# Appendix Vol III — Media General, Inc. v. Federal Communications Commission

> Briefs, arguments, decisions, and more.

URL: https://www.frixlaw.com/law-library/documents/brief%3Amicro_IA40386007_1337%3A4

## Record

- **Collection:** Supreme Court brief
- **Document type:** Appendix Vol III
- **Published:** January 1, 2005
- **Citation:** 545 U.S. 1123

## Text

04102 U JAN 2 8 2005
| ” |

IN THE

Supreme Court of the United States

No

MEDIA GENERAL, INC.

Petitioner,
Vv.

FEDERAL COMMUNICATIONS COMMISSION AND
UNITED STATES OF AMERICA,

Respondents.

On Petition for Writ of Certiorari to the United States
‘Court of Appeals for the Third Circuit

APPENDIX VOLUME III

GEORGE L. MAHONEY CHRISTOPHER LANDAU
MEDIA GENERAL, INC. Counsel of Record
333 E. Franklin St. SUSAN ENGEL
Richmond, VA 23219 MICHAEL SHUMSKY

| KIRKLAND & ELLIS LLP
JOHN R. FEORE, JR. 655 Fifteenth St., N.W.
MICHAEL D. Hays Washington, DC 20005
M. ANNE SWANSON (202) 879-5000

Dow, LOHNES &

ALBERTSON, PLLC
1200 New Hampshire Ave., N.W.
Washington, DC 20036

January 2005

WILSON-EPES PRINTING Co., INC. — (202) 789-0096 -— WASHINGTON, D. C. 20001

CONTENTS

Document Page

VOLUME I

United States Court of Appeals for the Third Circuit

Majority Opinion (June 24, 2004) ........ceseesesseeneeteeeeees la

DISSORTING OI ..scccsssceccsossccnscssesocsccsansassecessoncennases 107a

Order Denying Petition for Rehearing

QO, FTO ssccsicenstnuittsnsiinssintiisasvinintninneiasinsaiinsnnndaaneanes 19la

Order Granting Stay (Sept. 3, 2003)...........ssscsssseesneeees 194a

Order Denying Transfer (Sept. 15, 2003).........:eseeeee 197a
VOLUME II

Federal Communications Commission

Report and Order .............scccssocssssrccsenssssessessssnsssosseaneess 206a
VOLUME III

Report and Order (CONUINUEA).........:cseeeseeesereeneeeseeeeeees 486a

Notice of Proposed Rulemaking..........ccccseeseereeseeeees 652a

Statement of Chairman Powell..............scsssccsssseeeseeeees 663a

Statement of Commissioner Abernathy ...........:sesseeeeees 666a

Statement of Commissioner Martin ..........:ccceseceeeseeeeees 673a

Dissenting Statement of Commissioner Copps anna 677a

Dissenting Statement of Commissioner Adelstein........ 714a

486a

b. Choice of Media

401. We have determined which media to include in the
Diversity Index based on the survey information derived from
the “Consumer Survey on Media Usage” prepared by Nielsen
Media Research (FCC MOWG Study No. 8). This survey tells
us how consumers perceive the various media as sources of
news and information. The key threshold implication of this
study is that consumers use multiple media as sources of news
and current affairs, and hence that different media can be
substitutes in providing viewpoint diversity. For example,
consider a citizen who acquires local news from television,
newspapers, and radio. Suppose that a group of citizens in the
consumer’s home town wishes to oppose a bond issue for a new
sports stadium, and that the local newspaper and television
stations favor the bond issue and choose not to cover the
position of opponents. If the opponents nevertheless get radio
coverage for their position, they would be able to reach this
particular citizen. Indeed, one might think that part of the radio
coverage might address the fact that other media are “ignoring”
the story. This could then raise the profile of the story to a
level that might attract newspaper or television coverage. We
put forward this hypothetical sequence of events not because
we think that it describes a process that will happen with
respect to any particular controversy. Rather it is a useful
illustration of the process by which markets with multiple
independent media outlet owners operate, particularly in an
environment in which citizens generally do not depend on a
single medium for their local news and current affairs.

402. FCC MOWG Study No. 8 asked respondents to identify
the sources, if any, “used in the past 7 days for local news and
current affairs.” The same question was posed for national
news and current affairs. The choices offered were television,
newspaper, radio, Internet, magazines, friends/family, other,
none, don’t know, and refuse. The survey then asked follow-up
questions regarding the first five choices. For each one of the
five sources, respondents who did not mention a source were

487a

asked specifically if they used that source for local news and
current affairs. The survey posed analogous questions with
regard to national news and current affairs. Based on the initial
and follow-up questions, the survey presents “summary data”
on sources of local and of national news and current affairs
information.

403.In an ex parte communication filed May 28, 2003,
Media General submitted a critique of MOWG Study No. 8 by
Prof. Jerry A. Hausman. Hausman argues that the Nielsen
Survey has a number of serious flaws and questions its
usefulness in any rule-making concerning cross-ownership of
newspapers and broadcast stations.””’ First, he claims that the
low response rate of the survey may lead to biased results.”
Second, he argues that survey questions about hypothetical
‘future circumstances are unreliable. He cites anumber of cases
where respondents, presented with relatively simple questions,
are unlikely to consider the full, complete implications if a
particular form of media were to become unavailable.” Third,
he argues that the survey asks no questions that address the
newspaper/broadcast cross-ownership issues.”

404. We recognize Professor Hausman’s concerns, but we
believe that the Nielsen survey sample of 3,136 households
provides us with useful information. In addition, Professor

903 ; ee :
Statement by Jerry Hausman, Media General Notification of Ex Parte

Communication (May 28, 2003)

YO4
See Hausman at 3.

905 . ;
For example, he cites where the Nielsen survey asks whether

respondents would be more likely to use cable or satellite news channels for
news if broadcast TV channels were not available. He notes that the
question does not ask the respondent to consider a number of important
factors, such as whether VHF and UHF news programs would cease to exist
or whether only cable and satellite news programs would remain. Since it is
not clear what hypothetical world the respondents are assuming. Hausman
argues, the results of the survey are not reliable. Hausman, at 4-5.

906
Hausman at 6.

488a

Hausman provides no evidence that the sample is, in fact,
biased. Concerning Hausman’s second point, we agree that
answers to hypothetical questions are less useful than
information about actual behavior. MOWG Study No. 8
provides a substantial amount of information on reported actual
behavior. It is this information, not the hypotheticals, on which
we rely to conclude that media can be substitutes in providing
viewpoint diversity and to construct our Diversity Index.
Regarding Hausman’s third point, although the Nielsen survey
may not directly ask respondents for their views concerning
specific cross-ownership scenarios, we find that the results of
the survey are useful in a number of areas, such as which forms
of media are most heavily used for news. While questions could
have been posed that contained more specificity concerning
cross-ownership rules, we understand that such complexities -
could have made the survey design more difficult, as well as
possibly lowered the response rate. Overall, while Hausman
claims that the Nielsen survey does not “provide a basis for the
measurements necessary for the specification of policy,””” the
survey does, in fact, help us establish an “exchange rate”’ for
converting newspaper, television, radio, and other media into
common units so we can measure the extent of concentration in
the “market of ideas.” Finally, we emphasize that the
Commission has not relied solely on the results of the Nielsen
survey, but has used a number of studies and its own expert
judgment on media in reaching its decision.

405. The data in the Nielsen study indicate that television,
newspapers, radio, Internet, and magazines are the leading
sources of news and current affairs programming.” Indeed, the
summary data tables list only those five sources. In the initial
questions, less than one percent of respondents cite “other” as a
source. Based on the initial question, the average respondent
uses two of the five major sources for news and current affairs,

907
Hausman at 3.

°* MOWG Study No. 8 at Tables 97 and 98.

489a

whether the category is local or national. Taking account of the
follow-up questions, the average respondent uses three of the
five major sources for news and current affairs, again regardless
of whether the category is national or local.” These data
strongly suggest that citizens do use multiple media as sources
of viewpoint diversity, and that media can be viable substitutes
for one another for the dissemination of news, information and
viewpoint expression. On the basis of this finding, we proceed
to an analysis of local media markets and whether there are
particular kinds of cross-media transactions in particular kinds
of markets that would likely result in high levels of
concentration. To assist in making that determination, we rely
in part on our Diversity Index.

406. Our Diversity Index focuses on availability of sources
of local news and current affairs. As we explained in our
policy goals section above, we are concerned with promoting
viewpoint diversity in local media markets. Owners of media
outlets clearly have the ability to affect public discourse.
Consumers have numerous sources of national news and
information available to them. Three major commercial
broadcast networks, ABC, NBC, and CBS, provide this
material and are available to 98% of US television
households.’'” Several nonbroadcast networks also provide
national news and current affairs information also are widely
available. Subscribership to CNN is 77.4% of US television
households.’'' The comparable figures for Headline News, Fox
News, MSNBC, and CNBC are 74.1%, 71.8%, 68.4%, and

909 - —
The average respondent uses 2.93 different media for local and 2.71

different media tor national news and current affairs.

910 :

See OPP Working Paper 37 at 48. In January 2002, there were 105.5

million television households in the U.S. See Television Advertising

Bureau, Trends in Television at www.tvb.org.

oT See Kagan World Media, Cable Television Investor (July 29, 2002) at

14.

490a
74.9%, respectively.’'> Local newspapers generally provide
information on national issues, and a variety of major
newspapers have national footprints. They include USA
Today, the Wall Street Journal, the New York Times, and
otfiers. Moreover, a wide range of newspapers are available
on-line at no charge. National news magazines, such as Time,
Newsweek, US News and World Report, and more specialized
political journals, such as Weekly Standard and New Republic
are also widely available.”'’ Therefore we do not believe that
governmental regulation is needed to preserve access to
multiple sources of national news and public affairs
information.

407.The Diversity Index incorporates information on
respondents’ usage of television, newspapers, radio, and the
Internet. Respondents also reported getting local news and
information from magazines.’'* We exclude magazines,
however, from our Diversity Index. First, as the description
above makes clear, most (but not all) news magazines have a
national rather than a local focus. Although there are
exceptions (e.g., the Washingtonian and Texas Monthly), the
figures in MOWG Study No. 8 on magazines use appear to be
overstated. This simplification and assumption is supported by
other aspects of the study. For example, unlike newspapers,
radio, and television, almost no one cited magazines as their
primary source of news and current affairs. MOWG Study No.
8 includes a question asking respondents to identify their single
primary source of local or national news and current affairs.

9! Td. The total television households figure (105.5 million) is for January
2002 and is from Television Advertising Bureau, Trends in Television at
www.tvb.org.

913 ;, : ea a ’
See Appendix B for a summary list of major national news sources. We

note that some of the sources for national news and information are owned

by the same companies but we continue to believe that consumers have

numerous independently owned sources of national news and information.
4

14 .. , .
Six percent of respondents answered that they received their local news
and information from magazines.

49 la

The figure for magazines is 0.6%.”'> A 2000 study by the Pew
Research Center for the People and the Press provides similar
results.”’° The study examines “Trends in Regular News
Consumption” and finds that 12% of respondents, but only
4.2% of responses cite news magazines. An even lower share,
5% of respondents and 1.7% of responses, cite business
magazines. Moreover, these figures include consumption of
national as well as local news. The share of the sample
utilizing magazines for local news is smaller, perhaps
considerably so. The Pew Center data support our inference
that magazines play a negligible role overall as a source of local
news. We must also note that, although the actual local news
figure is small, because both the MOWG Study No. 8 and Pew
Center figures combine local and national news, the precise
magnitude of the local news figure is uncertain. Hence we are
unable to assign to magazines a weight (even a small weight) in
which we would have confidence. Nonetheless, the decision to
exclude magazines will be re-examined in the next biennial
review, and we will take the opportunity to gather additional
survey data at that time on magazine usage.

408. For similar reasons, we also exclude cable from our
Diversity Index. As discussed in the following section, we are
concerned that some consumers may have confused broadcast
and cable television. Thus, we believe some consumers who
replied that they receive their local news from cable may have
been viewing broadcast channels over the cable platform. We
also recognize, however, that cable systems do provide local
news and current affairs information through PEG channels
and, in some markets, local news channels. However, we do
not have accurate data for this measure. Because we do not

9!9 MOWG Study No. 8, Table 20.

