373 F. 3d 372 - Prometheus Radio Project v. Federal Communications Commission

                                 Table of Contents

  I.  Background .........................................................................382
      A.  The 1934 Communications Act and Early Broadcast Ownership
            Regulation ...................................................................382
      B.  Deregulation Initiatives .......................................................383
      C.  The Telecommunications Act of 1996 .............................................384
      D.  Regulatory Review Since 1996 ...................................................384
      E.  The Commission's 2003 Report and Order .........................................386
      F.  The Order's Modification of Broadcast Media Ownership Rules ....................386
          1.     Local Television Ownership ..............................................386
          2.     Local Radio Ownership ...................................................387
          3 & 4. Newspaper/Broadcast and Radio/Television Cross-Ownership ................387
          5.     National Television Ownership ...........................................388
          6.     Dual Network Rule .......................................................388
      G.  Procedural History of the Current Appeals ......................................388
      H.  Subsequent Legislation .........................................................389

 II.  Jurisdiction and Standard of Review ................................................389
      A.  Standard of Review Under the Administrative Procedure Act.......................389
      B.  Standard of Review Considerations Under Section 202(h)..........................390

          1.     "Determine whether any such rules are necessary in the public
                   interest." ............................................................391
          2.     "Repeal or modify any regulations it determines to be no longer
                   in the public interest." ..............................................394
      C.  Conclusion .....................................................................395

III.  Mootness and the National Television Ownership Rule ................................395

 IV.  Cross-Ownership Rules ..............................................................397
      A.  Regulatory Background and the 2002 Biennial Review .............................397
      B.  The Commission's decision not to retain a ban on newspaper/broadcast
            cross-ownership is justified under § 202(h) and is supported by record
            evidence......................................................................398
          1.     Newspaper/broadcast combinations can promote localism....................398
          2.     A blanket prohibition on newspaper/broadcast combinations is
                   not necessary to protect diversity ....................................399
      C.  The Commission's decision to retain some limits on common ownership of
            different-type media outlets was constitutional and did not violate
            § 202(h) .....................................................................400
          1.     Continuing to regulate cross-media ownership is in the public
                   interest...............................................................400
          2.     Continuing to regulate cross-media ownership does not violate
                   the Fifth Amendment....................................................401
          3.     Continuing to regulate cross-media ownership does not violate
                   the First Amendment. ..................................................401
      D.  The Commission did not provide reasoned analysis to support the specific
            Cross-Media Limits that it chose..............................................402
          1.     Overview of the Commission's Diversity Index Methodology ................403
          2.     The Commission did not justify its choice and weight of specific
                   media outlets..........................................................404
          3.     The Commission did not justify its assumption of equal market
                   shares.................................................................408
          4.     The Commission did not rationally derive its Cross-Media
                   Limits from the Diversity Index results ...............................409
          5.     The Commission should provide better notice on remand....................411

  V.  Local Television Ownership Rule ....................................................412
      A.  Regulatory Background and the 2002 Biennial Review .............................413
      B.  We uphold several threshold challenges to the Commission's overall
            regulatory approach...........................................................413
          1.     Limiting local television station ownership is not duplicative of
                   antitrust regulation ..................................................413
          2.     Media other than broadcast television may contribute to viewpoint
                   diversity in local markets ............................................414
          3.     Consolidation can improve local programming..............................415
          4.     The Commission adequately noticed its decision to allow
                   triopolies.............................................................416
      C.  We uphold the Commission's decision to retain the top-four restriction..........416
      D.  We remand the specific numerical limits for the Commission's further
            consideration ................................................................418
      E.  We remand the Commission's repeal of the Failed Station Solicitation
            Rule .........................................................................420

 VI.  Local Radio Ownership Rule .........................................................421
      A.  Regulatory Background and the 2002 Biennial Review .............................421
      B.  We uphold the Commission's new definition of local markets......................423
          1.     The Commission justified using Arbitron Metro markets....................423
          2.     The Commission justified including noncommercial stations................425
      C.  We uphold the Commission's transfer restriction.................................426

          1.     Transfer restriction is "in the public interest." .......................427
          2.     Transfer restriction is reasoned decisionmaking..........................427
          3.     Transfer restriction is constitutional...................................428
      D.  We affirm the attribution of Joint Sales Agreements.............................429
          1.     Attribution of JSAs is "necessary in the public interest."...............429
          2.     Attribution of JSAs is reasoned decisionmaking ..........................429
          3.     Attribution of JSAs is constitutional....................................430
      E.  We remand the numerical limits to the Commission for further
            justification.................................................................430
          1.     The Commission's numerical limits approach is rational and in
                   the public interest ...................................................431
          2.     The Commission did not support its decision to retain the
                   existing numerical limits with reasoned analysis.......................432
                 a.  The Commission did not sufficiently justified "five
                       equal-sized competitors" as the right benchmark ...................432
                 b.  The Commission did not sufficiently justified that the
                       existing numerical limits actually ensure that markets will
                       have five equal-sized competitors..................................433
                 c.  The Commission did not support its decision to retain the
                       AM/FM subcaps......................................................434

VII.  Conclusion .........................................................................435

In these consolidated appeals we consider revisions by the Federal Communications Commission to its regulations governing broadcast media ownership that the Commission promulgated following its 2002 biennial review. On July 2, 2003, the Commission announced a comprehensive overhaul of its broadcast media ownership rules. It increased the number of television stations a single entity may own, both locally and nationally; revised various provisions of the regulations governing common ownership of radio stations in the same community; and replaced two existing rules limiting common ownership among newspapers and broadcast stations (the newspaper/broadcast cross-ownership rule and the radio/television cross-ownership rule) with a single set of "Cross-Media Limits." See Report and Order and Notice of Proposed Rulemaking, 18 F.C.C.R. 13,620, 2003 WL 21511828 (2003) (the "Order").

Several public interest and consumer advocacy groups (collectively, the "Citizen Petitioners")1 petitioned for judicial review of the Order in various courts of appeals, contending that its deregulatory provisions contravened the Commission's statutory mandates as well as the Administrative Procedure Act, 5 U.S.C. §§ 551 et seq. (the "APA"). Associations of networks, broadcasters, and newspaper owners2 also challenged the Order, arguing that pro-regulatory revisions as well as the absence of further deregulation violate the Telecommunications Act of 1996, Pub. L. No. 104-104, 110 Stat. 56 (1996) (the "1996 Act"), the APA, and the United States Constitution. The Judicial Panel on Multidistrict Litigation, acting pursuant to the random selection procedures of 28 U.S.C. § 2112(a), consolidated the petitions in this Court. On September 3, 2003, we stayed implementation of the rules pending our review.

For the reasons stated below, we affirm the power of the Commission to regulate media ownership. In doing so, we reject the contention that the Constitution or § 202(h) of the 1996 Act somehow provides rigid limits on the Commission's ability to regulate in the public interest. But we must remand certain aspects of the Commission's Order that are not adequately supported by the record. Most importantly, the Commission has not sufficiently justified its particular chosen numerical limits for local television ownership, local radio ownership, and cross-ownership of media within local markets. Accordingly, we partially remand the Order for the Commission's additional justification or modification, and we partially affirm the Order. The stay will continue pending our review of the Commission's action on remand.3

A. The 1934 Communications Act and Early Broadcast Ownership Regulation

In 1934 Congress authorized the Commission to grant licenses for private parties' exclusive use of broadcast frequencies. Recognizing that the finite radio frequency spectrum inherently limits the number of broadcast stations that can operate without interfering with one another, Congress required that broadcast licensees serve the public interest, convenience, and necessity. Communications Act of 1934, 47 U.S.C. § 309(a); see also id. §§ 307(a), 310(d), 312.

"In setting its licensing policies, the Commission has long acted on the theory that diversification of mass media ownership serves the public interest by promoting diversity of program and service viewpoints, as well as by preventing undue concentration of economic power." FCC v. Nat'l Citizens Comm. for Broad., 436 U.S. 775, 780, 98 S.Ct. 2096, 56 L.Ed.2d 697 (1978) ("NCCB"). The Commission's early regulations reflected its presumption that a single entity holding more than one broadcast license in the same community contravened public interest. See Genesee Radio Corp., 5 F.C.C. 183, 186-87 (1938). In 1941, the Commission announced that it would not license more than one station in the same area to a single network organization. See Nat'l Broad. Co. v. United States, 319 U.S. 190, 206-08, 224-27, 63 S.Ct. 997, 87 L.Ed. 1344 (1943) (upholding the rule). At the same time, the Commission prohibited common ownership of stations within the same broadcast service (AM radio, FM radio, and television) in the same community. See Rules Governing Standard and High Frequency Broadcast Stations, 5 Fed. Reg. 2382, 2384 (June 26, 1940) (FM radio); Rules Governing Standard and High Frequency Broadcast Stations, 6 Fed. Reg. 2282, 2284-85 (May 6, 1941) (television); Rules Governing Standard and High Frequency Broadcast Stations, 8 Fed. Reg. 16065 (Nov. 27, 1943) (AM radio). Regulations limiting an entity to the common ownership of seven AM radio stations, seven FM radio stations, and seven television stations survived judicial scrutiny in 1956. See United States v. Storer Broad. Co., 351 U.S. 192, 76 S.Ct. 763, 100 L.Ed. 1081 (1956). In the 1970s the Commission adopted its first cross-ownership bans, which prohibited, on a prospective basis, the common ownership of television and radio stations serving the same market, as well as combinations of radio or broadcast stations with a daily newspaper in the same community. Amendment of Sections 73.35, 73.240 and 73.636 of the Commission Rules Relating to Multiple Ownership of Standard, FM and Television Broadcast Stations, 22 F.C.C.2d 306, ¶ 5, 1970 WL 18044 (1970); Amendment of Sections 73.34, 73.240 and 73.636 of the Commission's Rules Relating to Multiple Ownership of Standard, FM, and Television Broadcast Stations, 50 F.C.C.2d 1046, 1975 WL 30457 (1975). The Supreme Court upheld the newspaper/broadcast cross-ownership ban as a "reasonable means of promoting the public interest in diversified mass communications." NCCB, 436 U.S. at 802, 98 S.Ct. 2096.

The 1980s saw a deregulatory trend for media ownership. The Commission raised its national ownership limits to permit common ownership of 12 stations in each broadcast service (though still prohibiting station combinations that would reach more than 25% of the national audience). Amendment of Section 73.3555 (formerly 73.35, 73.240, and 73.636) of the Commission's Rules Relating to Multiple Ownership of AM, FM, and Television Broadcast Stations, 100 F.C.C.2d 74 ¶¶ 38, 39, 1985 WL 260060 (1985). The Commission also determined that UHF television stations should not be deemed to have the same audience reach as VHF stations,4 due to "the inherent physical limitations of [the UHF] medium," and therefore applied a 50% "discount" to UHF audiences as counted under the national audience limitation. Id. ¶¶ 42-44. In other words, UHF stations count only half of their audiences in determining compliance with the national television ownership rule.

