustrate, consider the following exercise I often use in my classroom. I stand up on a table in front ofthe room. I ask the class to individually and silently assess the distance f'rom the tabletop to my kneecap and write it down. I follow this by asking them to do the same thing for the distance f'rom my kneecap to my waist, my waist to my shoulder., and my shoulder to the top ofmy head. Then I ask them how tall I am. Finally, I ask them to add up the 6rst four judgments. The result is always the same. Most students coiIredtly assess that I am within an inch or two ofmy correct height, 5'10". When asked about the sum ofthe four components that should total up to the same thiing, I am often either below five feet or above seven feet.

The lessons are clear. Asking people to make judgments about unfamiliar. constructs produces invalid and urireli.able results. People do not think iin those terms. As such, when faced with an unfamiliar construct, respond ents must construct their own mental model of:it "on the spot." Research has shown that suchjudgments are transient and do not persist beyond thIe specific measurement. 'urthermore, brealdng down an entity Such as a cable network into its components — stations carried, specific programs, and to which con, tructed program category each program belonged — exacerbates that lack of validity~ and reliability. I shall return to the topic ofunfamiliar construct measurement in di.scussing the specific constant sum money question. There the effect ofunfamiliarity is accentuated'by the'question's complexity.

Nevertheless, the lack of famiharity in and ofitselfprevents the question from being clear and ~ therefore constitutes a violation of the 6fth aforementioned MCL'riterion.

Payne, J.W., J.R. Bettman, and D.A, Schkade (1999), "Measm'ing Constructed Preferences;, Toward a Building Code," Journal ofRisk and Uncertai!nty, 19:1-3, 243-270.

Simon, D., D.C. Krawczyk, A. Bleicher, and K. Hoiyoak (2008), "The Transience of Constructed Preferences," Journal ofBehavioral Decision Making, 21 (January), 1-14.

24 C. The Constant Sum Resource Allocation in the Mone uestion is Ambi uous. Furthermore that Ambi i Introduces Inconsistencies In the Unit o Anal sis and Biases Results in Favor o Smaller Cable S stem ~Oevators

The Bortz and Horowitz Survey questions are framed as if each respondent were responsible for only one system, When the word 'system'ppears in the text of a survey question it is always in the singular form. However, we know that many (ifnot most) respondents were responsible for multiple systems. The Horowitz Testimony demonstrates that each respondent was responsible for between seven and ten cable systems on average depending on the year. The Bortz Report is silent on this issue. Nevertheless, it would be beyond the bounds of credibility to presume that each Bortz respondent was responsible for one and only one system.

Responsibility for multiple systems raises two problems. I'irst, it is not clear how a respondent responsible for multiple systems is supposed to mentally process and answer a question framed in the singular. Is s/he supposed to pick the largest? Pick one at random? Or use an average of all the systems s/he is responsible for? Without clarity on this point, all questions about the programming on a system are ambiguous. Different respondents could be answering in different ways. The ambiguity in the questions on the two surveys represents a violation of the fifth aforementioned MCL criterion.

Second, the framing ofthe surveys biases their results in favor ofthe small cable operators. The data are collected and tabulated with the unit of analysis being the respondent cable system executive, not the cable system. Programs are viewed and customers are acquired within the context of a system and that should be the unit of analysis. The data &om an operator

Bortz Report, Appendix B; Horowitz Testimony, Appendix A.

" Horowitz Testimony, Appendix B.

25 responsible for ten c'able systems are inappropriately treated the same as the data from an operator responsible for only one in the Bortz Survey And thei H6rowitz Survey. Such a bias represents a violation ofthe fourth aforementioned MCL criterion.

D. ?'he Constant 'Sunz Resource Allocation in the Monequestion Does Pot Elicit Data that Corre~sond to the Relative ktarket!Val~u~e7uestion.

For the constant sutn resource, allocations to provide data useful for the Bortz/Horowitz! stated research objectives (i.e... relative market value) total marketplace returns for each program category would have to exltibit a one-to-one correspondence with the resources allocated to that category. Even if cable operators were able to accurately forecast marketplace returns oftheir programming investments {which they clearly cannot) and therefore render the stated research objectives equivalent to the statutory requirements., those returns are unlikely to exhibit a one-to- one correspondence with the resources allocated to each investment.

The reason resource allocations are not likely to correspond to ruarketpiace value is a matter ofbasic economic theojry. Basic economic theory demonstrates that resources are optimally allocated across .investments (in tins case programming category) so that marginal returns are equal. In other words, resources should be allocated so that the return from an additional dollar ofinvestment into each programming type would be the same. In contrasts, arty presumed equivalence between resource allocations and marketplace value rests on total return, not marginal return.

Consider the following stylized example. A cable'phratior has to invest 500 dollars in movies and non-network news. Suppose movies are much more popular and bring a greatel market return because people will watch any movie, but consumer viewership is much more

" This principle is a consequence of the basic mathematical theory of constrained optimization (cf. Carter, Michael (2001), Foundations ofMati'zem&ztical Economics, Cambridge, MA: MIT!Pre."&s, Chapter 5, Section 3), It results from the simple case where the objective to be optimized (e.p., returr& or revenue) is separable across categories and the constraint (e.g., budget) is additive across the separable categories (cf. Example 5.29 in Carter's text).

26 sensitive to which channel the news is coming &om and therefore investing in more news programrrung will likely bring greater incremental return than investing in more movie

programming. The following table details the hypothetical return that different levels of

investment in the two categories would bring:

I I $0 $ 100,000 $0 $ 10,000 $100 $100,250 $100 $ 11,250 $200 $100,400 $200 $ 12,500 $300 $100,500 $300 $ 13,500 $400 $100,550 $400 $ 13,750 $500 $100,575 $500 $13,825

The payoffs in this table conform to the typical expectation that greater investments bring greater

returns, albeit at different increments.

The above table can also be used to compute the optimal allocation ofthe 500 dollars

across the two programmmg categories. The required calculations are reflected as follows:

Movies Non-network News Movies 'on-network News $0 $500 $ 100,000 $ 13,825 $113,825 $ 100 $400 $ 100,250 $13,750 pl14 000 $200 $300 $ 100,400 $13,500 $113,900 $300 $200 $100,500 $12,500 $112,750 $400 $ 100 $ 100,550 $11,250 $ 111,800 $ 500 $0 $100,575 $ 10,000 $ 110,575

The first two columns of this table present the possible allocations of the 500 dollars in hundred

dollar increments. The next two columns simply rearrange and reformat the payoffs in the prior

table for the allocations presented in the first two columns, The fifth and final column computes

the total return for each allocation by adding the returns &om the previous two columns, The

27 highest total return is $ 114,000 and so the optimal allocation of .investment resources is $ 100

(20%) to movies and $400 (80%) to non-network news. These investments have equal marginal return ($250) for a $ 100 investment increase. This is how a rational cable operator would respond to how s/he would allocate resources. On the other hand, a resource allocation according to marketplace value would. be approximately $ 100,000/$ 110,000 =- 91% to movies, very, very different. Clearly, resource allocations do not in general correspond to relative marketplace value.

In survey research parlance, the constant smn allocati'on .lacks construct validity with respect to measuring marketplace value. Construct validity refers to the question ofwhether the

survey measure is designed to measure what it:is supposed to.51 Here, there is a fundamental mismatch between the foundation of resource allocations, marginal renlrn (that is, what the,

survey in these proceedings needs to measure) and total return (what the Bortz and Horowitz

Surveys purport to measure). This lack of equivalence demonstrates the violation of another MCL criterion, the fourth. As measured against the objective ofmeasuring market value, the 'onstant

sum allocation does not allow for the data to be analyzed in accordance with accepted statistical principles.

The Bortz/Horowitz Survey format presents other threats to construct validity, mostly clue to the complexity of'the const mt,sum money question.

The Constant Sum Mone~uestion is Too Co~miLex to Produce VaLid L)ata

The more complex a question is, the more difficult it will be to elicit a valid answer. As discussed earlier, respondents will satisfice and look for shortcuts called heuristics to reduce the

'oyle, Rick H,, Monica J. Harris, and Charles M. Judd (2002) I Research Methods in Sociai Reiatians, Independence, KY: Wadsworth, p. 32.

28 cognitive effort used for interpreting the questions and formulating easily defensible (at least to themselves) answers. People use heuristics to allow them to cope with complexities ofthe environment in which they make judgments. In survey research, respondents find heuristics useful in that they allow them to ignore the complexities ofthe question they are being asked.

As such, any responses formulated from these heuristics are potentially biased and wholly unreliable,

It would be hard to find a survey question more complex than the constant sum money question. Consider the following cognitive steps a respondent must go through in arriving at an answer to that question in the latest Bortz Surveys:

i) Recall the station(s) carried by the cable system

ii) Recall all types ofprogramming offered by the station(s) &om short term

memory;

iii) Mentally separate out all programming &om ABC, CBS, and NBC networks,

remember that Fox is not considered a network for compulsory license

purposes, and only simultaneously retransmitted programs on WGNA are

compensable;

iv) Organize the remainder ofthe programming on the stations carried into the

program categories required by the survey;

v) Identify &om the types ofprogrammmg organized in item iv) the particular

ones that were featured in subscriber acquisition and retention advertising and

promotion;

Kahneman, Daniel, Paul Slovic, and Amos Tversky (1982), Judgment Under Uncertainty: Heuristics and Biases: Cambridge, UK: Cambridge University Press.

29 vi) Retrieve a11 progrsrnrning acquisition costs in 2004 (2005) from short term,

memory;

vii) Map the unit ofacquisition (e.g., channel or network) to the categories af

progralxImmg offered;

viii) Allocate the costs in step vi) according to lthel map derived in step: vii);

ix) Add up the costs in step vi) or step,viii); 3

x) Divide each ofthe costs in step vii) byi the total in step ix); and

xi) Review steps i) — x) as demanded by question 4b.

In order to provide thoughtful answers to the constant sum money question, a respondent must go through all eleven steps (or equivalently the first ten twice). The sheer number. of calculations is daunting in aud of itself, but what makes it more iso are steps ii), vii), and viii).

Step iv) complicates things because it imposes a restriction on the remaining calculations that the. respondent must keep in mind throughout. In other ~ops, proxy step iii) on the respondent can never think ofhis business as a whole. To the extent that the cable operator does not think:ofhis: business in terms ofthe component isolated in step iv), he isibeilng asked to make an unfamiliar judgment. Similar issues infect steps vii) and viii). Cable operators make decisions about acquiring stations, not program categories. To the extent that stations have more than one: program category, asking respondents to separate their programming into categories again puts them on unfamiliar ground. As such, somewhat like the person being asked to judge how )ong an arm is, cable operator respondents 6nd no information in their memories that directly matches the task they are being asked to perform. However, their task is much more diKcult. Cable operators have to perform other calculations to translate what they do know into other

These totals should be the same since one is simply a reallocation ofthe other.

30 dimensions, At least we know from the measurement exercise I conducted with my students what an arm is and it is something we see every day.

Without directly relevant information stored in memory, the structure of the constant sum money question gives respondents no choice but to satisfice in constructing their answers. In searching their memories for information, cable operators will not retrieve all potentially relevant information that comes to mind to form a mental model of the situation that they are being asked about. Instead, they will invoke heuristics that allow them to truncate the search process as soon as they have identified enough information to allow them to make a defensible judgment.

Extensive information search will only occur when respondents are highly motivated because theirjudgments have important personal consequences. This is surely not the case in the

Bortz/Horowitz Surveys. Heuristics are the rule with few, if any, exceptions.

Three of the most important heuristics in survey research are described below:

~ Availabili Heuristic: People judge the frequency, probability, or likely

causes of an event by the extent to which instances ofthat event are readily

"available" in memory;

Sudman, Seytnour, Norman M. Bradburn, and Norbert Schwartz (1996), Thinking About Answers: The Application ofCognitive Processes to Survey Methodology, San Francisco: Jossey Bass, Chapter 3.

Bodenhausen, G. V. and Robert S. Wyer (1987), "Social Cognition and Social Reality: Information Acquisition and Use in the Laboratory and the Real World," In H.J. Hippler, N. Schwartz, and S. Sudman (eds.), Social Information Processing and Survey Methodology, New York: Springer, pp. 6-41.

56 I~glanski, Arie W. (1980), "Lay epistemo-logic—process and contents: Another look at attribution theory," Psychological Review, Vol 87(1), Jan, pp, 70-87.

Torangeau, Roger, Lance J. Rips, and Kenneth Rasinski (2000), The Psychology ofSurvey Response, Cambridge: Cambridge University Press, Chapter 5.

Tversky, Amos and Daniel Kahneman (1973), "Availability: A Heuristic for Judging Frequency and Probability," Cognitive Psychology, Vol. 5, No. 2, 207-32.

31 ~ Reoresentative Heuristic: When making @judgment about an individual,

object, or event, people tend to look for traits that individual, object, or evexrt

may have in common with previously formed stereotypes. In other words, t4,

speci6c individual, object, or event, is assumed to be representative Rom a

group to which it belongs; and

~ Anchorine and Adiustment Heuristic: iPeople asked:to make a numerical

judgment settle on an initial value, either an arbitrary one or perhaps the result ofa partial computation, and then adjust it to account for subsequent events.'t

is easy to see how any or all of these heuristics could reinforce each other and lead to biasedjudgments. For example, the complexity of a numerical judgment task could lead to respondents invoking an anchor and adjustment heuristic. In order to choose the anchor, a respondent could rely on availability and/or representativeness heuristics. Suppose that a cable operator respondent just acquired a single programming source, say the ABC suite of channels

(e.g., WABC New York, KABC Los Angeles, WLS Chicago). He has some idea ofwhat hejust bought; how much news, how much sports, etc. (Availability Heuristic). But he does not know how to integrate those numbers with the rest ofhis progrsrnrnIrIg portfolio, He assumes that the

ABC package for the most part is similar to general progrsrnrni ng overall (Representativeness

Heuristic). He also knows that this portfolio is a bit light on movies relative to the rest ofhis usual programming so he adjusts the percentage ofmovies uy and that for news down

(Adjustment). He does not know how large these adjustments need:to be so he has to essentially: guess. He has some estimate of time allocation here, but does not know how this translates pinto

Kahneman, Daniel and Amos Tvezsky (1973), "On the Psychology ofPrediction," Psychological Review, 80, 237- 51.

Tversky, Amos and Daniel Kahnemann (1974), "Judgment under Unoartamty: Heuristics and Biases," Science, 185, 1124-31. either total return or resource allocation dollars. Without that knowledge, he assumes the ratio of time to resource allocation is simply one-to-one. That leads to the final response to the constant sum money question. Given how this process is loaded with heuristics and short cuts, there is absolutely no reason its outcome would be anywhere in the neighborhood ofthe outcome ofthe earlier-outlined eleven step rational process.

The judgment and decision making literature is filled with literally dozens, ifnot hundreds, more ofthese heuristics and their resulting biases. Some may lead to correct judgments with some frequency, but there is no telling how often. It follows then

that the availability heuristic, representativeness heuristic, and the anchoring and adjustment heuristic can separately or together lead to erroneous and biased judgments.

The use of these heuristics and the resulting biases are not restricted to naive judges and decision makers. Experts are vulnerable to them as well. For example, in one study, neurosurgeons (a high mortality specialty) estimated entire in-hospital mortality rates to be more than double those estimates made by plastic surgeons (a low mortality specialty). 'ince one' own experience is more available than others', this result demonstrates the availability heuristic.

Overall estimates are more consistent with the surgeons'ost available information. In another landmark study, the researchers found that scientifically trained research psychologists employed the representativeness heuristic and were vulnerable to the resulting biases.62 The prevalence of heuristics and their resulting biases in other highly trained professionals strongly suggests that cable operator respondents are vulnerable to them as well, especially when the questions are as

" Detmer, D.E., D.G. Fryback, and Ik, Gassner (1978), "Heuristics and Biases in Medical Decision-Making," Journal ofMedical Education, Vol 53, 682-3.

Tversky, Amos and Daniel I&abneman (1974), "Judgment Under Uncertainty: Heuristics and Biases," Science, New Series, Vol. 185, No. 4157. (Sep. 27), pp. 1124-1131.

33 complex as the constant sum questions in the Bortz and Horowitz Surveys. This complexity contributes to the degree to which the: surveys violate the fifth MCL criterion.

C.'onstant Sum Questions in General Do 2V'ot (~ature Real World .Behavior

Both the Bortz and Horowitz Surveys ask respondents what percentage (adding up to a total of 100) of a fixed budget they would allocate to each of!a set ofprogram categories considering the value of each type ofprograrrurung to their respective cable systems in attracting and retaining subscribers..Respondents are,asked to allocate a set of chips or points according to the relative importance for or preference ofthe categories to which the chips or points are being allocated. Constant sum scales are relatively popular because they are simple to implement.

However, that does not make them reliable or valid.

I understand that in a prior proceeding a Professor Reid has extolled the virtues of constant sum scales. I also understand that he, bases that view largely on two early research articles published in marketing journals, one wiitten by Mr. Joel Axelrod and another written Uy

Professor Russell Haley and htr. Peter Case, I also underktaijtd that Bortz based its choice on a:

1998-99 CARP decision that notes that "uncontroverted testimony and years ofresearch indicate rather conclusively that constant sum methodol!ogy, as utilized iIi the Bortz survey, is highly predictive of actual marketplace behavior." The 1998-99 CARP decision was wrong to rely on these assertions in the Bortz Report.

Professor Alan Rubin has pointed out flaws in the mere application ofthe prior literature

,65 to the current matter. In particular, he correctly points out that the papers cited 'by Professor !

Testimony ofAlan M.. Rulbin, Ph.D. to the Copyright Royalty Tribunal for Ithe 1989 Cable Royalty Distribution Proceedings, November 1991 (Rubin 1991 testimony), p„5.

" Bortz Report (p.37) quoting the 1998-99 CARP report at 21.

Rubin 1991 testimony, pp. 6-7'.

34 Reid studied constant sum use in a face-to-face interview context, In fact, the Axelrod paper cautions against using constant sum measures in a telephone interview. 66 Mr, Axelrod explicitly recommends five measures over constant sum when the interviewing is done by telephone (as is the case in the Bortz Survey and the Horowitz Survey).

There are other flaws in Professor Reid's application of the Axelrod results. Specifically, the Axelrod studies were done in a very different context, purchase behavior as opposed to resource allocation. Mr. Axelrod himself acknowledges that the validity of the technique will vary across contexts. In particular, he writes that his "results do not necessarily apply to testing alternate stimuli..." Indeed, he even demonstrates that with seemingly slight variations in context, the validity of the technique can even vary, He shows that the effectiveness of constant sum measures varies between predicting repeat purchase and brand switching, with the constant sum scale being much better in predicting repeat purchase than brand switching.68 He also acknowledges that the validity of these measures may vary between price brands and premium brands.

Furthermore, recent literature has not been as kind to the external validity of constant stun questions and their ability to predict marketplace behavior. Constant sum questions reflect direct measures of category importance because they are based on direct questions. A recent study by Jordan Louviere and Towhidil Islam highlights the principle that indirect measures based on choices outperform direct measures of category importance because they provide

Axelrod, Joel N. (1968), "Attitude Measures That Predict Purchase," Journal ofAdvertising Research, Vol. 8, No. 1, pp. 3-17, p. 8.

Id., p. 16.

Id., p. 12.

IQ., p. 16 — 17,

35 "richer insights into tradeoffs; and as they have a natural linla with real choices ofmanagerial interest, they should be more externally valid."

Significant research has gone into developing and testing methodologies to estimate relative importance and relative values of categories to a given population. The Louviere and,

Islam study has examined several common existing methodologies in this regard. More. recently

Professors Oded Netzer of Columbia University and V. Srinivasan of Stanford have develope@a i measurement approach based on paired comparisons 'hat has been shown to out-predict

constant sum scales by a score of 82% correct to 60%. The message is clear. Constant sum . questions are simple, but they do not exhibit the strongest predictive validities. Consistent with that, sophisticated practicing marketing researchers (and academics) are using different methodologies. The incorporation ofconstant sum questiionsi in the Bortz and: Horowitz Surveys constitutes the use of an improper interview procedure and therefore violates the sixth MCL~ criterion.

G. The Bort'/Horowitz Survev Should Not Be Adnrinistered Bv Televhone

Both the Bortz and Horowitz Surveys were conducted by telephone, and without making the survey questionnaire available to respondents in advance ~of er during the interviews. In particular, surveys that contain complex questions and/or ,'large numbers ofresponse categories

Louviere, Jordan J. and Towhidil Islam (2008), "A comparison ofimportance weights and willingness-to-pay measures derived from choice-based conjoint, constant sum scales and best—worst scaling," Journal ofBusiness Research, 61, 903-911, p. 910.

'etzer, Oded and V. Srinivasan (2011), "Adaptive Self-Explication of 'Multi-Attribute Preferences," Journal of: MarketingResearch, 48 February (1), 140-156.