7! Pew Research Center for the People and the Press, /nternet Sapping
Broadcast News Audience, http://people-
press.org/repo:ts/display.php3?PageI D=203.

rr

492a

have reliable data on this point, we exclude cable from the DI
to simplify our general analysis.?"’
c. Weighting Different Media

409. We have concluded that various media are substitutes in
providing viewpoint diversity, but we have no reason to believe
that all media are of equal importance. Indeed the responses to
the survey make it clear that some media are more important
than others, suggesting a need to assign relative weights to the
various media. In view of our focus on local news and current
affairs, we choose to base our weights on survey responses to
the question asking respondents to identify the sources, if any,
“used in the past 7 days for local news and current affairs.”?"*
We recognize that this is not a perfect measure, and that it
requires some adjustment. We justify these adjustments and
assumptions, however, by emphasizing that we are using the DI
only to inform us of general market trends, not for precise
measurements.

410. As noted above, the average respondent uses three
different media for local news and current affairs information.
It is likely that, for a given respondent, the three are not all of
equal importance. If media differ in importance systematically
across respondents (e.g., if television were most important to
everyone, and everyone made only minor use of radio to
acquire news and current affairs information), then it would be
misleading to weight all responses equally.

411. Unfortunately, we do not have data on this question
specifically with regard to local news and current affairs. The
available “primary source” data address local and national news
together and do show that different media have different
importance, in the sense that primary usage differs across

917 : es ae
As with magazines, we will review this issue in the next biennial review,

and may collect at that time more accurate survey data on consumers’ use of

cable for local news and current affairs.

918

MOWG Study No. 8, Table 97.

493a

media.”'” Because “primary source” data are not available for
local news and current affairs alone, we use the data identifying
sources of local news and public affairs programming to weight
the various media to reflect relative usage. As noted below,
this leads to lower shares for television and higher shares for
radio than the “primary source” shares reflect.

412. The local response summary data, Table 97 of MOWG
Study No. 8, include five categories of media—Internet,
magazines, radio, newspaper, television. Magazines account
for 6.8% of responses to the questions on source of local news
and current affairs. We exclude magazines as explained above
and normalize the shares of the four remaining media to sum to
100%. The resulting weights are television (33.8%),
newspapers (28.8%), radio (24.9%), and Internet (12.5%).?”"
The local response summary data do not break down the
television responses between broadcast television and
cable/satellite television. Nor do these data separate out usage
of daily and weekly newspapers. We make use of other FCC
MOWG Siudy No. 8 questions to apportion the newspaper
shares further.

413. Although the responses to one question in MOWG
Study No. 8 suggests that cable is a significant source of local
news and current affairs, other data from che study casts some
doubt on this result. The following discussion explains the
reasoning that leads us to exclude cable/satellite television from
the current analysis of local news and current affairs for
diversity purposes. As a threshold matter, DBS currently
provides little or no local nonbroadcast content. We do,
however, recognize that cable provides some such content and
that it is becoming a more important source of local news and
information. Some markets do have commercial local news

9! 1d., Table 20.

| Sa ae ‘ ; . ; os
The “primary use” weights, excluding magazines, are television (57.8%),

newspapers (25.8%), radio (10.3%), and Internet (6.1%). When magazines
are included their weight is 0.6%. /d.

494a

channels on cable.””' Moreover, at least one national cable
news service (CNN Headline News) provides a five-minute
local “cut-in” every half hour in some markets. Additionally,
local public, educational, and governmental (“PEG’’) channels
provide some local news and information, although the extent
of their impact is enclear. MOWG Study No. 8 asked
respondents who get local news and current affairs from
television (table 8) to indicate if the source is “broadcast
television channels,” cable or satellite news channels,” “some
other channel,” “don’t know,” or “refuse.” Virtually all
responses fell into the first two categories, with 46.4% of
respondents who get local news from television identifying
cable as their source.’””

414.Our experience suggests that the local cable news
response is too high. A review of the responses reported in
tables 6, 16, and 18 of MOWG Study No. 8 support this
assumption. Table 18 provides responses from all who get
news (local or national) from cable to the question “what are
the names of the news channels you watched in the past 7 days
on cable or satellite for local or national news or current
affairs?” The list from which respondents can choose includes
CNN, Fox News Channel, MSNBC, Local Cable News
Channels, Headline News, CNBC, Other, Don’t Know, and
Refuse. The last two choices get minimal response, but 27.5%
of responses are “Other.” This suggests that some people may
be counting retransmitted broadcast signals on cable or satellite
as cable or satellite channels.”**> Moreover, joint examination

= Roughly one-third of cable subscribers, 22.3 million, had access to a
local or regional news channel in J uly 2002. See OPP Working Paper 37 at

126.

- The corresponding figure for national news (from table 16) is 51.1

percent.

923

Because all cable systems carry local broadcast stations pursuant to our
signal carriage rules, and because DBS carriers provide local broadcast
signals in many markets, also pursuant to our signal carriage rules, it is

495a

of the responses reported in tables 6, 16, and 18 make it
possible to infer that 94.3% of those who get news from cable
(the table 18 universe) claim to get at least some local news
from cable. However, only 6.1% of responses mention local
cable news channels.’ This disparity makes us question the
responses regarding local news via cable and satellite channels
and supports our conclusion that weighting cable 46.4% is too
high. An additional reason that leads us to question cable as a
local news source is that, of those local cable channels that
meet Nielsen’s minimum reporting standards, they are the least
watched of any broadcast or cable stations in the market.”
Given the low viewing of PEG channels and the facts that only
one-third of cable subscribers have access to a local cable news
channel and we do not have an accurate cable figure to use, we
believe excluding cable from the Diversity Index on a national
basis is a reasonable assumption. We will review the status of
cable as a local news provider in the 2004 biennial review. Our
review will include a follow-up to MOWG Study No. 8, which
will include more detailed questions regarding the use of
nonbroadcast video media for local news and current affairs.

415. With regard to newspapers, MOWG Study No. 8
indicates that 61.5% of those who cite newspapers as a source
of local news and current affairs acquire that information from

possible, even likely, that the “other” category actually reflects viewing of
retransmitted broadcast signals. If we assume that viewers are likely to be
familiar with local broadcast signals, it is not likely that the “don’t know”
category includes broadcast signal viewing.

%°4 Local cable news channels are, unlike the Internet, not available
everywhere, but only in select markets. Only approximately one-third of
cable subscribers have access to such channels. See OPP Working Paper
37 at 126. The ownership limits apply nationwide, and the diversity index
is intended to help us define these ownership limits. This was an
additional reason for excluding cable from the DI while counting the
Internet.

Y

-
= Nielsen Television Index (Nov. 2000).

496a

dailies only, 10.2% from local weeklies only, and 27.3% from
both.” This works out to a share of 70.3% daily and 29.7%
weeklies. We use these weights to divide the total newspaper
share (28.8%) among daily and weekly newspapers. Our next
biennial review will provide an opportunity for re-examination
of the role of weekly newspapers. Accounting for the
additiona! information on newspapers results in a revised set of
weights. They are: broadcast television 33.8%, daily
newspapers 20.2%, weekly newspapers 8.6%, radio 24.9%, and
Internet 12.5%.

416. Various commenters agree that MOWG Study No. 8
S.pports the conclusion that citizens do, in fact, see different
media as alternative sources of news. For example, NAA
opines that the study, “a comprehensive survey...shows that the
public makes ample use of a broad assortment of outlets” and
that it “demonstrates that the public relies heavily ona range of
alternative media.”””’ Fox opines that the study “demonstrates
that consumers are utilizing the wide variety of media available
to them to obtain both local and national news and
information.” Later, this commenter states that this study,
along with another Study discussed in the comments
“demonstrates that consumers are adept both at using various
sources to obtain information and at using multiple sources
simultaneously.”””* Critics of MOWG Study No. 8 include
AFL-CIO and AFTRA, both of whom rely on a paper by Baker,
attached to the AFL-CIO comments.””” The Baker submission
refers to the fact that MOWG Study No. 8 reports responses to
a number of hypothetical questions regarding how respondents
would behave if the availability to them of certain media were
to change. Baker observes that the study “looks at what people

*°° MOWG Study No. 8, Table 7.
= NAA Comments at 8.
aa Fox Comments at 11, 25.

_ See AFL-CIO Comments at 12 and Baker Study at 12-14 ; AFTRA
Comments at 8.

497a

say they will do” and goes on to assert that “[E]conomists
usually prefer looking at what people do.””*? We agree that
answers to hypothetical questions are less useful than
information about actual behavior. MOWG Study No. 8
provides a substantial amount of information on reported actual
behavior. It is this information, not the hypotheticals, on which
we rely to conclude that media can be substitutes in providing
viewpoint diversity and to construct our Diversity Index. It is
worth noting in this connection that much of the information we
have on radio listening and television viewing is also based on
reports by listeners and viewers of their behavior. Moreover,
the information in MOWG Study No. 8 on the range of media
that citizens use for news and information is quite similar to the
results of a recent independent survey by the Pew Research
Center.”

417.The most detailed analysis of MOWG Study No. 8
comes from the Consumer Federat.on of America.”” CFA
agrees that citizens get viewpoint diversity from multiple
media. Their comments refer to the “two dominant political
media—daily newspapers and television,” although CFA
asserts that these media “appear to play very different roles.”
As noted above, television has the largest weight in the Di
(33.8%) and daily newspapers also loom large at 20.2%.
Although the radio weight is somewhat higher at 24.9%, the
fact that markets generally have far more radio stations than
daily newspapers make our weights consistent with CFA’s
conclusion that newspapers are among the two most influential
media. CFA finds that the Internet plays a small but growing
role in citizen acquisition of news and information, a finding
not inconsistent with the relatively low weight of Internet in our

°) Baker Study at 14.

Pew Research Center for the People and the Press, Sources for
Campaign News, Fewer Turn to Broadcast TV and Papers, (Apr. 27, 2003)
- ht p://people-press.org/reports/display.php3?PageID=243.

931

~ See generally CFA Comments at 94-147.

498a

DI. CFA quotes statistics on daily use of television,
newspapers, radio, and Internet that yield usage shares not too
different from our DI weights. Drawing on two surveys, CFA
suggests that people spend 4 minutes per day on average
gathering news from the Internet, 25 minutes reading
newspapers, 15 minutes listening to radio news, and “over half
an hour” watching television news.’ Ascribing half an hour to
television leads to shares of 40.5% for television, 33.8% for
newspapers, 20.3% for radio, and 5.4% for Internet. These are
fairly close to our DI weights of 33.8%, 28.8%, 24.9%, and
12.5% for television, newspapers, radio, and Internet,
respectively.

418. Although CFA does not dispute the proposition that
different media address the same issues and stories, it asserts
that they do so in different ways, suggesting, inter alia, that
television is “the primary source for breaking news,” that
newspapers have a larger role in “the follow-up function,” and
that talk shows are a new and significant element of radio’s role
in disseminating viewpoints.’ Although CFA does not
discuss the role of radio as a source of breaking news, we
acknowledge that different media do present information in
different ways. CFA also argues that, particularly for “high use
respondents” (the one-third of respondents in MOWG Study
No. 8 whose total media use was above the sample average)
there is evidence that the media are complements rather than
substitutes, j.e., people who use more of one medium tend to
use more of the others. For the “low use respondents” (the one-
half of respondents whose total media use was below the
sample median), in many cases there are negative correlations
in usage across many pairs of media. CFA suggests that this is
consistent with substitution.”*> Thus, CFA appears to conclude

933 1. at 109.
934 1a at 112, 100.
935 1d. at 142-145.

499a

that media are substitutes for some citizens and complements
for others.