In 1989 the Commission eased its "one to a market" radio/television cross-ownership rules by allowing waiver requests for radio/television cross-ownership in the 25 largest television markets. The Commission stated that it would look favorably upon requests for waivers where there would be 30 independently owned broadcast "voices" remaining in the market after consolidation. Amendment of Section 73.3555 of the Commission's Rules, the Broadcast Multiple Ownership Rules, 4 F.C.C.R. 1741, ¶ 1, 1989 WL 510875 (1989). In 1992, the Commission relaxed local and national radio ownership restrictions and adopted a tiered approach to radio concentration that allowed a single entity to own more radio stations in the largest markets (up to three AM and three FM stations, subject to a local audience reach limitation of 25% and a national cap of 30 AM stations and 30 FM stations) and fewer in the smallest markets. Revision of Radio Rules and Policies, 7 F.C.C.R. 6387, ¶ 27, 1992 WL 690638 (1992).

In 1996 Congress overhauled the Communications Act by enacting the Telecommunications Act of 1996. The 1996 Act contemplated a "pro-competitive, deregulatory national policy framework designed to accelerate rapidly private sector development of advanced telecommunications and information technologies and services to all Americans by opening all telecommunications markets to competition." S.Rep. No. 104-230, at 1-2 (1996). The 1996 Act eliminated all limits on national radio ownership and raised the national television audience reach cap from 25% to 35%. 1996 Act §§ 202(a), (c)(1)(B), 110 Stat. at 110-11. Congress also eased local radio ownership limits, establishing a four-tier sliding scale limit of numerical caps that allowed for as many as eight co-owned radio stations in the largest markets. Id. § 202(b)(1), 110 Stat. at 110.

The 1996 Act did not contain a new local television rule, but it directed the Commission to "conduct a rulemaking proceeding to determine whether to retain, modify, or eliminate" its existing local television ownership limitations. Id. § 202(c)(2), 110 Stat. at 111. It also expanded the applicability of the one-to-a-market radio/television cross-ownership restriction waiver to the fifty largest markets. Id. § 202(d), 110 Stat. at 111.

Finally, the 1996 Act instructed the Commission to review biennially its broadcast ownership rules "to determine whether any of such rules are necessary in the public interest as the result of competition." Id. § 202(h), 110 Stat. at 111-12. Section 202(h) also required the Commission to "repeal or modify any regulation it determines to be no longer in the public interest." Id.

In 1999 the Commission responded to Congress's directive under § 202(c) of the 1996 Act to review its local television rule and announced that it would relax its prohibition on the common ownership of television stations with overlapping signals. The Commission's new rule would allow two commonly owned television stations (a television station "duopoly") in the same Designated Market Area ("DMA" or "market") as long as (1) neither station was ranked among the four largest ("top-four") stations in the market and (2) eight independently owned stations remained in the market post-merger. Review of the Commission's Regulations Governing Television Broadcasting, 14 F.C.C.R. 12,903, ¶ 8, 1999 WL 591820 (1999) ("1999 Television Rule Review"). In the same rulemaking, the Commission also relaxed the one-to-a-market radio/television cross-ownership restriction and allowed radio/television station combinations to exist within three-tiered limits that depend on the size of the market. Id. ¶ 9.5

Meanwhile, in 1998 the Commission began the first biennial review of its broadcast ownership regulations as required under § 202(h). In 2000 it announced that it would retain the national television ownership rule (the 35% limit provided in the 1996 Act) and its cable/broadcast cross-ownership rule after determining that both rules remained "necessary in the public interest." 1998 Biennial Regulatory Review — Review of the Commission's Broadcast Ownership Rules and Other Rules Adopted Pursuant to Section 202 of the Telecommunications Act of 1996, 15 F.C.C.R. 11,058, ¶ 4, 2000 WL 791562 (2000) ("1998 Biennial Regulatory Review").6 On appeal, however, the United States Court of Appeals for the D.C. Circuit held that the Commission had not sufficiently explained its reasons for retaining either of these rules. Fox Television Stations v. FCC, 280 F.3d 1027, 1043-44, 1051-52 (D.C.Cir.2002) ("Fox I"), modified on reh'g, 293 F.3d 537 (D.C.Cir.2002) ("Fox II"). As for the national television ownership rule, the Court determined that the agency had taken a "wait and see" approach in evaluating its competition and diversity effects, which was impermissible in light of § 202(h)'s mandate to repeal or modify rules found no longer necessary in the public interest. Id. at 1042. It remanded the rule to the Commission for additional justification. Id. at 1049. The Court vacated the cable/broadcast cross-ownership rule, however, finding that the Commission's decision to retain it was arbitrary and capricious and contrary to § 202(h). Id. at 1053.

A few months later, the same Court reviewed the local television multiple-ownership rule. Sinclair Broad. Group, Inc. v. FCC, 284 F.3d 148 (D.C.Cir.2002). The petitioner challenged the "eight independent voices" exception, contending that it lacked foundation or connection to the Commission's goal of promoting diversity in local markets. Id. at 152. The Court determined that the Commission had adequately justified its decision to retain the local television rule under the APA and § 202(h), but that it had not provided a rational basis for the exclusion of non-broadcast media from the eight voices exception. Id. at 165. It remanded the rule for the Commission's further justification. Id. at 169.

In September 2002 a Notice of Proposed Rulemaking (the "Notice") announced that the Commission would review four of its broadcast ownership rules pursuant to § 202(h): the 35% national audience reach limit remanded in Fox; the local television rule remanded in Sinclair; the radio/television cross-ownership rule; and the dual network rule. 2002 Biennial Regulatory Review Notice of Proposed Rulemaking, 17 F.C.C.R. 18,503, ¶ 6, 2002 WL 31108252 (2002). The Notice also advised that the Commission was incorporating into its biennial review pending proceedings on two additional rules: its rule limiting radio station ownership in local markets and its rule prohibiting newspaper/broadcast cross-ownership. Id. ¶ 7.

The Commission established a Media Ownership Working Group ("MOWG"), which commissioned twelve studies ranging from consumer surveys to economic analyses of media markets. These reports were released for public comment in October 2002. Interested parties filed thousands of pages of comments, consisting of legal, social, and economic analyses, empirical and anecdotal evidence, and industry and consumer data to respond to the issues identified in the Commission's Notice. Notably, nearly two million people weighed in by letters, postcards, e-mails, and petitions to oppose further relaxation of the rules. Statement of Commissioner Jonathan S. Adelstein, Dissenting, 18 F.C.C.R. 13,974, 13,977, 2003 WL 21251887 (July 2, 2003). The Commission also heard public comment at a February 2003 "field hearing" in Richmond, Virginia.7

On June 2, 2003, the Commission adopted the Order modifying its ownership rules to provide a "new, comprehensive framework for broadcast ownership regulation" by a vote of 3-2.8 Order ¶ 3. The Order was released on July 2, 2003.

F. The Order's Modification of Broadcast Media Ownership Rules

After reaffirming the Commission's three traditional policy objectives in promoting the public interest — competition, diversity, and localism — Order ¶ 8, the Commission considered whether each of the six rules remained in the public interest and proposed modifications where it believed necessary. With respect to each of the rules, the Commission determined as follows.

The existing local television ownership rule allowed television station duopolies, so long as at least one of the stations was not ranked among the market's four largest stations and so long as at least eight independently owned and operated full-power television stations would remain in the market post-merger. 47 C.F.R. § 73.3555(b). The Order modified this rule to permit television station triopolies in markets with 18 or more television stations and television station duopolies in markets with 17 or fewer television stations. Order ¶ 134. These limits are subject to the restriction (effectively a ban subject to a waiver provision) that a single firm may not own more than one top-four station in a market. This restriction forecloses common station ownership in markets with fewer than five television stations. Id. ¶ 186. The existing rule effectively precludes duopolies in most markets; only the largest 70 markets of the nation's 210 DMAs could comply with the "eight voices" test. Adelstein Dissenting, 18 F.C.C.R. at 13,998. The new rule would allow triopolies in the nine largest DMAs, which represent 25.2% of the population. Duopolies could exist under the new rule in the largest 162 markets, representing 95.4% of the nation's population. Id. at 13,997-98.

Although the Order retained the existing numerical limits on radio ownership that Congress established in § 202(b) of the 1996 Act,9 it modified other aspects of the rule. First, it changed the method for determining radio markets by replacing the "contour-overlap" method (described in detail in Part VI.B infra) with the geography-based market delineations created by Arbitron, a company that generates market data for radio advertisers. Order ¶ 239. Additionally, the Commission would now include noncommercial stations in the station count for each market. Id. The Commission grandfathered any existing radio station combinations rendered noncompliant under the newly defined markets, id. ¶ 484, but generally restricted transfer of these combinations, id. ¶ 487. The Commission also changed the local radio ownership rule by deciding to attribute "Joint Sales Agreements" (agreements under which a licensee sells advertising time on its station to a broker station for a fee) toward the brokering entity's numerical limit. Id. ¶ 239.

3 & 4. Newspaper/Broadcast and Radio/Television Cross-Ownership

The Commission has prohibited common ownership of a full-service television broadcast station and a daily public newspaper in the same community since 1975. Amendment of Sections 73.35, 73.240, and 73.636 of the Commission's Rules Relating to Multiple Ownership of Standard, FM, and Television Broadcast Stations, 50 F.C.C.2d 1046, 1975 WL 30457 (1975). Additionally, the Commission regulates the number of television and radio stations that may be commonly owned with limits that vary with the size of the market. 47 C.F.R. § 73.3555(c).

The Order announced the Commission's decision to repeal both cross-ownership rules (television/newspaper, radio/television) and replace them with a single set of Cross-Media Limits. The Commission determined that neither cross-ownership prohibition remained necessary in the public interest to ensure competition, diversity, or localism. Order ¶¶ 330, 371. The new Cross-Media Limits prohibit newspaper/broadcast combinations and radio/television combinations in the smallest DMAs, i.e., those with three or fewer full-power commercial or noncommercial television stations. Id. ¶ 454. In contrast, in the largest markets — those with more than eight television stations — common ownership among newspapers and broadcast stations is unrestricted. Id. ¶ 473. In medium-sized markets — those with between four and eight television stations — one entity may own a newspaper and either (a) one television station and up to 50% of the radio stations that may be commonly owned in that market under the local radio rule or (b) up to 100% of the radio stations allowed under the local radio rule. Id. ¶ 466.

In structuring its Cross-Media Limits, the Commission drew upon a methodological tool named the "Diversity Index," which the Commission developed as a measure of viewpoint diversity in local markets to identify those "at-risk" markets where consolidation would have a deleterious effect. Id. ¶¶ 391, 442. The Diversity Index, explained more fully in Part IV.D.1 below, is a highly modified version of the formula for measuring market concentration — the Herfindahl-Hirschman Index — applied by the Department of Justice and Federal Trade Commission to analyze mergers. Id. ¶ 428.

The national television ownership rule prohibits entities from owning television stations that in the aggregate reach a certain percentage of our country's households. 47 C.F.R. § 73.3555(e)(1). Section 202(c) of the 1996 Act directed the Commission to delete the then-existing twelve-station cap and raise the audience reach limit from 25% to 35%. After the D.C. Circuit Court remanded the Commission's decision in the 1998 biennial review to retain the limit at 35%, Fox I, 280 F.3d at 1049, the Commission decided to increase the audience reach limit to 45%. Order ¶ 499. The Commission also declined to repeal or modify its existing 50% discount for UHF stations' audiences as counted toward the audience reach limit. Id. ¶ 500.