V. Srinivasan and Gordon A. Wyner (2009), "An Improved Mhtho8 foi'hd Quhntitative Assessment of Customer Priorities," Stanford Graduate School ofBusiness Research Paper No. 2028. Available at SSRN: htto://sam.corn/abstract=.1435094 or htto://dx.doi.ore/10.2139/asm.1435094.,

36 benefit from written presentation to respondents.73 Written presentations enable respondents to focus and concentrate on what the question actually means and keep the response categories in mind while processing the question, For these reasons, telephone surveys are generally inferior for complex questions or questions, such as were asked in the Bortz/Horowitz studies in this matter. The complexity of the questioning and the context ofthe questions suggest those complications can only be exacerbated by the use ofthe telephone to administer it without respondents having the opportunity to have a written document to help structure their task.

Satisficing (and unreliable and invalid data) are the likely result. Such use ofimproper interview procedures constitute a violation ofthe sixth MCL criterion

H. The Lesser o Two Evils

The previous discussion makes it clear that I believe both the Bortz and Horowitz

Surveys do not provide an appropriate basis for the measurement of the relative marketplace values of the various programming categories. Nevertheless, counsel has asked me to render an opinion as to which survey is better, or, to put it differently, more persuasive. If I absolutely must render an opinion on the question, that opinion would have to be that the Horowitz Survey is better.

Both surveys investigate an artificial construct and presume behavior that does not exist.

They have similar constant-sum money questions, neither ofwhich corresponds to the statutory marketplace return requirement. So even ifthe data produced by either survey were reliable, they would not be valid for the purposes for which they were intended (i.e., estimating marketplace values). Furthermore, both surveys suffer from the limitations ofthe constant-sum allocation and the use oftelephone surveys described above.

Dillman, Don A., Jolene D. Smyth, and Leah Melani Christian (2009), Internet, Mail, andMixedMode Surveys: The Tailored Design Metltod, 3" ed., New York: Wiley, p. 321.

37 That said, the seven changes Horowitz Research made to the Bortz Survey allow the

Horowitz Survey to somewhat better cope with the complexity ofthe constant sum money . question. In particular, the provision ofprogram category descriptions:(change 2), the customization ofprogram category lists (change 3), and continually. reminding respondents about the speci6c broadcast stations at issue (change 4) supp'ortithei Grst 6ve steps ofthe eleven step process involved in answering the constant sum money question that I described earlier. They provide structure for the cognitive activity needed to get to the sixth step.

In addition, the breaking out of sports into two'ategories, live college: and professional team sports and all other (change 1), represents a significant improvement in the survey by

reducing ambiguity in the constant sum money question.:To:the extent that the JSC are only entitled to royalties ftom live college and professional team sports, the omission ofall other

sports in the constant sum money question in the Bortz Survey is problematic. It raises the. question as to how cable system operator responses would have (and should have) treated the resources they allocated to other sports in answering the Que4tioh about "all the progrnrn~ing 'ctually

broadcast during 2004(5) by the stations...listed.": The Bortz study constant sum budget allocations summed to 100 percent. The money allocated to other sports had to have been allocated to another category (or categories). The logical place is to the only sports category mentioned, live professional and college team sports, thereby arti6cially and erroneously inflating the resources reflected in that category. Indeed, the Horowitz Survey. results illustrate this.

Finally, the addition of systems that carry only. PBS and/or Canadian broadcast signhls to i the study reduce coverage bias.

" Bortz Report, Appendix B. In sum, given the choice of two flawed surveys, the logical choice is the Horowitz version.

VI. Surveys of Cable Programming Executives are Poor Ways to Measure Relative Market Value in this Matter

The Horowitz Survey improves upon the Bortz Survey, but remains flawed. The natural

question at this point is what kind of study would come closer to satisfying the statutory requirements in this matter. Contrary to what has been accepted in the past, the answer would not be a survey of cable system executives, There are at least two better approaches.

A. Ob'ective Market Results are I're erable to Sub'ective Jud ments Es eciall F%en the Jud mentsAreF/avvedasDescribedAbove.

Economists generally view market data as the most valid source of information on

consumer preferences and behavior. Market data reflect the objective results of actual consumer preferences in the relevant marketplace, Consumers have spoken with their wallets and these

choices reflect their preferences and how various aspects ofproduct offerings. Indeed, such market transactions forzn the basis for many economic studies of consumer preferences and behavj.or.

In contrast, surveys of cable executives reflect their subjective judgments ofhow

consumer preferences operate. Furthermore, as I have previously stated, those judgments are

complex and biased. Actual marketplace results are objective and unbiased. As such, they

would provide better insight into marketplace return or value to the cable system ten out of ten

times as long as appropriate data were available.

For comprehensive treatments of the empirical approach to revealed preference analysis see Train, K.E. (2009), Discrete Choice Methods with Simulation, 2'd., New York: Cambridge University Press and Ben-Akiva, M. and S.R. Lerman (1985), Discrete Choice Analysis: Theoryand Application to Travel Demand, Cambridge MA: MIT Press.

39 Conceivably, there could be instances where marketplace values need to be assessed and marketplace data are not available. New product introductions provide the most common example. In those cases, survey data can be quite useful. However, the most useful survey data are not collected Rom managers (cable operators in this case), they are collected Rom their customers, the subscribers.

B. The Relevant Orpinions for Proiectine Marketolace Results are Not T'hose ofCable Executives: Thev are Those ofCable Customers.

Marketplace value is the result of subscriber (i,e., customer) behavior. The

Bortz/Horowitz Survey structure explicitly recognizes this with its focus on the role of different program categories in subscriber acquisition and retention. The best source ofinformation

(other than marketplace results) is the cable system's customers& In prior testimony,:Professor.

Alan Rubin writes,

To assess the value ofprogr~mmi~g on distant siyxals to attract and retain subscribers, we need to examine what subscribers actually prefer to watch, not someone else's perceptions of subscribers'references. Perceived value lies within people who, themselves, act or behave in a manner that provides the value; here, the perceived value ofdifferent program categories in

attracting subscribers lies within the subscribers themselves. It is . the individual subscriber who is or is not attracted to certain types ofproNmns and is or is not retained by a cable system. It is the individual subscriber who can tell us more directly, and: de6nitively, what type ofprogram js nlorei likelylto make that subscriber take and continue to pay for. cable service. While Professor Rubin's writing is very insightful, there really is nothing new in it. In factl the basic idea behind it has been the cornerstone ofthe Marketing Research discipline since itsy inception. Ifyou want to know if customers will buy a product,i ask them.: Ifyou want to know

Testimony ofAlan M. Rnbin, Ph.D. to the Copyright Royalty Judges for the 2004-2005 Cable Royalty, Distribution Proceeding, June 1, 2009, Corrected September 28, 2009 (Rubin 2009 Testimony), pp. 7-8.

40 why customers are not buying a product, ask them. Ifyou want to know what customers (i.e., the market) value, ask them. Ifmanagers really understood what their customers value, every product would be a success. In fact, we know over half ofnew industrial products fail.

The bottom line here is that management is not the best judge ofmarketplace value and the customer preferences that drive it; customers are the best judges ofwhat customers want, value, and will do.

'ollins, Mike (2015), "Reducing the Failure Rate OfNew Products," Forbes, April 30; accessed from htt://aww.t'nrhes.corn/cmites/mikccnllins'15'04/30/reducin -the-failure-rate-nf-new- rnducti9/561fhlh061a4, January 22, 2015.

41 DECLARATION OF JOEL STECKEL. PH.D.

I declare under penalty ofperjury that the foregoing testimony is true and correct, and of my personal knowledge.

Executed on December ~~2016

Joel Steckel, Ph.D. APPENDIX A

JOEL HOWARD STECKEL

New York University 812 Tisch Hall New York, NY 10012-1126 Tel: (212) 998-0521 EMail: JSTECKEL STERN.NYU.EDU

EDUCATION

UNIVERSITY OF PENNSYLVANIA, THE WHARTON SCHOOL

Doctor of Philosophy Degree (Marketing/Statistics) awarded, May 1982. Dissertation Title: "A Game Theoretic and Experimental Approach to the Group Choice Phenomenon in Organizational Buying Behavior;" Professor Yoram Wind, advisor.

Master of Arts Degree (Statistics) awarded May 1980.

Master of Business Administration Degree (Management Science) awarded with Distinction, May 1979.

Elected to Beta Gamma Sigma, May 1979.

COLUMBIA UNIVERSITY

Bachelor ofArts (Mathematics) awarded Summa Cum Laude, May 1977.

Elected to Phi Beta Kappa, May 1977.

ACADEMIC POSITIONS

Vice Dean for Doctoral Education, Stern School ofBusiness, New York University, August 2012-Present.

Director PhD Programs, Stern School ofBusiness, New York University, May 2007-July 2012

Marketing Department Chairperson, Stern School of Business, New York University, July 1998- June 2004.

Professor and Associate Professor, Stern School ofBusiness, New York University, January 1989 - present. Taught courses in Business Strategy, Marketing Management, Marketing Research, Corporate Reputation and Branding, Models ofPricing and Promotion, Field Studies in the New

43 Economy, Marketing Engineering, and Analytic Marketing for Management Consulting. Also taught Doctoral Seminars in Mathematical Models.in Marketing and Research Methods.

Visiting Professor, Wharton School, University of Pennsylvania, January 1995 — December 1995. Taught Core Marketing course.

Visiting Professor, Escola de P6s-Graduaqao em Ci8ncias Econoruicas e Empresariais, Universidade Catolica Portuguesa, May - June 1992, May.- June 1993. Taught Industrial Marketing and Marketing Strategy.

Associate Professor and Assistant Professor, Graduate School ofBusiness, Columbia University, July 1981 - December 1988. Taught MBA-level courses in Industrial Marketing, Marketing Planning, and Marketing Research. Taught three Ph.D.-le'vel Marketing Seminars and Applied~ Multivariate Statistics.

Visiting Associate Professor, School ofOrganization and Management, Yale University, September - December 1988. Taught graduate course in Marketing Strategy.

Visiting Assistant and Associate Professor, Graduate School ofManagement, University ofi California at Los Angeles, July 1984 - June 1985, January'- March 1987. Taught Advanced Marketing Management, Marketing Research, and'Strategic Marketing Planning.

Assistant Instructor, Department of Statistics, University ofPennsylvania, July 1979 - June 1980 Assisted in undergraduate and MBA-level courses tin Statiktick. Taught undergraduate oourise in Calculus.

Teaching Assistant, Department ofMathematics, Columbia University, September 1976 - May 1977. Assisted in courses in Number Theory and Bi6breritiall Equations.:

PROFESSIONAL INTERESTS

Marketing Strategy and Marketing Research. In particular, marketing research methodology, marketing and branding strategies, electronic commerce, approaches for one-to-one marketing,', and managerial decision making.

PUBLICATIONS

Books

Marketine Research (with D. Lehmann and S. Guyta),'Boston: Addison-Wesley Longman, 1998.

Analvsis for Stratemc Marketine (with V. Rao), Boston: Addison-Wesley Longman, 1998..

The New Science ofMarketine: State of the Art Tools'for'Anticipatiuu and Trackine the Market Forces that will Shape Your Companv's Future (with V. Rao), Chicago: Irwin Professional Publishers, 1995.

44 Journal Articles

"Behavioral Reasons for New Product Failure: Does Overconfidence Induce Over-forecasts?" (with D. Markovjonitch, A/ Michaut-Denizeau, D. Philip, and W. M. Tracy), Journal of Product Innovation Mana ement Vol. 32, No. 5, September 2015.

"Modeling Credit Card Share of Wallet: Solving the Incomplete Information Problem," (with Y. Chen), Journal of Marketin Research Vol. 49, No. 5, October 2012.

"The Role of Consmner Surveys in Trademark Infringement: Evidence From the Federal Courts," (with R. Bird), Universi of Penns lvania Journal of Business Law Vol. 14, Issue 4, Summer 2012, 1013-1054.

"Do Initial Stock Price Reactions Provide a Good Measurement Stick for Marketing Strategies? The Case of Major New Product Introductions in the US" (with D. Markovich), Euro ean Journal ~of Marketin Vok 46, lss. 3, 2012, 406-421.

"When Do Purchase Intentions Predict Sales?" (with V. Morwitz and A. Gupta), International Journal ofForecastin, VoL 23, November 2007, 347-64.

"Dilution through the Looking Glass: A Marketing View ofthe Trademark Dilution Revision Act of 2005," (with R. Klein and S. Schussheim), The Trademark Re orter VoL 96, No. 3, May-June 2006.

"Choice in Interactive Environments," (with R. Winer, R.Bucklin, B. Dellaert, X. Dreze, G. Jfgeist, S. Jap. J.D.C. Little, T. Meyvis, A. Montgomery, and A. Rangasvvamy), ~Marketin Letters VoL 16, No.3/4, 2006.

"Using Capital Markets as Market Intelligence: Evidence &om the Pharmaceutical Industry," (with D. Markovich and B. Yeung), Mana ement Science October 2005.

"Marketing Science — Growth and Evolution," (with J. Hauser, G. Allenby, F.H. Murphy, J.S. Raju, and R. Staelin), Marketin Science Vol. 24, No. 1, Winter 2005.

"Supply Chain Decision Making: Will Shorter Cycle Times and Shared Point of Sale Information Necessarily Help?," (with S. Gupta and A. Banerji), Mana ement Science, Vol. 50, No. 4, April 2004.

"Choice and the Internet: From Clickstream to Research Stream," (with R. Bucklin, J. Lattin, A. Ansari, S. Gnpta, D. Bell, B. Conpey, J.D.C. Little, C. Mela, and A. Montgomery), M~arkettn Letters, Vol. 13, No. 3, Summer 2002.

"A Multiple Ideal Point Model: Capturing Multiple Preference Effects from within an Ideal Point Framework," (with J. Lee and K. Sudhir), Journal of Marketin Research, Vol. 39, No, 1, February 2002.

"2001: A Marketing Odyssey," (with E. Brody), Vol. 20, No. 4, Marketin Science Fall 2001.

"Consumer Strategies for Purchasing Assortments within a Single Product Class," (with Jack K.H. Lee), Journal ofRetailin, Vol. 75, No. 3, Pall 1999.

45

"The Max-Miu-Min Principle ofProduct Differentiation," (with A. Ansari and N. Bconomides), Journal ofRemonal Science, May 1998. .'f

"Dynamic Influences on Individual Choice Behavior,"'(with R. Meyer, T. Brdem, F. Feinberg, I. Gilboa, W. Hutchinson, A. Krishna, S. Lippman, C. Mela, A. Pazgal, and D. Prelic), Marketine Letters, Vol. 8, No. 3, Juiy 1997.

"Addendum to 'Cross Validating Regression Models in Marketing Research'," (with W. Vanhonacker), Marketinu Science, Vol. 15, No. 1, 1996.

"Selecting, Evaluating, and Updating Prospects in Direct Mail Marketing," (with V.:Rao), Journal Direct Marketine, Vol. 9, No. 2, Spring 1995.

"A Cross-Cultural Analysis ofPrice Responses to Envixoxnnental:Changes," (with V. Rao),i Marketinu Letters, Vol. 6, No. 1, January 1995.

"Cross Validating Regression Models in Marketing Research,".(with W. Vanhonacker), Marketinu Science, Vol. 12, No. 4, Fall 1993.

"Preference Aggregation and Repeat Buying in Households," (with S, Gupta), Marketinu Letters, Vol. 4, No. 4, October 1993.

"Roles in the NBA: There's Still Always Room for a Big Maxi, Btxt His Role Has Changed". (with A. Ghosh), Interfaces, Vol. 23, No. 4, July-August'1993.

"Introduction to 'Contributions ofPanel and Point of Sale Data to Retailing Theory and Practice'," Journal ofRetailinu, Vol. 68, No.3, Fall 1992.

"Explanations for Successful and Unsuccessful MaIrke6nglDecisions:: The Decision Maker's Perspective" (with M.T. Curren and V.S. Folkes), Journal ofMarketinu, Vol, 56, No. 2, April 1992.

"Locally Rational Decision Making: The Distracting Effect ofInformation on Managerial . Performance" (with R. Glazer and R. Winer), Management Science, VoL 38, No. 2, February 1992. "Prospects and Problems in Modelling Group Decisions" (with K.P. Corfman, D.J. ~, S. Gupta, and J. Shanteau), Marketinu Letters, Vol. 2, No. 3, July 1991.

"A Stochastic Multidimensional Scaling Methodology for'the Empirical Determination of Convex Indifference Curves in Consumer Preference/Choice Analysis" (with W.S. DeSarbo and K. Jedidi), Psvchometrika, Vol. 56, No. 2, June 1991.,

"A Polarization Model for Describing Group Preferences": (with V. Rao), Journal of Consumer Research, Vol. 18, No. 1, June 1991.

" On the Creation ofAcceptable Conjoint Analysis Experimental Designs," (with W.S. Degarbp and V. Mahajan), Decision Sciences, Vol. 22, No. 2, Spririg 1991.

"Longitudinal Patterns of Group Decisions: An B@lolatoIry Analysis" (with K.P. Cor&nani and D.R. Lehmann), Multivariate Behavioral Research,'ol. 25, No. 3, July 1990. "Investing in the Stock Market: Statistical Pooling ofIndividual Preference Judgments," (with N. Capon), Annals of0 erations Research, Vol, 23, 1990.

"Judgmental Forecasts ofKey Marketing Variables: Rational vs. Adaptive Expectations" (with R. Glazer and R. Winer), International Journal ofForecastin, Vol. 6, No. 3, July 1990.

"Committee Decision Making in Organizations: An Experimental Test of the Core," Decision Sciences, Vol. 21, No. 1, Winter 1990.

"Towards a New Way to Measure Power: Applying Conjoint Analysis to Group Purchase Decisions" (with J. O'Shaughnessy), Marketin Letters, Vol. 1, No. 1, December 1989.

"The Formation and Use ofKey Marketing Variable Expectations and their Impact on Firm Performance: Some Experimental Evidence" (with R. Glazer and R. Winer), Marketin Science, Vol. 8, No. 1, Winter 1989.

"A Heterogeneous Conditional Logit Model of Choice" (with W. Vanhonacker), Journal of Business and Economic Statistics, Vol. 6, No. 3, July 1988.

"Estimating Probabilistic Choice Models from Sparse Data: A Method and an Application to Groups" (with D.R. Lehmann and K. Corfman), Ps cholo ical Bulletin, Vol. 95, No. 1, January 1988.

"A Friction Model for Describing and Forecasting Price Changes" (with W.S. DeSarbo, V.R. Rao, Y.J. Wind and R. Colombo), Marketin Science, Vol. 6, No. 4, Fall 1987.

"Group Process and Decision Performance in a Simulated Marketing Environment" (with R. Glazer and R. Winer), Journal of Business Research, Vol. 15, No. 6, December 1987.

"Effective Advertising in Industrial Supplier Directories" (with D.R. Lehmann), Industrial Marketin Mana ement, Vol. 15, No. 2, April 1985.

"Dynamic Decision Making in Marketing Channels", with S. Gupta, and A. Banerji), in Ex erimental Business Research, A. Rapoport and R. Zwick (eds.), Boston, MA: Kluwer Academic Publishers, 2002.

Refereed Proeeedin s

"PIONEER: Decision Support for Industrial Product Planning" in Efficienc and Effectiveness i~nMarketm, Proceedings of the American Marketing Association Edncator's Conference, Vol. 54, 1988, G.L. Frazier and C.A. Ingene, eds., Chicago.

47 "Mathematical Approaches to the Study of Power: A Critical Review" in Advances in Consumer Research, Vol. XII, 1985,, E. Hirschrnan and M. Holbrook, eds., Provo, UT.

"On Obtaining Measures from Rants" in A&i Assessment offMark&:tine Th~ou ~ht and Practice Proceedings of the American Marketing As. ociation Educator's Conference, Vol. 48, B.J. Walker, ed., 1982,, Chicago.

Other

"Forecasting Online Shopping," Stern Business', Fall/Winter 2000, pp. 22-27.

"Method to Their Madness," The Industry Standard August 7„2000.

Book review of T]he A~lication of R~e~..ession Ana~lsis by D.R. Wittink Journal of Mar~ketin Research, Vol. 26, No. 4, November 1989.

Co-author (with many others) of The Statistics Problem Solver, Research and Education Association, New York, 1978.

CONFERENCE PRESENTATIONS

"Measuring Trademark Dilution", Conference on Empirical Analysis of Intellectual Property, NYU Law School, October 2014.

"Using Surveys in Intellectual Property Cases:; What's the Damage," AIPLA Spring Meeting, May 2013, Seattle W,A.

"Trademark D&ilution: An Elusive Concept in the Law," Conference on Brands and Branding in Law, Accounting, and Marketing Kanan Flagler School, University ofNorth Caroline, April 2012

"The Role of Consumer Surveys in Trademark Infringement Gases: Evidence &om the Federal Courts," (with R. Bird), A5/hk Summer Educator's Cot]tferj."ncaa, August 2010, Boston.

"Global Market Share Dynamics: Winners and Lo&rs lin a Tumultuous World," (with P. Golder and S. Chang), INFORMS Marketing Science Conference., June 2010, Cologne, Germany.