419.We disagree with CFA's conclusion that the DI is
invalid because some citizens may consider certain media
outlets complements rather than substitutes. In the technical
economic sense, two goods are substitutes if an increase in the
price of good A (which leads to a decrease in consumption of
good A) leads to an increase in the consumption of good B. In
the context of our diversity goal, we are concerned with the
question of what happens when one or more media outlets
refuses to transmit a particular viewpoint. If most citizens
accessed only one type of outlet, e.g., radio but not newspapers
or television, then our diversity goal would prompt us to
analyze separately the structure of the “radio marketplace of
ideas.” If, on the other hand, most citizens access multiple
media, then we can rely on the reasonable probability that, if,
e.g., the local newspaper refused to cover a particular story,
citizens would be exposed to that story via independently-
owned other media, such as radio or television. In other words,
evidence that media are complements in the sense that, for at
least some citizens, there is a positive correlation between use
of one medium and use of another, does not invalidate the
premise underlying the DI.

d. Weighting Outlets Within the Same Medium

420. Having decided on relative weights for the various
media, we next confront whether and how to weight different
media outlets within each category. The decision of whether to
do weighting turns on whether our focus is on the availability
of outlets as a measure of potential voices or whether it is on
usave (7.e., which outlets are currently being used by consumers
for news and information). We have chosen the availability
measure, which is implemented by counting the number of
independent outlets available for a particular medium and
assuming that all outlets within a medium have equai shares. In
the context of evaluating viewpoint diversity, this approach

500a

reflects a measure of the likelihood that some particular
viewpoint might be censored or foreclosed, i.e., blocked from
transmission to the public.

421. The underlying assumption here is that all qutlets have
at least similar technical coverage characteristics. This is a
good, but not perfectly accurate assumption. Our signal
carriage rules more or less equalize the covemge of all
television stations in a particular DMA,””° and it appears that
newspapers (even those with limited current circulation) can
expand their circulation area at relatively low cost. That is,
assuming that additional readers are interested in the content,
additional delivery personnel and vending machines are readily
available at low cost. However, the assumption is less certain
for radio. For example, a Class C FM station and a daytime
AM station, in fact, have different coverage characteristics.
The Class A station cannot expand its coverage to match that of
the Class C FM station and thus reach additional listeners who
might otherwise not have access to the views expressed on this
outlet. Nevertheless, we believe the assumption to be
reasonable across all cases. Arbitron radio metros are smaller
than many radio station service areas and so would have the
effect of truncating the service areas of more powerful stations.
In addition, even though radio’s total diversity share is 24.9%,
on average there are enough radio stations so that the per-
Station share is fairly small. Any distortion in share by
overestimating the reach of small radio stations is therefore
small.

422. Even though we choose to assign the same weight to
each outlet of a particular medium, we reiterate the importance
of assigning different weights to different media. As noted
above in ¥ 409 et seq., different media are of different

936 ae :
We make this assumption for the purposes of constructing the DI; the

actual differences in coverage are accounted for in the rules themselves, e.g.,
the UHF discount in the national rule, and our waiver policy in the local TV
and CML rules will look to the actual reach of stations.

SOla

importance. The differences in usage across media documented
in MOWG Study No. 8 are in part reflections of the differential
impact on the user of television, radio, newspapers, etc. We
believe that the overall impact of a medium is substantially .
determined by the physical attributes of its distribution
technology, along with user preferences. A radio station owner
is able to change format, say from classic rock to all-news, and
thus change its impact on the marketplace of ideas. But a radio
station switching to all-news does not thereby turn itself into
the equivalent of a television station nor does its impact on the
marketplace of ideas become that of a television station.
Conversely, if a home shopping television station began to
carry substantial local news programming, the impact on the
marketplace of ideas would be greater than that of the former
classic rock radio station.

423. The case for a usage measure is that it reflects actual
behavior. However, current behavior is not necessarily an
accurate predictor of future behavior. Moreover, in order to
implement a usage measure accurately, it would be necessary
for us to define which content should be considered local news
and current affairs. Current behavior, e.g., viewing or listening
to a broadcast station, is based on the content provided by the
station in question. However, media outlets can change the
amount of news and current affairs that they offer, perhaps in
response to competitive conditions in the “viewpoint diversity”
marketplace. Such changes are unpredictable, so current
market shares (e.g., of viewing or listening) may not be good
predictors of future behavior. Indeed, advocates of a
concentration approach to diversity analysis have noted the
weakness of the usage approach, pointing out that “[E]vidence
of past production does not, as a matter of logic, necessarily
give a proper picture of acompany’s future ability to compete.’
Only in examining ‘its structure, history and probably future,
does one provide’ the appropriate setting for judging the

502a

probably anticompetitive effect of the merger’”’ 7 ‘This point
has particular force when dealing with competition in the
marketplace of ideas because media outlets can rapidly expand
their distribution of content (including local news and current
affairs) at very low marginal cost. Of course, availability of
media can also change. However, this is less likely to occur
than is change in the program schedule or station format.
Moreover, availability is far more likely to increase than
decrease. Although a broadcast station owner could turn in the
station license and take it out of service, this happens rarely if
ever. A more likely scenario is an increase in media
availability as a new station enters the market.

424. If we were to adopt a usage measure designed to reflect
our concern with local news and current affairs, we would need
information on viewing/listening/reading of local news and
current affairs material. To implement this procedure, it would
be necessary first to determine which programming constituted
news and current affairs. We believe that this type of content
analysis would present both legal/Constitutional and data
collection problems. News and current affairs content is not
necessarily limited to regularly-scheduled news programs. So
we could be faced with deciding which other programs were
news and currevit affairs, whether some portion of a program
not primarily news should count as news, and, indeed, whether
portions of a news report devoted, e.g., to movie reviews should
count as news. Overall ratings or (in the case of newspapers)
subscription data would not suffice. Someone who subscribes
to a daily newspaper but only reads the (nationally-distributed)
comics and the classified advertisements is undoubtedly getting
a valuable service, but it is not clear that the service has
anything to do with news and current affairs. Similarly, a
television station that attracts large audiences by virtue of its

°37 Maaurice E. Stucke and Allen P. Grunes, Antitrust and the Marketplace of
Ideas, 69 ANTITRUST L. J. 249, 277 (2001) (quoting U.S. v. General
Dynamics Corp., 415 U.S. 486, 501 (1974)).

503a

movies and national sports programming provides an important
service, but it would be misleading to judge the station's
importance as a local news outlet by its overall ratings.

425. Ultimately, our goal is not to prescribe what content
citizens access, but to ensure that a wide range of viewpoints
have an opportunity to reach the public. This goal, the
limitations of current usage as a predictor of future usage, and
the content classification requirements for implementing a
usage measure all lead us to adopt an “equal share” approach to
weighting outlets within the same medium. ~

426. We deviate from this approach only in the case of the
Internet. We use subscription shares to divide the Internet
category among the two current significant sources of Internet
access—telephone companies and cable companies. In order to
determine the number of subscribers to telephone company
based Internet access, it is necessary to add together “dial-up”
and DSL subscribers. Dial-up service is available to anyone
with a telephone line and offers a low-capacity connection (up
to 56 kbps). DSL service offers much higher speed
connections, but, due to the requirements of the technology and
certain physical limitations of the telephone distribution
network, it is not available everywhere. Cable companies offer
high speed Internet access, and most cable plant has been
upgraded to support this service. Some applications, such as
viewing video clips of news and other content, are not fully
supported by dial-up services. Trade and industry sources
estimate that, as of the end of 2002, 85 million households had
access to cable high speed Internet service and 11.3 million
subscribed.” This leaves over 15 percent of households
without access to cable modem service. Moreover, it is not
clear how the areas in which cable modem service is
unavailable compare to the areas in which DSL is unavailable.
We therefore think it prudent to use subscriber figures to

3K. ated ;
See Cable and Telecommunications Industry Overview 2002 Yearend, at
http://www .ncta« ‘om/industry_overview.

504a

calculate how to divide the Internet category between cable and
telephone companies. _

427.Table 78 of FCC MOWG Study No. 8 provides
information on Internet access. Respondents who said they
have home access to the Internet were asked a follow-up
question regarding how they access the Internet. The answers
(in percentages) were as follows: cable line 18.9 percent, DSL
line 14.7 percent, telephone line 66.1 percent, other 3.5 percent,
don’t know 5.9 percent, and refuse 0.5 percent. The responses
sum to 109.6 percehi. If we take the 99.7 percent of
respondents who picked cable, DSL, or telephone line as the
base, and if we combine telephone and DSL, the resulting
shares are 19 percent cable and 81 percent telephone. We
recognize that, given the relatively small share of Internet in the
total diversity market (12.5% weight), using subscriber shares
rather than equal availability for Internet providers has a very
small impact on our Diversity Index calculation. 9% In this
regard, however, we reject the argument made by some
commenters that we should not include the Internet at all. They
argue that people only utilize the Internet to access their
newspapers’ and local broadcast stations’ websites and,

939 ; ; : ”
As explained in the next section, “Calculation Methodology,” our

diversity index is calculated by squaring relevant market shares. If we were
to assume that the two Internet sources had equal shares, the contribution to
the index of Internet would be 78 points. The assumption we use as
described in the text leads to a contribution to the index of 109 points. We
do not attribute common ownership to Internet Service Providers, e.g., even
if Cox owns a television station and the local cable ISP in a market, we will
not combine their market shares for the purposes of the Diversity Index
calculation. We will assume (subject to examination at the next biennial
review and to future findings we might make in our cable modem
proceeding), that iSPs do not restrict subscriber access to Internet content
based on the identity of the content provider. We also note that, as
explained above, we are looking at the availability of news and information
sources generally -- and websites particularly -- not their popularity.

S0Sa

therefore, the Internet does not add to diversity.” Although
many local newspapers and broadcast stations maintain
websites with news content, that does not begin to plumb the
extent of news sources on the Internet. Some websites compile
news from numerous sources, many of which an individual may
not have know of or known how to access (e.g., The Drudge
Report). Others are unique to the Internet (e.g., Salon).
Moreover, we include the Internet because, as previously
indicated, we are looking at availability of media, not the
popularity of specific publications, stations, cable channels, or
websites. There is a virtual universe of information sources on
the Internet and there are websites not maintained by existing
news media conveying information on everything from fringe
political groups to local civic events. We cannot pretend that
these are not in the “diversity” mix simply because only a small
number of people may visit them.

e. Calculation Methodology

428. The Diversity Index is structured like an HHI, i.e., it is
simply the sum of squared market shares. As explained above,
squaring market shares, unlike measures based on the “raw”
market shares, permits construction of an index that takes
account of the market shares of all providers in the “market” for
viewpoint diversity. As noted above, the geographic market we
are using is local. We currently define television markets in
terms of the Nielsen DMA. DMAs are exhaustive
Classifications, covering the entire United States, and it is
straightforward to count the number of television stations in a
DMA. Weare including public as weil as commercial stations.
Public stations provide viewpoint diversity; indeed that is a
specific part of their mandate. Although they do not have the

940

See, e.g., AFL-CIO Comments at 12-14, (citing Consumer Union
Comments in MM Docket No. 01-235, Douglas Gomery, The FCC's
Newspaper-Broadcast Cross-Ownership Rule: An Analysis (Econ. Policy
Inst., Feb. 2001)); AFTRA Comments at 10; UCC at al Comments at 23.

506a

same programming incentives that commercial stations do, their
partial reliance on viewer contributions means that they, like
commercial stations, must be sensitive to the demands of their
audience. We choose not to include television stations from
outside the DMA in question, even if they obtain a measurable
audience share in the DMA. Our focus is on local news and
current affairs and it is not reasonable to assume that stations
outside of the DMA in question will devote significant
resources to news and current affairs programming targeted to
that DMA. Our cable television signal carriage rules generally
permit a television broadcast station within a DMA to obtain
cable carriage throughout the DMA, and our DBS signal
carriage rules generally ensure that all television stations within
a DMA are treated the same with respect to satellite
retransmission.” For this reason, we assume that all television
broadcast stations in a DMA are available throughout the
DMA. As explained above, each broadcast television station
receives an equal share of the broadcast television weight.

429. We combine the television stations in each DMA with
the radio stations in the Arbitron radio metro with which the
DMA is paired. There are 287 Arbitron radio metros in the
country. Each one is smaller than the DMA within which it
lies.”** Arbitron radio metros do not cover the entire country.
More sparsely populated areas are not included in radio metros;
approximately one-half of radio stations are not in a metro
market. As explained below in the cross-media limits section
of this Order, we use the Diversity Index to help us identify
markets that are “at risk” for excessive concentration in the
“viewpoint diversity market.” Once those markets have been
identified, and cross-media limits imposed, the actual
implementation of the cross-media diversity limits will not
require information on a local radio market, only on the

941 coe 47 CER. § 76.56 (Cable) and § 76.66 (DBS).