Under the dual network rule, a television station may affiliate with more than one network except that it may not affiliate with more than one of the four largest networks, ABC, CBS, Fox, and NBC. 47 C.F.R. § 73.658(g). The rule effectively permits common ownership of networks to the exclusion of the top four. Order ¶ 592. The Commission determined that the dual network rule remained necessary in the public interest, and thus did not repeal or modify it. Id.10

G. Procedural History of the Current Appeals

Within days of the publication of the Order, several organizations filed petitions for review of the Commission's revised rules in various courts of appeals, some contending that the Commission had gone too far in revising the rules, and others asserting that the Commission had not gone far enough. Some of these organizations, including the Prometheus Radio Project, filed their petitions in this Court. Under 28 U.S.C. § 2112(a), petitions to review administrative orders filed in different circuit courts within the first ten days of the appeal period trigger a lottery conducted by the Judicial Panel on Multidistrict Litigation. On August 19, 2003, the Panel announced that our Court had been selected in the lottery and consolidated the appeals here. We entered a stay of the effective date of the proposed rules after a hearing on September 3, 2003. Prometheus Radio Project v. FCC, No. 03-3388, 2003 WL 22052896 (3d Cir. Sept. 3, 2003). We then denied the Deregulatory Petitioners' motion, joined by the Commission, to transfer venue to the D.C. Circuit Court on September 16, 2003. Prometheus Radio Project v. FCC, No. 03-3388 (3d Cir. Sept. 16, 2003) (order denying motion to transfer). After pushing back briefing and oral argument at the request of the parties, on February 11, 2004, we heard approximately eight hours of oral argument addressing the merits of Petitioners' claims.

In January 2004, while the petitions to review the Order were pending in this Court, Congress amended the 1996 Act by increasing from 35% to 39% the national television ownership rule's audience reach cap in § 202(c). Consolidated Appropriations Act, 2004, Pub. L. No. 108-199, § 629, 118 Stat. 3, 99 (2004). The legislation also amended § 202(h) in two ways: (1) making the Commission's biennial review obligation quadrennial; and (2) insulating from § 202(h) review "rules relating to the 39 percent national audience reach limitation." 118 Stat. at 100. Prior to oral argument, Petitioners filed letter briefs addressing the effect of these amendments on their challenges to the Order.

This is an appeal of an agency decision under the Communications Act of 1934, 47 U.S.C. §§ 151 et seq. Our jurisdiction is based on 47 U.S.C. § 402(a) and 28 U.S.C. § 2342(1). Our standard of review is governed by the APA and the 1996 Act provision authorizing the Commission's periodic regulatory review.11

A. Standard of Review Under the Administrative Procedure Act

Our standard of review in the agency rulemaking context is governed first by the judicial review provision of the APA, 5 U.S.C. § 706. Under it, we "hold unlawful or set aside agency action, findings, and conclusions" that are found to be "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with the law ... [or] unsupported by substantial evidence." Id. § 706(2)(a), (e); N.J. Coalition for Fair Broad. v. FCC, 574 F.2d 1119, 1125 (3d Cir.1978).

The scope of review under the "arbitrary and capricious" standard is "narrow, and a court is not to substitute its judgment for that of the agency." Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983) ("State Farm"). Nevertheless, we must ensure that, in reaching its decision, the agency examined the relevant data and articulated a satisfactory explanation for its action, including a "rational connection between the facts found and the choice made." Id. (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 9 L.Ed.2d 207 (1962)). Normally, we may find an agency rule is arbitrary and capricious where

the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise. The reviewing court should not attempt itself to make up for such deficiencies; we may not supply a reasoned basis for the agency's action that the agency itself has not given.

Id. (citing SEC v. Chenery Corp., 332 U.S. 194, 196, 67 S.Ct. 1575, 91 L.Ed. 1995 (1947)); see also Robert Wood Johnson Univ. Hosp. v. Thompson, 297 F.3d 273, 280 (3d Cir.2002). Put another way, we reverse an agency's decision when it "is not supported by substantial evidence, or the agency has made a clear error in judgment." AT&T Corp. v. FCC, 220 F.3d 607, 616 (D.C.Cir.2000) (citing Kisser v. Cisneros, 14 F.3d 615, 619 (D.C.Cir.1994)).

We will, however, "uphold a decision of less than ideal clarity if the agency's path may reasonably be discerned." State Farm, 463 U.S. at 43, 103 S.Ct. 2856 (quoting Bowman Transp., Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286, 95 S.Ct. 438, 42 L.Ed.2d 447 (1974)). But an agency that departs from its "former views" is "obligated to supply a reasoned analysis for the change beyond that which may be required when an agency does not act in the first instance" in order to survive judicial scrutiny for compliance with the APA. State Farm, 463 U.S. at 41-42, 103 S.Ct. 2856.

Finally, the traditional APA standard of review is even more deferential "where the issues involve `elusive' and `not easily defined' areas such as programming diversity in broadcasting." Sinclair, 284 F.3d at 159. Yet even when an administrative order involves policy determinations on such elusive goals, a "rationality" standard is appropriate. See NCCB, 436 U.S. at 796-97, 98 S.Ct. 2096 (finding that the Commission acted rationally in determining that diversification of ownership would enhance the possibility of increasing diverse viewpoints). Additionally, when an agency has engaged in line-drawing determinations and our review is necessarily deferential to agency expertise, see AT&T Corp., 220 F.3d at 627, its decisions may not be "patently unreasonable" or run counter to the evidence before the agency. Sinclair, 284 F.3d at 162.

B. Standard of Review Considerations Under Section 202(h)

The Order was promulgated as part of the periodic review requirements of § 202(h) of the 1996 Act. Consequently, our review standard is informed by that provision, which, at the time of the Order's release, read:

(h) Further Commission Review. The Commission shall review its rules adopted pursuant to this section and all of its ownership rules biennially12 as part of its regulatory reform review under section 11 of the Communications Act of 1934 and shall determine whether any of such rules are necessary in the public interest as the result of competition. The Commission shall repeal or modify any regulation it determines to be no longer in the public interest.

110 Stat. 111-12. Section 11 of the Communications Act, to which § 202(h) refers, was also added by the 1996 Act to ensure that the Commission review periodically its regulations governing telecommunications services to "determine whether any such regulation is no longer necessary in the public interest as a result of meaningful economic competition between providers of such service" and "repeal or modify any regulation it determines to be no longer necessary in the public interest." 47 U.S.C. § 161.

The text and legislative history of the 1996 Act indicate that Congress intended periodic reviews to operate as an "ongoing mechanism to ensure that the Commission's regulatory framework would keep pace with the competitive changes in the marketplace" resulting from that Act's relaxation of the Commission's regulations, including the broadcast media ownership regulations. 2002 Biennial Regulatory Review, 18 F.C.C.R. 4726, ¶¶ 16, 17 (2003) (citing preamble to the 1996 Act; H.R. Conf. Rep. No. 104-458 (1996)). Put another way, the periodic review provisions require the Commission to "monitor the effect of... competition ... and make appropriate adjustments" to its regulations. Id. ¶ 5.

As noted, the first sentence of § 202(h) requires the Commission to "determine" whether media concentration rules are "necessary in the public interest as the result of competition." The second sentence contains a separate instruction to the Commission: to "repeal or modify" those rules "no longer in the public interest." 110 Stat. 111-12. We analyze each of these instructions in turn.

1. "Determine whether any such rules are necessary in the public interest."

Recognizing that competitive changes in the media marketplace could obviate the public necessity for some of the Commission's ownership rules, the first instruction requires the Commission to take a fresh look at its regulations periodically in order to ensure that they remain "necessary in the public interest." This raises the question of what is "necessary."

In the context of § 11 of the Communications Act, 47 U.S.C. § 161 — which, like § 202(h), requires the Commission periodically to review its telecommunications regulations and determine whether they "remain necessary in the public interest" — the Commission has interpreted "necessary" to mean "useful," "convenient" or "appropriate" rather than "required" or "indispensable." Setting out its rationale for this interpretation in the 2002 Biennial Regulatory Review, 18 F.C.C.R. 4726, ¶¶ 14-22, 2003 WL 1192543 (2003), the Commission determined that the 1996 Act's legislative history indicated that Congress meant "no longer necessary" to mean "no longer in the public interest" and "no longer meaningful." 18 F.C.C.R. 4726, ¶ 17 (citing H.R. Conf. Rep. No. 104-458, at 185 (1996)).

Next, the Commission found that an "indispensable" construction of "necessary" as to § 11 would be unreasonably inconsistent with the Communications Act's grant of general rulemaking authority to the Commission. Id. at ¶ 18 n. 31. Under 47 U.S.C. § 201(b) the Commission is authorized to "prescribe such rules and regulations [regarding services and charges of communications common carriers] as may be necessary in the public interest to carry out the provisions of this Act." In AT&T Corp. v. Iowa Utilities Board, the Supreme Court interpreted this provision as a grant of "general rulemaking authority." 525 U.S. 366, 374, 119 S.Ct. 721, 142 L.Ed.2d 835 (1999). Characterizing this interpretation "not as a limitation on the Commission's authority, but a confirmation of it," the Commission concluded that the standard of review applicable to its rulemaking authority under § 201 is a "plain public interest" standard. 18 F.C.C.R. 4276, ¶¶ 18 n. 31, 22 (citing 525 U.S. at 374, 378, 119 S.Ct. 721). The Commission reasoned that the same standard must also apply to the review process required under § 11 in order to avoid absurd results. If the rulemaking and review standards were different, the Commission could promulgate any rule that is useful, but then, at the next periodic review, would have to revoke any of those rules that do not also meet a higher standard of "indispensable." Id. ¶ 18 & n. 33. Under such a system, periodic review would either be inefficient or irrelevant, as the Commission could effectively sidestep the more stringent review standard by subsequently reissuing any "useful" rule that it had to repeal for failing to be "indispensable." Id. ¶ 18.

Lastly, the Commission rejected arguments that there is controlling judicial precedent for an "indispensable" construction of "necessary." It acknowledged that the Supreme Court and the D.C. Circuit Court of Appeals have upheld such constructions, see id. ¶ 19 (citing Iowa Utils. Bd., 525 U.S. at 374, 378, 119 S.Ct. 721; GTE Serv. Corp. v. FCC, 205 F.3d 416 (D.C.Cir.2000)), but countered that these cases "simply demonstrate that terms such as `necessary'... must be read in their statutory context." Id. Furthermore, the Commission found judicial support for its interpretation of "necessary" in the D.C. Circuit Court's decision in Sinclair, 284 F.3d at 159, regarding the Commission's periodic review of its local television ownership rule under § 202(h). The Commission noted that the Sinclair Court did not expressly adopt any particular definition of "necessary," but, in affirming the Commission's § 202(h) finding that the rule furthers diversity and is thus necessary in the public interest, id. at 160, it "did not articulate a new or higher public interest yardstick." 18 F.C.C.R. 4726, ¶ 20.13 Nor did the D.C. Circuit Court's decision in Fox foreclose its interpretation of "necessary," the Commission determined, because on rehearing the Court deleted language in its initial decision that the Commission had applied too lax a standard in reviewing its broadcast media ownership rules under § 202(h).14 Id. ¶ 14 (citing Fox II, 293 F.3d at 540).

For these reasons, the Commission determined that § 11's requirement that it review its telecommunications regulations to determine whether they remain "necessary in the public interest" does not require it to employ a more stringent standard than "plain public interest" found in other parts of the Communications Act. Id. ¶¶ 18, 22 (citing 47 U.S.C. § 201(b) as an example).