"Use and Abuse of Consumer Perception Research in Antitrust and Advertising Cases," ABA Antitrust Section Spr].ng Meeting, March 2009, Washington, DC.

"New Product Development: The Stock Market as Crystal Ball," (with D. Markovich), INFORMS Marketing Sci&mce Conference, Atlanta, GA., June 2005.

"Modeling Credit Card Usage. Behav].or: Where is my VISA and Should I Use It?," (with Y. Chen), INFORMS Marketing Science Conference, College Park, Md., June 2003.

"Using Capita]l. Markets as Market Intelligence: Evidence &om the Pharmaceutical Industry," (with D. Markovich and B. Yeung), IM..ORMS Marketing Science Conference, College.'Park, Md., June 2003.

48 "Using Capital Markets as Market Intelligence: Evidence &om the P~eutical Industry," (with D. Markovich and B. Yeung), Share Price Accuracy and Transition Economies Conference, U. of Mich. Law School, Ann Arbor, Mi., May 2003.

"Modeling Internet Site Visit Behavior," (with E. Bradlow and O. Sak), Joint Statistical Meetings, Indianapolis, August 2000.

"Consumer Strategies for Purchasing Assortments within a Single Product Class," (with Jack K.H. Lee), INFORMS Fall Conference, Philadelphia, November 1999.

"When Do Purchase Intentions Predict Sales?" (with V. Morwitz and A. Gupta), AMA Advanced Research Techniques Forum, Santa Fe, NM, June 1999.

"Modeling New Product Preannouncements as a Signaling Game," (with H. Jung), University of Mainz Conference on Competition in Marketing, Germany, June 1999,

"A Multiple Idea Point Model: Capturing Multiple Preference Effects from within an Ideal Point Framework," (with J. Lee), Joint Statistical Meetings, Dallas, TX, Aug. 1998.

"Modeling New Product Preannouncements as a Signaling Game," (with H. J'ung), INFORMS Marketing Science Conference, Fontainbleau, France, July 1998.

"Dynamic Decision-Making in Marketing Channels: Traditional Systems, Quick Response, and POS Information," (with S. Gupta and A. Banerji), NYU Conference on Managerial Cognition, May 1998.

"When Do Purchase Intentions Predict Sales?" (with V. Morwitz and A. Gupta), INFORMS International Meetings, Barcelona, July 1997.

"Mental Models in Competitive Decision Making: A Blessing and A Curse," Conference on Competitive Decision Making, Charleston, SC, June 1997.

"When Do Purchase Intentions Predict Sales?" (with V. Morwitz and A. Gupt'a), INFORMS Marketing Science Conference, Berkeley, March 1997. " "Model Adequacy versus Model Comparison: Is the 'Best'odel Any 'Good', (with A. Ansari and P. Manchanda), INFORMS Marketing Science Conference, Berkeley, March 1997.

"Dynamic Decision-Making in Marketing Channels: Traditional Systems, Quick Response, and POS Information," (with S. Gupta and A. Banerji), First Conference in Retailing and Service Sciences, Banff, 1994.

"Dynamic Decision-Making in Marketing Channels: Traditional Systems, Quick Response, and POS Information," (with S. Gupta and A. Banerji), Behavioral Decision Research in Management Conference, Boston, 1994.

"Modeling Consideration Set Formation: The Role ofUncertainty," (with B. Buchanan and S. Sen), TIMS Marketing Science Conference,'uscon, 1994.

49

"A Cross-Cultural Analysis ofPrice Conjectures to Environmental Changes," (with V. Rao),

TIMS Marketing Science Conference, St. Louis, 1993.

"Decision-Making in a Dynamic Distribution Channel:Environment," (with S. Gupta and A. Banerji), TIMS Marketing Science Conference, St) Louis, i1993. 'Cross

Validating Regression Models in Marketing Research,t'with W. Vanhonacker), TIMS 'arketing Science Conference, London, 1992. "The Influence of Stock Price on Marketing Strategy," (with D. Gautschi and D. Sabavala),'IMS 'arketing Science Conference, Wilmington, DE, 1991.

"A Polarization Model for Describing Group Preferences" (with V. Rao), ORSA/TIMS National Fall Meetings, Philadelphia, 1990.

"A Polarization Model for Describing Group Preference," (with V. Rao), Behavioral Decision Research in Management Conference, Philadelphia, 1990.,

"Conflict Resolution and Repeat Buying" (with S. Gupta), TIMS Marketing Science Conference, Champai~, Ill., 1990.

"Variety Seeking at the Group Level" (with S. Gupta)/ Association for Consumer Research Fall Meetings, New Orleans, 1989.

"On Using Attraction Models to Allocate Resources in a Competitive Environment," TIME Marketing Science Conference, Durham, NC, 1989.

"Multidimensional Scaling with Convex Preferences" (with W.S.. DeSarbo), ORSA/TIMS National Fall Meetings, St. Louis, 1987.

"A Social Comparison Model for Describing Group Preference Evaluations" (with V. Rao), TIMS Marketing Science Conference, Jouy-en-Josas, France,:1987.

"The Day the Earth Stood Still," Association for Consumer Research Fall Meetings, Toronto, i 1986. 'An

"A Friction Model For Describing and Forecasting Price Movements" (with W. DeSarbo, V. Rao, Y. Wind, and R. Colombo), ORSA/TIMS National Fall Meetmgs, Miami Beach, 1986.

Eigenvalue Method for Measuring Consumer Preferences" (with E. Greenleaf and R. I Stinerock), TIMS Marketing Science Conference, Dallas, 1986.

"Creating Conjoint Analysis Experimental Designs wiitho4t Infeasible Stimuli" (with%'. DeSarbo and V. Mahajan), TIMS Marketing Science Conference, Dallas, 1986.

"The Mediating Role ofInformation in Marketing Managers'ecisions" (with R. Glazer and R. Winer), TIMS Marketing Science Conference, Dallas, 1986.

"Incorporating Interdependencies ofUtility Functions into Models of Bargaining" (with S. Gupta), ORSA/TIMS National Fall Meetings, Atlanta, 1985..

50 "The Formation ofKey Marketing Variable Expectations" (with R. Glazer and R. Winer), ORSA/TIMS National Fall Meetings, Atlanta, 1985.

"Does the Nash Equilibrium Really Describe Competitive Behavior?: The Case of Cigarette Advertising," TIMS Marketing Science Conference, Nashville, 1985.

"A Heterogeneous Conditional Logit Model ofChoice" (with W. Vanhonacker), ORSA/TIMS National Fall Meetings, Dallas, 1984.

"Using a 'Robust'esponse Function to Allocate Resources in a Competitive Environment," TIMS Marketing Science Conference, Chicago, 1984.

"Longitudinal Models ofGroup Choice Behavior," (with D. Lehmann and K. Corfman), ORSA/TIMS National Fall Meetings, Orlando, 1983.

"Considerations of Optimal Design ofNew Task Industrial Products," ORSA/TIMS National Fall Meetings, San Diego, 1982.

"Game Theoretic Choice Models in Organizational Buying Behavior," TIMS Special Interest Conference in Marketing Measurement and Analysis, Philadelphia, 1982.

OTHER RESEARCH IN PROGRESS

Marketing Research in the Courtroom vs. the Boardroom: What are the Differences and Do They Matter? (with R. Bird)

The Impact ofTrademark Litigation Outcomes on Brand Equity and Marketing Decision Making (with R. Bird)

Loss Aversion — Are Professional Tennis Players too Careful with the Second Serve? (with L. Nelson and S. Yang)

Modeling the Tradeoffs between Marketing Research and Flexible Manufacturing.

Modeling the Strategic Use ofList Rentals (with D. Schmittlein)

INVITED SEMINARS

Columbia University Spring 1991, Summer 1994 Cornell University Fall 1983, Spring 1989 Georgetown University Fall 2006 Pennsylvania State University Fall 1996, Fall 2006 Rutgers University Spring 1994 Temple University Fall 1995 University of California, Berkeley Spring 1990 University of California, Los Angeles Spring 1985, Spring 1996

51 University of California, San Diego Fall 2003 University ofFlorida Spring 1992 University ofMainz, Germany Summer 1998 University ofMichigan Spring 1993 University ofPennsylvania Spring 1992, Spring 1995, Spring 1998 University of Southern California Spring 1987 Washington University, St. Louis Spring 2003

EDITORIAL SERVICE

Editorshins

Co-Editor, Marketing Letters, July 2010 - Present

Guest editor, special section of Marketine Science.on the history of marketing science theory and! practice, 2001.

Consulting editor in marketing, Addison-Wesley!Longm!sn Academic Publishers, Boston, MA,! 1993-1999.

Guest editor, special issue ofJournal ofRetailing on the use ofpanel and point of sale data,! 1992. !

Other

Member ofEditorial Boards, Marketing Science, Review ofMarketinu Science, Journal of, Retailine.

Have served as ad-hoc referee for Journal ofMarketing, Journal ofMarketing Research,

Management Science, Journal of Consumer Research, Journal ofRetailine and Consumer Services, Manufacturine and Service Operations Manas.ement, Decision Sciences, Journal of Business and Economic Statistics, Journal ofEconometrics, Strategic Information Svstems,'oroorate Reoutation Review, and Journal ofBusiness Research.,

SERVICE

Dissertation Committees Chaired

Joseph Pancras (co-chair) (Marketing - New York University): Sergio Meza (co-chair) (Marketing — New York University): Dmitri Markovich (Marketing — New York University) Heonsoo Jung (Marketing - New York University) Jack Lee (Marketing - New York University) Asim Ansari (co-chair) (Marketing - New'ork University) Shahana Sen (co-chair) (Marketing - New, York University)

52 Dissertation Committees Served on

Tingting Fan (Marketing — New York University) Kei-Wei Huang (Information Systems — New York University) SherrifNassir Pufarketing — New York University) Jane Gu (Marketing — New York University) Orkun Sak (Marketing — University of Pennsylvania) Atanu Sinha (Marketing - New York University) Louis Choi (Marketing - Columbia University) Sunder Narayanan (Marketing — Columbia University) Carol Rhodes (Ed. Psych. — Columbia University) Rita Wheat (Marketing — Columbia University) Robert Stinerock (Marketing - Columbia University) Bruce Buchanan (Business Economics - Columbia University) Chen Young Chang (Marketing - University of Pennsylvania)

Other Disci line Related Service

Chairperson, Marketing Committee, INFORMS, January 2006 — June 2010.

Past President, INFORMS Society on Marketing Science, January 2004 — December 2005.

Founding President, INFORMS Society on Marketing Science, January 2003 — December 2003.

President, INFORMS College on Marketing, January 2002 — December 2002.

President Elect, INFORMS College on Marketing, January 2000- December 2001.

Secretary-Treasurer, INFORMS College on Marketing, January 1998-December 1999.

Association of Consumer Research, Annual Program Committee, 1999.

Co-Organizer of 1996 Conference on Consumer Choice and Decision Making, Arden House, Harriman, New York, June 1996.

Organized Marketing Sessions at Fall 1989 TIMS/ORSA Joint National Meetings, New York, October 1989.

Other Universi Related Service

Member, Research Resources Committee, Stern School ofBusiness, September 2009 — Present.

Chair, Statistical and Quantitative Reasoning Task Force, Stern School of Business, September 2005 — August 2007.

Member, Specialization Committee, Stern School ofBusiness, September 2004 — Present.

53 Member, PhD Oversight Committee, Stern School ofBusiness, January 2006 — May 2007.:

Member, Executive Committee, Digital Economy Initiative, Stern School of Business, Januaryi 2000 — August 2002.

Member, Board ofDirectors, Center for Information Intensive Organizations, Stern:School:of: Business, September 1998- December 1999.

Member ofMBA Committee, Stern School ofBusiness, New.York University, 1989-December 1998. Committee was responsible for supervising redesign ofMBA programs in 1991 and 1995, Chaimxan September 1997-August 1998.

Member of Stern MBA Cumculum Review Committee, September 1997-December 1998. Committee redesigned MBA Core.

Member of Stern School Committee on Improving Consulting Activities:, July 1998-December, 1998.

Member ofBuilding Committee, Stem School ofBusiness, New York University, 1990-1992.

Member ofResearch Committee, Stern School ofBusiness, New York University, 1990-1. i

Elected member of Columbia University Senate. Served on Budget Review and Alumni Relations Committees, 1986-1988.

AWARDS

Awarded the J. Parker Bursk Memorial Prize as the outstanding student participating in thei Department of Statistics, University of Pennsylvariia, 1979.

Dissertation was awarded Honorable Mention in the 1982 American Marketing Association Dissertation Competition.

Dissertation was named Winner ofthe 1983 Academy ofMarketing Science Dissertation Competition.

Invited speaker at the J. Parker Bursk Memorial Prize Luncheon, Department of Statistics, University ofPennsylvania, 1992.

Invited speaker at American Marketing Association Doctoral.Consortium, University of Southern California, 1999.

Cited for outstanding editorial support, Fordham University Pricing Center, Sept. 2002.

Named one ofthe inaugural winners ofthe Best Reviewer Award for the Journal ofRetailing, 2003.

Work recognized by West publishing as one of the outstanding 2012:law. review. articles on Intellectual Property. Work recognized with the Highly Commended Paper Award at the Literati Network Awards for Excellence 2013.

SELECTED CONSULTING AND OTHER PROFESSIONAL ACTIVITIES

AOL MovieFone, Inc., New York, NY. Performed general consulting on analyzing caller data for telephone movie information service; Consulted as expert in conjunction with damage assessment in legal proceedings.

Citicorp, New York, NY. Built choice model for bank services. Gave lectures on Marketing Strategy to CitiCards executives.

Directions for Decisions, Inc., New York, NY and Jersey City, NJ. Consulted on segmentation study of sports apparel market, designed and implemented "Construction Test", a concept design decision tool. Performed general consulting on marketing research practice on an ongoing basis.

eComplaints.corn, New York, NY. Member board of advisors.

Federal Trade Commission, Washington, D.C. Served as consultant on branding strategies in antitrust investigation.

J.C. Penney Co., New York, NY. Performed sales-advertising response analysis. Work was done on request for Management Decision Systems, Inc., Weston, MA.

The Open Center, New York, NY. Consulted on marketing strategy and direct marketing practices.

Pfizer Phannaceuticals, New York, NY. Conducted seminar on conjoint analysis.

Union Carbide Corporation, Danbury CT, Built econometric model to forecast prices .

Various Expert Witness Engagements in Intellectual Property Cases. Clients include AOL Moviefone, AT&T, Avon, Brother International, Dyson, Epson, Hershey's, BM, JP Morgan Chase, Gerber Products, Johnson & Johnson, K-Swiss, Mead Johnson, Microsoft, Monster Cable, McDonald's, Playtex, PNC Financial, Proctor & Gamble, Roche, Seagate, Sergio Garcia, Sharp, TiVo, Under Armour, Wal-Mart, Warnaco, and various plaintiffs in conquer class actions.

MEMBERSHIPS

American Marketing Association

American Statistical Association

Association for Consumer Research

The Institute for Operations Research and Management Science INFORMS)

55 APPENDIX B

JOEL STECKEL TESTIMONY IN THE LAST FOUR YEARS

DEPOSITIONS

~ Visteon Technologies, LLC. v. Garmin International, Inc., Civil Action No. 2:10-cv- 10578-PDB-MAR (United States District Court, Eastern District ofMichigan — Southern Division)

~ Margaret Korrow, on behalf ofherselfand others similarly situated, v. Aaron's Inc., also known as Aaron's Sales & Lease Ownership, Inc. and formerly known as Aaron Rents and John Does 1-25, Civil Action No. 10-cv-06317 (JAP) (LHG) (United States District Court, District ofNew Jersey)

~ Btkin & Company, Inc. v. SBD LLC, Dr. Arthur Agatston, SBD Trademark Limited Partnership, and SBD Holdings Group Corp., Civil Action No.: 1:11-cv-21321- Lenard/O'ullivan, United States District Court (Southern District ofFlorida)

~ Under Armour, Inc. v. Body Armor Nutrition LLC, Civil Action No. 1:12-cv-01283- JKB, United States District Court (District ofMaryland — Baltimore Division)

~ United States ofAmerica et. al. v. American Express Co., et. al., Case No. 10-CV-04496 (NGG) (RER), United States District Court (Eastern District ofNew York)

~ People ofthe State of California vs. Overstock.corn, Inc., Case No. RG10-546833. Superior Court of California, (County ofAlameda).

~ Denima6a, Inc. v. New Balance Athletic Shoe, Inc., Foot Locker, Inc., The Sports Authority, Inc., and Famous Horse, Inc., d/b/a V.I.M., Civil Action No. 12-cv-04112 (AJP), United States District Court (Southern District ofNew York).

Moroccanoil, Inc. v. Mare Anthony Cosmetics, Inc., Case No. CV 13-02747 DMG (ARGx), United States District Court (Central District of California, Western Division).

~ QS Wholesale, Inc. and Quiksilver, Inc.v. Rox Volleyball, Inc. and 1" Place Team Sales, Inc., Case No. SACV 13-00512 AG (JPRx), United States District Court (Central District of California, Southern Division).

~ Church & Dwight Co., Inc. v. SPD Swiss Precision Diagnostics GmbH, Civil Action No.: 14-CV-585 (AJN), United States District Court (Southern District ofNew York)

Twentieth Century Fox, et al. v. Empire Distribution, Inc. Case No: 2:15-cv-02158-PA- FFM (United States District Court for the Central District of California).

57 ~ United States ofAmerica, ex rel., Floyd Landis vs. Tailwind Sports Corporation et. al., Case No. No. 1:10-cv-00976 (CRC); United States District Court (District of Columbia)

~ Art Cohen, Individually and on Behalf ofAll Others Similarly Situated, v. Donald J. Trump, Case No. 13-CV-2519-GPC(WVG), United Statles District Court:(Southern District of California)

~ Kenneth Hobbs on behalf ofhimselfand all others similarly situated, v. Brother International and Does 1 thxough 10 inclusive, Case No. 2:15-cv-01866-PSG (MRWx), United States District Court, (Central District ofCalifornia)

~ Munchkin, Inc. v. Playtex Products, LLC. 2011 WL 2174383 (United States District Court, Central District ofCalifornia)

~ In Re Tobacco Cases II, JCCP No. 4042, Case, Na. 711400 (Superior Court ofthe State of California for the County of San Diego).

~ People ofthe State of California vs. Overstock.corn, Inc., Case No. RG10-546833. I

Superior Court of California, (County ofAlameda). I

Church Er, Dwight Co., Inc. v. SPD Swiss Precision Diagnostics GmbH, Civil Action No.: 14-CV-585 (AJN), United States District Coul (Souther& District ofNew York).

~ Dayna Craft (withdrawn), Deborah Larsen, Wendi Alper-Pressman, Individually and On. BehalfofAll Others Similarly Situated v. Philip Morxis Companies, Inc., a corporation, and Philip Morris Incorporated, a corporation,.Case No. 2202-00406-02, Division No..6 (Missouri Circuit Court, Twenty-Second Judicial Circuit City of St. Louis):

ARBITRA.TION

~ Bayer Healthcare, LLC v. Roche Diagnostics Corporation and F. Hofnnan- La Roche ~ Ltd., Case No. CPR G-08-378, CPR International Institute for Conflict Prevention Andi Resolution.

DAUBERT HEARING

~ Visteon Technologies, LLC. v. Garmin International,i Inc., Civil Action No. 2:10-cv- — 10578-PDB-MAR (United States District Court, Eastern District ofMichigan Southern . Division)

58 II I I I I

I I i ~ I I TESTIMONY OF JOHN MANSELL COPYRIGHT ROYALTY JUDGES 2010-2013 COPYRIGHT ROYALTY DISTRIBUTION PROCEEDING

THE MIGRATION OF LIVE TEAM SPORTS PROGRAMMING FROM BROADCAST TELEVISION TO CABLE-SATELLITE TV AND OTHER NEW MEDIA

PREPARED BY:

JOHN MANSELL ASSOCIATES, INC. 1093 LORAN COURT GREAT FALLS, YA 22066-1533 DECEMBER 22, 2016 MIGRATION OF SPORTS PROGRAMMING

TABLE OF CONTENTS

I. BACKGROUND AND EXPERIENCE . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ a ~ ~ ~ ~ 1

'. .' '. ..' II ~ PURPOSE OF TESTIMONY. ~ .. ~ ~ ~ ~ . ~ ~ ~ . ~ ~ ~ .'. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ a3

III. EXECUTIVE SUMMARY ...... l....l... l l ~ i...:..:....:.....:...... 4

A. LOCAL BROADCASTING STATIOINS VERSUS RSNS. ~ ~ . ~ ~ ~ ~ . ~ ~ ~ ~ ~ ~ ~ .5

B. GROWTH OF NATIONAL CABLlE NElTWORKS.J.. ~ ~ ~ .. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I. ~ ~ ~ls61

C NEW MEDIA COMPETITION ' ' 7

D. NHL-NBA WORK STOPPAGES REDUCE LOCAL

BROADCASTS.