942 ‘ , cs ‘ P ,
Most radio metros lie wholly within a single DMA; virtually all of the
others are predominantly within a single DMA.

507a

television market (DMA) within which the radio stations are
located that are part of a proposed merger. As detailed in the
cross-media limits section, the analysis that we use to identify
at-risk markets is based on examination of a substantial sample
of the 287 Arbitron radio metro markets.

430). Daily newspaper publication and circulation data are not
collected based on Arbitron radio metros. A different market
concept, developed by the Department of Commerce, is used by
the industry. The basic building block is the “Metropolitan
Statistical Area,” or “MSA.” The Department of Commerce
recognizes 318 metropolitan areas, which include 248 MSAs,
58 “PMSAs” (primary metropolitan statistical areas), and 12
“NECMAs (“New England county metropolitan statistical
areas”). For Diversity Index calculation purposes, these areas
are matched to Arbitron radio metros. Each daily newspaper
that is locally published in the metropolitan area is included in
the market. The daily newspaper share of the Diversity Index
is divided evenly among all daily newspapers included in the
market. In the absence of market-specific information on
weekly newspaper availability, we make the most conservative
assumption that there is one independently-owned weekly
newspaper in each local market, and assign to it the entire
weekly newspaper share.”

431. In terms of calculating the Index, within each medium
we combine commonly-owned outlets and calculate each
owner’s share of the total availability of that medium. We then
multiply that share by the share of the medium in question in
the total media universe (television plus newspaper plus radio
plus Internet). Once these shares in the overall “diversity
market” have been calculated, we add together the shares of
properties that are commonly-owned (e.g., a newspaper and a

943, . ' er

In fact, there were 7,689 weekly newspapers in 2000, so it is likely that
the average market has at least one weekly. See NAA, supra note 200 at
htp://www.naa.org/infolfacts02/13_ facts2002.him.

508a

television station), square the resultant shares, and sum them to
get the base Diversity Index for the market in question.”

4. Cross-Media Limits

432. In this Section we modify our rules by adopting a new
set of cross-media limits (“CML”) in lieu of our former
newspaper/broadcast and television/radio cross-ownership
rules.. The CML have been designed specifically to check the
acquisition by any single entity of a dominant position in local
media, markets -- not in economic terms, but in the sense of
being able to dominate public debate -- through combinations
of cross-media properties. Because we have traditionally relied
upon blanket prohibitions on certain cross-media combinations,
we have never before had to confront head-on the challenge of
identifying specifically which types of markets give us the
greatest cause for concern in terms of preserving diversity of
viewpoint, and which types of transactions are most
problematic in this regard. This effort is compiicated by the
nature of the public interesi we are seeking to protect —
diversity — which is as elusive as it is cherished.

433.Our modification of the newspaper/broadcast and
television/radio cross ownership rules into a set of cross-media
limits or CML is our first comprehensive attempt to answer this
difficult and complex set of questions. The CML derives from
data in the record regarding the relative reliance by consumers
of various types of media outlets for news and information. To
help us analyze that data, we use a methodological tool — a
diversity index or “DI” — that allows us to measure the degree
to which any local market could be regarded as concentrated
for purposes of diversity. Based on an analysis of a large
sample of markets of various sizes, the diversity index suggests
that the vast majority of local media markets are healthy, well-
functioning, and diverse.

944 Meare aoe
Appendix C, Diversity Indices in Ten Sample Markets, contains the
Diversity Index calculations for the ten markets examined in MOWG Study

No. 1, based on the market structure as of November 2002.

509a

434. Moreover, because we are adopting herein intra-service
competition caps for radio and television properties, those caps
will ensure that local markets will continue to be served by a
diversity of voices within each of these respective services. By
the nature of the exercise, markets defined for competition
purposes are no broader than, and generally are narrower than,
markets defined for diversity purposes. Thus, our radio and
television competition caps will not only serve to promote and
protect competition within the radio and television services,
they will also be protective of diversity interests when
television-only or radio-only transactions are at issue. For
example, in a market with 12 TV stations, our intra-service
caps guarantee at least six different owners of television
stations. If there are forty radio stations in the market, our
radio cap will ensure at least six different owners of radio
properties.

435. We recognize, however, that our intra-service caps will
not address diversity concerns that may result from cross-media
combinations. Although our local radio and television caps
will ensure a significant number of independent voices in larger
markets, cross-media combinations in very small markets might
result in problematical levels of concentration for diversity
purposes. Accordingly, we are herein supplementing our two
intra-service local rules with a narrowly drawn set of cross-
media limits to reach those combinations that are not already
prohibited by our television or radio caps, but which would
give rise to serious diversity concerns. The cross-media limits
are based on a set of assumptions drawn directly from the
record evidence in this proceeding and premises that are
consistent with past Commission policy and practice. Although
we rely in part on our data analysis to help define the CML, we
clearly respect that diversity is inherently subjective and cannot
be reduced to scientific formula. We do believe, however, that
greater use of empirical data and evaluation of that data can
significantly strengthen the reasoning that underlies our expert
judgment. The CML, therefore, ultimately rests on our

RR i te ee

510a

independent judgments about the kinds of markets that are most
at-risk for viewpoint concentration, and the kinds of
transactions that pose the greatest threat to diversity.

a. Competition Caps Protect Diversity

436. As set forth above, we have adopted a cap both on the
number of television stations that any one owner may hold ina
market, and on the number of radio stations that any one owner
may hold in a market. These caps were designed to promote
and protect competition within these two distinct services. The
caps are, therefore, based on product market definitions that
consider only those products or services that may be regarded
as reasonable substitutes for competition purposes. We
recognize, however, that although radio and television outlets
may not compete in economic terms with other types of speech
outlets, e.g., newspapers, they all inhabit the mass media
landscape that Americans turn to for news and information. In
that sense, whatever the confines of their markets for
competition purposes, many different outlets serve core
democratic functions as purveyors of ideas, outlets for opinion,
and distributors of news.

437. The data in the record evidence this difference. As set
forth above, radio and television compete in economic terms in
separate and distinct product markets. Both radio and
television outlets, however, inhabit the larger speech market, as
do several other types of entities. For example, MOWG Study
No. 8, a consumer survey on media usage, reveals that, when
asked to identify their primary source of all news and
information -- both local and national -- , approximately 40% of
Americans responded that broadcast television was their
primary source and approximately 10% of Americans
responded that radio was their primary source. > However,
nearly 24% of responderts identified daily newspapers as their
primary source of news and information, 18% identified cable

°45 MOWG Study No. 8, Table 20.

Slla

news networks, 6% identified the Internet, and 2% identified
weekly newspapers or magazines.’ These figures track
closely results from a Pew Research Center survey asking
similar questions about Americans’ use of media for news and
information. When asked where they turned for their primary
source of election news, 39% of respondents said broadcast
television, 24% said cable television, 24% said newspapers, 9%
said radio, and 5% said the Internet.”*’ Other studies confirm
that, today, Americans substitute among and between many
different sources for news and information on a regular basis.””*
The record reflects, in short, that the “viewpoint” market in
which television and radio stations participate is broader than
the economic product markets, as defined by standard
competition theory, in which either competes. ”” As a result,
intra-service caps designed to ameliorate competition concerns
necessarily also will protect against undue concentration of
speech outlets for diversity purposes.

438.Our diversity index helps to illustrate this point.
Pursuant to our new local radio rule, no single owner, even in
the smallest markets, will own more than 50% of the radio
outlets. In larger markets, the percentage of radio outlets that
can be held by any one entity is considerably smaller. Thus,
using the most extreme set of facts, and using Altoona,
Pennsylvania, as our test case, the diversity index focused on
local news and information alone (again, the most conservative
assumption) reveals a relatively minimal impact on viewpoint
diver: ty even should the radio outlets become split between
only .vo owners. The current base case DI for local news and

946 Id
a.

*” The Pew Research Center for the People and the Press, Sources for
Campaign News, Fewer Turn to Broadcast TV and Papers (Apr. 27, 2003)
at http://people-press.org/reports/display.php3? PagelD=243.

YAR
See, e.g., UCLA Internet Report.
y

9

49
Fox Comments, Owen Statement.

512a

information for Altoona is 960.”” If the local radio market
were to become restructured into a duopoly, the DI would rise
to only 1,156.”! Again, this hypothetical posits the most
extreme restructuring of radio outlets in the smallest market
among those in our test cases. The change in the diversity
index will be far smaller as a result of radio transactions in
larger markets or where the restructuring is less extreme.

439. Similarly, pursuant to our new local television rule, ne
single owner will be permitted to own more than two television
outlets in most markets. Using Altoona again, a two-TV
combination raises the base DI for local news and information
by only 64 points.”” Indeed, using a set of randomly sampled
markets of varying sizes, the average change in DI as a result of
an owner of one television property buying another to create a
television duopoly in a small market with only five licensed
television stations is 91.” In markets with twenty licensed
television stations the change in DI as a result of the creation of
a television duopoly is only six.” Thus, although our intra-
service television and radio caps are designed to protect and
promote competition, they have a corollary benefit of also
guarding against concentration in the viewpoint markets, at
least with respect to intra-service combinations.

9 See Appendix C, Diversity Indices in Ten Sample Markets.
951

Id.

on running this sample case, we assume that a duopoly would be
permitted in the market even though, in fact, a new duopoly would not be
permitted in Altoona under our new local television cap (Altoona has five
stations and one existing duopoly; a second therefore would violate our top
four restriction).

93 See Appendix D, Diversity Index Scenarios.

94 Id. We note, also, that our local television ownership cap includes a
prohibition on top-four combinations. This will have the effect of
prohibiting combinations of the local television stations most likely to
produce and carry significant local news programming. Thus, although the
top-four restriction is based on competition theory, the rule will also have
beneficial effects on local diversity.

Sl3a

440. We recognize, however, that cross-media combinations
that may impact the range and diversity of voices in local
markets will not be captured by our television and radio caps.
We therefore adopt, as described below, new cross-media limits
targeted specifically and solely at the types of transactions that
would give us the most concern and which are not already
prohibited by our intra-service caps.

b. Foundations of the Cross-Media Limits

441. We begin with the proposition that, because this rule
will limit the speech opportunities not only for broadcasters,
but also for other entities that may seek to own and operate
broadcast outlets (including those with the fullest First
Amendment protection — newspapers), we should draw the rule
as narrowly as possible in order to serve our public interest
goals while imposing the least possible burden on the freedom
of expression.” We also recognize that the tools that we are
using to evaluate market diversity involve as much art as
science. “Diversity” is not susceptible to microscopic
examination; it cannot be mapped with any known formal
system or reduced to mathematical equations. Although we
attempt to measure it and assign some quantitative value to it in
order to understand relative diversity of different types of
markets, we recognize that this process is_ inherently
approximate.” ° We must exercise great care, therefore, before
categorically prohibiting any particular transaction or set of
transactions as a prophylactic matter.

442. Nonetheless, it is apparent, based on the record in this
proceeding, that certain types of transactions in certain markets

” See FOEF Comments at 41-42; WVRC Comments at 43-44.

Using the Diversity Index allows us to see different market
characteristics in markets of different sizes. We have also found, however,
that differentiating markets by the number of newspapers present is too blunt
while differentiating markets by the number of radio stations is too fine.
Therefore, we use the number of television stations as an identifier of market
size.