Recently, the D.C. Circuit Court upheld, in the context of § 11, the Commission's interpretation of "necessary" contained in the 2002 Biennial Regulatory Review. Cellco P'ship v. FCC, 357 F.3d 88 (D.C.Cir.2004). Recognizing that "necessary" is a "chameleon-like" word whose "meaning... may be influenced by its context," the Cellco Court determined that it would uphold any reasonable interpretation that did not contravene the express provisions of the Communications Act. Id. at 94, 96 (citing Chevron U.S.A., Inc. v. Natural Res. Def. Council, Inc., 467 U.S. 837, 842-43, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984)). It went on to determine that the Commission's interpretation is both reasonable and consistent with the Communications Act, endorsing the Commission's view that "necessary" must mean the same thing in the periodic review context as in the rulemaking context in order to avoid absurd results. Id. at 98.

Cellco also acknowledged that the Commission's interpretation of "necessary" is consistent with the many courts that have endorsed a "useful" or "appropriate" interpretation over an "essential" or "indispensable" one. Id. at 97 (citing, inter alia, NCCB, 436 U.S. at 795-96 98 S.Ct. 2096; McCulloch v. Maryland, 17 U.S. (4 Wheat.) 316, 413, 4 L.Ed. 579 (1819); Cellular Telecomm. & Internet Ass'n v. FCC, 330 F.3d 502, 510 (D.C.Cir.2003) (specifically rejecting an "indispensable" connotation of "necessary" as used in the Communications Act's enforcement forbearance provision, § 10(a))).

Finally, the Cellco Court rejected suggestions that the Commission's interpretation was inconsistent with its prior decisions in Sinclair and Fox. As noted above, Sinclair did not expressly adopt any particular definition of "necessary" and Fox I's suggestion of a heightened standard was expressly retracted by Fox II, 293 F.3d at 540. Cellco limited Fox I's statement that "necessary" implied a presumption in favor of modification or elimination of existing regulations, see 280 F.3d at 1048, to the context in which it was made: discussing whether vacating or remanding the national television ownership rule was the appropriate remedy. Cellco, 357 F.3d at 98. And while Sinclair apparently endorsed this language from Fox I, see 284 F.3d at 159, the Cellco Court characterized Sinclair as merely "piggyback[ing]" on Fox I without "adopt[ing] a general presumption in favor of modification or elimination of regulations when considering a substantive challenge to the adequacy of the Commission's determinations." Cellco, 357 F.3d at 98. In sum, the D.C. Circuit Court determined that the definition of "necessary" was not constrained by either its Fox or Sinclair decision. It remained an open issue for the Commission to decide in the first instance, as it did when it released the 2002 Biennial Regulatory Review. Id.

For the same reasons proffered by the Commission and endorsed by the D.C. Circuit Court to reject the "indispensable" definition of "necessary" under § 11, we do so under § 202(h). Though § 11 and § 202(h) are separate statutory provisions, they are both periodic review provisions from the same statute. As evidence of their relatedness, § 202(h) imposes periodic review obligations "as part of its regulatory reform review under section 11 of this Communications Act of 1934," 110 Stat. at 111.15 We see no reason to adopt a different definition of "necessary" under § 202(h) than under § 11. Moreover, interpreting § 202(h)'s first sentence to require the Commission to review its rules to determine whether they are indispensable in the public interest would lead to incongruous results when compared to the instruction in § 202(h)'s second sentence, which requires the Commission to "repeal or modify any regulation it determines to be no longer in the public interest."16 For the "determine" instruction to be meaningful, "necessary" must embody the same "plain public interest" standard that Congress set out in the "repeal or modify" instruction.17 Lastly, as explained by the Cellco Court, the "convenient," "useful," or "helpful" definition of "necessary" is not foreclosed to the Commission by any judicial precedent, including Fox and Sinclair. So in interpreting the Commission's obligation under § 202(h) to review its broadcast media ownership rules to determine whether they are "necessary in the public interest," we adopt what the Commission termed "the plain public interest" standard under which "necessary" means "convenient," "useful," or "helpful," not "essential" or "indispensable."

2. "Repeal or modify any regulations it determines to be no longer in the public interest."

Turning to the second instruction of § 202(h), the Commission is required to "repeal or modify" rules that are "no longer in the public interest." Having concluded that the first instruction requires the Commission to determine whether any existing rule fails to satisfy the "plain public interest" standard, the relationship between the first and second instruction is evident. Under the second instruction, the Commission must repeal or modify the regulations that it has determined under the first instruction do not satisfy that same standard.

While we acknowledge that § 202(h) was enacted in the context of deregulatory amendments (the 1996 Act) to the Communications Act, see Fox I, 280 F.3d at 1033; Sinclair, 284 F.3d at 159, we do not accept that the "repeal or modify in the public interest" instruction must therefore operate only as a one-way ratchet, i.e., the Commission can use the review process only to eliminate then-extant regulations. For starters, this ignores both "modify" and the requirement that the Commission act "in the public interest." What if the Commission reasonably determines that the public interest calls for a more stringent regulation? Did Congress strip it of the power to implement that determination? The obvious answer is no, and it will continue to be so absent clear congressional direction otherwise.18

What, then, makes § 202(h) "deregulatory"? It is this: Section 202(h) requires the Commission periodically to justify its existing regulations, an obligation it would not otherwise have. A regulation deemed useful when promulgated must remain so. If not, it must be vacated or modified.

Misguided by the Fox and Sinclair Courts' "deregulatory presumption" characterization and lacking the benefit of Cellco's subsequent clarification, the Commission concluded that § 202(h) "appears to upend traditional administrative law principles" by not requiring it to justify affirmatively a rule's repeal or modification. Order ¶ 11. This overstates the case. Rather than "upending" the reasoned analysis requirement that under the APA ordinarily applies to an agency's decision to promulgate new regulations (or modify or repeal existing regulations), see State Farm, 463 U.S. at 43, 103 S.Ct. 2856, § 202(h) extends this requirement to the Commission's decision to retain its existing regulations. This interpretation avoids a crabbed reading of the statute under which we would have to infer, without express language, that Congress intended to curtail the Commission's rulemaking authority and to contravene "traditional administrative law principles."19

Though our standard of review analysis is lengthy, it is in the end amenable to a straightforward summing-up: In a periodic review under § 202(h), the Commission is required to determine whether its then-extant rules remain useful in the public interest; if no longer useful, they must be repealed or modified.20 Yet no matter what the Commission decides to do to any particular rule — retain, repeal, or modify (whether to make more or less stringent) — it must do so in the public interest and support its decision with a reasoned analysis. We shall evaluate each aspect of the Commission's Order accordingly.

III. Mootness and the National Television Ownership Rule

The national television ownership rule caps the number of television stations that a single entity may own on a national basis. In the 1996 Act, Congress limited the number of commonly owned stations to those reaching no more than 35% of the national audience.1996 Act § 202(c)(1)(B), 110 Stat. 111. In its first biennial review, the Commission retained the 35% cap as "necessary in the public interest," but the D.C. Circuit Court held that the Commission had not adequately justified this decision. Fox I, 280 F.3d at 1048. On remand and in connection with its 2002 biennial review proceeding, the Commission increased the cap from 35% to 45%. Order ¶ 583. The Commission also decided to retain its method of discounting by 50% the audiences of UHF stations toward the cap. Id. ¶ 586.

Subsequently, however, Congress enacted a new national television ownership cap. In its 2004 Consolidated Appropriations Act, it modified § 202(c)(1)(B) of the 1996 Act to provide that "[t]he Commission shall modify its rules for multiple ownership ... by increasing the national audience reach limitation for television stations to 39%." See Pub. L. No. 108-199, § 629, 118 Stat. 3, 99 (2004). Because the Commission is under a statutory directive to modify the national television ownership cap to 39%, challenges to the Commission's decision to raise the cap to 45% cap are moot.21 Cf. PLMRS Narrowband Corp. v. FCC, 182 F.3d 995, 1002 (D.C.Cir.1999) (challenges to Commission's later-modified order are moot).

Although the 2004 Consolidated Appropriations Act did not expressly mention the UHF discount, challenges to the Commission's decision to retain it are likewise moot. Congress instructed the Commission to "increase the national audience reach limitation for television stations to 39%." 118 Stat. at 99. Since 1985 the Commission has defined "national audience reach" to mean "the total number of television households" reached by an entity's stations, except that "UHF stations shall be attributed with 50 percent of the television households" reached. 47 C.F.R. § 73.3555(e)(2)(i); Multiple Ownership of AM, FM and Television Broadcast Stations, 50 Fed. Reg. 4666, 4676, 1985 WL 260060 (Feb. 1, 1985). We assume that when Congress uses an administratively defined term, it intended its words to have the defined meaning. See, e.g., Bragdon v. Abbott, 524 U.S. 624, 631, 118 S.Ct. 2196, 141 L.Ed.2d 540 (1998). Furthermore, because reducing or eliminating the discount for UHF station audiences would effectively raise the audience reach limit, we cannot entertain challenges to the Commission's decision to retain the 50% UHF discount. Any relief we granted on these claims would undermine Congress's specification of a precise 39% cap.

As additional evidence of the mootness of challenges to the UHF discount, we note that the 2004 Consolidated Appropriations Act also added a sentence to § 202(h): "This subsection does not apply to any rules relating to the 39% national audience limitation." 118 Stat. at 100. The UHF discount is a rule "relating to" the national audience limitation. See 47 C.F.R. § 73.3555(e)(2) (providing for the UHF discount in a section qualified "for the purposes of this paragraph (e)," the national television ownership rule paragraph). Congress apparently intended to insulate the UHF discount from periodic review, a position that is consistent with our reading of the legislation as endorsing the almost 20-year-old regulatory definition of "national audience reach" that provides for the UHF discount.

Although we find that the UHF discount is insulated from this and future periodic review requirements, we do not intend our decision to foreclose the Commission's consideration of its regulation defining the UHF discount in a rulemaking outside the context of Section 202(h). The Commission is now considering its authority going forward to modify or eliminate the UHF discount and recently accepted public comment on this issue. 69 Fed. Reg. 9216-17 (Feb. 27, 2004). Barring congressional intervention, see, e.g., S. 1264, 108th Cong. § 12 (2003) (proposing phase-out and 2008 sunset of the UHF), the Commission may decide, in the first instance, the scope of its authority to modify or eliminate the UHF discount outside the context of § 202(h).

The Commission's decision to repeal its newspaper/broadcast cross-ownership rules22 in favor of new Cross-Media Limits has been attacked on all fronts. Some petitioners support the repeal but argue that the Cross-Media Limits are too restrictive. Others challenge the repeal decision and argue that the new limits are too lenient. We conclude that the Commission's decision to replace its cross-ownership rules with the Cross-Media Limits is not of itself constitutionally flawed and does not violate § 202(h). But we cannot uphold the Cross-Media Limits themselves because the Commission does not provide a reasoned analysis to support the limits that it chose.

A. Regulatory Background and the 2002 Biennial Review

Since the 1970s, the Commission has enforced two separate limits on the common ownership of different-type media outlets in local markets. One cross-ownership rule prohibits the common ownership of a full-service television broadcast station and a daily public newspaper in the same community. 47 C.F.R. § 73.3555(d). The other limits the number of television and radio stations to the following combinations: (1) in markets where at least 20 independently owned media voices would remain post-merger, two television stations and six radio stations or one television station and seven radio stations; (2) in markets where at least 10 independent voices would remain, two television stations and four radio stations; and (3) in other markets, two television stations (subject to the local television ownership rule) and one radio station. Id. § 73.3555(c).