E. DIGITAL VIDEO RECORDERS REDUCE

LIVE SPORTS VIEWING. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 7

IV. LOCAL BROADCASTING VERSUS CABLE REGIONAL SPORTS

NETWORKS (RSNS)...... ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ s8

A. INTRODUCTION .. ~ ~ ~ ~... ~ ~ ~ ~ ~... ~ ~ ~ ~ ssJ ~ ass'~ ~ sas ~ ~ ~ ~ ~ ~ ~ sss ~ sa ~ ~ ~ ~ assa ~ asas ~ s8

B. DECLINE OF LOCAL TELECASTS AND INCREASE

.' ..''~ .' ..' IN RSN TELECASTS. ~ ~ ~ ~ . ~ ~ ~ ~ ~ ~ . ~ ~ ~ ~ ~ ..'. ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ a1 0

V. NATIONAL BROADCASTING NETWORKS VERSUS NATIONAL

CABLE SPORTS NETWORKS . ~ ~ ~ ~ ~ I, i i J ~ ~ i...'...... '...i ~ ~ .'".".....12

A. MLB NATIONAL BROADCAST TV GAMES VERSUS

NATIONAL CABLE GAMES ~ ~ ~ ~ ~ ~ ~ ~ 1 3

B. NBA NATIONAL BROADCAST TV GAMES VERSUS

NATIONAL CABLE GAMES ...... 15 MIGRATION OF SPORTS PROGRAMMING

C. NHL NATIONAL TV COVERAGE .

D. NFL LAUNCHES NFL NETWORK GAMES

E. COLLEGE FOOTBALL DISTRIBUTION. 18

F. COLLEGE BASKETBALL 19

Vl. OUT OF MARKET CABLE-SATELLITE VIDEO-ON-DEMAND ...... 20

A. NFL FOOTBALL 20

B. MLB 21

C. COLLEGE FOOTBALL AND BASKETBALL

D. NCAA BASKETBALL TOURNAMENT . ~ . 22

Vll. INTERNET SPORTS PROGRAMMING ON THE RISE

A. NCAA BASKETBALL TOURNAMENT 24

B. MLB . 26

C. NFL ~ 27

D. NBA. 28

E. NHL 29

F. COLLEGE SPORTS 30

Vill ~ MOBILE AND OTHER DIGITAL DEVICES

A. NFL . 31

B. MLB. . 32

C. NBA.

D. NHL . 33

IX. SUMMARY. NIIGRATION OF SPORTS PROGRAMMING

APPENDIX I-RSN SUBSCRIBERS...... i ~ ~ .~ ~ . ~ ~ i...~...... 38

APPENDIX II-LOCAL MLB TELECASTS .. s J ..40

APPENDIX III-LOCAL NBA TELECASTS ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ll ~ ~ I ~ ~ ~ ~ ~ ~ ~ ~ ~ sa ~ a 41

APPENDIX IV-LOCAL NHL TELECASTS . i i i ~ ' ~ ~ ~ ~ ~ ~ . 42 APPENDIX V-NATIONAL CABLE SPORTS NETWORKS ...... '... 43','PPENDIX

Vl-NATIONAL CABLE NETWORKS WI ItH SPORTS ...... '... Vll-COLLEGE FOOTBALL TELECASTSl~.... ~ . i . 4645'PPENDIX i

'. .' .'. ' APPENDIX Vill-CURRICULUM VITAE . ~ . ~ . ~ ~ . ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ I ~ ~ a 49 MIGRATION OF SPORTS PROGRAMIIIIING

I ~ BACKGROUND AND EXPERIENCE

I am John Mansell, Jr., President/CEO of John Mansell Associates, Inc., a company established over 20 years ago. Until April 2007, I worked exclusively for Kagan Research and related successor companies for over 32 years. Kagan is one of the leading media research organizations in the U.S. Its vast array of publications covers many arenas of media and entertainment, including broadcast and cable television. I began my career as a newsletter editor in 1975 and was promoted to senior analyst in 1986. When I left the company, I was

Kagan's lead analyst for cable TV overbuilds and competition, sports media rights and franchise valuations, communications law, and a seasoned appraiser of sports business, digital media, wireless, satellite and communications properties. I was responsible for writing, editing and contributing to several

Kagan books, special reports and newsletters, including Kagan's Wireless

Broadband, Cable TV Law Reporter and Media Sports Business. I served as moderator at Kagan events such as the Kagan Digital Media Summit and have been invited to speak at industry association conferences, including the Wireless

Communications Association, National Cable TV 8 Telecom Association, and the

Western Cable Show.

From 1984 to 2007, I wrote and edited the monthly Media Sports Business.

This newsletter regularly reported on the evolving relationship between sports teams and the electronic media, franchise valuations, transactions, media rights fees, advertising, viewership, subscription revenue, labor relations and MIGRATION OF SPORTS PROGRAMMING stadium/arena financings. In addition to coverage of these issues in newsletters,,

I headed a Kagan team and contributed to a series of special reports, including

The Business ofBaseball, The Business of Football, The Business of Basketball,

The Business of Hockey, and the U.S. TV Sports Databaok.

Over the years, I have done consulting and/or valuations for numerous entertainment and telecommunications organizatioi!Is, lincluding sports leagues, teams, and financial entities involved in transactions with teams and leagues. I have more than 25 years of experience in analyzing the sports business and

appraising sports, media and communications properties. During that period, I

have appraised over $10 billion worth of media properties on contract assignment. In addition, many more billions of dollars'orth of such properties'ave

been valued through regular informal analyses in Kagan Research's Media

Sports Business, Cable TV investor: Deals & IFirIanceJ Motion Picture Investor,,

Broadcast Investor: Deals & Finance, Wireless Broadband and Wireless Telecom

investor newsletters. I have based my analysis in part on personal knowledge of i the media and communications industries acquired over more than three decades.

I have a B.A. in economics from the University 6f Michigan, 1974, received my J.D. degree in 1978 from Thomas M. Cooley Law School, and I am a member of the Michigan Bar, District of Columbia Bar and Federal Communications Bar

Association. MIGRATION OF SPORTS PROGRAMMING

II ~ PURPOSE OF TESTIMONY

The purpose of my testimony is to analyze the changes in the carriage of

"live professional and college team sports" programming (as such term is used in this proceeding) in light of distribution and technology options that evolved through 2013 to compete for the attention of the consumer of live team sports I programming.

In addition to the research I conducted during my 33 years with Kagan

Research, for this assignment I have relied upon the broadcasttmedia calendars on the websites of each MLB, NBA and NHL team prior to the start of regular- season games as well as other websites, including those of the sports leagues and programmers.

In the broadest sense, sports programming is defined as the dissemination of audio and/or video of a game or contest involving individual skill or physical prowess, including baseball, football, basketball, soccer, hockey, golf, tennis, bowling, track & field, horseracing, swimming-diving, ice-skating, skiing, auto racing, boxing, wrestling, softball, lacrosse, volleyball, sailing, extreme sports, fishing and other activities.

For the purpose of the instant proceeding to allocate cable compulsory license royalties, however, I focus on live over-the-air television broadcasts of professional and college team sport games involving teams belonging to Major

League Baseball (MLB), the National Football League (NFL), the National

Basketball Association (NBA), the National Hockey League (NHL), and the MIGRATION OF SPORTS PROGRAIIIIIIIIING

National Collegiate Athletic Association (NCAA), the sports.programming that, falls within the Joint Sports Claimants program category ("JSC Sports"). The

JSC Sports programming is a narrower subset of the broader sports programming.

I have analyzed the changes in live professional and college team sparta games on television. Based on that analysis, I conclude that over the past 30 years, the number of live professional and college team sports games on local over-the-air TV stations has significantly declined. At the same time, the number of professional and college team sports games on cable networks and regional sports networks (RSNs) has dramatically incrtealsed. In effect, live professional and college team sports games in general, and JSC Sports programming in

particular, have shifted dramatically from local olvei-thle-air TV stati6ns to regional

sports networks and basic cable sports networks. furthermore, the trend has since 2005 and there is no reason to believe that this trend will rIot'ccelerated continue.

III ~ EXECUTIVE SUMMARY

By any measure, from 2005 through 2613, live regular season MLB, NBA, l

NHL and NCAA basketball and football games increasingly aired on cable TV i national networks and RSNs, and not on broadcast television networks and local

TV stations. In addition, since 2005, more MIJ.B, I NBA lanG NHL playoff games have migrated from national broadcast to national cable networks. MIGRATION OF SPORTS PROGRAMMING

A. Local broadcasting stations versus RSNs: Since 1990, RSNs have grown in number, in subscribership, and cover a larger percentage of live MLB,

NBA, and NHL games televised. At the end of 2005, 33 RSNs, including overlapping networks in 10 regions, served a combined total of 142.0 million cable subscribers. By 2010, 38 RSNs increased the number of subscribers served to 173.7 million subscribers, a 22A% increase, and by the end of 2014,

51 RSNs were serving 199.0 million subscribers up, another 14.6% (a 40% increase in subscribers served between 2005 and 2014).

As shown in the chart below, between 2010 and 2013, the number of live game telecasts on local broadcast television declined, while the number carried by cable television RSNs increased for each of the major sports leagues. Each year, more games were televised on RSNs than were televised on local TV stations, with the trend toward RSNs accelerating over time, except for those years in which there were labor disputes (NBA 2011-12 and NHL 2012-1 3).

LOCAL BROADCAST VS. REGIONAL SPORTS NETWORK NUMBER OF MAJOR LEAGUE GAMES—2010-to-2013

--2009-10--- --2010-2011-- --2011-2012-- --2012-2013-- Broad- Cable Broad- Cable Broad- Cable Broad- Cable League cast RSN cast RSN cast RSN cast RSN

MLB 498 3,902 449 3,997 406 4,082 338 4,115

NBA 219 1,887 180 1,968 77 1,654 86 2,113

NHL 80 1,564 77 1,640 48 1,643 7 979 John Mansell Associates, 2016 MIGRATION OF SPORTS PROGRAMIMIING

IV. LOCAL BROADCASTING VERSUS CABLE REGIONAL SPORTS NETWORKS (RShlS)

A. Introduction

RSNs are cable television networks whose primary, if not exclusive, content is sports programming. RSNs offer the live games of "home" teams for the region. Their increased presence in the distribution market over more than 20 years has had a significant impact on sports telecasts of the major sports leagues. RSNs, all of which were carried on the basic or expended-basic tier, generate revenue from subscription IFees and advertising. They cover sever@i regions with multiple and/or overlapping networks, such as Atlanta, Baltimore,

Chicago, Colorado, Florida, Los Angeles, New England, Ohio, New York, and

Washington, D.C. Over the past '15 years, two trends have developed,: (1) professional teams and colleges have been establishing their own RSNs to broadcast their games throughout the region of most interest by viewers; and (2)', there is an increasing amount of cross-ownership between multichannel video program distributors (MVPDs), sports franchises and RSNS.

RSNs typically fell into several ownership categories in 20'l3-14.

Approximately two dozen were owned by Fox and seven were owned by

Comcast. In about a dozen cases, one or more teams owned the majority of the network or the network owned the teams; in at least seven cases teams held a minority stake in the network; and the rest were owned by other distributors, including Time Warner Cable, DirecTV and ATBT. MIGRATION OF SPORTS PROGRAMMING

At the end of 2010, there were 38 RSNs which had a combined total of approximately 173.7 million subscribers, a 24% increase over 2005. Of the 38

RSNs, 19 were Fox-affiliated RSNs and carried 51 of 81 pro sports teams with approximately 83 million subscribers; Comcast's 12 RSNs totaled 29 million subscribers; and another seven RSNs totaled nearly 62 million subscribers. By

2014, there were 51 RSNs with approximately 199 million subscribers - in other words, 55% more RSNs and 42% more RSN subscribers than there were in

2005. This increase reflects the fact that cable subscribers were interested in watching their home teams'ames. See Appendix I.

By 2005, RSNs were generating an estimated $3.1 billion in revenue—

$2.5 billion in affiliate fees paid by cable TV and l3BS companies, plus nearly

$600 million in advertising revenue. By 2010, according to SNLKagan, RSN affiliate fees had risen to $4.2 billion, 68% higher than in 2005 and were rising at a 10A% compound annual growth rate.

The growth of RSNs has affected the number of locally televised broadcasts of MLB, NBA and NHL. From 1990 through 2013, the number of local telecasts of MLB, NBA and NHL regular-season games on over-the-air flagship

TV stations plunged, both in absolute terms and as a percentage of the total telecasts of over-the-air TV stations and RSNs. The reasons for this decline include the following:

'ee Kagan, Media Sports Business, Feb. 28, 2006.

MIGRATION OF SPORTS PROGRAMMING

Broadcast networks effectively restrict their. affiliates from carrying

non-network programming; RSNs are able to outbid local TV stations for the local rights to'LB,

NBA and NHL live team sports and NCAA basketball and

football programming because RSNsl halve. multiple reveriue

streams from cable/satellite/teloo TV company affiliate fees Plds ItV I

and website advertising;

RSNs often lock-up major league sports programming, including

digital (Internet) rights, by paying very high rights fees in exchange

for exclusive and long-term agreements with the home teams in a

region; and

Sports teams and college conferences are launching their own

RSNs and participating in ownership of new RSNs.

B. Decline of local telecasts and increase in RSN telecasts

The divergence between an increasingly smaller number of local broadcasts and a rapidly rising number of local RSN telecasts since 1990 is dramatic.

Maior Leaaue Baseball Local Telecasfs

In 1990, 59% (1,5?7) of MLB games wbre'oll'tel'ecasts. By 2005, however, those telecasts had dwindled to about 26% (1,066), and further downward to about 8% (338) by 2013. In the same period, the number of total

MLB games on RSNs increased from 41%(1,097) i'n 1'990 to 74% (3,067) in

10 MIGRATION OF SPORTS PROGRAMMING

2005, and to approximately 92'/o (4,115) in 2013. For that period, MLB local broadcast telecasts declined at a compound annual growth rate (CAGR) of 6.5% per year, while RSN carriage rose at a 5.9'/0 annual clip. See Appendix II.

NBA Basketball Local Telecasts

Forty-five percent (724) of all 1990-1991 regular season NBA games were local telecasts. That number declined to 26'io (558) in 2004-05 and to 3.9% (86) in the 2012-13 season. During that same period, regular season games carried by RSNs increased from 55'/0 (885) during the 'I 990-1991 to 74% (1,561) in the

2004-2005 season, and then to 96'/0 (2113) in the 2012-13 season. For that period, local broadcasts of NBA games declined at a CAGR of 9% while RSN telecasts increased at a CAGR of 4%. See Appendix III.

NHL Hocke Local Telecasts

NHL hockey has experienced the same trend as the NBA and MLB.

During the 1989-1990 hockey season, 28.7'/0 (251) of NHL games were local telecasts. That number plummeted to 11.8% (194) in the 2003-2004 season. A lockout in 2004-05 resulted in the cancellation of the entire NHL season. After the lockout, the decline in local telecasts continued such that by the 2012-2013 season, only 0.7% (7) of NHL games were local telecasts. During that same period, the share of total games carried by RSNs went from 71.3%, to 88.2% to

99.3%. For that period, local broadcasts declined at a CAGR of 14% while RSN telecasts increased at a CAGR of 2%. See Appendix IV. IVIIGRATION OF SPORTS PROGRAIIJIMING

V. NATIONAL BROADCASTING NETWORKS, VERSUS NATIONAL. CABLE SPORTS NETWORKS

Besides televised sports games on RSNs, cable subscribers have sports,

available on full-time national cable sports networks. i By the end of 2004, 18 full- time national cable sports networks, including 12 that had launched since 1999, had a combined total of 538.3 million subscribers. Between December 2004 and

December 2009, the number of national cable sports networks rose to 26, and by: the end of 2013 had risen to 30. These national cable sports services had a combined 1.37 billion subscribers, up 155% from December 2004.

Full-time cable sports networks like ESPN2, ESPNU,. College Sports ItV, I

The Golf Channel, The Speed Channel (motorsports), The Tennis Channel and multiple soccer channels all continued to compete for ~the attention of sports fans during this period. Also, the NBA launched its own, channel in January 2003.

The NFL followed suit by launching the NFL Network in November 2003.

From 2005 to 2013, ten new national sports services were launched including the MLB, NHL, Big-10, and Pac-12 hetwdrk5. In addition, since 2010,

Fox, NBC, and CBS each transformed other channels into 100% sports channels—Fox Sports 1 and Fox Sports 2, NBC Sports Network, and CBS

Sports Network. While some of the sports prbgIarhming'ori these new national cable sports networks was previously carried by local TV stations, other programming was being televised for the first time or had migrated from other national broadcast networks. ln either case, the abundance of new full-time;

12 MIGRATION OF SPORTS PROGRAMIWING national cable sports networks during this period spurred competition for viewers who might otherwise watch major league team sports programming on over-the- air TV networks.

Appendix V shows the national cable sports networks, their launch dates and the discussed subscriber levels.

In addition, many general interest national cable networks with sigriificant subscribership carried sports programming. These networks, which are identified in Appendix Vl, experienced an aggregate subscriber growth from 897.3 billion subscribers in 2004 to 1.17 billion subscribers in 2013.

A. IILB National Broadcast TV Games Versus National Cable Games

In September 2005, ESPN agreed to a new eight-year deal for a package of about 80-90 regular-season games on ESPN and ESPN2 from 2006 through

2013. From 2001 through 2006, Fox retained broadcast rights for Saturday afternoon games, plus the All-Star Game and all post-season rights. Separately,

ESPN agreed to a deal for digital multimedia rights—exclusive in-progress highlights and live cut-ins as well as rights for ESPN.corn, ESPN Deportes,

ESPN 360 (now ESPN3) and ESPN Mobile.

In October 2006, MLB contracted with Fox and Turner Broadcasting for new seven-year deals. The American League Championship Series ("ALCS") and the National League Championship Series ("NLCS"), for the first time, shifted from broadcast television to cable, with TBS carrying the NLCS in 2007,

2009, 2011 and 2013 and the ALCS in 2008, 2010, and 2012. TBS also carried

13

MIGRATION OF SPORTS PROGRAMMING

most post-season division series under the 2007-to-2013 Fox and Turner In addition, TBS obtained rights to 26 Sunday afternoon games'ontracts. from

2008-13. Fox obtained rights to the NLCS in 2068,'2010 and 2012 and the

ALCS 2007, 2009, 2011 and 2013. Fox also obtained rights to the All-Star Game,

26 Saturday games and the World Series from 2007-1 3.

Since launching in 2009, MLB Network has been carrying about 130 to 150 live games per season plus several post-season division series games and up to

150 pre-season games. In 2014, national broadcast games on Fox declined from 26 games to 12, as Fox agreed to carry 40 games on Fox Sports1.

The national telecast of MLB games aniiong broadcast networks and full- i time and general interest national cable networks is summarized below.

MLB NATIONAL TELECASTS

2005 2010 2011 2012 2013 Fox TV 18 26 '26' 26 26 ESPN 160 90 90 90 90 WTBS 72 26 '26','26 26 WGN 99 87 94 92 87 WOR 21 21 20 17 21 WPIX 50 23 25' 25 23

MLB Network 130 1I301 11 33 133 TOTAL 420 403 4'11 '409 406 % Network TV 4'/o 6% 6% 6% 6 % Cable 38% 61% 60% 61% 61% %Su perstation 58% 33% 34% 33'/o 32%

ESPN in 2004-05 was 160+; MLB Networki 2010 $2011 is an: estimate. MLB website says about 150 games John Mansell Associates, 2016

14 MIGRATION OF SPORTS PROGRAMMING

B. NBA National Broadcast TV Games Versus National Cable Games

In the 2004-05 season, cable networks TNT, ESPN and NBA TV televised over 280 games. That was 82% of the total national telecasts compared to about

10% on broadcast network TV. The number of games televised on broadcast network TV declined from 58 games on NBC in the 1998-99 season, when there was a lockout, to about 26 games on ABC in the 2003-04 season. From the

1998-99 season to the 2003-04 season, the percentage of games on broadcast network TV declined by two-thirds from 34% to 10% while the percentage of games on national cable networks doubled from 40% to 82%. Since then, there has been little change. ln the 2012-13 season, 85% of national games, including playoffs, were on national cable networks, 8% on broadcast network TV and 7% on WGN.

NBA NATIONAL TELECASTS

2004- 2012- Network 2005 2013 Chg. ABC 36 29 -19% TNT 98 96 -2% ESPNlESPN2 90 9S 6% NBA TV 96 103 7% WGN 25 23 -8% TOTAL 345 346 % Network TV 10% 8% % Cable 82% 85% %Superstation 7% 7%

John Mansell Associates, 2016 MIGRATION OF SPORTS PROGRAMMING

The decline in broadcast network telecasts of NBA games is directly tied to a couple of deals the NBA made with national cable networks. In 2002, the NBA made pacts with Disney (ABC/ESPN) and AGL Time Warner (TNT) for two six- year national TV contracts that covered four TV outlets..TNT gained carriage rights to 45 regular-season and 45 playoff games each year, plus the All-Star

Game and All-Star Saturday. For the first time, the All-Star Game and two conference finals migrated to cable. TNT also negotiated exclusivity for

Thursday nights and Spanish language rights.