Sl4a

present an elevated risk of harm to the range and breadth of
viewpoints that may be available to the public.””’ Using our
diversity index analysis and our independent judgment
regarding desired levels of diversity, we first identify “at-risk”
markets that might already be thought to be moderately
concentrated for diversity purposes. We then identify the types
of transactions that pose the greatest risk to diversity, and
impose specific limits on those transactions in at-risk markets.
Finally, because certain transactions in less concentrated
markets pose a high risk of rapid concentration, we impose
separate restrictions on transactions outside of the at-risk
markets.

c. Identifying At-Risk Local Markets

443. We begin by identifying those markets most susceptible
to high levels of viewpoint concentration; i.e., those markets
where our diversity concerns cut most deeply. At the outset,
consistent with our past practice and precedent, we focus in this
regard on local, not national, viewpoint market(s).””* Evidence
in the record before us supports the conclusion that the number
of outlets for national news and information is large and
growing, and that government regulation is thus unnecessary to
protect lai

444. With respect to local markets, our ten city study and our
DI test cases reveal that most local markets today are well-
functioning, healthy markets for speech.” For example, as of
2000, the largest media market in the country, New York City,
had 184 different media outlets owned by 114 different

asa ‘f. 1975 Multiple Ownership Second Report and Order, supra note 33
(in which we required divestiture in “egregious” newspaper/broadcast
cases).

998 See Policy Goals Section III, supra, and the Diversity Index, Section

VI(C\(3), supra.
9 See Appendix B, National News Sources.
960 cee MOWG Study No. 1.

96 . .
owners.” Perhaps —_ more impressively, ihe

Burlington/Piattsburgh market — market 141 out of 287 — had
53 outlets owned by 34 different owners.’ Even Altoona,
Pennsylvania, market 255, had 23 outlets owned by 15 different
owners.” That is, in the 255th ranked market, there currently
are fifteen different independent voices.

445. Not all voices, however, speak with the same volume.
Using our Diversity Index, we have examined the concentration
of media outlets in the ten markets that were the subject of our
Ten City Study using weighted voices. New York has a base
DI for local news and information of 373; Lancaster,
Pennsylvania, has a DI of 939; and Myrtle Beach, South
Carolina, has a DI of 989." Indeed, the average DI for all ten
markets, which range from the largest to near the smallest, is
758.” A DI of 758 is the equivalent of 13 equally-sized firms.

446. Moreover, to ensure that the results of our ten city study
were not anomalous, we have calculated the average DI for a
different set of randomly selected markets, both large and

961 Td. Even though both MOWG Study No. | and Appendix C (Diversity
Indices in Ten Sample Markets Study) used the same media markets, the
number of outlets and owners in individual markets as described in MOWG
Study No. | are different from the number of outlets and owners in
Appendix C, Diversity Indices in Ten Sample Markets, for two reasons.
First. MOWG Study No. | used outlet and ownership data that was current
in 2000, in order to make a comparison between 1960 and 1980. The
Diversity Indices in Ten Sample Markets Study used more current outlet and
ownership data from 2002, in order to be more up-to-date. In addition,
MOWG Study No. | included the “embedded” radio metro markets that are
physically in the NYC metro, for illustrative purposes. The Diversity
Indices in Ten Sample Markets Study used only the radio stations assigned

to the NYC metro, for analytical purposes.

962
Id.

as Id.
— Id.

ne

5l6a

small.”°° The average DI for markets in which there are 20
television stations is 612; the average DI for markets in which
there are 15 television stations is 595; the average DI for
markets in which there are 10 television stations is 635; and the
average DI for markets in which there are 5 television stations
is 911 — all well below the point at which one would
characterize them as highly concentrated if one were using the
analogous HHI to measure competition in the market.””’

447. We believe the analogy to the HHI is apt. The HHI is
an indicator of economic concentration; it provides an
analytical framework for determining when and if an entity or
group of entities is likely to wield market power in an economic
market. Our DI, which was inspired by and modeled after the
HHI, similarly is an indicator of viewpoint concentration.
Using the DI as an analytical tool, we can assign approximate
weights to different types of media outlets, account for the
diversity effects of commonly-owned properties, and measure
relative concentration between and among markets. The DI can
help us, therefore, identify the point at which an entity or group
of entities is likely to wield inordinate power in the marketplace
of ideas.

448. Although competition theory does not provide a hard-
and-fast rule on the number of competitors necessary to ensure
that the benefits of competition are realized, a market that has
ten or more equally- sized firms normally can be considered
fully competitive.” A 1000 DI correlates to market in which
there are roughly ten firms with approximately equal market
power. An 1800 DI would correspond to a market with SiX

70 See Appendix D, Diversity Index Scenarios.
967 14

968 ‘4 market with 10 or more equally-sized firms has an HHI of 1000 or
less. DOJ/FTC regards markets in this region to be unconcentrated.
Mergers resulting in unconcentrated markets are unlikely to have adverse
competitive effects and ordinarily require no further analysis. See DOJIFTC
Merger Guidelines § 1.51.

517a

roughly equal voices. Using our DI analysis of sample markets,
we note that it is not until we reach markets with three or fewer
licensed television stations that the average DI exceeds 1000,
the point at which the market normally would be characterized
as moderately concentrated for competition purposes.”

449. Our DI analysis of these sample markets, however, is
not the end of our inquiry. Because of the importance we
associate with maintaining diversity among the three principal
platforms — newspaper, radio and television — for the
expression of viewpoint at the local level, and because these
same three outlets produce a large share of local news
content,’”” we previously have used a “voice test” focused on
one or more of these outlets for measuring diversity. Indeed,
the Sinclair court suggested that our choice of an eight-voice
test, then used in conjunction with the local television rule, was
an exercise of agency discretion entitled to some deference.””'
Although we no longer are willing to base our rules upon the
comparatively rudimentary eight-voice test, we continue to
believe that unacceptable diversity losses can occur in very
small markets when the principal distribution platforms for
local news content come under common ownership and control.
In larger markets, we expect that the number of distribution
outlets for local news content will be larger, and that consumers
will have greater access to secondary outlets for news and
information.”

9 See Appendix D, Diversity Index Scenarios. The average DI for markets
with three television stations is 1027; the average DI for markets with two
television stations is 1316; and the average DI for markets with a single
television station is 1707.

ite CFA Comments at 32-39; UCC Comments at 23.

97) Sinclair, 284 F.3d at 162.

972 as ne
E.g., Deployment of Advanced Telecommunications Capability to All

Americans in a Reasonable and Timely Fashion, 15 FCC Red 20913, 20918
(2000) (broadband access in rural areas limited); 200] Price Survey Report,
17 FCC Red 6301, 6318 (2002) (low capacity cable systems in rural areas

Sees =

OS A

a5. Rie Tt Mee cE I MEER SPS Phe RE Rohe

518a

450. Finally, we are concerned not merely with the absolute
level of diversity that might already exist in any market or type
of market, but also with the degree to which diversity might be
sacrificed as a result of likely transactions. Accordingly, in
defining “at-risk” markets, we have used our DI and sampled
the effect of transactions, in large and small markets, involving
heavily used sources of local news and information.””* In so
doing, we have focused on the types of transactions that most
likely will lead to large DI changes and rapid concentration.
Our line-drawing effort is informed by the approach the DOJ
has taken in assessing competition issues. Although DOJ
policy is to review any transaction in a moderately concentrated
market that would result in a change in HHI of 100 points or
more, we have found no case in many years in which DOJ has
filed suit to block a merger that produced less than a 400 or
more point HHI change.”’* Based on our analysis, cross-media
combinations involving newspaper and television, newspaper
and radio, or radio and television properties do not produce a

offer fewer channels and are less likely to have stand-alone local or regional

cable news channels).

97 See Appendix D, Diversity Index Scenarios.

974 Under the FTC/DOJ Merger Guidelines, an HHI between 1000 and 1800
suggests a moderately concentrated market, and an HHI above 1800
suggests a highly concentrated market. Where the post-merger market
would be in the moderately concentrated range, the Guidelines suggest that a
merger that increases the HHI by more than 100 points will, absent other
factors, present antitrust concerns. Where the post-merger market would be
in the highly concentrated range, the Guidelines suggest that a merger
producing an increase in the HHI of more than 100 points, absent other
factors, is presumed to create or enhance market power or facilitate its
exercise. FTC/DOJ Merger Guidelines § 1.51. However, in the cases we
found over the past 15 years, the FTC or the DOJ has filed suit to block a
merger only when the change in the HHI is at least four times greater than
the Guideline’s standards. See, e.g., FTC v. Illinois Cereal Mills, Inc., 691
F. Supp. 1131 (N.D. Ill., E. Div. 1988); U.S. v. Georgia-Pacific Corp., 1996
W.L. 634212 (D. Del. 1996). In the majority of cases, the proposed merger
would have resulted in a change in the HHI in excess of 1,000 points.

519a

change in the DI of anything even approaching that magnitude
other than in markets with three or fewer television stations.’””
For example, a newspaper/radio combination in markets with
only two licensed television stations produces a DI change of
more than 300 points, a television/radio combination in markets
of that size produces a DI change of 301 points, and a
newspaper/television combination in markets of that size
produces a DI change of 731 points. A newspaper/television
combination in a market with three licensed television stations
produces a DI change of 331 points.””°

451. These changes, of course, reflect approximations based
upon sample data and are provided only to be illustrative of the
diversity losses that can occur as a result of cross-media
combinations in small markets. Nonetheless, based on all of
the foregoing, we conclude that a market with the equivalent of
ten or more equally-sized firms cannot be regarded as even
moderately concentrated for diversity purposes. In light of that
conclusion, and in consideration of the properties of small
markets and on our analysis of potential transactional impacts
in those markets, we conclude that markets with three or fewer
licensed television stations should be regarded as “at-risk”
markets for purposes of diversity concentration. Markets of
that size, we expect, will be moderately concentrated and
subject to rapid concentration if cross-media combinations are
created involving radio, television and/or newspaper
properties.”’’ Accordingly, we will prohibit certain cross-

— See Appendix D, Diversity Index Scenarios.

°”° The calculated changes in the Diversity Index for these markets are
premised on the assumption that the radio markets have consolidated to the
maximum extent permissible under our new local radio ownership rule. On
this basis, this is a “worst case” estimation of the impact of newspaper/radio

and television/radio combinations under the Diversity Index.

977 § market with an HHI of more than 1800 is regarded as highly

concentrated. We noted above that a DI of 1800 would correspond to six
equally-sized “voices.” Because of the amorphous nature of diversity as an
interest and the difficulty of measuring it with precision, we decline to draw

Sa ark te

Riis Soret Wine ut Nea coe oe

“ri Ty Tite 8 Nag
SERA RTS

*&).

be OE Ts Fe

Tr a

Bs a

520a

media combinations involving those properties in markets with
three or fewer television stations.’”*

d. Local Cross-Media Limits in At-Risk Markets

452. With respect to the limits themselves, we tread lightly in
view of the sensitive First Amendment interests at stake and the
deregulatory purpose of Section 202(h). Our intent is to draw
our rules narrowly, focusing on those transactions that are
likely to have a substantial impact on the diversity of voices
available in the market. The record shows that broadcast
television, daily newspapers, and broadcast radio are the three
media platforms that Americans turn to most often for local
news and information.””” They are, accordingly, the focus of
our diversity concerns, and we decline to impose any cross-
media limit on transactions involving media properties other
than radio, television, and newspaper outlets.

453. Further, we are establishing rules of nationwide
applicability. We desire, therefore, to provide the industry and
the public with clear, easy to administer rules reflective of
common market trends and characteristics. We recognize that,
in any given market, the lines we draw here may appear under-
or over-inclusive. Indeed, that quality inheres in the nature of

an absolute line prohibiting transactions that would take a market beyond the
1800 DI (i.e., six voice) level. The rules we are adopting herein, however,
are intended to protect against markets becoming highly concentrated — in a
qualitative sense — for diversity purposes.

978 When we originally crafted the newspaper/broadcast rule we required
divestiture of either a newspaper or a broadcast station in a limited number
of so-called “egregious” cases. We defined the relevant market in those
cases as the area encompassed by the city-grade signal of the relevant
broadcast station. Divestiture was required where the only daily newspaper
was published in a community within the city-grade signal of the only
commercial television (or only commercial radio station in cases where no
local TV station also existed) where the newspaper and the broadcast station
were commonly owned. See generally 1975 Second Report and Order,
supra note 33.