The Commission considered both cross-ownership rules during its 2002 biennial review under § 202(h). In the Order, the Commission announced that because neither rule remained necessary in the public interest, it was repealing them and replacing them with a single set of Cross-Media Limits. The three-tiered Cross-Media Limits regulate common ownership depending on the size of the market: small (those with three or fewer full-power commercial or noncommercial television stations), mid-sized (between four and eight television stations), and large (more than eight television stations). In small markets, newspaper/broadcast combinations and radio/television combinations are prohibited. Order ¶ 454. In medium-sized markets, an entity may own a newspaper and either (a) one television station and up to 50% of the radio stations that may be commonly owned in that market under the local radio rule or (b) up to 100% of the radio stations allowed under the local rule. Id. ¶ 466. In large markets, cross-ownership is unrestricted. Id. ¶ 473.

B. The Commission's decision not to retain a ban on newspaper/broadcast cross-ownership is justified under § 202(h) and is supported by record evidence.

The Commission determined that the rule prohibiting newspaper/broadcast cross-ownership was no longer necessary in the public interest for three primary reasons: (1) the ban is not necessary to promote competition in local markets because most advertisers do not view newspapers and television stations as close substitutes, Order ¶ 332; (2) the ban undermines localism by preventing efficient combinations that would allow for the production of high-quality local news, id. ¶ 343; and (3) there is not enough evidence to conclude that ownership influences viewpoint to warrant a blanket cross-ownership ban, thus making it unjustifiable on diversity grounds, id. at ¶ 364, (and moreover, the presence of other media sources — such as the Internet and cable — compensate for the viewpoint diversity lost to consolidation, id. ¶ 365). The Citizen Petitioners object to the localism and diversity components of the Commission's rationale. We conclude differently, as reasoned analysis supports the Commission's determination that the blanket ban on newspaper/broadcast cross-ownership was no longer in the public interest. Part II.C supra.

1. Newspaper/broadcast combinations can promote localism.

The Commission measured the promotion of localism by considering "the selection of programming responsive to local needs and interests, and local news quantity and quality." Order ¶ 78. Evidence that existing (grandfathered) newspaper-owned broadcast stations produced local news in higher quantity with better quality than other stations convinced the Commission that the ban on newspaper/broadcast combinations undermined its localism interest. The Commission principally relied on the findings of its MOWG study that newspaper-owned television stations provide almost 50% more local news and public affairs programming than other stations, an average of 21.9 hours per week. Id. ¶ 344 (citing Thomas C. Spavins et al., The Measurement of Local Television News and Public Affairs Programs (MOWG Study No. 7) at 3 (Sept.2002)). The Commission also found corresponding advantages in quality of local coverage provided by newspaper-owned stations, as shown by ratings (measuring consumer approval) and industry awards (measuring critical approval). Id. ¶ 344-45 (citing, among other things, findings by the Project for Excellence in Journalism that newspaper-owned stations "were more likely to do stories focusing on important community issues and to provide a wide mix of opinions, and they were less likely to do celebrity and human interest features").

The Citizen Petitioners argue that the MOWG study was flawed because it examined all newspaper/broadcast station combinations, including "intermarket" combinations (entities that own a newspaper and broadcast stations in different cities), as well as "intramarket" combinations (entities that own a newspaper and a broadcast station in the same city). But the Citizens Petitioners do not suggest that a study entirely focused on intramarket combinations would have different results. The six intramarket combinations that were included in the study (grandfathered exceptions to the cross-ownership ban) averaged more local news and public affairs programming as compared to the overall average (26 weekly hours compared to 21.9) and higher ratings for their 5:30 p.m. and 6:00 p.m. news programs (9.8 and 11 compared to 7.8 and 8.2). MOWG Study No. 7 app. A; Comments of Newspaper Association of America, MB Docket No. 02-277 at 15 (Jan. 2, 2003).

The Citizen Petitioners also protest the Commission's reliance on anecdotal evidence of pro-localism combinations and its disregard of anecdotal evidence to the contrary. Significantly, however, the Commission used anecdotal evidence23 merely to illustrate its statistical findings — it did not rely on anecdote as the sole basis for its conclusions about localism. Moreover, the Commission properly discounted anecdotal evidence of a Canadian newspaper conglomerate's detrimental effect on localism. Can West, which controls 30% of Canada's daily newspaper circulation, requires its newspaper editors to publish editorials from headquarters, which it forbids local editorials to contradict. But the Commission explained that the Can West example shows the peril of national ownership and corporate centralization of media services, which is not relevant to the Commission's regulations on local combinations. Order ¶ 352. In summary, the Citizen Petitioners' arguments do not unsettle the Commission's conclusion that the newspaper/broadcast cross-ownership ban undermined localism.

2. A blanket prohibition on newspaper/broadcast combinations is not necessary to protect diversity.

The Commission offered two rationales for its conclusion that a blanket prohibition on newspaper/broadcast combinations is no longer necessary to ensure diversity in local markets. First, it found that "[c]ommonly-owned newspapers and broadcast stations do not necessarily speak with a single, monolithic voice." Id. ¶ 361. Given conflicting evidence in the record24 on whether ownership influences viewpoint, the Commission reasonably concluded that it did not have enough confidence in the proposition that commonly owned outlets have a uniform bias to warrant sustaining the cross-ownership ban. Id. ¶ 364.

Second, the Commission found that diverse viewpoints from other media sources in local markets (such as cable and the Internet) compensate for viewpoints lost to newspaper/broadcast consolidations. Id. ¶ 366. We agree record evidence suggests that cable and the Internet supplement the viewpoint diversity provided by broadcast and newspaper outlets in local markets.25 As discussed more fully below, we believe that the Commission gave too much weight to the Internet in deriving the Cross-Media Limits. But separate from the question of degree, we conclude that it was acceptable for the Commission to find that cable and the Internet contribute to viewpoint diversity.

C. The Commission's decision to retain some limits on common ownership of different-type media outlets was constitutional and did not violate § 202(h).

The Deregulatory Petitioners support the Commission's repeal of the newspaper/broadcast cross-ownership ban but object to its decision to retain any restriction on the common ownership of newspaper and broadcast media outlets. First they argue that any limits on newspaper/broadcast cross-ownership violate § 202(h), because, as the Commission acknowledges, the evidence suggests that the cross-ownership restrictions are not in the public interest. They also argue that the Cross-Media Limits violate the First and Fifth Amendments of the United States Constitution. We disagree on all counts.

1. Continuing to regulate cross-media ownership is in the public interest.

The Commission's finding that a blanket prohibition on newspaper/broadcast cross-ownership is no longer in the public interest does not compel the conclusion that no regulation is necessary. See Order ¶ 364. As described above, the Commission found evidence to undermine the premise that ownership always influences viewpoint,26 but it did not find the opposite to be true. And while the Commission found that other media sources contributed to viewpoint diversity in local markets, it could not have found that the Internet and cable were complete substitutes for the viewpoints provided by newspapers and broadcast stations. See MOWG Study No. 3 (finding that the Internet and cable rank as sources of local news, but they do not outrank newspapers and broadcast television). Given the Commission's goal of balancing the public's interests in competition, localism, and diversity, it reasonably concluded that repealing the cross-ownership ban was necessary to promote competition and localism, while retaining some limits was necessary to ensure diversity.27

2. Continuing to regulate cross-media ownership does not violate the Fifth Amendment.

The Deregulatory Petitioners argue that the Cross-Media Limits violate the Equal Protection Clause of the Fifth Amendment because they single out newspaper owners for special restrictions that do not apply to other media outlets. This argument is foreclosed, however, by the Supreme Court's decision in NCCB that endorsed the constitutionality of the 1975 newspaper/television cross-ownership ban. 436 U.S. at 801-02, 98 S.Ct. 2096. Rather than concluding that newspaper owners were singled out for different treatment, the Court determined that "the regulations treat newspaper owners in essentially the same fashion as other owners of the major media of mass communication" by imposing limits on them as well as owners of television and radio stations. Id. at 801, 98 S.Ct. 2096. We decline the Deregulatory Petitioners' invitation to disregard Supreme Court precedent because of changing times. Surely there are more media outlets today (such as cable, the Internet, and satellite broadcast) than there were in 1978 when NCCB was decided. But it cannot be assumed that these media outlets contribute significantly to viewpoint diversity as sources of local news and information. See Part IV.D.2 infra. Even if it could, it is the Supreme Court's prerogative to change its own precedent. See State Oil Co. v. Khan, 522 U.S. 3, 20, 118 S.Ct. 275, 139 L.Ed.2d 199 (1997) (stating that only the Supreme Court can overrule its own precedent); Am. Civil Liberties Union of N.J. v. Black Horse Pike Reg'l Bd. of Educ., 84 F.3d 1471, 1484 (3d Cir.1996) (stating that we are obliged to follow Supreme Court precedent "until instructed otherwise by a majority of the Supreme Court").

3. Continuing to regulate cross-media ownership does not violate the First Amendment.

The Deregulatory Petitioners argue that the Commission's decision to retain restrictions on the common ownership of newspapers and broadcast stations contravenes the First Amendment because it limits the speech opportunities of newspaper owners and broadcast station owners, and hence limits the public's access to information. Yet again their challenge is foreclosed by NCCB, where the Supreme Court affirmed the Commission's authority, despite a First Amendment challenge, to regulate broadcast/newspaper cross-ownership in the public interest. Due to the "physical scarcity" of the broadcast spectrum, the Court scrutinized the regulation to discern a rational basis. 436 U.S. at 799, 98 S.Ct. 2096. The Commission's action, it held, was "a reasonable means of promoting the public interest in diversified mass communications." Id. at 802, 98 S.Ct. 2096.

The Deregulatory Petitioners suggest, as they did in mounting their Fifth Amendment challenge, that the expansion of media outlets since NCCB's day requires a rethinking of the scarcity rationale and the lower level of constitutional review it entails. Again we decline their invitation to disregard precedent, and we are not alone. See FCC v. League of Women Voters of Cal., 468 U.S. 364, 376 & n. 11, 104 S.Ct. 3106, 82 L.Ed.2d 278 (1984) (upholding the scarcity rationale until Congress speaks to the issue); Fox I, 280 F.3d at 1046 ("First, contrary to the implication of the networks' argument, this court is not in a position to reject the scarcity rationale even if we agree that it no longer makes sense. The Supreme Court has already heard the empirical case against that rationale and still `declined to question its continuing validity.'" (citing Turner v. FCC, 512 U.S. 622, 638, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994))).

Even were we not constrained by Supreme Court precedent, we would not accept the Deregulatory Petitioners' contention that the expansion of media outlets has rendered the broadcast spectrum less scarce. In NCCB, the Court referred to the "physical scarcity" of the spectrum — the fact that many more people would like access to it than can be accommodated. 436 U.S. at 799, 98 S.Ct. 2096. The abundance of non-broadcast media does not render the broadcast spectrum any less scarce. See, e.g., Ruggiero v. FCC, 278 F.3d 1323, 1325 (D.C.Cir.2002), rev'd en banc, 317 F.3d 239 (D.C.Cir.2003) (citing the Commission's statement that "[n]ow ... radio service is widely available throughout the country and very little spectrum remains available for new full-powered stations.").