ESPN acquired rights to 75 regular-season games (one Wednesday and a doubleheader on Friday), but no blackout rights for local telecasts. ESPN also gained rights to 15 to 24 playoff games, two 6onlferlen6e finals, the NBA draft, video-on-demand rights and Spanish language rights. In addition, ESPN2 gained rights to Tuesday Night Games in progress during a two-hour window.

The NBA established a deal with Time Warner to develop NBATV, a new cable channel to carry up to 96 games per season.

Similar to the 2002 pact, the 2007 eight-year extension through the 2015-,

16 season called for TNT to carry 62-regular-season games, up to 52 playoff games and the All-Star Game. In addition, TNT was granted rights to exclusively cover the conference semifinals and one conference finals round every year.

ABC retained rights to at least 15 regular-season andi 16 post-season games, i including the best-of-seven finals. ESPN and ESPN2 retained rights to 76- regular-season games and ESPN was granted up to 29 playoff games, including MIGRATION OF SPORTS PROGRAMMING the conference semifinals and one conference final. NBA TV continues to carry about 96 regular-season games and about a half-dozen playoff games.

In terms of new digital media, ESPN obtained rights to simulcast its games on ESPN360.corn and ESPN Mobile TV. TNT obtained rights to simulcast online its Thursday night doubleheader. ESPN and TNT have rights to stream games live and on-demand as well as other content. This enables numerous enhancements, such as different camera angles, statistic streams, highlights and studio show content for broadband.

C. NHL National TV Coverage

Starting with the 2005-06 season, Versus began carrying 50-60 games per season while NBC carried a half-dozen regular-season games and about 10 playoff games on weekends. From 2009-10 to 2012-13, NBC carried 10-15 regular-season games, NBCSN carried 50 to 100 games per season and NHL

Network carried 78 regular-season games. During that period only 7% to 10% of national telecasts were on national over-the-air TV. Since 2006, at least two

Stanley Cup Finals Games have been on NBCSN.

NHL NATIONAL TELECASTS

2009- 2010- 2011- 2012- Network 2010 2011 2012 2013 NBC 10 10 11 14 NBCSN 54 78 94 58 NHLNet 78 78 78 78 TOTAL 142 166 183 150 % NBC 7.0% 6.0% 6.0% 9.3% 17 MIGRATION OF SPORTS PROGRAMNIING

D. NFL Launches NFL Network Ghmlesl

The NFL has had significant changes to national carriage of its flagship programs. In 2006, the NFL switched its prized Monday Night Football (17 games) from ABC to ESPN and began offering eight games on NFL Network

(Thursday nights and a few Saturdays). In 2612, NFL'etwork's ThursdaIIt'ight

package was expanded to 15 games. (Note: NFI Network games were siinullcalst l in home-team cities).

E. College Football Distribution

For the years 2005-06, 2009-10, and 2013-14, we analyzed the carriage of college football games by the leading eleven conferences, independents, and

Bowl Games. See Appendix Vll. During that period, the national broadcast networks carried between 98 and 104 games per season. ln 2013, there were at least seventeen national cable networks carrying college football compared to

2005 when there were only seven national cable networks with college football.,

From 2005 to 2009, the number of college football games on national cable increased 42% from 220 games per season to 313 games and the number of games increased another 20% from 2009 to 2013 when there were 377 national telecasts.

In 2005-06, there were 28 Bowl games, of which 71% were on national cable, ESPN or ESPN2. In 2009-10, there wbre 33 Bowls, with 73% on national cable networks. By 2013-14, 89% of 35 games had migrated to national cable.

18 MIGRATION OF SPORTS PROGRAIIIMING

As shown in the table below, from 2009 to 2013, the college football broadcast syndication market declined by 53%. During that same period the

RSN syndication market grew 10?'/0 from?6 games to 15?. There has also been an explosion in Internet exclusive games, mainly due to carriage on ESPN3

(formerly ESPN 360), where there were 25 games in 2005-06 versus 112 in

2009-10, an increase of 348'/c. By 2013-14, there were 146 Internet exclusive games, a rise of another 30'k. Starting in 2011, the three-game BCS

Championship Series migrated from Fox to ESPN.

COLLEGE FOOTBALL TELECASTS: 2005, 2009, 2013

Change Change Network 2005 2009 2013 2005%9 2009-13 Natl. Best 104 98 101 3'/o Natl. Cabte 220 313 377 42% 20% Best synd. 129 134 63 4% -53% RSN synd 44 76 157 73/o 107/o Internet 25 112 146 348% 30% Total 522 733 844 40% Source: http//mattsamspcrts.corn/ John Manseil Associates, 2016

F. College Basketball

Likewise, with college basketball, while the number of games on national broadcast television stayed roughly the same, the telecasts available on national cable networks and RSNs increased dramatically from 2005 to 2009 and from

2009 to 2013 due mainly to the launch of new cable networks and reconfiguration of others, including ESPNU, FS1, FS2, NBCSN and CBSSN, Big 10, and Pac 10. MIGRATION OF SPORTS PROGRAMMING

Similar to NCAA football, far more games are syndicated to RSNs than to ITV stations. By 2013, ESPN3 typically carried at least two dozen games per week.

Vl. OUT-OF-MARKET CABLE-SATELLITE VIDEO-ON-DEMAND

Since the mid-1990's, the NFL, MLB, NiBA, NHL., NASCAR and ESPN

(college football and basketball) have all licensed eIIerIts,to be viewed on a pay- per-view, season-ticket and/or partial seasonlticket! balsi5. Cable and/or satellite

TV subscribers, even if subject to local blackout restrictions, can use these services to watch nearly every NFL, MLB, NBA, and NHL regular-season game

as well as hundreds of college football and basketball 'games. To the extent )hat sports seasons overlap, the leagues are competing more vigorously against each other and reducing the value of exclusive national and local programming

windows for sports programming. Viewership of sports programming on local TV stations is reduced and fragmented as audiences look to these pay services and other out-of-market cable and satellite alternatives.. A. NFL Football ,'FL

During the years 2003-07, DIRECTV struck a new deal with the NFL for

Sunday Ticket, a premium package that allows subscribers access to all

NFL Games subject to home-game blackouts. That agreement was replaced in

2004 by a contract covering 2006 through 2010. In 1999, the Sunday Ticket subscribership was about 900,000. By 2004, itherei were. approximately 1.7

million subscribers who paid for the right to view NFL games other than those

20 MIGRATION OF SPORTS PROGRAMMING carried by their local TV stations. Since 2004, the number of NFL Sunday Ticket subscribers is believed to be about 10% to 12% of DirecTV customers, or approximately 2.0 million to 2.4 million per season.

B. MLB

MLB's Extra Innings package of out-of-market games has been available to cable and satellite TV subscribers since 1996. Nearly every out-of-market game is available. Subscriber numbers are rarely made public, but Extra Innings attracted 150,000 cable customers and 285,000 satellite subscribers in 2006. As with other out-of-market packages, this offers subscribers games that would not be available to them on local TV stations, RSNs, national broadcast networks and national cable networks.

By 2008, dual feeds of home and away productions were available. DISH carried the package from 2004-06 but ceased carriage until 2011. In 2010, MLB

Advance Media (MLBAM) reportedly sold approximately 500,000 Extra Innings subscriptions for $ 120 apiece.

In 2009, MLB would not sell its 80-game per week out-of-market Extra innings package to distributors unless they agreed to carry the new MLB Network and MLB agreed to bundle Extra Innings with MLB.TV. Extra Innings subscribers paid $ 195 for both packages, or $ 130 for MLB.TV by itself.

C. College Football and Basketball

In 2004, ESPN Game Plan offered about 100 football games from the Big-

10, Big 12, ACC, SEC, Pac-10, Big East, and Mountain West conferences plus a

21 MIGRATION OF SPORTS PROGRAMMING selection from several independent schools. In 2005, ESPN Game Plan increased its package of out-of-market games to 150,,or an increase from about

10 games per weekend in 1999, to a dozen games in 2004-05. Cable TV and satellite companies offered weekly and season-ticket plans and allocate about a

half-dozen channels to the service. ESPN Full Court, launched in 200T, offered

hundreds of college basketball games for about $140. However, since 2009, with

national cable networks, RSNs, and ESPN360 offering more games, Game Plan

and Full Court experienced decline in subscribership and fewer cable companies

distributed both services. In 2015, ESPN replaced them.with ESPN Co//ege

Extra.

D. NCAA Basketball Tournament i

In 1999, DIRECTV began carrying all garnes otherwise not shown on local

TV, providing subscribers to its Mega March Madness premium package with 37

games from outside the region of the local CBS affiliate station's telecasts. All 64

would teams were carried, enabling subscribers to view 30 plus games that not'therwise

be available on the local CBS affiliate. In 2002, the package of 30, plus,

games recorded 55,000 to 60,000 buys, about the same as in 2001. Mega

March Madness was exclusive to DirecTV from 2002 to 2010. Over time,

enhancements were added, including a mosaic screen enabling fans to view four

games at one time.

Until 2011, the NCAA basketball tournament was exclusively on CBS.,

Under its new contract, which runs from 2011-24, a majority of the games have

22 MIGRATION OF SPORTS PROGRAMMING moved to national cable. In 2011, CBS broadcast network carried 26 games, national cable networks, TBS, TruTV, and TNT carried 16 games, 13 games, 12 games, respectively. The split was the same for 2012 and 2013. Starting in

2014, the Final Four shifted to Turner and in 2016 the Final Championship Game will migrate from CBS to Turner.

Vll. Internet Sports Programming on the Rise

Streaming of live sports events is attractive to advertisers because fans generally want to view sports events in real time, rather than storing them on

DVRs. At the same time, however, there is evidence that the vast majority of multichannel video subscribers who watch sports and have DVRs, use their

DVRs to record sports events. Sports sites that stream live games can jam advertisements into timeouts. Both usage of TV and usage of video on the

Internet and mobile phones have risen in the past few years. According to a

Nielsen "A2M2 Three Screen Report," as of Q1 2009, the average person in the

U.S. watched approximately 153 hours of TV per month, while the 131 million

Americans who watched video on the Internet averaged three hours of viewing video. People who used a mobile phone spent 3.5 hours a month watching mobile video.

According to Nielsen's "Year in Sports Media Report: 2013," from April

2012 to April 2013, the number of subscribers who accessed video on a sports website via a smartphone rose from 35.5 million to 61.7 million (+74%) and the average amount of time spent per month watching sports programming

23 MIGRATION OF SPORTS PROGRAMMING increased by 56%. Watching videos from sports sites on a computer increased by 36% while viewing sports on TV climbed 27%.

According to the 2014 Global Media Sports Consumption Report,

published by Sporting News Media, Kantar Media Sports and the Sports

Business Group, there are 168 million adults in the U.S. who follow sports. In

2013, fans spent an average of 8.3 hours per week following sports content, up from 6.2 hours in 2011. The primary driver of the increase was "less time watching sports on TV." Thirty-three percent of sports fans watched via live,

streaming in 2013, mostly via tablet.

In 2013, NBC announced that all m@rqijied eveI1ts would be streamed,

including programming on NBCSN and Golf Channel. FoxSportsGo, a

comprehensive mobile streaming app, announced that it will offer content from

Fox, Fox Sports1 and Fox'SNs. Similar to the WatchESPN app, which was launched in 2010 to stream simulcasts of ESPN's networks, FoxSportsGo allow's 'obile

viewing of more than 1,000 live games and events. In 2014, WatchESPN i

announced that it was giving 15 college conferences their own dedicated

streamed channels stocked with live events alndl onl-demand replays.

A. NCAA Basketball Tournament i

In 2002, FinaiFour.net, the official site of the NCAA men's and woments I

college basketball tournaments, registered 4.0 nlIillibnlvisits, the same as in 2001.

Users spent over 1.1 million hours on the site, averaging 19 minutes per visit,

notching 200 million impressions. In 2004I live video was available for the first,

24 MIGRATION OF SPORTS PROGRAMMING time for 37 games (first three rounds). In 2005, CBS registered 23 million total visits for the first week of the 2005 men's tournament, an all-time high. Daily unique visitors for the first two days of the tournament increased 23% and 24%, respectively, over 2004. In 2005, for the first time, 56 NCAA Division One

Basketball Championship games—through the regional semi-finals—were streamed live over the Internet along with pre- and post-game press conferences and video highlights on NCAAsports.corn and CBS.Sportsline.corn.

In 2006, March Madness On Demand (MMOD) shifted from a pay model to ad-support, moving from about 25,000 subscribers to more than 1.3 million unique users. CBSSportsLine streamed the first 56 games live and free of charge. In response to five million visits and 19 million video streams in 2006,

CBS Sportsline doubled its bandwidth capacity in 2007 and offered free live streaming of the first three rounds of the Tournament.

In 2008, CBSSports.corn provided the first live online broadcast of the

NCAA men's Final Four and National Championship. MMOD attracted 4.3 million unique visitors, up 147% from 2007. All CBS telecasts were available online. A new platform was launched allowing more than 200 websites to carry live video, including ESPN.corn, Yahoo, Sl.corn, YouTube and Facebook. In

2009, there were 7.5 million unique MMOD users (up 58%) and in 2010, there were 8.3 million unique users (up 10.4%).

In 2011, for the first time, MMOD was available free to iPad and iPhone users as well as online via NCAA.corn. Despite the availability of games on TV,

25 MIGRATION OF SPORTS PROGRAMMING

2008, for the first time, the NFL and NBC began carrying the 17 Sunday Night

Football games on NFL.corn and NBCSportslcorn. Iln 2009,. the NFL introduced the Red Zone Channel to carriers of NFL Netwark. It tracks every touchdown and is nearly identical to the Red Zone Channels laonched by DirecTV. In 2010,

NFL Sunday Ticket was offered online for $350 (including Red Zone Channel) to non-DirecTV subscribers.

The 2013 Super Bowl on CBSSports.corn attracted 3 million viewers, 43% more than in 2012 when NBCSports.corn had 2.1 million for the first Super Bowl to be streamed live. The 2013 Super Bowl had 114 million total minutes of

streaming, up 46% over 2012. The 2014 Super, Bowl, averaged 47 miriutes per

stream, 25% higher than in 2013. NFL Now,i an over~the-top streaming service was launched in 2014 and could be a future vehicle for live games. The 2015,'uper

Bowl on NBC was streamed without authentication and in 2015 there were nearly 34 million streams for the exclusive Yahoo carriage of the regular season i game from London between the Buffalo Bills:and Jacksonville Jaguars.

D. NBA

NBA League Pass for TV was launched in 1995-96. By 2013, League

Pass for broadband was combined with League Pass for Mobile. The TV and

broadband service were bundled for $ 189. Dual home and away game feeds i were introduced along with live DVR pause and rewind and the ability to view four games simultaneously. By 2015-16, fans had mare purchase options,

28 MIGRATION OF SPORTS PROGRAMMING including $59/team games or $6.99/single game and $99 for broadband only.

The mobile 549.99 app was available for 18 different platforms.

In 2008-09, the NBA became the first league to authorize its teams to launch regional video streaming, interactive TV and video-on-demand by the start of the season. In 2009-10, there were authorization tests by Comcast

Sportsnet in Chicago and Philadelphia testing free and paid local streaming services respectively. During the 2010-11 season, the Portland Trailblazers streamed 15 games. By 2013, Fox RSNs were testing TV Everywhere and Fox

SportsGo. All of the Fox 17 RSNs along with a half dozen independent RSNs were streaming and by 2014 all six Comcast RSNs were streaming as well.

Local games of 20 of the 30 teams were available to all distributors with TV

Everywhere deals.

E. NHL

The NHL has been slower than the other leagues to embrace digital offerings, but during the 2005-06 season, the NHL allowed Comcast to stream up to 300 live regular-season NHL games to Comcast's 7.7 million broadband customers.

ln 2008, the NHL reworked is digital offerings for the second time since

2006 and relaunched NHL Center Ice as NHL Game Center. Following research showing that 80% of hockey fans are also baseball fans, in 2009 MLB Advanced

Media and the NHL began selling a combined NHL and MLB package for

$ 139.95 covering a full season of baseball and a half season of the NHL.

29 MIGRATION OF SPORTS PROGRAMMING

Approximately 40 live garnes per week are available. Starting in 20'l3, all NHL

playoffs plus the Stanley Cup Finals, for the first time, became available online,

up from twelve playoff games in 2012.

F. College sports

CSTV.corn was launched in 2003. ln 2006„CSTV.corn offered live

streaming of more than 10,000 events from 250 schools covering 35 men's and

women's sports, includling basketball and football. ESPN360, launched in 2001

as ESPN Broadband, charging cable companies a license fee for carriage, which

included much of ESPIU/ESPN2 proclramming and additional college

football/basketball games.

Much has changed since. then. In 2008, CBS began streaming its SEC football schedule and adcled basketball in 20()9 and the Big Ten .treamedl over'00

events, including some men's non-conference basketball games. By 2013„

the Big Ten was streaming football, men's and women's basketball, hockey and

volleyball. The ACC launched its own diigital network in 2011, to carry 50 live

football and basketball games produced by Raycom along with 25 to 100

Olympics sports events not carried by ESPN or lFox Sports Net. ln 2012,

Oklahoma State became the first IBig-12 school to launch its own online

streaming network for football, basketball and other sports.

Starting in 2010, ESPN360 was rebranded ESPN3, with 40% of its content

simulcast from ESPN'. linear TV services, ESPN, ESPN2 and ESPNU and 60%

original programming. That year, all world Cup matches were on ESPN3,

30 MIGRATION OF SPORTS PROGRAMMING available to 51 million of 77 million broadband homes plus 20 million students on college campuses and 3.5 million stationed on military bases. Most of ESPN3's programming is exclusive, including college football and hundreds of college basketball games, pro and college lacrosse, Arena Football, cricket, Grand Slam tennis matches, alternate camera angles of simulcasts, and some simulcasts from ABC, RSNs and syndicated programming.

Vill. Mobile and Other Digital Devices

While a number of services have long provided game highlights, audio services and archived video programming for mobile devices, streamed video of live major league sports programming has grown dramatically since 2005. NBA

League Pass, for example, can be streamed via at least 17 different devices: iPhone, iPad, android tablet, Amazon Fire tablet, Amazon Fire TV, Amazon Fire

TV stick, Apple TV, Roku, PS3, PS4, PS Vita, Sony BluRay, Xbox 360, Xbox

One, Samsung Smart TV, Android TV and Chromecast.

A. NFL

From 2005-2010, the NFL had an exclusive sponsorship agreement with

Sprint for mobile audio exclusivity. In 2010, live Sunday Ticket and NFL Network games were available for a Verizon "V-Cast" service costing $10/mo. In 2011,

"NFL 2011" for Verizon had 140,000 downloads the first week for NFL Network video. In 2012, the Super Bowl, Pro Bowl and wild card games were streamed live via NFL Mobile for Verizon. NFL.corn and CBSSports.corn streamed the

Super Bowl, NBC's three post-season games were streamed on NFL.corn and

31 MIGRATION OF SPORTS PROGRAMMING

NBCSports.corn, including camera angles and highlights. In 2013, instantaneous

highlights (as many as 20/game), were introduced and Verizon signed a four- year extension to stream every regular-season and playoff game to mobile

phones. Its prior deal was for ESPN, NFL.Network and NBC games. The new

deal added CBS, Fox and playoff games. Iln 201l4, Vehzon charged $4.99/mo. for

its app and in 2015 NFL games were free from Verizon and the NFL introduced a

new mobile app.

B. MLB

Since 2008, MLB.corn At Bat has been, among the most popular of iTunes

downloads. Mobile traffic represented more than 50%'of.MLB.corn's overall traffic i

in 2011 compared to 37% in 2010 and a mlerh 8% I!n 2008 when MLBAM

introduced its iPhone app. In 2011, there were 3.3 million downloads of At Hat,.

of which 1.1 million were the paid version with additional features.

In 2009, for the post-season, MLB introduced online camera angles fromi

both dugouts, down foul lines and behind home plate. MLB.corn At Bat ended

2010 as the top-grossing iPhone app on all of ITUnes.l That year, MLB also

began streaming to several new platforms] inbludi6g the iPad, Playatation3 andi

interconnected TVs. Starting in 2014, MLB.TV began carrying the All-Star Game

and World Series.

C. NBA

NBA League Pass Mobile was first launched in 2009—a 40 live-game iper

week package, in partnership with MobiTV, for $39.95. In 2013, mobile gam'es'2 MIGRATION OF SPORTS PROGRAMMING were $49.99/season if purchased separately from the League Pass TV- broadband bundle. In 2014, for $ 149.95 non-Sprint subscribers could opt to watch five teams on digital platforms only. Unlimited access across TV, broadband and mobile was $ 199 in 2014.