979 Soe MOWG Study No. 8, Table 97.

S2la

proscriptive rules themselves. Nonetheless, our analysis of the
record in this proceeding gives us confidence that our rules will
prevent the transactions that would seriously impair the
availability of diverse viewpoints in any local market while
permitting efficiency enhancing combinations. Again, although
they have a methodological foundation in the DI, these
judgments are based on agency expertise and experience
dealing with broadcast markets and the media industries
generally. Accordingly, except as specifically prohibited
herein, cross-media combinations will not be subject to
anything other than routine Commission review, 1.e., unless the
transaction is barred by the CML or our other ownership rules,
the combination is permissible under our rules, and we will not
apply the DI to it.”

454. As explained below, combinations of daily newspaper
and broadcast properties in at-risk markets present a serious
threat to local viewpoint diversity.”"! We therefore, adopt a
rule prohibiting common owuership of broadcast stations and
daily newspapers, and TV/radio combinations, in markets with
three or fewer television stations. In order to determine which
markets have 3 or fewer broadcast television stations, we will
rely on Nielsen television Designated Market Areas (DMAs).
We include for these purposes, commercial and noncommercial
television stations assigned to the DMA. This is consistent
with our overall measurement of the DI, explained above, as we

9X0 ‘ . ‘ ‘ . idea .
Bright lines provide the certainty and predictability needed for

companies to make business plans and for capital markets to make
investments in the growth and innovation in media markets. Conversely,
case-by-case review of even below-cap mergers on diversity grounds would
lead to uncertainty and undermine our efforts to encourage growth in
broadcast services. Accordingly, petitioners should not use the petition to
deny process to relitigate the issues resolved in this proceeding.

oy See, e.g., NABOB/Rainbow, PUSH Comments at 23-24; Gray
Comments at 16-19.

a

5§22a

assume that all television stations in the DMA are viewable in
the radio metro with which it is paired.”*”

455. A number of parties have questioned whether a cross-
ownership rule applicable to entities other than broadcasters,
e.g., newspaper owners, would be constitutional.”** We
continue to believe that a narrowly-drawn rule prohibiting or
limiting common ownership of broadcast properties and daily
newspapers is consistent with our constitutional framework.
Our current newspaper/broadcast cross-ownership rule has been
upheld by the Supreme Court against constitutional challenge”
and, as discussed above, broadcast/newspaper and
radio/television cross-ownership rules, like broadcast
ownership rules, are reviewed under the rational basis
standard.”*° We believe that our new cross-media limits satisfy
this standard because they are “a reasonable means of
promoting the public interest in diversified mass
communications,”””” and they are founded on a substantial
record. Nevertheless, we are mindful of the court’s concern in
another context, where a higher standard of constitutional
scrutiny applied, that our rules should focus on those markets
and transactions that are likely to result in substantial, rather
than only incremental, changes in diversity.” Our new cross-
ownership rules accomplish this because they are narrowly
tailored to restrict cross-ownership only in select markets.

456. Television-Newspaper. Nielsen survey data reveal that
daily newspapers and broadcast television remain the two most

982 .
See 4 428, supra.

Qx3 ; é
Media General Comments at 37; Tribune Comments at 17-28; Fox

Comments at 50-51.
7B See NCCB, supra, note 20.
8 See Legal Framework, Section II, 4§ 13-16, supra.
986

Id.
87 NCCB, 436 U.S. at 802.

988 Time Warner II, 240 F.3d at 1135.

523a

important sources of local news and information.’ The
importance of these outlets is reflected in our DI. As noted
above, a combination of a daily newspaper and a television
station in a market with only three television stations leads to
an average DI change of 331 points. These combinations in
markets with only two or one television station lead to DI
changes of 731 and 910 DI points, respectively. In these at-risk
markets, a single combination of a daily newspaper and a
television station could quickly jeopardize the range of
viewpoints available to consumers in the market. We therefore,
adopt a rule prohibiting the combination of a daily newspaper
and a broadcast television facility in any market with three or
fewer television properties. To trigger the rule, we will count
all television stations assigned to the DMA that contains the
newspaper's community of publication. We presume that
broadcast television stations are generally carried throughout
the DMA to which the station is assigned. Our rules will not,
however, bar a broadcast television station in such a market
from starting a new newspaper, as that would expand, not
decrease, diversity.

457. One additional issue in the cross-interest context is the
definition of “daily newspaper” for the purposes of
newspaper/broadcast cross-ownership. Currently, Note 6 to the
multiple ownership rule defines a daily newspaper as “one

~ which is published four or more days per week, which is in the
English language and which is circulated generally in the

ORY ‘ . :
Approximately 28.8 percent of Americans rely on newspapers as a

source of local news and information, and 33.8 percent use broadcast
television for this purpose. These figures are derived from normalizing the

_ figures in MOWG Study No. 8, Table 097. Because respondents were
asked what sources they had used in the previous 7 days for local news an
information, and because many respondents listed more than a single source,
the totals in the Table add up to more than 100%. Also, magazines were
excluded from the normalizing process as they typically are not sources of
local news.

524a

community of publication. *” Commenters raised the issue of

the English language requirement when applied in Puerto Rico
where the Spanish language is the dominant language.”
Caribbean argues that the Commission expressly rejected
requests to exempt Puerto Rico from the rule at the time of its
adoption and recognized that the goals underlying the rule were
of equal concern in Puerto Rico as on the mainland.” Both
Caribbean and Arso argue that the exclusion of foreign
language newspapers also allows for the exercise of market
power by the dominant newspapers in Puerto Rico which, due
to the exclusion of non-English newspapers, could be owned in
tandem with broadcast stations in the market.”

458.The exclusion of non-English language daily
newspapers in areas where the dominant language of the market
is not English creates a discrepancy in treatment that must be
ended. As Caribbean notes, in adopting the original
newspaper/broadcast cross-ownership rule, the Commission
recognized that the need for diversity in Puerto Rico was the
same as that elsewhere. Since the definition of a daily
newspaper was adopted in 1975, the percentage of households
in which Spanish has spoken has approximately doubled.” It
is appropriate, therefore, at this point in time, that we apply the
CML to non-English daily papers in markets in which the
language that they are printed in 1s the dominant language of
their market.”” While the example of Puerto Rico was

9% 47 CER. § 73.3555 Note 6.

™ Arso Comments at 1-4; Caribbean Comments in MM Docket No. 01-235
at 22-35.

992 : ‘ ;

Caribbean Comments in MM Docket No. 01-235 at 22.
993 ,

Id. at 30-38. Arso Comments at 3-4.

4 . —_—

In 2000, Spanish was the language spoken at home in 10.5 % ot
American households. See www.census.gov. In 1980, the percentage was
5.3%. This is derived from data contained in INFORMATION PLEASE

ALMANAC (Otto Johnson ed., Houghton Mifflin Co. 1995) at 835.

995
As previously indicated, to trigger the rule, we will count all television

5§25a

addressed in the comments, there may be other communities to

which this will apply now or in the future. Those whose

primary language is not English deserve the same protections of

diversity and competition as do English speakers. Accordingly,

for purposes of applying the CML to newspaper/broadcast
transactions we will change the definition of daily newspapers

to include non-English dailies printed in the primary language 7
of the market.

459. Radio-Newspaper. Although broadcast radio generally
has less of an impact on local diversity than broadcast
television, according to the results of our Nielsen survey,
discussed above, in at-risk markets the combination of a daily
newspaper with one or more broadcast radio facilities can
nonetheless have significant negative implications for the range
of viewpoints available. Indeed, markets with three or fewer
television stations have, on average, only 21 radio stations.’”°
Under our radio cap, a single owner in a market with 21
stations could own six stations, or 29% of all the radio outlets
in the market. Combining such a station group with, perhaps,
the only daily newspaper could, therefore, seriously impair the
range of independent viewpoints available in the market”’
Again, based on a sample of markets with three or fewer
television outlets, we find that the change in DI as a result of a
newspaper-radio combination, assuming that the radio owner
has reached the radio ownership cap under our new local rules,

stations assigned to the DMA that contains the newspaper’s community of
publication. For the purposes of evaluating whether the non-English daily is
printed in the primary language of the “market,” however, the market shall

be defined as the newspaper's community of publication.
996 |

BIA Master Access Data Base (Nov. 2002).
99

Although any given market may have more than one daily newspaper,
and of course every radio owner does not buy stations up to the regulatory
limit, we are adopting general rules of nationwide applicability.
Accordingly, we are positing for these purposes that the market is as
concentrated as possible consistent with our other local rules.

526a

would be 242 points or higher.” Given that markets of three
television outlets begin with an average DI of 1027, which we
regard as the beginning of the moderately concentrated range, a
242 point DI increase moves the market substantially toward a
highly concentrated state. We therefore, adopt a rule
prohibiting the combination of a daily newspaper and a
broadcast radio facility in any market with three or fewer
television properties.” To trigger the rule for newspaper/radio
combinations we will retain our current standard. That
standard requires complete encompassment of the newspaper's
community of publication by the requisite signal strength
contour of the commonly owned radio station(s).”

8 See Appendix D (Diversity Index Scenarios).

Again, we note that this rule does not apply in the event that a broadcast

licensee seeks to found a new daily newspaper in the market.

1000 :' ; :
For AM radio stations that standard is complete encompassment of the

newspaper’s community of publication by the predicted or measured 2mV/m
contour computed in accordance with § 73.183 or § 73.186 of the
Commission’s Rules. For FM radio stations the standard is complete
encompassment of the newspaper’s community of publication by the |
mV/m contour computed in accordance with § 73.313 of the Commission's
Rules. Previously, we discussed the inherent flaws in defining radio markets
using a contour-based definition, and decided to move to a geographic based
definition. Specifically, we found that a contour based definition for
defining radio markets can create inconsistencies in counting stations that
comprise a market, counting stations that an entity owns in a market, and
determining a radio market’s size and geographic area. See Local
Radio/Problems with the Existing Radio Market Definition and Counting
Methodologies, Section VI(B)(1)(a)(ii)(a), supra. However, such problems
do not arise in the context of using contours to determine whether the cross-
media limits rule is triggered. Here, we are concerned with the physical
proximity of the broadcast station and the newspaper's community of
publication, or in the case of radio/television cross-ownership, we are
concemed with the relative distance between two specific stations. Because
the cross-media rule relies, in part, on a geographic location, i.e. the
community of publication or the communities of license, parties cannot take
advantage of such discussed inconsistencies to circumvent the rules.
Moreover, we are not relying on a contour-based definition to define a cross-
media market; we are only using it to determine whether the rule is

cee

5§27a

460. Television-Radio. Combinations involving daily
newspapers and broadcast properties are not the only cross-
media combinations that present diversity concerns in at-risk
markets. Approximately one-fourth of Americans rely on radio
as a source of local news and information, and one-third use
broadcast television for this purpose.'”’' Cross-media
combination* involving television and radio properties also,
therefore, are likely to give rise to systematic diversity concerns
in at-risk markets. Our DI analysis confirms this fact.'"’’ We
therefore adopt a rule prohibiting the combination of broadcast
radio and broadcast television facilities in any market with
three or fewer television properties. In such markets, we will
not permit an owner of a TV station to own any radio stations
in the market, and vice versa. Although this modification is
more stringent than our current radio/T V cross-ownership rule
in a limited number of markets,'””* the overall thrust of our
CML approach has been to eliminate regulatory restrictions
where they are unnecessary."

461. The television/radio cross-ownership rule is triggered
when the radio station’s community of license is in the
commonly owned television station’s DMA. Similar to
requests for waiver of the newspaper/broadcast cross-
ownership rule, parties seeking waiver of the television/radio

triggered.

l MOWG Study No. 8, Table 097. The figures above are derived from
normalizing the figures in Table 097. Because respondents were asked
what sources they had used in the previous 7 days for local news and
information, and because many respondents listed more than a single source,
the totals in the Table add up to more than 100%. Magazines were excluded
from the normalizing process because they typically are not used for local
news.

1002 ,. , : , a :
See Appendix D, Diversity Index Scenarios.
1003

47 C.F.R. § 73.3555(c).

, 1004 ; ' Aree nee :
We discuss grandfathering of existing combinations in these markets

below. See Grandfathering and Transition Section VI(D), infra.

528a

cross-ownership rule can rebut this by showing that the
stations’ signals do not overlap and the television station is not
carried on cable systems in the radio station’s market.