In this context, we will apply a rational basis standard to the Commission's restrictions on the common ownership between newspaper and broadcast stations, and uphold them if they are rationally related to a substantial government interest. See NCCB, 436 U.S. at 799-800, 98 S.Ct. 2096; see also Am. Family Ass'n v. FCC, 365 F.3d 1156, 1168 (D.C.Cir.2004). In NCCB, the Supreme Court endorsed a substantial government interest in promoting diversified mass communications. Id. at 795, 802, 98 S.Ct. 2096. The Supreme Court held that the Commission had "acted rationally in finding that diversification of ownership would enhance the possibility of achieving greater diversity of viewpoints." Id. at 796, 98 S.Ct. 2096. Here, as in NCCB, the Commission justified its continued restrictions on common ownership of newspapers and broadcast stations as promoting the public interest in viewpoint diversity. Order ¶ 355. The Court has said that limiting common ownership is a reasonable means of promoting the public interest in viewpoint diversity. NCCB, 436 U.S. at 796, 436 U.S. 775. Therefore, applying NCCB, we hold that the Commission's continued regulation of the common ownership of newspapers and broadcasters does not violate the First Amendment rights of either.

D. The Commission did not provide reasoned analysis to support the specific Cross-Media Limits that it chose.

The Commission concluded that cross-ownership limits were necessary in specific situations to guard against "an elevated risk of harm to the range and breadth of viewpoints that may be available to the public." Order ¶ 442. But recognizing that ownership limits impede the speech opportunities for both broadcasters and newspapers, the Commission endeavored to craft new limits "as narrowly as possible." Id. ¶ 441. In that vein, the Commission sought to identify "at risk" local markets — those with high levels of viewpoint concentration — where continued regulation was necessary. By focusing its regulation on those markets, the Commission hoped to avoid needlessly overregulating markets with already ample viewpoint diversity.

But for all of its efforts, the Commission's Cross-Media Limits employ several irrational assumptions and inconsistencies. We do not object in principle to the Commission's reliance on the Department of Justice and Federal Trade Commission's antitrust formula, the Herfindahl-Hirschmann Index ("HHI"), as its starting point for measuring diversity in local markets. In converting the HHI to a measure for diversity in local markets, however, the Commission gave too much weight to the Internet as a media outlet, irrationally assigned outlets of the same media type equal market shares, and inconsistently derived the Cross-Media Limits from its Diversity Index results. For these reasons, detailed below, we remand for the Commission to justify or modify further its Cross-Media Limits.

1. Overview of the Commission's Diversity Index Methodology

The Commission designed a methodology called the Diversity Index to identify "at risk" markets where limits on cross-media ownership should be retained. The Diversity Index was "inspired by" the HHI, which the Department of Justice and the Federal Trade Commission use to measure proposed mergers' effect on competition in local markets. Id. ¶ 394. A market's HHI score is the sum of market shares squared. A highly competitive market will have a lower HHI score than a concentrated one. For example, compare the HHI score of a market with 10 equal-sized competitors (10 [competitors] × 10 [each competitor's market share percentage]² = 1000) to the HHI score of a market with only five equal-sized competitors (5 × 20² = 2000).28 The Department of Justice and the Federal Trade Commission categorize markets with HHI scores above 1800 as "highly concentrated." If a proposed merger would exceed that level of concentration, the agencies believe that it would be harmful to the competition in that market. At its core, the Diversity Index here uses the same sum of market share-squared formula.

First, the Commission selected which media outlets to include in its analysis of viewpoint diversity in local markets based on consumers' reported preferences for sources of local news and information. The Commission determined that broadcast television, daily and weekly newspapers, radio, and Internet (via cable connection and DSL, dial-up, or other connections) are the relevant contributors to viewpoint diversity in local markets. See Neilsen Media Research, Consumer Survey on Media Usage (MOWG Study No. 8) (Sept. 2002). Based on the popularity of each media source, the Commission assigned each a relative weight: broadcast TV 33.8%, daily newspapers 20.2%, weekly newspapers 8.6%, radio 24.9%, Internet (cable) 2.3% and Internet (DSL, dial-up, or other connection) 10.2%. Order ¶¶ 412, 415, 427.

Next, the Commission selected many sample markets29 for which it would determine a Diversity Index score. In each of those markets, it counted the number of outlets within each media type and assigned each outlet an equal market share. (For example, the New York City market has 23 television stations, so each one was attributed an equal 4.3%30 market share.) The Commission calculated the ownership share by multiplying the number of outlets owned by an entity by the market share. (Univision owns three television stations in New York, so its ownership share was 3 × 4.3% = 13.0%.) Each ownership share was then given its relative weight by media type. (Univision's 13.0% share was thus subject to the 33.8% multiplier for television stations). The Commission then squared all of the weighted ownership shares; their sum was the market's total Diversity Index score. But in markets with cross-owned shares (outlets of different media types owned by the same entity) the entity's weighted ownership shares were summed together before they were squared. (ABC owned one television station and four radio stations in New York, so the Commission added its weighted ownership share for the television station (4.3% × 33.8% = 1.45%) to its weighted ownership share for the radio stations (6.7% × 24.9% = 1.67%) for a combined ABC ownership share of 3.12%. The square of 3.12 was included in the summation). See generally Order app. C.

After the Commission calculated Diversity Index scores for markets of different sizes, it determined how those scores would change in several different consolidation "scenarios." To illustrate, the Commission determined that markets with five television stations had an average Diversity Index score of 911. If a newspaper and television station combined in a market of that size, that score would increase by 223, to 1134. The other combination scenarios were: (1) one television station and all of the radio stations allowed to be commonly owned under the local radio rule; (2) one newspaper and all of the radio stations allowed to be commonly owned under the local radio rule; (3) one newspaper, one television station, and half of the number of radio stations allowed to be commonly owned under the local radio rule; (4) two television stations; (5) one newspaper and two television stations; and (6) one newspaper, two television stations, and all of the radio stations allowed under the local radio rule. Order app. D.

Finding that all of the consolidation scenarios resulted in relatively high increases to the average Diversity Index scores for the smallest markets (those with three or fewer television stations), the Commission prohibited newspaper/television, newspaper/radio, and radio/television combinations in those markets. Order ¶¶ 456, 459, 460. In the large markets (nine or more stations), all of the consolidation scenarios resulted in acceptable increases to the average Diversity Index scores, so the Commission imposed no limits on cross-media ownership in those markets. In the mid-sized markets (between four and eight television stations), the Commission found that all of the scenarios, except the two involving a newspaper and television duopoly, should be allowed, based on their modest increases to the average Diversity Index scores for those markets. Id. ¶ 466.

2. The Commission did not justify its choice and weight of specific media outlets.

Petitioners from both sides of the regulatory spectrum attack the Commission's selection of media outlets for inclusion in the Diversity Index formula. The Citizen Petitioners argue that the Commission gave too much weight to the Internet at the expense of television and daily newspapers, thereby understating the level of concentration and overstating the level of diversity in a market. The Deregulatory Petitioners, on the other hand, argue that the Diversity Index understates the actual amount of diversity in a market by ignoring important media outlets, primarily cable television. As explained below, we affirm the Commission's reasoned decision to discount cable. But we think that the same rationale also applies to the Internet. Therefore, its decision to count the Internet as a source of viewpoint diversity, while discounting cable, was not rational.

The Commission properly excluded cable because of serious doubts as to the extent that cable provided independent local news — the Commission's recognized indicator of viewpoint diversity in local markets. Order ¶ 394 ("News and public affairs programming is the clearest example of programming that can provide viewpoint diversity .... [and] the appropriate geographic market for viewpoint diversity is local."). While the responses to one question in the Commission's survey suggested that cable is a significant source of local news, see MOWG Study No. 8 tbl. 8, the Commission did not find this credible because the responses to other survey questions suggested that respondents were counting broadcast signals that are transmitted as cable channels as sources of local news on cable. See id. tbl. 18 (from a list of popular national cable news channels and a choice for "local cable news channels," almost half of the respondents chose "other" as their source for cable network news, indicating that cable news channels were probably confused with broadcast networks' news). The survey's indication that cable is an independent source of local news was also undercut by external evidence, including Neilsen ratings, that local cable channels are the least watched of any broadcast or cable stations in the market. Order ¶ 414. Furthermore, the Commission noted that local cable channels are not available everywhere, but only in select markets. Id. ¶ 414 n. 924; see also UCC Comments at 30-31 (reporting that only 10 to 15% of cable systems include channels that provide local and public affairs programming — i.e., public, educational, and governmental access channels — and that there are only 22 local news cable channels in the country, five of which serve the New York City area). Thus, the Commission justifiably excluded cable from its Diversity Index calculations.

Similarly, the responses to one of the Commission survey questions suggested that the Internet is a source of local news. See MOWG Study No. 831 tb1.1 (18.8% of respondents listed the Internet as a source of local news). But the survey did not identify which websites respondents used as sources of local news.32 There is a critical distinction between websites that are independent sources of local news and websites of local newspapers and broadcast stations that merely republish the information already being reported by the newspaper or broadcast station counterpart. The latter do not present an "independent" viewpoint and thus should not be considered as contributing diversity to local markets. Accordingly, the Commission should have discounted the respondents who primarily rely on these websites from its total number of respondents who indicated that they use the Internet to access local news.33

Furthermore, just as the Commission discounted responses indicating that cable was an independent source of local news because this finding conflicted with other record evidence, it should have discounted the Internet responses as well. The Commission does not cite, nor does the record contain, persuasive evidence that there is a significant presence of independent local news sites on the Internet. According to the record, most sources of local news on the Internet are the websites for newspapers and broadcast television stations. See, e.g., UCC Comments at 33 (62% of Internet users get local news from newspaper websites, 39% visit television station websites). And the examples the Commission does cite — the Drudge Report and Salon.com — have a national, not local, news focus.34 Order ¶ 427.

The Commission suggests that "the virtual universe of information sources" on the Internet qualifies it as a source of viewpoint diversity. Order ¶ 427. But to accept this rationale we would have to distort the Commission's own premise that local news is an indicator of viewpoint diversity. E.g., id. ¶ 391 (Diversity Index measures "the relative importance of these media as a source of local news").35 Search-engine sponsored pages such as Yahoo! Local and about.com,36 which were suggested by commenters as sources of local news and information, may be useful for finding restaurant reviews and concert schedules, but this is not the type of "news and public affairs programming" that the Commission said was "the clearest example of programming that can provide viewpoint diversity." Id. ¶ 394.

To accept its "universe of information" characterization of the Internet's viewpoint diversity, we would also have to disregard the Commission's professed intent to focus its consideration of viewpoint diversity on media outlets. Id. ¶¶ 20, 391. In terms of content, "the media" provides (to different degrees, depending on the outlet) accuracy and depth in local news in a way that an individual posting in a chat room on a particular issue of local concern does not. But more importantly, media outlets have an entirely different character from individual or organizations' websites and thus contribute to viewpoint diversity in an entirely different way. They provide an aggregator function (bringing news/information to one place) as well as a distillation function (making a judgment as to what is interesting, important, entertaining, etc.).37 Individuals (such as political candidates) and entities (such as local governments or community organizations) may use the Internet to disseminate information and opinions about matters of local concern (such as the extension of a bike path on the Schuylkill River in Center City Philadelphia, see dissent Part IV.A.1), but these individuals and organizations are not, themselves, media outlets. We agree that the Internet "helps citizens discharge the obligations of citizenship in a democracy," Order ¶ 393, when someone can go cityofglenfalls.com to find the city council's next agenda or use sfgov.org to learn how to get a marriage license in San Francisco. See Comments of the Hearst Corporation, MM Docket No. 01-235 at 11 & n. 36, apps. A & C (Dec. 3, 2001) ("Hearst Comments") (listing these website addresses, among others, as examples of local government websites). But local governments are not, themselves, "media outlets" for viewpoint-diversity purposes. Like many entities, they just happen to use a particular media outlet — the Internet — to disseminate information.