D. NHL

Since 2010, the NHL has had an exclusive wireless deal with Verizon, including full-game video, live audio, in-game highlights, and condensed game video. Local games are blacked out. The NHL's GameCenter Premium Upgrade for $9.99 offers radio broadcasts, in-game video highlights and condensed game replays. Verizon customers get live broadcasts of NBC national games.

IX. SUMMARY

From 2005 through 2013, the number of national broadcast TV network telecasts and local TV station broadcasts of MLB, NBA, and NHL games plunged, both in absolute terms and as a percentage of the total number of national and local games distributed by broadcasters and cable TV networks.

During that time period, local broadcasts for MLB declined by 68% while national broadcasts and superstation games declined 61%. Similarly, for the NBA, local broadcasts declined by 85% while national broadcasts and superstation games were down by 15%. In contrast, the number of national and regional cable MLB televised games were up 35% and NBA local and regional cable telecasts rose by 12% from 2004-05 through 2012-13. The trend has continued since 2012-13 and there is no reason to believe that this trend will not continue.

33 MIGRATION OF SPORTS PROGRAMMING

For NHL, the number of regular-season local broadcast games in 2004-06 ~ and 2012-1 3 are not representative. There were lockouts in both seasons. As a

result, there were no games in 2004-05 and the 2012-13 regular-season was .

reduced from 82 games to 48 games. In 2011-1,2, however, there were 11

regular-season games on national broadcast and 48 on local broadcast TV. That season, there were 172 games on national cable networks and 1,643 on RSNs.

Thus, only 3% of NHL regular-season games'w6re'on'roadcast TV in 2011-12.

Not only have games migrated from 'broadcast to cable, since 2005, thle I

number of national and regional cable sports networks and subscribership to

cable networks have significantly increased while national and local broadcast

TV households have barely changed.

In 2005, 33 RSNs reached approximately 140.0 million subscribers. By

2010, there were 38 RSNs with 174.0 million subscribers. IrI 2011-12, 10 new

RSNs were launched adding 18.8 million new RSN subscribers. By 2014, th'ere were 51 RSNs with 199 million subscribers- 55% more RSNs.and 42% morh

RSN subscribers than in 2005.

By December 2004, there were 18 f'ull-time national cable sports

networks, including a dozen that had launched since 1999. These cable sports

networks had a combined total of about 540.6 million subscribers. By 2013 there

were 30 full-time national cable networks carrying sports programming, with a i

combined total of 1.37 billion subscribers compared to 1.11 billion subscribeIrs in

December 2009. The trend has continued. Iri 2013 there were 14 other

34 MIGRATION OF SPORTS PROGRAMMING established national cable networks with a combined 1.17 billion subscribers that carried sports programming.

The number of NCAA college football garnes carried by national broadcast networks, RSNs, and in syndication remained flat from 1999 to 2005, but the launch of ESPNU and College Sports TV resulted in the number of national cable telecasts doubling from 132 games in 1999 to 274 games in 2005.

From 2005 to 2013, there continued to be little change in the number of national broadcast and local broadcast/syndication of NCAA football games. In contrast, from 2005 to 2009, national and cable RSN/syndication of NCAA football were up 42% and 73% respectively. From 2009 to 2013, the cable increase was even more dramatic. Cable RSN syndication was up 107% while broadcast syndication declined by 53%. The most significant change, however, was ESPN3/over-the-top Internet, which was up 348% from 2005 to 2009 and another 30% from 2009 to 2013.

The trend in regular-season NCAA basketball is similar to NCAA football.

There are many more games on cable in light of the launch of new national cable sports networks and the shift from broadcast syndication to RSN syndication. In addition, there has been a tremendous increase in the number of games on

ESPN3 and other online networks.

The availability of out-of-market/video-on-demand MLB, NBA and NFL packages provides another dimension to the dissemination of sports MIGRATION OF SPORTS PROGRAIIIMING programming, as does the availability of live games on ESPN3, Internet

streaming websites, and mobile devices. Since 2005, MLB, the NBA, the NHL, the NFL, and the NCAA have.'go'ne.'eyond

offering national subscription video-on-dernan'd out-of-market packages and expanded their live-game offerings on broadband and mobile to more devices. More games are available and more distributors offer games. In addition, there has been an increase in innovative features (i.e., dual feeds,

camera angles, split screens) and broader bundling choices for subscribeis. I

Improvements are being made every year. Since 2013, leagues have begun to offer individual games, and in 2016 many RSNs are expected to begin offering authenticated local NBA and MLB home games on broadband within their home territories.

Similarly, since 2005, college conferences have expanded and enhanced their broadband and mobile offerings, often going beyond simulcasts to provide original programming with innovative features. Since 2009, all NCAA TouI'nament college basketball games have been streamed livel ovkr the Internet. Siri6sXMi has expanded its distribution, too, adding audio streaming of NFL and NBA games in 2011, NASCAR and NHL in 2012, plus there are apps for mobile devices.

The bottom line for 2010-13 and continuing through present day: A growing amount of the value of sports programming and sports viewership is attributable to new media. Sports programmi'ng'is valuable, but for over a

36 MIGRATION OF SPORTS PROGRAMMING decade much of that value has migrated from national broadcast networks, non- network local broadcasting, and so-called superstations to national cable networks, RSNs, the Internet and wireless platforms. That trend continues and is accelerating.

In addition to the migration of programming, the overall sports pie is growing. Much more sports programming is available, but not on broadcast networks and TV stations. While the number of live broadcast sports games has mainly declined or been static, the number of live games on cable, the Internet and wireless platforms has exploded over the past decade. The growth of new distribution outlets means there is more competition for traditional broadcasting of sports games. As a result, more and more sports programming value is attributable to new media and cable.

While some live games are simulcast on multiple media outlets, there has also been a boom in the number of live sports games that are distributed exclusively over cable networks (national and regional), as well as over the

Internet (wireline and wireless). This, too, cuts into the viewership of live team sports broadcast on TV stations during 2010-13. MIGRATION OF SPOR'TS IPROGRAMIUIING

MIGRATION OF SPORTS IPROGRAMIHING

Appendix II

LOCAL MAJOR LEAGUE BASI=BALL TELECASTS,,1990-2013 % Share % Share Local of Total of Total Total Year Biroadcast Games RSN Games Garnes 1990 1,577 59.0% 1,097 41.0'/o 2,674 1991 1, 714 57.4% 1,274 42.6% 2,988 1992 1,632 56.6% 1,251 43.4% 2,883 1993 1,815 58.1% 1,310 4'1.9% 3,125 1994 1,799 57.0% 1,355 43.0% 3,154 1995 1,847 56.8% 1,404 43.2% 3,251 1996 1,852 56.7% 1,416 43 3o/ 3,268

1997 1,686 49 6% 1,715 i 50.4% 3,401 1998 1,I55'1 44.5% 2,057 55.5% 3,708 1999 1I556 43.4% 2,160 56.6% 3,816 20CIO 1,559 40.8% 2,262 59.2% 3,82'1 2001 1,'508 38.5% 2,406 61.5% 3,914 2002 1,356 3'5.3% 2,483 64.7% 3,839 2003 1,236 31.3% 2,707 68.7% 3,943 2004 1,150 28.4% 2,906 71.6% 4,056 2005 1,066 25.8% 3,067 74.2% 4,'1 33 2006 994 23.2% 3,290 76.8% 4,284

20CI7 940 i3,450 i 78.6% 4,390 2008 7?2 1?.2% 3,725 82.8% 4 497 2009 512 11.3% 4,024 88.7% 4,536 2010 498 11.3% 3,902 88.7% 4,400 2011 449 10.1 /o 3,997 89.9% 4,446 2012 406 9 Qo/ 4,082 91.0% 4,488 2013 338 7 6% 4,115 92.4% 4,453

CAGR -6.5'/o 5.9% 2.2%

See: http:i'/wv ~.nmia.corn/-roberts/media07.htmi http://www.nmia.corn/-roberts/media0ti.htinl, Media Sports Business Newsletters MLB team, websites at start of season John Mansell Associates, 2016

40 MIGRATION OF SPORTS PROGRAMMING

Appendix lll

LOCAL NBA BASKETBALL TELECASTS (1990-2013)

Local Share Share Year Broad- of Total of Total Total cast Games RSN Games Games 1990-91 724 45.0% 885 55.0% 1,609 1991-92 700 43.2% 919 56.8% 1,619 1992-93 705 42.9% 937 57.1% 1,642 1993-94 684 44.2% 862 55.8% 1,546 1994-95 764 46.9% 866 53.1% 1,630 1995-96 831 47.6% 913 52.4% 1,744 1996-97 823 45.7% 978 54.3% 1,801 1997-98 837 45.6% 1,000 54.4% 1,837 1998-99 534 46.0% 626 54.0% 1,160 1999-00 781 39.5% 1,197 60.5% 1,978 2000-01 786 40.0% 1,178 60 0% 1,964 2001-02 783 39.5% 1,198 60.5% 1,981 2002-03 726 37.2% 1,224 62.8% 1„950 2003-04 790 39.0% 1,236 61.0% 2,026 2004-05 558 26.3% 1,561 73.7% 2,119 2005-06 583 26.9% 1,582 73.1% 2,165 2006-07 488 22.2% 1,706 77.8% 2,194 2007-08 412 18.8% 1,783 81.2% 2,195 2008-09 341 16.1% 1,778 83.9% 2,119 2009-10 219 10.5% 1,870 89.5% 2,089 2010-11 180 8.3% 1,998 91.7% 2,178 2011-12 77 4.3% 1,719 95.7% 1,796 2012-13 86 3.9% 2,113 96.1% 2,199

CAGR -9.2% 4.0% 1.4%

Source: Media Sports Business Newsletters NBA team websites at start of season John Mansell Associates, 2016

41 MIGRATION OF SPORTS I ROGRAMhllING

Appendix IIV

LOCAIL hlHL TELECA iTS (1990-2013)

Local Share Share Y'ear Broad- of Total of Total 'Total cast Games RSN Games Games 1989-cJO 251 28.7% 624 71.3% 875 1990-CJ1 2'31 28.9 618 71.1% 869 1991-CJ2 272 30.1% 632 69 9o/ 904 1992-CI3 283 28 7% 703 '1.3% 986 1993-CI4 3'I 6 27'.6% 828 72.4% '1,144 1994 CI5 165 23?% $31 76.3% 696 1995 CJ6 308 25 9% 881 74.1% '1,189 1996-CI? 337 2~I 3% 996 74.7% '1,333 1997-CJ8 g4~) '1,251 308 24 6% '5.4'%9.1% 1998 CI9 269 20.9% I,'016 1,285 1999-00 273 18.7% 1I187 '1.3'% '1,460 2000-01 262 16.3o/o 1 )343 83.7% 1,605 2001-01 283 17'.8% 1,311 82.2% '1,594 2002-03 228 14.2% 1,'374 '5.8% "l,602 2003-04 194 11.8% 1 ~ 44'7 '8.2% 1,641 2004-05 0 n.a. 0 tl.a. 0 2005-06* 169 10.3o/o 1,467 89.7% '1,636 2006-07 183 11.2% 1,456 88.8% '1,639 2007-08 106 6.6'/o 1,490 '3.4% 1,596 2008-09 126 7.5'/o 1',54'6 '2.5% '1,672 2009-11 0 80 4.1o/o 1,874 95.9% 1,954 2010-111 77 4.5'/o 1,640 95.5% 1,717 2011-11 2 48 2.8% 1',643 '7.2o/o "l,691 2012-1I 3 7 0.?o/o 979 99.3% 986

CAGR 14.4% 2,0 /o 0.5%

Media Sports Business Newsletters NHL team websites at start of:season John Mansell As. ociates, 2016

42 MIGRATION OF SPORTS PROGRAMINING

Appendix V

NATIONAL CABLE SPORTS NETWORK SUBSCRIBERS

Launch Date 2004 2009 2013 (mil.) (mil.) (mil.) Sep-79 ESPN 89.1 99.1 97.0 Oct-93 ESPN2 87.9 98.8 96.9 Jul-94 TVG Network (e) 14.0 30.0 35.0 Jan-95 Golf Channel 66.9 81.8 81.7 May-95 ESPN-Classic 54.8 60.4 29.7

Jan-96 Fox Sports 1 (formerly SPEED) 63A 74.7 88.0 Nov-96 ESPNews 42.9 70.9 74.4 Nov-96 Fox Deportes 6.1 11.5 13.3 Nov-97 Fox Soccer Channel (now FXX) 26.7 35.6 70.3 Jul-02 Fox Sports 2 (formerly Fuel TV) 29.3 36.0 Jan-03 NBA TV 13.8 46.2 59.5 Jan-03 Horseracing TV (e) 12.0 18.0 19.0 Feb-03 Gol TV 5.5 13.1 14.1 Jun-02 CBS Sports Network (CSTV) 7.6 34.8 51.0 May-03 Tennis Channel 11.5 24.4 33.9 Nov-03 NFL Network 24.0 54.8 73.0 Jan-04 ESPN Deportes 3.0 11.1 16.1 Feb-04 siTV 9.1 n.a. n.a. Sep-04 Fox College Sports 39.8 49.6 Oct-04 MAVTV 6.1 32.8 Oct-04 Blackbelt TV (e) 8.0 9.6

Source: SNLKagan database, (e)-estimate John Mansell Associates, 201 6

43 MIGRATION OF SPORTS PROGRAMMING

Appendix V (continued)

NATIONAL CABLE SPORTS NETWORK SUBSCRIBER.S

Launch Date 2004 2009 2013 (mil.) (mil.,) (mil.) Mar-05 ESPNU 66.9 '4.6

Dec-05 World Fishing Network ~ 3.5 ~ 6.9 Sep-06 NBC Sport. Net (Versus) 63.4 77.6 Aug-07 Big-Ten Network 53.8 Oct-07 NHL Network 25.5 38.4 Jan-09 MLEI Network 52.3 69.6 Apr-12 IJnivision Deportes n.a. 30.3

Aug-12 belP J Sport. n.a. 11.9 Aug-12 Pac-1 2 Networks n.ai 10.9 Aug-13 One World Sports n.a. 20.0 Total 538.3 1,,105.1 1,374„9

Source: Sl JLKagan database, (e)-estimate John Mansell Associates, 2016

44 MIGRATION OF SPORTS PROGRAMMING

Appendix Vi

NATIONAL CABLE NETWORKS WITH SPORTS PROGRAMS Launch ------Subscribers------Dec. Dec. Date Network 2004 2009 Dec. 2013 (mil.) (mil.) (mil.) Dec-76 TBS 88.5 100.1 98.6 Nov-78 WGN America 72.0 ?2.3 73.0 Oct-79 Galavision 56.0 68.2 Apr-80 USA 88.7 99.2 98.2 Mar-83 Spike TV (former TNN) 88.2 98.8 97.0 Jun-85 Discovery 99.8 98.3 Feb-87 Travel Channel 77.7 95.1 93.9 Oct-88 TNT 99.1 97.5 Apr-89 CNBC 24.8 97.6 95.5 Jul-91 truTV n.a. 92.2 90.8 Apr-93 The Outdoor Channel n.a. 38.7 Jun-94 FX 86.6 95.8 96.7 Jul-95 Outdoor Life (Versus) 6'l.6 63.4 n.a. Oct-96 Animal Planet 86.4 96.3 95.9 Apr-03 The Sportsman Ch. 7.6 19.5 31.9 TOTAL 89?.3 1219.3 1174.2 truTV and Outdoor Ch. did not have sports in 2004. Versus was all-sports NBCSN in 2013 Source: SNLKagan database John Mansell Associates, Inc. 2016

45 MIGRATION OF SPORTS PROGRAMMING

Appendix Vll (continued)

COLLEGE FOOTBALL BROADCAST, CABLE AND STREAMING, 2009-10

Big Big Big- Pac ACC East 12 10 CUSA lndep MAC MWC 12 SEC Sunbelt WAC Bowls Totals ESPN 27 19 6 24 11 1 16 5 35 3 17 24 188

Syndication 12 17 1 20 12 33 19 20 134

Internet 3 1 16 33 16 5 77 ABC 13 6 17 15 RSN 10 11 28 15 BTN 39 ESPN3 26 4 35 CBSSN 'i5 9 32 FSN 18 13 31 NBCSN 8 5 19 CBS 15 17 RSN NBC Fox Fox Col Sports NFL Net Total 733

Source: http://mattsarzsports.corn/ O.lnhn Msnsell Assnriates 9018

47 MIGRATION OF SPORTS PROGRAIIIMING

Appendix Vll (continued)

COLLEGE FOOTBALL BROADCAST, CABLE AND STREAMING, 2013-2014

Big Big- Pac ACC Amer. 12 10 CUSA Indep MAC NIWC 12 SEC Sunbelt Bowls, Totals ESPN 25 26 9 21 16 20 17 44 6 215 ESPN3 43 5 25 120 RSN 15 17 6 63 Syndication 16 10 14 7 63 ABC 15 2 6 20 FS1 12 15 56 BTN 47 47 Pac-12 35 CBSSN 24 RSN (Fox) 17 34 Internet 17 RSN (Comcast) 17 6 23 CBS 16

Fox 10 1 19 SEC 14 14 Fox Col Sports 10 10 NBC Altitude

Aggievision

HDNet NBCSN NFL Net Other RSN TBS TOTAL 644

Source: http://mattsarzsports.corn/ John Mansell Associates, 2016

48 MIGRATION OF SPORTS PROGRAMMING

APPENDIX VIII

John Mansell Curriculum Vitae

John M. Mansell is president of John Mansell Associates, an independent research and consulting firm. In 2011-15, Mr. Mansell has completed consulting assignments relating to the economics ofregional sports networks, the dynamics of broadcast retransmission consent/reverse compensation negotiations, U.S. Sports rights, and cable TV bundling. Most recently, in April 2015, Mr. Mansell completed a valuation of 26 EBS licenses in Maine for the Camden National Bank.

He was previously a senior analyst at Paul Kagan Associates, Kagan Research, LLC and Kagan Media Appraisals, a media and communications consulting firm based in Monterey, California. He began his career with Kagan as an analyst in 1975 and remained at the firm until 2007. Mr. Mansell is 63 years old and resides in Great Falls, Virginia, outside ofWashington, D.C. He has a bachelor of arts in economics &om the University ofMichigan, a law degree from Thomas M. Cooley Law School in Lansing, Michigan, and is a member ofthe Michigan Bar, District of Columbia Bar, and the Federal Communications Bar Association.

Mr. Mansell's interest in multichannel video services dates back to the industry's formative growth years in the early 1970s, when he was attending college. In the late 1970s and early 1980s, as &anchising became popular, he began closely monitoring and analyzing cable TV &anchising and regulation to identify trends within the industry. At the same time, he began writing his first newsletter, CABLE TV REGULATION, which analyzed the federal, state, and local regulation of cable TV. As public interest in the industry began to develop, he also started to monitor rate changes, pole attachment disputes, overbuilding, and copyright.

Over the years, Mr. Mansell has concentrated on analysis of financial, technical and legal-regulatory developments affecting multichannel video program distribution technologies, including cable TV, wireless cable PADS), direct broadcast satellite as well as Satellite Master Antenna TV. He has also developed specific expertise in the legal and regulatory aspects of the cable television industry and related electronic media

As a Senior Analyst, Mr. Mansell was responsible for the writing and editing of specialty newsletters that monitor and analyze developments in the cable and electronic media. The production of these newsletters requires continual research, including the review and analysis of all pertinent legal developments that affect the industry. Mr. Mansell was responsible for writing, editing and/or contributing to the following monthly newsletters:

49

MIGRATION OF SPORTS PROGRAMMING

CABLE TVLAP'REPORTER which monitors, and evaluates signi6cant lawsuits involving the cable industry, including antitrust, First Amendment, franchising, taxation, copyright, rate regulation, and mandatory access to property..

SIRELESS BROADBAND (formerly WITLESS/PRIVATE CABLE

INVESTOR) which covers the economic, 6nancial, technical, legal and: operating issues that affect SMATV, MMDS, LMDS and other wireless cable services. A comprehensive, is data base of system sales and wireless cable transactions maintained and constantly 'nd updated.

MEDIA SPORTS BUSINESS which reportsl od th6 6 olving relationship between sports teams and the electronic media and focuses; upon regional sports networks, franchise valuations, transactions, media rights fees, advertising, subscription revenue, stadium/arena 6nancings.

THE DBS REPORTwhich consists of analgsid ofmarketing, technology and . regulatory developments that affect the C-band, Ku-band and Ka-band direct-to-home satellite service worldwide.

DIGITAL TELEVISION tracks the 6nancial, regulatory and technical aspects;of the transition &om analog technology to digital.

In addition, Mr. Mansell has contributed to other publications and special reports, including: the F/'ireless Cable Databook MediaCast,i The Business ofBaseball, The Business Football, The Business Basketball, The Business Hockey and the U.S.: of of of 'f TVSports Business Databook.