5. Additional Cross-Media Limits in Small to
Medium-Size Markets

462. Although markets with four or more licensed television
stations do not qualify, in our judgment, as at-risk markets, a
combination of a daily newspaper with a television duopoly and
a significant radio presence can, in small to medium-size
markets result in substantial changes in the level of diversity.
For example, assuming that owners of broadcast properties are
constrained only by our local radio and television caps (i.e.,
they may acquire stations up to the cap in either service), a
newspaper owner might attempt to acquire a television duopoly
and several radio properties within the same market. Referring
again to our sample markets we find that, in a five-television
market, a combination of a newspaper, a television duopoly,
and as many radio stations as permitted by the applicable local
radio cap results in an average DI change of 846 points.
Indeed, even in an eight-television market, the resulting
average DI change from such a newspaper/TV duopoly/radio
combination DI is 734 points. Given that eight-television
markets begin, on average, with a DI of almost 900 points,
changes of this magnitude can lead quickly to a highly
concentrated market.

463. We notice a dramatic difference, however, in the base
DI, and in the DI changes that result from a combination
involving a newspaper, a TV duopoly, and a radio station
group, between our sample markets that have four to eight
television stations and those that have nine or more television
stations. The base DI for markets with eight television stations
is still almost 900 points — nearly in the moderately
concentrated range; there is almost a 200 point difference
between these markets and those with nine television stations,
which, in our sample, have a base case DI of 705 points. In

5§29a

addition, although a newspaper/TV duopoly/radio combination
produces a change of over 700 points in an eight television
market, bringing the DI up to approximately 1600 points, the
change is fewer than 500 DI points in a nine television market,
bringing the DI up to only 1200 points. These numbers accord
with our experience and judgment regarding the operation of
small to medium-size markets, and are supported by other
evidence in the record.'"”

464. We also note significant differences between the DI
changes that result from newspaper/TV combinations in
markets with between four and eight television stations and
those with nine or more television stations. Using our sample
markets, a newspaper combining with a television duopoly ina
market with only five television stations leads to an increase in
the DI of 376 points. Even in markets with eight television
stations, the average DI increase as a result of such a
combination is over 300 points. In markets with nine television
stations, however, the DI increase from a merger of a
newspaper with a television duopoly is only 172 points; it is
about 100 points in markets with ten televisions.'"° The
potential for rapid concentration that may result from a
combination of a newspaper with a television duopoly in
markets with between four and eight licensed television stations
(“small markets”’) leads us to conclude that it would be prudent,
in these markets, to impose additional local ownership
restrictions as part of our CML.

1005

See, e.g. Buckley Comments at 4-5; UCC Comments at 16-17, 40-41.

"° Because of the number of radio stations in the markets observed for our
sample of seven-television-station markets, the DI increases in those markets
are smaller than those in eight TV markets. This deviation does not
undermine, in our judgment, the more general conclusions that we draw
from the data and from our DI methodology regarding the markets most at
risk for viewpoint concentration (i.e., we do not deem markets with seven
television stations, in general, to be less at risk than markets with eight
television stations).

530a

465. We are cognizant, however, of the fact that substantial
public interest benefits may flow from broadcast/newspaper
combinations. As discussed above, television stations that are
co-owned with daily newspapers tend to produce more, and
arguably better, local news and public affairs programming
than stations that have no newspaper affiliation. Because of the
news resources available to local newspapers, we expect similar
benefits to be associated with newspaper ownership of radio
stations (e.g., radio stations affiliated with a local newspaper
may have an enhanced ability to produce local, all-news radio
programming and to cover local political and cultural events in
greater depth than stations unaffiliated with a newspaper).
Accordingly, we are not inclined to prohibit outright
newspaper/ broadcast combinations in markets with 4 — 8
television stations (referred to below as “small to medium size
markets’’).

466. Balancing these interests, we believe it appropriate, in
small to medium size markets (those with between four and
eight television stations) to allow the following: 1) one entity
may own a combination that includes radio, television and
newspaper properties, but the entity may not exceed 50% of
either of the applicable local radio or the local television caps
in the market: 2) a radio station group owner that also owns a
newspaper in the market, but which does not own any
television properties in the market, may acquire radio stations
up to 100% of the applicable radio cap. In these small to
medium size markets, therefore, we will prohibit: television
broadcasters that also own a daily newspaper in the market
from having a television duopoly in that market; a broadcaster
with a duopoly from obtaining a daily newspaper in the same
DMA; a newspaper owner from purchasing more than a single
television station within the DMA; and a radio station owner
that also owns a daily newspaper and a television station in the

S3la

market from exceeding 50% of the applicable radio cap for the
market.”

467. We believe that this CML achieves an appropriate
balance in small to medium size markets between fostering the
production of high quality local programming and protecting
diversity. To begin with, the public interest benefits of
newspaper ownership (the benefits of cross-fertilization
between media) likely are realized primarily in the first
broadcast station co-owned in either service. Although there
may be economic benefits to the owner from more extensive
combinations, it is not as clear that those benefits will accrue to
the public in any meaningful way; at least the public interest
component of these benefits is likely to decline incrementally
as the number of stations increases. Given that no owner will
be permitted, in accordance with our local television cap, to
hold more than two television stations in a small to medium
size market, a limit of one station in these markets for owners
of local newspapers will maximize the public interest benefits,
while reducing any loss of diversity. Although the loss of
diversity that might result were that owner to add a significant
radio presence in the market warrants a further 50% limit in the
number of radio properties that owner might hold, such is not
the case if the combination does not include any television
properties.

468. Again, our DI and a set of sample markets help to
illustrate the fact that our modified 50% CML for newspaper

1007 ,
For these purposes, we use the Arbitron or contour-overlap market

definitions discussed above in determining whether the newspaper and a
radio station serve the same market. We are not imposing a limitation that
would preclude a top four television station in a market from being
combined in common with a newspaper or radio station similar to the
restriction imposed in the local television rule context. . The top four
restriction imposed under the local TV ownership rule ts specifically
designed to protection competition, as fully discussed in that section. The
cross-media limit, on the other hand, is designed to protect viewpoint
diversity, not economic compeiition.

oS a ee Nag Moe te

See Sia acetate a til tla SU ecard icra

532a

combinations in small to medium size markets will significantly
reduce any loss of diversity that might result from efficiency-
enhancing newspaper/broadcast combinations. In a five-
television station market, a combination involving a
newspaper, a TV duopoly and a radio station group at the radio
cap would result in an average DI increase of 846 points, which
would take the market to 1757 points, near the highly
concentrated range.” If the combination is limited to a single
television station and no more than 50% of the applicable radio
cap, the DI change is 393 points, a decrease of 453 points. In
an eight-television market, a combination involving a
newspaper, a TV duopoly, and a radio station group at the cap
results in an increase in the average DI of over 700 points. By
limiting the combination to 50% of both the television cap and
the radio cap, the DI increase is reduced to 314 points.

469. Similarly, whereas a combination involving a
newspaper and a television duopoly alone will, on average,
raise the DI of a five-television station market by 376 points, a
combination involving a newspaper and a single television
station in a market of that size will raise the DI, on average,
only 223 points. The difference is more dramatic in markets
with eight licensed television stations, where the average DI
increase drops from 308 points to only 152 points for a
newspaper/TV duopoly combination. Newspaper/radio group
combinations result in significantly lower levels of viewpoint
concentration wken the combination does not include any TV
properties. Accordingly, we will permit newspaper/radio
combinations in small to medium size markets, provided they
comply with the local radio rule.

470. Similarly, our DI analysis indicates that radio/television
combinations in small to medium size markets result in
relatively small DI changes. For example, in a market with
only four television stations, a radio television combination,

1008 1 ader the DOJ/FTC Merger Guidelines, an HHI above 1800 suggests a
highly concentrated market.

533a

even assuming the radio owner holds the maximum number of
stations permitted under our local radio cap, results in a DI
change of fewer than 150 points.” Such a combination in a
market with eight television stations results in a DI change of
fewer than 100 points.'°"”

471. We have engaged in this analysis using our DI and a
randomly selected sample of markets not with the idea of
slavishly following the numbers that our index generated, but to
confirm and support the judgments we make regarding the
kinds of markets that are most susceptible to viewpoint
concentration, and the kinds of transactions that are most likely
to have a significant impact on the level of diversity available
in any given market. As noted above, we do not believe that
markets with between four and eight television stations can be
regarded as moderately concentrated for viewpoint purposes or
otherwise “at risk.” We do, however, believe, and our DI
confirms, that these markets are approaching a level of
viewpoint concentration that we would regard as moderate, and
we are concerned that some combinations involving the three
major sources of local news and public affairs information in
these markets would lead to inordinate diversity losses.
Accordingly, we will permit television/radio combinations in
small to medium size markets, provided they comply with the
local radio and television rules.

472. In markets with 9 or more TV stations, we will permit
any newspaper and broadcast cross-media combinations that
comply with our local TV ownership rule and local radio rule.
These tiers are derived from our DI analysis and our judgment
as to what markets are sufficiently diverse so that combined
newspaper/broadcast ownership would not unduly harm
diversity.

i” See Appendix D, Diversity Index Scenarios.

vo on

534a

473. With respect to markets with nine or more TV stations
(“large markets”), we impose no cross-media restrictions To
begin with, markets of this size today tend to have robust media
cultures characterized by a large number of outlets and a wide
variety of owners. New York City, for instance, which has 23
licensed television stations, 61 radio stations, and 21 daily
newspapers, had 61 different owners of broadcast stations and
daily newspapers as of November 2002.'"'' Using our diversity
index as a measure, New York City today has a base DI of only
373.'"'* More striking, perhaps, is the example provided by
Kansas City, Missouri, which has only nine licensed television
stations. Our Ten City Study reveals that Kansas City had 35
different owners and our Diversity Index analysis shows that
Kansas City has a base DI today of only —

474. Again, to ensure that the results of our Ten City Study
were not anomalous, we conducted a DI analysis on a random
sample of markets of various sizes, including markets with nine
licensed television stations, markets with ten television stations,
markets with fifteen television stations, and markets with
twenty television stations. Among our sample markets, the
average DI for those with nine television stations is 705; the
average DI for those with ten television stations is 635; the
average for those with fifteen television stations is 595, and the
average DI for those with twenty television stations is 612.'°""
That is, markets with nine or more television stations today are
very much un-concentrated.

475. The local radio and local television caps adopted herein
will help to ensure that large markets continue to be served by a
large number of different local media owners. For example,

OM See Appendix C, Diversity Indices in Ten Sample Markets.

1012
Id..
1013

See MOWG Study No. | and Appendix C, Diversity Indices in Ten
Sample Markets.

14 See Appendix D (Diversity Index Scenarios).

535a

positing Kansas City, Missouri, again as a typical market of
nine television stations, and assuming that four television
duopolies could in fact be created in that market, and further
assuming maximum radio consolidation under our new local
radio rule, there should still remain five different owners of
television stations and seven different owners of radio
stations.'”'” There currently also are five daily newspaper
owners serving the market. Therefore, even assuming that, in
the absence of any cross-media limit in the market, the owners
of the radio, television, and newspaper properties combine to
the maximum extent possible, there would remain at least seven
different owners of local media in the market, each with a
significant presence. In accordance with the mandate of
Section 202(h), we do not believe that we can justify a
restriction in a market where the worst case scenario (indeed,
one that may not even be possible given existing combinations
in the market), still results in a market with at least seven
different owners of the major sources of local news and
information.

476.More_ realistically, although some _ cross-media
combinations are likely to occur in the absence of a restriction,
constraints imposed by existing groups and the presence of
public stations that cannot be acquired by commercial entities
make it highly unlikely that Kansas Gity>or any market, will
consolidate to the level described in the preceding paragraph.
In order to get a better sense, therefore, for the actual affect of
various cross-media combinations in markets with nine or more

Ol See Appendix C (Diversity Indices in Ten Sample Markets). That is, in

a market with nine television stations, four duopolies can, in theory, be
created, leaving one singleton station, so that five owners of television
stations would remain. If there are forty-four radio stations in the market,
and group owners assembled the largest combinations possible under the
radio cap (seven), there would remain at least seven group owners; six with
groups of s'x stations and one with a group of two stations.