Similarly, advertiser-driven websites such as hvnet.com and sfadvertiser.com, see Hearst Comments at apps. A & C, hardly contribute to viewpoint diversity. Like local governments, the sponsors of these websites are not "media outlets" just because they have "local information" that they want the public to have — if they were, their advertisements in telephone directories would also have to count toward viewpoint diversity. Compare Order ¶ 424 (explaining that the Commission would not count toward viewpoint diversity a local newspaper's comics and classified ads because, while the subscriber is "undoubtedly getting a valuable service, it is not clear that the service has anything to do with news and current affairs").

The Commission attempts to justify different treatment for cable and the Internet by suggesting that local cable news channels are only available in select markets, while the Internet is available everywhere. Not only is this distinction demonstrably false (as even the Commission acknowledged that almost 30% of Americans do not have Internet access, Order ¶ 365), it is irrelevant. That the Internet is more available than local cable news channels does not mean that it is providing independent local news. On remand the Commission must either exclude the Internet from the media selected for inclusion in the Diversity Index or provide a better explanation for why it is included in light of the exclusion of cable. See Sinclair, 284 F.3d at 163-65 (unexplained inconsistency was arbitrary and capricious).

3. The Commission did not justify its assumption of equal market shares.

Both the Citizen Petitioners and the Deregulatory Petitioners object to the Commission's decision to assign all outlets within the same media type (that is, television stations, daily papers, or radio stations) an equal market share. The assumption of equal market shares is inconsistent with the Commission's overall approach to its Diversity Index and also makes unrealistic assumptions about media outlets' relative contributions to viewpoint diversity in local markets.38

The Commission's decision to assign equal market shares to outlets within a media type does not jibe with the Commission's decision to assign relative weights to the different media types themselves, about which it said "we have no reason to believe that all media are of equal importance." Order ¶ 409; see also id. ¶ 445 ("Not all voices, however, speak with the same volume."). It also negates the Commission's proffered rationale for using the HHI formula in the first place — to allow it to measure the actual loss of diversity from consolidation by taking into account the actual "diversity importance" of the merging parties, something it could not do with a simple "voices" test. Id. ¶ 396. Finally, assigning equal market shares to outlets that provide no local news almost certainly presents an understated view of concentration in several markets, thus contravening the Commission's goal of making "the most conservative assumption possible" about viewpoint diversity. Id. ¶ 400.

Additionally, there is no dispute that the assignment of equal market shares generates absurd results. For example, in New York City, the Dutchess Community College television station and the stations owned by ABC each receive an equal 4.3% market share. Or compare the Dutchess Community College station's weighted share of 1.5% (4.3% times the 33.8% multiplier for television) to a mere 1.4% weighted, combined share assigned to the New York Times Company's co-owned daily newspaper and radio station. Order app. C. A Diversity Index that requires us to accept that a community college television station makes a greater contribution to viewpoint diversity than a conglomerate that includes the third-largest newspaper in America39 also requires us to abandon both logic and reality.

The Commission's attempt to justify its failure to consider actual market share of outlets within a media type is not persuasive. It suggests that actual-use data use data is not relevant because "current behavior is not necessarily an accurate predictor of future behavior." Order ¶ 423. But this truism did not prevent the Commission from preferring actual-use data in assigning relative weight to the different media types. The Commission also suggests that, compared to consumer preferences for a general media type (which are generally stable), consumers' preferences for particular media outlets are fluid because they depend on the media outlet's chosen format or quantity or quality of content, which are easily changed. Id. ¶ 422. But while variance in the local news content of individual media outlets is conceivable — e.g., the home shopping television station could start carrying local news — the Commission does not provide any evidence that media outlets actually undergo any such radical content change, let alone that they regularly do so. Simply put, the Commission needs to undergird its predictive judgment that stations can freely change the level of their news content with some evidence for that judgment to survive arbitrary and capricious review. Fox I, 280 F.3d at 1051.

Lastly, the Commission attempts to justify its refusal to employ actual-use data by arguing that collecting "information on viewing/listening/reading of local news and current affairs material" would make it "necessary first to determine which programming constituted news and current affairs," which in turn "would present both legal/Constitutional and data collection problems." Order ¶ 424. With respect to "legal/Constitutional" problems, the Commission is apparently concerned about categorizing programming as news or "non-news." But the Commission obtained actual-use data in MOWG Study No. 8 without any "legal/Constitutional problems." There it avoided having to make content-distinguishing judgments simply by asking respondents where they got their local news. And the Commission's reference here to data collection problems is vague and unexplained; there is no suggestion that obtaining actual-use data for outlets within a media type would be prohibitively more onerous than obtaining the same data for the media types themselves, as it did in MOWG Study No. 8.

Because the Commission's reasons for eschewing actual-use data in assigning market shares to outlets within a media type and assuming equal market shares are unrealistic and inconsistent with the Commission's overall approach to the Diversity Index and its proffered rationale, we remand for the Commission's additional consideration of this aspect of the Order.

4. The Commission did not rationally derive its Cross-Media Limits from the Diversity Index results.

After the Commission calculated the average Diversity Index scores for markets of various sizes, it set out to determine whether the increase in those scores resulting from different consolidation scenarios would have an acceptable or unacceptable effect. The Commission's results, contained in appendix D of the Order, are replicated here for ease of reference. (We have used boldface type to indicate those increases in Diversity Index scores that the Commission found acceptable in crafting the Cross-Media Limits.)

------------------------------------------------------------------------------------------------------------
|     Base Case       |         Average Change in Diversity Index, Resulting from Mergers                  |
|---------------------|------------------------------------------------------------------------------------|
| TV        Average   | 100%      Newspaper     Newspaper    Newspaper    2 TV      Newspaper    Newspaper |
| stations  Diversity | Radio +   +             +            + 1          stations  +            +         |
| in        Index     | 1 TV      100%          1 TV         TV                     2 TV         100%      |
| market    score     | station   Radio         station      station +              stations     Radio +   |
|                     |                                      50%                                 2 TV      |
|                     |                                      Radio                               stations  |
|---------------------|------------------------------------------------------------------------------------|
| 1         1707      | 651       271           910          1321         -         -            -         |
|                     |                                                                                    |
| 2         1316      | 301       335           731          1009         -         -            -         |
|                     |                                                                                    |
| 3         1027      | 190       242           331          515          -         -            -         |
|---------------------|------------------------------------------------------------------------------------|
| 4         928       | 138       236           242          408          -         -            -         |
|                     |                                                                                    |
| 5         911       | 111       263           223          393          91        376          846       |
|                     |                                                                                    |
| 6         889       | 79        239           200          340          63        357          688       |
|                     |                                                                                    |
| 7         753       | 73        171           121          247          47        242          533       |
|                     |                                                                                    |
| 8         885       | 79        299           152          314          36        308          734       |
|---------------------|------------------------------------------------------------------------------------|
| 9         705       | 64        198           86           207          28        172          473       |
|                     |                                                                                    |
| 10        635       | 56        107           51           119          23        101          292       |
|                     |                                                                                    |
| 15        595       | 43        149           48           145          10        97           302       |
|                     |                                                                                    |
| 20        612       | 49        222           40           128          6         80           350       |
------------------------------------------------------------------------------------------------------------

As illustrated by the chart's figures that are not in boldface type, the Commission prohibited all cross-media consolidation in the smallest markets where, under any of their sample consolidation scenarios,40 the increase in Diversity Index would be excessive. The Commission decided to permit all of the scenarios in the largest markets, finding the Diversity Index increases to be acceptable. In the mid-sized markets, the Commission allowed most of the consolidation scenarios but prohibited consolidations involving a newspaper and a television duopoly.

The Deregulatory Petitioners suggest that the Commission should have prohibited only cross-ownership mergers that would lead to an increase of more than 400 points, consistent with the practice of the Department of Justice to block mergers that would increase the local market's HHI score by over 400 points. As illustrated by the chart above, this approach would have generated far more permissive Cross-Media Limits. The Citizen Petitioners argue, on the other hand, that the Commission should have prohibited cross-ownership mergers that lead to an increase of more than 100 points, the increase that triggers the Department of Justice's "further review." But the decision of where to draw the line between acceptable increases and unacceptable increases is almost always the Commission's to make. Deference to the Commission's judgment is highest when assessing the rationality of the agency's line-drawing endeavors. See NCCB, 436 U.S. at 814-15, 98 S.Ct. 2096; AT&T Corp., 220 F.3d at 627. The Commission rationalized its decision to make conservative assumptions in order to "protect[its] core policy objective of viewpoint diversity." Order ¶ 399; see also Sinclair, 284 F.3d at 153 (Commission determined that antitrust merger guidelines "might be too low as their purpose lay in defining the point at which antitrust scrutiny is required, and not in encouraging a wide array of voices and viewpoints").

Although the Commission is entitled to deference in deciding where to draw the line between acceptable and unacceptable increases in markets' Diversity Index scores, we do not affirm the seemingly inconsistent manner in which the line was drawn. As the chart above illustrates, the Cross-Media Limits allow some combinations where the increases in Diversity Index scores were generally higher than for other combinations that were not allowed. Consider the mid-sized markets (four to eight stations), where the Commission found that a combination of a newspaper, a television station, and half the radio stations allowed under the local radio rule would increase the average Diversity Index scores in those markets by 408 (four stations), 393 (five), 340(six), 247 (seven), and 314 (eight) points respectively. These permitted increases seem to belong on the other side of the Commission's line. They are considerably higher than the Diversity Index score increases resulting from other combinations that the Commission permitted, such as the newspaper and television combination, 242 (four stations), 223 (five), 200(six), 121 (seven), and 152 (eight). They are even higher than those resulting from the combination of a newspaper and television duopoly — 376 (five stations), 357(six), 242 (seven), and 308 (eight) — which the Commission did not permit. The Commission's failure to provide any explanation for this glaring inconsistency is without doubt arbitrary and capricious, and so provides further basis for remand of the Cross-Media Limits.41

5. The Commission should provide better notice on remand.

Our decision to remand the Cross-Media Limits also gives the Commission an opportunity to cure its questionable notice. Under the APA, an agency must publish notice of either the terms or substance of the proposed rule or a description of the subjects and issues involved. 5 U.S.C. § 553(b)(3). We have held that "the adequacy of the notice must be tested by determining whether it would fairly apprise interested persons of the `subjects and issues' before the agency." Am. Iron & Steel Inst. v. EPA, 568 F.2d 284, 293 (3d Cir.1977). The Citizen Petitioners argue that, in order to "fairly apprise" the public, the Commission is also obligated to provide notice of its underlying methodology, the reasoning from which it derived the proposed rule. The Commission's notice only indicated that it was considering "creating a new metric" to "reformulate [its] mechanism for measuring diversity and competition in a market," and that it was contemplating "design[ing] a test that accords different weights to different outlet types." Notice, 17 F.C.C.R. 18,503, ¶¶ 113-15. The Citizen Petitioners argue that the Commission should have publically noticed the Diversity Index once it determined that was the methodology on which it would rely to derive the Cross-Media Limits. See McLouth Steel Prods. Corp. v. Thomas, 838 F.2d 1317 (D.C.Cir.1988) (failure to describe a particular "model" for computing contamination levels was not adequate notice); United States v. Nova Scotia Food Prods. Corp., 568 F.2d 240, 251 (2d Cir.1977) (agency's failure to provide notice of the data from which it derived regulation foreclosed "criticism of the methodology used or the meaning to be inferred from the data").