Mr. Mansell has also been responsible for writing'nd contributing to other newsletters, including, CABLE TV TECHNOLOGY, CABLE TV SECURITY, CELLULAR INVESTOR, SMATV NEWS, and MEDIA MERGERS 8'c.ACQUISITIONS. He keeps

abreast oftechnological developments in the cable industry, such as 6ber optics, new build and upgrade construction trends, headend technology', converter-descramblers,'igital

set-tops, and cable modem technology for high-speed access to the Internet. l

He has authored several reports on cable TV overbuilds for the National Cable 8.

Telecommunications Association. In 2007, he prepared. a report for Mediacom on the ~ economics ofhigh-speed Internet in rural areas that was submitted to a unit ofthe U.S. Department ofAgriculture and in 2006 a report for Bright House Networks on th'e ixbphct competition in Manatee County, FL. He also writes for other publications and engages in consulting and valuation work relating to cable TV, private:cable, wireless spectrum, and the sports business.

Over the years, Mr. Mansell has done consulting and/or valuations for numerous. entertainment and telecommunications organizations including the National Cable 4 50 MIGRATION OF SPORTS PROGRAMMING

Telecommunications Association, Major League Baseball, National Football League, Boston Red Sox, Minnesota Twins, Virginia Stadium Authority, Fox Sports, Cablevision Systems, Microband, Continental Cablevision, Tele-Communications, Viacom International, Time, Inc., Sprint, Goldman Sachs, J.P. Morgan, Nucentrix Broadband Networks, Bell Atlantic, Transworld Wireless, Tribune Co., General Instrument, and many others. Mr. Mansell has completed &anchise valuations and/or media rights valuations of MLB, NBA, NHL and NFL franchises.

In the course of his association with the Kagan Group, he has been wholly responsible for writing reports appraising over 100 MMDS (BRS/EBS) propeities and has also worked on valuations of another dozen wireless properties outside the United States.

He has appraised ~S/ITFS channels in over numerous communities across the United States including: Buffalo, NY; Cleveland, OH; Dallas-Fort Worth, TX; Hartford, CT; Las Vegas, NV; Minneapolis, MN; New York, NY; Pittsburgh, PA; Washington, DC. numerous cities in Florida including Sarasota, Bradenton, and Fort Pierce; numerous cites in California including San Francisco, San Jose, Anaheim, San Bernardino and Riverside; Traverse City, MI; numerous cities in Maine; numerous cities in North Carolina including Asheville, Burlington, Charlotte, Elizabeth City, Fayetteville, Gastonia, Goldsboro, Kinston, Greensboro, Greenville, Hickory, Jacksonville, New Bern, Raleigh, Roanoke Rapids, Rockingham, Rocky Mount, Shelby, Wilmington, and Winston-Salem.

In 2009, Mr. Mansell appraised the value of South Carolina's 60+ EBS 2.5 GHz licenses covering 500 million MHz pops and assisted in negotiations with potential bidders for the spectrum.

He has also appraised WCS frequencies and worked on appraisals outside the United States including a number of LMDS and MMDS interests in New Zealand (Auckland, Christchurch, and Wellington, NZ) and South America (Guatamala City, Caracas, Maracaibo, Maracay, Valencia, Barquisimeto, San Jose, Panama City, Parana-Santa Fe, and Aracaju).

In WIRELESS BROADBAND (formerly WIRELESS/PRIVATE CABLE INVESTOR) newsletter, Mr. Mansell regularly wrote about valuations ofMMDS mergers and acquisitions and valuations ofpublicly-traded companies.

Mr, Mansell has also been a speaker and moderator at numerous Kagan seminars, meetings, national cable TV trade shows, and private/wireless cable business conferences, and is often quoted in trade publications as well as popular business media.

He was responsible for organizing and moderating one- and two-day seminars relating to the sports business, including &anchise valuations, labor relations, stadium-arena finance, TV rights fees, ratings, advertising, new leagues, new technology, and new revenue 51 MIGRATION OF SPORTS PROGRAMMING

opportunities. He has speci6cally organized and moderated one- day seminars relating.to . the economics ofthe wireless cable (MV1DS) business, including valuations, finance, legal, regulatory, marketing, competitive, and technical developments. These meetings involve wireless communications industry leaders, beth ias speakers and: attendees. Since: he has been writing 8VRELESS BROADBAND, he has been invited to moderate or speak i as a panelist nearly every year at the annual convention of the Wireless Communications Association.

Mr. Mansell has been a &equent speaker at industry trade shows such as the National Cable TV Assn. Convention, the Western Cable Show, the Private 8r, Wireless Cable Show, and the Wireless Communications Assn. Convention. Mr. Mansell has been quoted in the Wall Street Journal,CAUSA Today, Washington Post,'ew York Times, Newsday, and many other daily newspapers along with Business

Week, Forbes, and other business magazines. Heihas also been quoted in media trade . publications such as CableWorld, Multichannel News, and.Hollywood Reporter.

In 2004, Mr. Mansell was named to the Board ofDirectors oftwo George Mason University non-pro6t companies that own 2.5 GHlz @ectruin and:in 2006 assisted iq thy valuation of leases assigned to Cleanvire and Sprmt. In 2013, Mr:, Mansell was named Chairman ofthe George Mason University Instructicnal Foundation. In 2009 he also served as a director on the George Mason University Intellectual Properties Board. He has been Treasurer ofthe Timberlake Estates South Homeowner's Association since 2010.

During the course ofhis employment he has testi6ed or:been deposed as an expert witness in numerous legal proceeding relating to cable TV, &anchising,. private cable (SMATV), sports rights, and MMDS in federal and state court proceedings. He has: testi6ed before a U.S. House subcommittee and been involved in arbitrations.

Expert Witness/Arbitration Cases: l. 1982 dispute between Teleprompter and Pittsburgh, PA, over the terms of Teleprompter's &anchise application—Expert witness for Teleprompter.

2. Rate hike arbitration, Viacom v. City ofN&h48e—IPaiticipated as:expert for Viacom.

3. Cable TV overbuild litigation between Americable and Tele-Communications, Inc. over right to serve Homestead AFB, FL.

52 MIGRATION OF SPORTS PROGRAMMING

4. Federal court suit against the Kansas City Royals for breach of contract and other allegations claimed by a disgruntled regional sports network promoter.

5. Arbitration involving a disgruntled shareholder of an SMATV company, who claimed other shareholders wrongfully deprived him ofhis share of the company. Participated as expert for plaintiff.

6. Contract dispute between Amplica and Arvin Industries relating to components for Amplica's TVRO receiver—1990.

7. Dispute between partners ofWestern Pay TV and Boettcher over allegedly fraudulent economic projections.

8. Shea Properties v. Heritage Cable TV—access to property/overbuild dispute in San Jose, CA—1990.

9. Sacramento Cable disputes with Pacific West Cable Co., City of Sacramento and others—Overbuild/First Amendment/unfair trade/price discrimination disputes—1991-95.

10. U.S. v, Playboy Entertainment Group, 98-1682 (U.S. Supreme Court, May 22, 2000)—First Amendment challenge to audio/video scrambling requirements of the 1996 Telecommunications Act. Analysis of cable TV encryption costs.

11, City of San Jose v. Gill Industries, Inc., et al., CV 735578, (Santa Clara Superior Court, 1997)—Cable TV &anchise fee dispute,

12. Warner v. Wireless Broadcast Systems ofAmerica, 96-2214-CIV-T-25 (M.D.Fla. 1998)—Dispute over valuation of several Florida wireless cable markets.

13. Showtime Networks, Inc. v. Comsat Video Enterprises, 95-600849 (New York County Supreme Court, 1998)—Hotel pay-TV contract dispute.

14. Children's Cross-Cultural Communications Foundation v. Bay Area Cablevision, Case No. 216081 (Sonoma County, CA, Super. Ct.)—Dispute over a putative contract and value of ITFS &equencies in San Francisco.

15. SportSouth Network d/b/a Fox Sports South v. Time Warner Inc., 1999CV10083 (Fulton County, GA, Super. Ct., Sept. 7, 1999)—Covenant not to compete dispute over the definition of regional sports network and valuation of southeastern RSN.

16. John Hansen v. Transworld Wireless TV-Spokane, Inc., 99203108-8 (Spokane County, WA, Super. Ct., filed May 27, 1999)—Contractual dispute over value of Spokane MMDS &equencies.

53 MIGRATION OF SPORTS PROGRAMMING

17. USA Media Group v. Truckee Donner PUD (E.D. Cal.)—2003 antitrust- declaratory judgment suit relating to cable operator's access to utility poles for purpose of rebuild in a community alleged to be planning a municipal overbuild.

18. Zenith Electronics Corp. v. WH-TV Broadcasting Corp., 04-1635 (7th Cir. Jan. 20, 2005)—Expert for Zenith in dispute over damages for San Juan, Puerto Rico digital MMDS operator who purchased allegedly flawed set-tops &om Zenith.

19. Sprint Las Vegas MMDS Partnership arbitration—Valuation ofMVG3S &~uencies in Las Vegas—Arbitrator appointed by Sprint in 2003 dispute.

20. Houston McLane Co., Inc. et al v. Af61iated Regional Communicaiions, Ltd.l d/b/a Fox Sports Southwest, No. 2003-10943 (~Harris~ County, TX Dist. Ct., 2004)— Worked for Fox Sports Southwest in litigation~relating to "covered offer" to acquirelthe Houston Astros and Houston Rockets home televilsiolI1 rights.:

21. YankeeNets v. Cablevision Systems (2004)—Private arbitration over the pricing and tiering ofa regional sports network distributing the New York Yankees and New Jersey Nets.

22. The Maryland Jockey Club ofBallimore City v. ODS Technologies, L.P., WNM

03-CV-2124 (D. Md. 2004)—Testified for TVG in dispute ~over exclusive rights to . televise and provide electronic wagering on horse; racing at; Pixnlico.

23. Interactive Management Services v. AGS Telecom, Inc., et al. (Dec. 2004)-, Retained as expert witness for AGS Telecom in arbitration over the value of 218-219 MHz spectrum in Boston, New York, Houston'nd L'os Angeles. 24. In the Matter of Distribution ofthe 2004 and 2005 Cable Copyright Royalty 'unds, Docket No. 2007-3 CRB CD 2004-05 (June 1, 2009)—Testi6ed as expert witness on behalfofMotion Picture Assn. ofAmerica in distribution dispute with sports claimants—MLB, NBA, NHL, NFL, and NCAA—'relating'o distant signal royalties.

25. Anson, et al. v. Triple Action Entertainment, 34-2010-00079813, Sacramento County Super. Ct., (settled March 2015)—Retained as expert in leased access channel dispute relating to the economic viability of the business model for an extreme sports network.

Contact: John Mansell 1093 Loran Court Great Falls, VA 22066-1533 Phone: 703-433-0571 Mobile: 703-932-9719 e-mail:[email protected] 54 DECLARATION OF JOHN NIANSELL

I declare under penalty of perjury that the foregoing testimony is true and correct, and of my personal knowledge.

Executed on December /~, 2016

~.'ohn M. Mansell, Jr. I

I

1

I Before the COPYRIGHT ROYALTY JUDGES Washington, D.C.

In the Matter of

Distribution of the Docket No. 14-CRB-0010-CD (2010-13)

2010, 2011, 2012, and 2013 Cable Royalty Funds

DIRECT TESTIMONY OF SONDA K. MARTIN

DECEMBER 22, 2016 DIRECT TESTIMONY OF JONDA K. MARTIN

I. BIOGRAPHICAL INFORMATION

My name is Jonda K. Martin. I am the President of Cable Data Corporation

("CDC"), which is located in Mount Airy, Maryland. In my more than 25 years at CDC,

I have been actively involved in all aspects ofthe company, including research, data entry, report generation, and administration. I received a Bachelor of Science/Business

Administration degree from American University in Washington, D.C., with concentrations in international business and management of information systems. I also received an MBA from University ofMaryland. I have previously testified before the

Copyright Arbitration Royalty Panel ("CARP") regarding CDC's operations in connection with the CARP's distribution of 1993 and 1999 cable compulsory license royalties, and before the Copyright Royalty Judges ("Judges") in connection with Phase I and Phase II proceedings regarding the distribution ofthe 2000-2003 cable royalty funds, a Phase I proceeding regarding the distribution ofthe 2004 and 2005 cable royalty funds, and a Phase II proceeding regarding the 2004-2009 cable and 1999-2009 satellite royalty funds.

H. PURPOSE OF TESTIMONY

The purpose ofmy testimony is to provide the Judges with an overview of CDC's operations and describe its data collection process and methodologies in relevant detail. I will describe the carriage data that CDC supplied to Program Suppliers in connection with this proceeding, which I understand were utilized by Dr. Martin Frankel and Dr. Jeffrey S. Gray in their sample selection and analyses, and by Howard Horowitz in connection with Program Suppliers'able operator survey. I will also describe the local county analysis that CDC performed for Program Suppliers in connection with this proceeding, which I understand was provided to Nielsen.

IH. CABLE DATA CORPORATION

CDC was established in 1979 to collect and analyze information on Statements of

Account ("SOAs") that cable systems file with the Licensing Division ofthe Copyright

Office ("Licensing Division"). CDC makes the collected information available to users by purchase, either on an as needed basis or by subscription. CDC is the only company providing such a service. Numerous parties involved in the cable and satellite industries rely on data collected by CDC. This is particularly true for parties involved in copyright compulsory license proceedings. As a result, CDC data have been presented over the years to the Copyright Royalty Tribunal, the CARP, and the Judges in virtually all ofthe cable and satellite copyright royalty distribution proceedings and rate adjustment proceedings. In this section ofmy testimony, I will provide an overview of CDC's operations and its data collection methodologies.

Data collection is an integral part of CDC's operations. CDC has two full-time employees who spend the vast majority of each work day on-location in the Licensing

Division ofthe Copyright Office. Those employees record data and other information from each publicly-available, filed SOA on laptop computers.'he employees return to

'nce SOAs are filed at the Copyright Of6ce they are subject to review by the Licensing Division before they are made available for public inspection. The Licensing Division's initial CDC's office periodically to transfer the data collected at the Copyright Office on laptops to CDC's HP3000 minicomputer. Once the data are transferred to CDC's minicomputer,

CDC produces standard reports and customized reports which summarize the SOA data.

To keep CDC data as consistent as possible with the SOAs on file with the Licensing

Division, CDC performs regular system updates to account for modifications made to a

system's filing, for reasons such as additional royalty payments and refunds issued by the

Licensing Division.

After the SOA data is entered in CDC's computer system, CDC produces both

standardized and customized reports of aggregated cable system data for its clients.

Both CDC's standardized and customized reports are derived from the same database and

rely on the same CDC data protocols.

IV. CDC CARRIAGE DATA PROVIDED TO PROGRAM SUPPLIERS IN CONNECTION WITH THEIR CABLE OPERATOR SURVEY

A. CDC Data Supplied Prior to Sample Selection

In early 2011, at Program Suppliers'equest, CDC provided a data report of all

publicly-available SOAs filed by Form 3 cable systems, which reported at least one

distant signal on their SOA filing for the first accounting period ofthe previous royalty

year (i.e., 2010-1), and the corresponding royalties paid by each cable system. Program

Suppliers made similar requests, and CDC provided similar reports, in early 2012 relative

review typically causes a several month delay between the date that the SOAs are filed with the Office and the date that the filings are available for CDC's on-site employees to access the information.

"- For the 2010-13 cable royalty years, Form 3 cable systems contributed the majority ofthe cable royalties that were collected by the Licensing Division. Form 3 cable systems were responsible for 98.4% of the cable royalties paid in 2010, 98.6% in 2011, 98.7% in 2012, and 98.8% in 2013. to all Form 3 cable systems who submitted. an SOA in the first accounting period of 2011

(i.e., 2011-1), in early 2013 relative to all Form 3 cable systems who submitted an SOA in the first accounting period of 2012 (i.e., 2012-1), and in early 2014 relative to all Form

3 cable systems who submitted an SOA in the first accounting period of 2013 (i.e., 2013-

As directed by Program Suppliers, CDC sent the data report for each accounting period described above directly to Dr. Frankel. CDC understood that the data CDC provided in these reports would be utilized by Dr. Frankel to select a sample of cable systems that carried distant signals in the years 2010, 2011, 2012, and 2013.

B. CDC Data Supplied Following Sample Selection

After Dr. Frankel completed the sample selection process for a given royalty year,

CDC received a second request from Program Suppliers for more detailed information about the Form 3 systems in Dr. Frankel's sample and the universe of all Form 3 systems previously reported by CDC for 2010, 2011, 2012, and 2013. Program Suppliers requested that the detailed information be broken down into the following five carriage categories: (1) PTV-only systems (cable systems carrying only PTV distant signals); (2)

Canadian-only systems (cable systems that carried only Canadian distant signals); (3)

WGN-only systems (cable systems that carried only WGN as a distant signal); (4) network affiliates plus systems (cable systems that carried network-affiliated stations only or with other types of stations as distant signals); and (5) independent stations only systems (cable systems that carried two or more independent distant signals, but are not

PTV-only, Canadian-only, or WGN-only systems). As requested by Program Suppliers, CDC sent the detailed information directly to

Howard Horowitz and Nuria Riera at Horowitz Research, Inc. CDC understood that this data was to be utilized by Mr. Horowitz and Ms. Riera in connection with Program

Suppliers'able operator surveys for 2010-13.

V. CDC CARRIAGE DATA PROVIDED TO PROGRAM SUPPLIERS IN CONNECTION WITH THEIR NIELSEN STUDIES

At the request of counsel, CDC also provided Program Suppliers with another set of customized data reports for each ofthe 2010-2013 cable royalty years. For each royalty year, these data reports listed all commercial and non-commercial broadcast stations carried as full-time distant signals by cable systems, the number of distant subscribers to which each station was available, and CDC's calculation ofthe distant fees generated by that station. I understand that Dr. Jeffrey Gray utilized these data to select the sample stations for each royalty year at issue in this proceeding for Program

Suppliers'ielsen distant viewing studies. I also understand that Dr. Gray utilized these

CDC data in his economic analysis.

VI. CDC's LOCAL COUNTY ANALYSIS

After Dr. Gray selected sample stations for each ofthe 2010-2013 cable royalty years, Program Suppliers'ounsel sent the lists of Dr. Gray's sample stations to CDC.

CDC then analyzed each ofthese stations in order to determine which counties fall within the station's local service area. We refer to this exercise as local county analysis. CDC based identification ofthe counties local to each ofthe 2010-2013 cable sample stations on the FCC signal carriage rules. The process CDC followed was different for commercial stations, noncommercial educational stations, and Canadian stations. I explain how CDC conducted its local county analysis for each ofthese types of stations below. In addition, Appendix A to my testimony provides illustrations ofhow the local county analysis was performed for each category of stations for 2010-2013.

A. Local County Analysis Of Commercial Stations

Por our local county analysis of commercial stations, CDC employed the following general steps: Grst, we identified the counties that constituted each station's

Designated Market Area ("DMA"). All such counties are considered local for that station. Second, we identified the counties in which each station was deemed

"significantly viewed" per the PCC. All such counties are considered local for that station pursuant to the FCC's signal carriage rules. Lastly, we looked at other factors that would qualify a county as local to the station in question.

B. Local County Analysis OfNoncommercial Educational Stations

In order to determine the local service area of a noncommercial station, CDC applies the two FCC criteria that define whether a PTV station must be carried by a cable system - the Grade 8 contour and/or a 50-mile plotted radius. If a cable system's primary headend. is located either (1) inside the station's Grade B contour; or (2) within a 50-mile

'he signal carriage rules, now rescinded, were found at Sections 76.57 through 76.63 ofthe regulations ofthe FCC. 47 C.F.R. f$ 76.57-76.63 (1976). radius plotted around the city to which the station is licensed, the cable system must carry the PTV station as a local signal.

C. Local County Analysis Of Canadian Stations

CDC used multiple criteria to identify local counties for the Canadian stations in the sample. Some Canadian stations are significantly viewed (SV) in U.S. counties; all such counties were designated as local to the station. Where a station's Grade B included counties located outside all television markets, those counties were designated as local.

Once CDC completed the local county analysis, I sent the results to Program

Suppliers'ounsel. I understand that the results of CDC's local county analysis were provided to Nielsen.

Thank you for the opportunity to present this information in this proceeding. I hope that it will assist you in your deliberations. DECLARATION OF JONDA K. MARTIN

I declare under penalty of perjury that the foregoing testimony is true and correct, and of my personal knowledge. Executed on December ~L, 2016

Jon a . Martin APPENDIX A

I. LOCAL COUNTY ANALYSIS OF COMMERCIAL STATIONS

A. Local County Analysis Illustration For Commercial Station WKBD, Channel 50, Licensed To Detroit, Michigan

Described below are the steps CDC employed for the county analysis of commercial stations. The steps consist principally ofDesignated Market Area ("DMA") analysis, significantly viewed ("SV") analysis, 35-mile zone analysis, and the Grade B Contour analysis. I selected for the purpose of illustration, commercial station WKBD, channel 50, licensed to Detroit. The analysis applied to WKBD is repeated for all ofthe commercial stations in each sample for which CDC conducted a county analysis.