536a

television stations, we use our DI in sample markets and test
. . . 016
hypothetical combinations.'”!

477. Beginning in markets with nine licensed television
Stations, we see that, on average, the change in DI that would
result from a television owner acquiring a radio group
consisting of the maximum number of radio stations
permissible under our local radio rule is only 64 points.'"!” If
instead it were the owner of a daily newspaper acquiring that
radio group, the DI change would be 198 points, leaving the
market below 1000 DI.'°'* If the owner of a daily newspaper
were to purchase a television station instead of a large radio
group in a market of this size, the DI would increase only 86
points.'”'” Indeed, the largest combination possible in the
market — a combination that would include a daily newspaper, a
television duopoly, and a large radio group — would result in a
DI increase of 473 points, taking the average nine television
market to a base DI of under 1200 points, only marginally in
the range that we would consider moderately concentrated. ‘°””

478. As detailed in Appendix D (Diversity Index Scenarios),
in markets with ten television stations, the average base DI is
635 and the increase that would result from the assemblage of
the largest media combination possible would be 292 points —
leaving the market un-concentrated.'°*' In markets with fifteen
television stations, the average base DI is 595 and a
newspaper/television duopoly/large radio group combination
would increase the DI only 302 points.'’’* Similar results
obtain in markets with twenty television stations.

cae See Appendix D (Diversity Index Scenarios).

1017 Id.
Id.
Id.
Id.
Id.

1022
ld.

1018
1019
1020

1021

S37a

479. This analysis is premised on the creation of very large
combinations of media properties at the local level. Even so,
the results show that markets with nine or more televisiorr
stations are un-concentrated today and are unlikely to become
highly concentrated even in the absence of cross-media limits.
Section 202(h) requires that we justify broadcast ownership
limits on more than supposition or inchoate fears; our
governing law requires that we target our structural limits at
real and demonstrable harms. Based on the foregoing, we
cannot, therefore, justify cross-media restrictions in markets
with nine or more licensed television stations.

480. The tiers adopted above — “at-risk” markets, “small to
medium size” markets, and “large” markets — are derived from
our DI analysis and our independent judgment regarding market
operation and the effect of various combinations on diversity.
Our diversity concerns are greatest in at-risk markets and we
have accordingly prohibited all forms of cross-media
combinations in those markets. In small to medium markets we
have imposed specific limitations on particular kinds of
combinations that would, in our estimation, most likely result in
unacceptable harm to viewpoint diversity. In large markets, our
analysis indicates that no cross-media limit is necessary, nor
can one be justified, given the large number of outlets and
owners that typify these markets and the operation of our intra-
service television and radio caps.

481. Conclusion. Although we generally prohibit television-
radio, and newspaper-broadcast, cross-ownership in at-risk
markets. and we limit newspaper-broadcast combinations in
small to medium size markets, we recognize that special
circumstances may render these cross-media limits unnecessary
or counter-productive in particular markets. Accordingly, we
will continue to entertain requests for waiver of these cross-
media limits and, in particular, will give special consideration
to walver requests demonstrating that an otherwise prohibited
combination would, in fact, enhance the quality and quantity of

538a

broadcast news available in the market.'"?’ In addition, of
course, we will review our entire local broadcast ownership
framework, including our new cross-media limits, beginning
next year, in our 2004 biennial review. We will not, however.
permit collateral attack upon our rules in individual cases on
diversity grounds based upon more particularized showings
using the DI in a given market. The rules we adopt herein are
rules of general applicability. The lines that have been drawn
and the judgments that have been made reflect our conclusions
regarding the probable effects of given transactions in the run
of cases. Those conclusions necessarily rely upon
generalizations, approximations, and assumptions that will not
hold true in every case. Indeed, many of these assumptions
would not be true in a particular context or specific market. As
we stated above, the Diversity Index itself is a blunt tool
capable only of capturing and measuring large effects and
general trends in typical markets. It is of no use, therefore, for
parties to attempt to apply the DI to a particular transaciion in a
particular market. ;

D. Grandfathering and Transition Procedures
1. Grandfathering Provisions
482. Existing Combinations. There may be some existing
combinations of broadcast stations that exceed the new
ownership limits due to the modifications of both the local TV
and the local radio ownership rules. Because the modified local
TV rule permits increased common ownership of local TV
Stations, we expect few existing ownership combinations to

1023 ,_. ae
As is the case with our new local television ownership rules, we will

require that a licensee who obtains a waiver of our cross-media limits show
at renewal time the benefits that have accrued to the public as a consequence
of the waiver. At the end of the broadcast station’s (or stations’) license
term(s), the licensee of the station(s) must certify to the Commission that the
public interest benefits of the Commission's grant of the waiver are being
fulfilled. This certification must include a specific, factual showing of the
program-related benefits that have accrued to the public. Cost savings or
other efficiencies, standing alone, will not constitute a sufficient showing.

539a

violate the rule adopted herein. However, some existing same-
market combinations may not comply with the modified TV
ownership rule because of the elimination of the Grade B
overlap exclusion that is in the current rules. In addition, there
may be instances in which a party currently owns a
radio/television combination that may not comply with the new
cross-media limits."""*

483. As for radio, we are modifying the definition of many
radio markets, replacing the existing signal-contour based
definition with a geographic based market definition.'””? This
may result in a different number of stations being considered as
participating in a local radio market. Because our radio
ownership rule is based on a tiered system, if fewer stations
comprise the radio market, and the market falls into a smaller
tier, then the number of stations an entity may own would
decrease. We also are attributing in-market radio JSAs, which
could increase the number of radio stations that count toward
an entity's numerical ownership limit.

484. We are persuaded by the record to grandfather existing
combinations of radio stations, existing combinations of
television stations, and existing combinations of
radioftelevision stations.""’° As such, we will not require

024... ais
While we are not aware of any existing newspaper/broadcast

conibinations that have been previously grandfathered or approved by the
Commission that would be barred under the new rules, to the extent such
combinations do exist, they will be subject to the grandfathering and
transferability provisions described in this section.

02> We are retaining a modified contour-based definition outside of
Arbitron markets until we have completed a rulemaking to define geographic
radio markets in these areas. The grandfathering and transition procedures
adopted herein apply to Arbitron and non-Arbitron areas. In areas not
defined by Arbitron, through the completion of the rulemaking, licensees
should apply the modified contour-based market definition for these
purposes.

1026 We requested comment on grandfathering issues in the Radio NPRMs:
Radio Market Definition NPRM, 15 FCC Red at 25081-82 4 11; Local Radio

540a

entities to divest their current interests in stations in order to
come into compliance with the new ownership rules.'"*’ As
suggested by commenters, doing so would unfairly penalize
parties who bought stations in good faith in accordance with the
Commission’s rules.'"** Also, we also are sensitive to
commenters’ concerns that licensees of current combinations
should be afforded an opportunity to retain the value of their
investments made in reliance on our rules and orders. We also
agree with the commenters that argue that compulsory
divestiture would be too disruptive to the industry."""? On

Ownership NPRM, 16 FCC Rcd at 19888 4 65.

1027 ' ;
Secret proposes that we grandfather general radio station ownership

limits for markets rather than grandfathering specific ownership
combinations. In the alternative, it proposes that we permit any broadcaster
to own at least as many stations as the largest group owns presently in the
specific market. Secret Comments in MM Docket No. 00-244 at 4. Secret’s
approach is administratively problematic, requiring the Commission to
create and monitor a range of numerical limits in all of the Arbitron metros,
as well as in non-Arbitron areas. Moreover, it would create disparate
treatment in radio markets, not based on competitive analysis or public
interest assessment, but based solely on existing combinations. Because
these existing combinations were created using the current contour-based
market definition, which we find does not promote our competition goals,
some combinations may raise competition concerns and may violate the new
rules. To allow additional groups to obtain the same \.umerical limits would
only exacerbate such concerns.

18 See, e.g., NAB Comments in MM Docket No. 01-317 at 50: WVRC
Comments in MM Docket No. 0i-317 at 35; Cumulus Comments in MM
Docket No. 01-317 at 20; Eure Comments in MM Docket No. 01-317 at 5:
HBC Comments in MM Docket No. 01-317 at 13, n.2; MBC Comments in
MM Docket No. 01-317 at 11-12; Clear Channel Reply Comments in MM
Docket No. 01-317 at n.5; MBC Reply Comments in MM Docket No. 01-
317 at 4; Zimmer Comments in MM Docket No. 00-244 at 7; Weigle
Comments in MM Docket No. 00-244 at 6; NAB Comments in MM Docket
No. 00-244 at 29-30; Entercom Comments in MM Docket No. 00-244 at 7.

‘°° NAB Comments in MM Docket No. 01-317 at 50; MBC Reply
Comments in MM Docket No. 01-317 at 4; Zimmer Comments in MM
Docket No. 01-317 at 7-8; NAB Comments in MM Docket No. 00-244 at
29-30. We disagree with the commenters that support divestitures of current
combinations. See Dick Broadcasting Comments in MM Docket No. 01-317

¢
'

54la

balance, any benefit to competition from forcing divestitures is
likely to be outweighed by these countervailing considerations.

485. While commenters overwhelming support
grandfathering existing combinations, many nonetheless argue
that grandfathering will create competitive imbalances which
favor existing group owners - those that assembled
combinations under the current rules - and disfavor those that
cannot assemble competing combinations because of new

, ae 30 F : ‘ or
ownership restrictions.'”"” Like all grandfathering decisions, ‘
some disparity will exist between grandfathered owners and

non-grandfathered owners. We do not believe this fact

outweighs the equitable considerations that persuade us to
grandfather existing combinations.

486. We expect that the issue of grandfathering existing {
combinations will affect predominately radio group owners
because of the changes we make herein to the radio market
definition. We recognize that a geographic based radio market
definition may result in a fewer number of stations in certain
markets. In those instances, parties may not be able to acquire
the same number of stations as the largest owner in a particular

1031 Siific }
market. However, those combinations were created based

at 6-7; Idaho Wireless Comments in MM Docket No. 01-317 at 8; NABC
Comments in MM Docket No. 01-317 at 17. The Commission has required
divestitures of existing combinations pursuant to changes in media
ownership rules in “egregious cases.” 1975 Multiple Ownership Second

Report and Order, 50 F.C.C.2d at 1049.

030 NAB Comments in MM Docket No. 01-317 at 48; WVRC Comments in

MM Docket No. 01-317 at 26; Blakeney Comments in MM Docket No. 01-
317 at 2; Cox Comments in MM Docket No. 01-317 at 12; Daugherty
Comments in MM Docket No. 01-317 at 4; Davis Comments in MM Docket
No. 01-317 at 2; MBC Reply Comments in MM Docket No. 01-317 at 3; }
NABOB Reply Comments in MM Docket No. 01-317 at 8; Secret
Comments in MM Docket No. 00-244 at 3; NAB Comments in MM Docket
No. 00-244 at 8, 9, n.15; Brill Comments in MM Docket No. 00-244 at 1;
Aurora Comments in MM Docket No. 00-244 at 27; Great Scott Reply ,

Comments in MM Docket No. 00-244 at 2.
1031 |
At the same time, however, we believe that the impact on radio owners

542a

upon the contour-based definition that we find herein fails to
adequately address our competition goals in local radio
markets. To allow additional broadcasters to obtain such
combinations would disserve our goals. Our decision to
grandfather existing combinations simply reflects the
substantial equitable considerations discussed above,
considerations that we conclude outweigh our interest in
improving the precision of our radio market definition in these
particular cases.

487. Transferability. We also asked for comments on
whether to allow licensees to assign or to transfer control of
grandfathered combinations that violate of the new ownership
rules.'°** In general, we will prohibit the sale of existing
combinations that violate the modified local radio ownership
rule, the local television ownership rule, or the cross media
limits."°* Therefore, parties must comply with the new
ownership rules in place at the time a transfer of control or
assignment application is filed. However, as discussed earlier,
in order to

[Text truncated at 120,000 characters. The full text is on the page linked above.]

---

Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/brief%3Amicro_IA40386007_1337%3A4. Public record. Not legal advice.