1. DMA Analvsis

Nielsen groups counties by DMAs.'ach DMA consists of a group of counties forming an exclusive geographic area in which Nielsen has determined that the home market television stations hold a dominance ofviewing. Although a few counties are split between DMAs, as a rule each county is assigned to one and only one DMA.

Attachment 1 is a page with the Detroit market (among others) from Nielsen's 2010 DMA report titled "U.S. TV Household Estimates" ("the DMA book"), published September 2009, which shows all DMAs and the counties associated with each. This page provides a good example ofhow DMAs are used to identify local cable system carriage for station WKBD. Again, the objective in determining the counties where a station is local is to enable Nielsen to exclude cable viewing &om those counties, with the result that only distant viewing for WKBD will be captured. The Detroit DMA market consists ofnine Michigan counties: Lapeer 8anilac Livingston St. Clair Macomb Washtenaw Monroe Wayne Oakland Because WKBD is licensed to Detroit, a cable system serving communities in any of these nine counties must carry WKBD to its subscribers as a local signal. CDC concluded that these nine counties were within WKBD's local service area. 2. Sienificantlv Viewed Analvsis

Besides the DMA criterion, stations are considered local in counties and/or communities in which the Federal Communications Commission ("FCC") has deemed the station is "significantly viewed" (SV) meaning the station reaches certain FCC-defined viewing thresholds

'he definition of "local service area" in Section 111(f) and 47 C.F.R. $ 76.55(e)(2) defines a station's market as its Nielsen DMA. APPENDIX A within the county or community. Because a cable system serving County X must carry stations that are significantly viewed in County X (or Community X), such carriage is considered local. Attachment 2 lists selected counties in Michigan and Ohio and the TV stations significantly viewed in each as reported on the FCC's website: httos://transition.fcc.eov/mb/sienificantviewedstations041916.pdf. In the case of WKBD, four counties outside the DMA are considered SV: Genesee, MI Lucas, OH Lenawee, MI Wood, OH Ifa cable system serves communities located in these SV counties, that system must carry WKBD as a local signal and consequently, those SV counties were considered local to WKBD. B. Other Criteria For Determining Whether a Station Is Local

Besides the DMA and SV criteria, which identify the vast majority of local counties, CDC also examined other criteria to see ifthere are any additional counties that would be considered local. These criteria include the station's 35-Mile Specified Zone and Grade 8 Contour. 1. 35-Mile Specified Zone

For all television markets, major and smaller, a cable system's carriage ofa TV station to subscribers located inside the station's 35 mile specified zone means the station is local to those subscribers.

A copy ofthe 35-Mile Specified Zone for Detroit is shown as Attachment 3. FCC rules require a cable system serving communities located within that specified zone to carry WKBD as a local signal. Review ofthe zone indicated that the counties within the specified zone had already been classified as local due to the SV/DMA criteria. 2. The Grade B Contour

Another criterion, in some situations, is the Grade B contour. The contour is an irregular and oddly-shaped circle surrounding the TV station's transmitter site. The Grade B is a measure of estimated signal strength based on the station's antenna size, power, and direction. The Grade B, in other words, is a measure ofhow clear a picture can be expected to be on a person' television set.

The Grade B contour can be used as a criterion in two circumstances relevant to the local county analysis CDC performed for 2010-2013. The first circumstance has to do with stations licensed to smaller markets. Ifa cable system serves communities located in a smaller market, (i.e., located within the 35-mile zone of such a market), the system can carry as local any station f'rom another smaller market whose Grade B encompasses the communities served by the system. The second circumstance relates to systems serving communities located outside all television markets. In the case of a system serving such an area, the system may carry as local all stations whose Grade Bs encompass the communities served by the cable system. In both APPENDIX A circumstances, if the station is local per the Grade 8 criterion, the Form 3 cable operator does not have to account for the station in its royalty calculation. Attachment 3 further shows the Grade 8 contour of WKBD that CDC downloaded from the FCC's website shaded in yellow: htt s://www.fcc. ov/media/television/tv-service-contour- ~datn- oints. Cable systems serving communities that fall "outside all (television) markets" must carry as a local signal any station whose Grade 8 encompasses the communities ofthe system. To identify any such counties, CDC looked to see ifthere were any additional counties within WKBD's Grade 8 contour that were outside all markets and not already classified local by a previous standard. That review indicated that the counties within the Grade 8 contour had either already been classified as local due to the SV/DMA criteria or were not outside all markets. Once CDC identified all the local counties for WKBD, I provided those counties to Program Suppliers'ounsel. WKBD's local counties were the nine counties in the Detroit DMA, plus the four SV counties, for a total of 13 counties in which WKBD was a local signal. C. County Analysis For "Partially-Local" Stations

In the course of CDC's analyses, we may find that an entire county is neither wholly distant nor wholly local. An example would be a county that is neither SV nor DMA for a station, but which falls partially within the 35-mile zone ofthe station's market. Another example might be a county located outside all television markets and partially covered by a station's Grade B. In these few cases, CDC relies on the location of a majority ofthe county's population to designate the county as local or distant. Because the entire county must be classified as either local or distant for purposes ofthe Nielsen Studies, it is reasonable to assume that viewing will track with population. For example, are more people (i.e., viewers) located inside the 35-mile zone (or Grade 8) or outside? If, in our example, most ofthe population is within the station's Grade 8 contour, we consider the county local. CDC relies on mapping, distance calculations and census data to measure coverage of a county in relation to the Grade 8 (or 35-mile zone). II. LOCAL COUNTY ANALYSIS OF NONCOMMERCIAL STATIONS

A. Local County Analysis Illustration For Public Television ("PTV") Station KCTS, Channel 9, Licensed To Seattle, Washington

Described below are the steps CDC employed for the county analysis ofnoncommercial stations. The steps consist principally of conducting a 50-mile zone analysis and/or the Grade 8 Contour analysis. I selected for the purpose of illustration, PTV station KCTS, channel 9 licensed to Seattle, WA. The analysis applied to KCTS is repeated for all of the noncommercial stations in each sample for which CDC conducted a county analysis.

As set forth above, a station's Grade 8 contour is an irregular and oddly-shaped circle surrounding the TV station's transmitter site, and is a measure of estimated signal strength based on the station's antenna size, power, and direction.

To identify the Grade 8 contour for KCTS, we downloaded the KCTS Grade 8 map &om the FCC website: APPENDIX A

htt s://www.fcc. ov/media/television/tv-service-contour-data- pints

Next, CDC mapped a 50-mile radius around the reference coordinates (76.53) for Seattle, the city to which KCTS is licensed. As Attachment 4 shows, KCTS's Grade 8 is larger than its 50-mile radius, so the Grade 8 was the sole determining factor for identifying the local counties. However, had the 50-mile radius been larger, any additional counties partially or totally within the radius would have been marked as local.

Three Washington counties are entirely within KCTS's Grade 8:

Island Kitsap

King

Through the process of mapping these contours in relation to the census block population data from the Census Bureau website, we identified five more Washington counties had the majority of their population within KCTS's Grade 8:

Jefferson Pierce Thurston

Mason Snohomish

We classified these five counties as local for purposes ofthe Nielsen analysis.

III. LOCAL COUNTY ANALYSIS FOR CANADIAN STATIONS

A. Local County Analysis Illustration For Canadian Station CFTO, Channel 9, Licensed To Toronto, Canada

As set forth in my testimony, we employed multiple criteria to identify local counties for the Canadian stations in the sample. Where a Canadian station is SV in U.S. counties, all such counties were designated as local to the station. Where a Canadian station's Grade 8 included counties located outside all television markets, those counties were designated as local.

Described below are the steps CDC employed for the county analysis of Canadian station, CFTO, channel 9, licensed to Toronto, Canada. The analysis applied to CFTO is repeated for all of the Canadian stations in each sample for which CDC conducted a county analysis.

CFTO is significantly viewed (SV) in Niagara County, NY, so I designated that county as local for CFTO. Because CFTO is SV in some communities located in Erie County, NY, I analyzed the county population as descri.bed above for KCTS to determine that CFTO was SV to only 39.1% of the Erie County population, so I did not designate Erie County, NY as a local county for station CFTO. MARTI% APPENDIX A

ATTACHMENT 1 U.S. TV HOUBEHOLD ESTIMATES BY COUNTY)uf)f)lTHlN DESlGNATED MARKET AREA (DMA)

CESCN)!TED A9!IB!EC PPuFA %TV %CF CEBCNATED III)NEET ARE'A %TV %0= SIATE CCUNIY TC)CAL TV PEPE US TV S'ATE CCLMY TCHÃL 7V PH& TV " US CGPIIY 8ZE %USE)ClCS HLKK1ls TPAT)CN ICIEE5~ CCU4)Y SZE " 1CU&MS PCIEEHlCE) TRAT)CN 81EEHXIS

6'79 DBS MOIIIES-MES )CC!Pi'9) 891 KGB!JE 247,790 241,739 98 .219

ZCWA OREGON MAHASKA D 9, CQ". 8, 916 Br'!TQN 33,8QO 32, 686 MARION D 12„406 12,330 CO63 27,360 26, 610 MARSFALL D 15, 50". 15,430 LOSGLAS 42,806 41, 890 MQ'IPCE D 3, 10" 3,066 LAtr 143,800 140!550 P6OIL!)ONTAS D 3,200 3,19C PQLF, M B 174,360 173,5CC PQWESHIEK D 7,40i0 'l,33C 63,909 61,999 95 .053 P.Iti D 10." CGC 33016 2, 2, CALIFCRtlXA STORY M C 32,603 32,360 DEL Fi RTP, 10,000 59u0 TAYLOR D 2,500 2,470 9, ECMBQLIPi 53,9u0 51,50C PIXON D 5,403 5,36" WARPEt) E 17,600 16,976 WAY!IE D 2,663 2,550 EFANh IIL' 293,699 291,83Cu99 .254 h'STER D 15, 103 15,630 WGIGAi D 5,36: 5,25. ILLINOIS EDWAPSG 2,706 2,693 WABASH 4,930 4, 87" 595 DETROIT 1.646 1,999,900 1,89C,220 99 WA)3!E 7,003 6,900 MXCBIGA!i WBITE 6,360 6,260 L&iDZA%i LAPSER M A 32,300 32,16C DGROZS LIVINGS 6:I M A 67,56 67,3,90 16,103 16,030 GIBSON 13, 100 13,030 FACCH M A 336,860 336,096 PERRY 500 MQHRCZ A 58,16", 57,93C 7, 7,470 PZKZ S, 000 QAKLMiJ M A 483,6CC 481,796 4,950 POSER 900 GANZLAC D 16,560 9, 9,856 16,3CP SPEtiiER 606 GT CLALR M A 65i066 64,'73i 7, 7,546 t!T!h7!ERRTRGB WAGhuLXIAW A 136,4CC 134,910 72,806 72,530 WARRICK 22,500 WAYttG M A 'l63„8uu 22,396 699,12i KP!ITCCKY DANZESS 39,260 39,650 696 192!60". 191,849 99 .989 HANCOCK 3,506 3,460 EENDERGCN 19,CCO 18,923 Ai!PAPJf POPKL!iS 19,466 19,240 CQFPEE M D 20,36'. 2i", 130 MCLEAN 4,000 3,963 DALE M C 19,36." 19,140 MuuLIL!IS ERG 12,56i" 12,270 GBii!AA N D 10,903 16,863 OHXO 9,666 9,526 BSNPY M D 6, GC." 6,850 CutJION 5,600 5,583 Eu'JSTCN M C 46,600 40, 36'0 WRPSTER 5, 400 5,320 GEORGIA EARLY 4,666 4,560 PAIRPJQIKS 37,999 36,25i9 96 ,932 DIZBTE-SCPERIQR 176,39'. 174,369 99 .152 ALASKA FAIRS!IKS-PLQS 37,90C 36,250 MICHIGAN GOGEGXC 6,606 6,460 MIÃtilE SOYA 724 FARPJ-VALLEY CITY 242,399 240,339 99 .299 CARLTO'I M D 13,403 13,330 i!Q!CIEGOTA COOK D 2,600 2,490 BECKER D ITASCA D 18, 76.0 18,560 13,100 12 ~ 92 0 CLAY KQOCBZCHZ!IG D 5,803 5,716 C 21,400 21,343 CLHEWATZR LA!!E D 4, 660 5'l0 D 3,460 3,303 4, KZTiSC'.I GT LQQZS N C 82,460 81,680 D 1,800 1,793 LAKE WX -CCNIZtt QP WOO 8 D li700 1,666 A'BUQ "J 6 6,70 6, 5'76 !OfrNQNEN D 2,000 li 990 MARSHALL GARFIELD D 6,636 6,456 3,966 3, 88i! t/Qut!JL!I D ZQCulAS M C 1S,506 18,416 2,700 2,680 TAZL XP"'I D 3,03u 2,996 OTTBR D 23,0i0 22,786 PELTIXNGTG!I GAW'CER D 7,40u 7,200 D 5,706 5,666 POLK C 12, ZC0 12,026 RED LL!IE D 1,700 1,650 765 EL PASO ILAS CRECEB) 313,')Q0 319, ')60 99 . 271 ROGGAQ D 6,206 6, 123 WZLKIN D 2,466 2,393 '~d tuEtfICO NvRTH DAFOTP. DQt!A ANA 71,96C 76,550 BAliiiEG D 4,506 4,48J T!~G Ru!iSQ!I D 2,306 2,2S3 CLIDERGC.'I D 90C 89'0 CAGS C 62, 666 62,660 EL PAGO M 8 239,800 238,26Q Cia)LISP. D 1,6CO 1,59i EDSPrÃut D 1,10u 1,660 DXCKEY D 2,600 1,990 EDDY D 1,C60 996 FOSTER D 565 EL!IIRA (CQPJIINS) 97,909 95,799 99 ."0!83 1,400 1,39u GELID FORKS C 25,906 25,750 ttad YORK GRIGGS D 1,060 1,030 CB~iG 34,763 34,520 LA EQURE D 1, 706 1,690 'LER 7,403 7,306 NELSON D 1,406 1,390 GT:"QSEtt 38, 8" 3 38,240 PPMBXNA D 3i 102 3,030 Pi INuYL)7fNXA RIL!BEY D 4, 700 4,683 iuIOuh 16, 10.0 15,730 PhÃQN D 2! 300 2,280 RZCSLL'ID D 6,400 6,360 GARGENT D 1,700 1, 680 ERIE 157,700 156,520 99 .136 STRRLB D 700 696 GTOTG)!JAN D 8,400 8,340 PEtiEi G)7 FANIA TCh&'R 900 890 CRAWPOPJ) C 34,200 33,690 TPJfILL 3,160 3,086 ERIE N C 106,863 106,330 WALSE D 4,560 4,4SO BARRL! D 16,703 16,500 513 PLltii-SAGIHAN-BAY ClullP 469,200 458,920 109 .599 MZCBXGMl AuPtEIAC 6,460 6,356 BAY d4i2u6 44,C40 GGNGurEE 168,360 167!606

M METRO COD!JiY OF DMJ) MARKET 34 & SHE PAGE A FOR CQLIITY SIZE DZFIIIITIQ'IB NM ttETRO CQD!ITY GP t!Oli-DMA MARKET +u SEE PAGE A FQP. DEL CODE At)D tiAME DEFIIIITIQN Sienificantlv-Viewed Counties—-WKBD. Detroit

Michigan

Genesee WNEM-TV, 5, Bay City, MI WJRT-TV, 12, Flint, MI +WSMH, 66, Flint, MI ¹WJBK, 2, Detroit, MI ¹WDlV, 4, Detroit, MI (formerly WWJ) ¹WXYZ-TV, 7, Detroit, MI ¹WKBD-TV, 50, Detroit, MI WLNS-TV, 6, Lansing, MI (formerly WJIM)

Lenawee WJBK, 2, Detroit, MI WDIV, 4, Detroit, MI (formerly WWJ) WXYZ-TV, 7, Detroit, MI CBET, 9, Canada (formerly CKLW) WKBD-TV, 50, Detroit, MI WTOL-TV, 11, Toledo, OH WTVG, 13, Toledo, OH (formerly WSPD) WNWO-TV, 24, Toledo, OH (formerly WDHO) +WUPW, 36, Toledo, OH

Ohio

Lucas WTOL-TV, 11, Toledo, OH WTVG, 13, Toledo, OH (formerly WSPD) WNWO-TV, 24, Toledo, OH (formerly WDHO) +WUPW, 36, Toledo, OH ¹WJBK, 2, Detroit, MI ¹WXYZ-TV, 7, Detroi, MI +WKBD-TV, 50, Detroit, MI

Wood WTOL-TV, 11, Toledo, OH WTVG, 13, Toledo, OH (formerly WSPD) WNWO-TV, 24, Toledo, OH (formerly WDHO) +WUPW, 36, Toledo, OH WKBD-TV, 50, Detroit, MI MARTIN APPENDIX A ATTACHMENT 3 g/

7j Mecosta Isabella Iwldland /-'ay Nevpaygo Tuscola Sanilac 88glnwv kegon — -- — ------5honfcalrn Graflot ~peer---' OftaNa Cenesee Digital Service ionia Shiawassee .. Contour jGrade Bj

St. Clail', / I,'nk

AIIegan Barry Eaton ingham ', Livingston

WKBD-TV

/ Kalalnazoo Calhoun VVashtenavr Wayne

/ 35-mile radius

Case . St. Joseph, Hlllsdalo Lenallvee IHonl'oo

p Ashtahula L86range Luc88 coulton l Q Williams ! I j ~v-" Ge8uga i Noble DeKalb aandusky II I Defl8nce Henry Trulnbull

Kosciusko I MARTIN APPENDIX A ATTACHMENT 4 KCTS-DT Ch.9 PBS Seattle, WA Cable Data Corp

Nhatcom

Okanogan Digital Service Contour (sr deB) //' island'nohomisk

Chelen Jeffer on

I' Kitsap Douglas KCTS-TV t

I King 50-mila radius ',Masan Grant

Grays Harbor Kittitas

Pierce il T'hiirston

Lewis Yakima Pacific LgN CERTIFICATE OF SERVICE

I hereby certify that on this 22nd day of December, 2016, a copy of the foregoing pleading was sent by Federal Express overnight mail to the parties listed on the attached service list,

Lucy Holmes Plovnick SERVICE LIST

JOINT SPORTS CLAIMANTS SETTLING DEVOTIONAL CLAIMANTS

Robert Alan Garrett Clifford M. Harrington Sean Laane Matthew J. MacLean Michael Kientzle Victoria N. Lynch-Draper Bryan L. Adkins PILLSBURY WINTHROP ARNOLD k PORTER LLP SHAW PITTMAN LLP 601 Massachusetts Avenue, N.W. 1200 Seventeenth Street NW Washington, D.C. 20001 Washington, DC 20036

SETTING DEVOTIONAL CLAIMANTS/PROFESSIONAL BULL RIDERS

Arnold P. Lutzker Benjamin Sternberg LUTZKER k, LUTZKER LLP 1233 20th Street, NW Suite 703 Washington, DC 20036

PUBLIC TELEVISION COMMERCIAL TELEVISION CLAIMANTS CLAIMANTS NATIONAL ASSOCIATION OF BROADCASTERS

Ronald G. Dove, Jr. John I. Stewart, Jr. Lindsey Tonsager Ann Mace Dustin Cho David Ervin COVINGTON k, BURLING LLP CROWELL k MORING LLP One City Center 1001 Pennsylvania Ave., NW 850 10th Street NW Washington, DC 20004-2595 Washington, DC 20001 MUSIC CLAIMANTS

AMERICAN SOCIETY OF BROADCAST MUSIC, INC. COMPOSERS, AUTHORS AND PUBLISHERS

Samuel Mosenkis Joseph J. DiMona Jackson Wagener BROADCAST MUSIC, INC. ASCAP 7 World Trade Center One Lincoln Plaza 250 Greenwich Street New York, NY 10023 New York, NY 10007-0030

Michael J. Remington Brian Coleman Jennifer T. Criss DRINKER BIDDLE 0 REATH LLP 1500 K Street, NW — Suite 1100 Washington, DC 20005

SESAC, INC.

John C. Beiter SHACKELFORD, BOWEN, MCKINLEY % NORTON, LLP 47 Music Square East Nashville, TN 37203 CANADIAN CLAIMANTS

L. Kendall Satterfield Victor Cosentino SATTERFIELD PLLC LARSON 2 GASTON LLP 1629 K Street NW, Suite 3000 200 S. Los Robles Avenue, Suite 530 Washington, DC 20006 Pasadena, CA 91101

MULTIGROUP CLAIMANTS NATIONAL PUBLIC RADIO SPANISH LANGUAGE PRODUCERS

Brian D. Boydston Jonathan D. Hart PICK 8r BOYDSTON LLP Gregory A. Lewis 10786 Le Conte Avenue NATIONAL PUBLIC RADIO Los Angeles, CA 90024 1111 North Capitol Street, NE Washington, DC 20002

MAJOR LEAGUE SOCCER

Edward S. Hammerman HAMMERMAN PLLC 5335 Wisconsin Ave., Suite 440 Washington, DC 20015-2054