The Commission argues that the Diversity Index was formulated as a response to comments, and it need not seek additional public comment on data formulated as a response to earlier comments.42 We are mindful that the APA's notice obligations are not supposed to result in a notice-and-comment "revolving door." "Rulemaking proceedings would never end if an agency's response to comments must always be made the subject of additional comments." Cmty. Nutrition Inst. v. Block, 749 F.2d 50, 58 (D.C.Cir.1984). But courts have also suggested that an agency may withhold notice of its comment-derived data only in the absence of prejudice. Id. ("The response may, moreover, take the form of new scientific studies without entailing the procedural consequences appellants would impose, unless prejudice is shown."); see also Solite Corp. v. EPA, 952 F.2d 473, 484 (D.C.Cir.1991) (allowing agency to use "`supplementary' data, unavailable during the notice and comment period, that `expand[s] on and confirm[s]' information contained in the proposed rulemaking and addresses `alleged deficiencies' in the pre-existing data, so long as no prejudice is shown" (citing Cmty. Nutrition Inst., 749 F.2d at 57-58) (alteration in original)). As the Diversity Index's numerous flaws make apparent, the Commission's decision to withhold it from public scrutiny was not without prejudice. As the Commission reconsiders its Cross-Media Limits on remand, it is advisable that any new "metric" for measuring diversity and competition in a market be made subject to public notice and comment before it is incorporated into a final rule.

Both the Citizen Petitioners and the Deregulatory Petitioners challenge the Commission's modification to the local television ownership rule, which would allow triopolies in markets of 18 stations or more and duopolies in other markets, subject to a restriction on combinations of the four largest stations in any market. We uphold the top-four restriction but remand the numerical limits for the Commission to harmonize certain inconsistencies and better support its assumptions and rationale. We also remand for the Commission to reconsider or justify its decision to expand the rule's waiver provision — applicable to sales of failed, failing, or unbuilt television stations — by eliminating the requirement that waiver applicants notice the station's availability to out-of-market buyers.

A. Regulatory Background and the 2002 Biennial Review

As part of the 1996 Act, Congress directed the Commission to conduct a rulemaking to determine whether to "retain, modify, or eliminate" its local television ownership rule, which at the time prohibited the common ownership of two television stations with overlapping Grade B signal contours.43 § 202(c)(2), 110 Stat. 111; Amendment of Sections 73.35, 73.240, and 73.636 of the Commission's Rules Relating to Multiple Ownership of Standard, FM and Television Broadcast Stations, 45 F.C.C. 1476, ¶ 3 (1964) (establishing the restriction on the common ownership of televisions stations).

In response to Congress's directive, the Commission promulgated a rule allowing an entity to own two television stations in the same DMA, provided that: (1) the Grade B field-strength contours of the stations do not overlap; and (2)(a) at least one of the stations is not ranked among the four highest-ranked stations in the DMA, and (b) at least eight "voices" — that is, independently owned, operational, commercial or noncommercial full-power broadcast television stations — would remain in the DMA after the proposed combination. 1999 Television Rule Review, 14 F.C.C.R. 12,903, ¶ 8; 47 C.F.R. § 73.3555(b). The D.C. Circuit Court reviewed this so-called "duopoly rule," and in 2002 remanded it for the Commission to justify its decision to count only broadcast television stations as voices to the exclusion of other non-broadcast media. Sinclair, 284 F.3d at 162.

The Commission consolidated Sinclair's remand order with its 2002 biennial review under § 202(h). The Order announced the Commission's decision to abandon the duopoly rule in favor of a rule that would permit common ownership of two commercial stations in markets that have 17 or fewer full-power commercial and noncommercial stations, and common ownership of three commercial stations in markets that have 18 or more stations. Both limits are subject to a restriction44 on the common ownership of stations ranked among the market's largest four based on audience share. Order ¶ 186. The Commission also decided to expand its criteria for waiving the rule's ownership restrictions for proposed combinations that involve failed, failing, or unbuilt stations by eliminating the requirement that waiver applicants provide notice of the sale to out-of-market buyers. Id. ¶ 225.

B. We uphold several threshold challenges to the Commission's overall regulatory approach.

Before we consider the specific modifications to the local television rule, we address some threshold challenges to the Commission's regulatory approach.

1. Limiting local television station ownership is not duplicative of antitrust regulation.

We reject the Deregulatory Petitioners' contention that the Commission's local television rule is duplicative of antitrust enforcement (by the Department of Justice and the Federal Trade Commission) and, thus, not in the public interest. The Commission ensures that license transfers serve public goals of diversity, competition, and localism, while the antitrust authorities have a different purpose: ensuring that merging companies do not raise prices above competitive levels. See, e.g., Clayton Act, § 7, 15 U.S.C. § 18 (restraining mergers that would lessen competition in a market); Dep't of Justice and Federal Trade Comm'n, Horizontal Merger Guidelines § 0.1 (1997 rev. ed.) ("Merger Guidelines") (seeking to protect consumers by ensuring mergers do not result in anticompetitive prices).

The Deregulatory Petitioners contend that the Commission's new regulations are not derived from its professed reliance on audience preference, which the Commission suggests distinguishes its regulatory approach from that of the antitrust authorities. Order ¶ 65 (pointing out that it considers audience preferences plus advertising data as indicators of competition, while the antitrust authorities focus on prices). But the Commission's local television ownership rule does reflect audience preferences in at least three ways. First, the Commission decided to focus its competition analysis on the delivered program market (as opposed to the video programming market or the video advertising market) because that is the market that "directly affects viewers." Id. ¶ 141. Second, as discussed in Part V.C below, the Commission used audience preference data to support its conclusion that common ownership of local television stations can improve program quality. Id. ¶ 150. Third, as also discussed in Part V.C, the Commission justified the top-four restriction on evidence of an audience share "cushion" between the top-four stations and the fifth-ranked station in most markets. Id. ¶ 195. Thus, we reject the Deregulatory Petitioners' suggestion that the local television ownership rule does not reflect the Commission's concern for audience preferences.

Finally, we note that the Commission reviews all license transfers, 47 U.S.C. § 310(d), while the antitrust agencies typically review only large mergers. See 15 U.S.C. § 18a(a) (a merger must only be reported to the FTC and DOJ if the size of the transaction exceeds $200 million or if the assets of one party exceed $10 million and the assets of the other party exceed $100 million). Eighty-five percent of station mergers that have taken place since 2000 would not have been subject to antitrust review because the parties' assets fell below these thresholds. Br. of Intervenor UCC at 37 (citing BIA Financial Network, Television Market Report (2d ed. 2003)). In this context, it hardly seems that the Commission's local television station ownership rule is duplicative of other agencies' antitrust enforcement.

2. Media other than broadcast television may contribute to viewpoint diversity in local markets.

Recognizing that allowing more television concentration in local markets could detract from viewpoint diversity, the Commission rationalized its decision to deregulate with its finding that media other than broadcast television contribute to viewpoint diversity. Order ¶ 171. This is a departure from the Commission's rationale for the existing rule — the issue remanded by the Sinclair Court — that only television stations are relevant to its diversity analysis.

We agree with the Commission's conclusion that broadcast media are not the only media outlets contributing to viewpoint diversity in local markets. Yet because we remand the Commission's numerical limits, as explained in Part V.D below, we need not decide the degree to which non-broadcast media compensate for lost viewpoint diversity to justify the modified rule. Rather, we leave it for the Commission to demonstrate that there is ample substitutability from non-broadcast media to warrant the particular numerical limits that it chooses on remand.

We note that the record contains only weak evidence that cable can substitute for broadcast television as a source of viewpoint diversity. For example, the Commission found that among cable subscribers (a class that already omits one-third of American households) only 30% have access to local cable news channels. See Order ¶ 414 n. 924; see also supra Part IV.D.2 (describing why the Commission concluded that a majority of survey respondents' purported reliance on cable as a source of local news was probably based on confusion with broadcast news); Reply Comments of the UCC, MM Docket No. 01-235 at 5 (Feb. 15, 2002), UCC Comments at 30-31. With respect to the Internet, while record evidence indicates a negative correlation between respondents' reliance on broadcast television and the Internet as news sources (suggesting that people who use the Internet for local news do so at the expense of television), the Internet is also limited in its availability and as a source of local news. Compare MOWG Study No. 3 at 3, 34; with UCC Reply Comments at 10 and MOWG Study No. 8 tbl. 097. Therefore, it seems that the degree to which the Commission can rely on cable or the Internet to mitigate the threat that local station consolidations pose to viewpoint diversity is limited.

3. Consolidation can improve local programming.

The Commission supported its decision to relax the existing "eight voices" rule with findings that common ownership of television stations in local markets can result in "consumer welfare enhancing efficiencies" by eliminating redundant expenses and increasing opportunities for cross-promotion and related programming. Order ¶ 147. To promote localism, the Commission found that these efficiencies translated into improved local news and public interest programming.45 Id. ¶ 164. Evidence supporting this conclusion included findings that commonly owned television stations are more likely to carry local news than other stations and air a similar quality and quantity of local news as other stations. See Bruce M. Owen et al., Effect of Common Ownership or Operation on Television News Carriage, Quantity and Quality, in Comments of Fox Entertainment Group, Inc. et al., MB Docket 02-277 (Jan. 2, 2003). And a study of seven commonly owned broadcast television stations indicates that consolidation generally improved audience ratings — the studied stations increased their audience shares by an average of 3.2% over the share they enjoyed prior to entering into co-ownership with another station. Order ¶ 150 & n. 295 (citing Mark R. Fratrik, Television Local Marketing Agreement and Local Duopolies: Do They Generate New Competition and Diversity? (Jan. 2003), appendix to Comments of Coalition Broadcasters et al., MM Docket 02-277 (Jan. 2, 2003)). In light of this evidence, we reject the Citizen Petitioners' contention that the Commission's finding of localism benefits from consolidation was unsupported.

4. The Commission adequately noticed its decision to allow triopolies.

We also reject the Citizen Petitioners' contention that the Commission failed to provide adequate notice that it was considering a rule that would allow triopolies. The APA requires agencies to publish a notice of proposed rulemaking that contains "either the terms or substance of the proposed rule or a description of the subjects and issues involved." 5 U.S.C. § 553(b). The Notice advised parties that any reevaluation of the television ownership rule would take account of the remand ordered by Sinclair, which left it to the Commission to reexamine its "voices" test as well as "the numerical limit, given that there is a relationship between the definition of voices and the choice of a numerical limit." Notice, 17 F.C.C.R. 18,503, ¶ 76 (citing Sinclair, 284 F.3d at 162). Specifically, the Commission asked for comment on "different economic incentives" relating to diverse viewpoints in newscasting that might exist "among stand-alone stations, duopolies, or triopolies." Id. ¶ 80. This leaves little doubt that the Notice provided a sufficient "description of the subjects and issues involved" in the Commission's decision to allow triopolies.

C. We uphold the Commission's decision to retain the top-four restriction.

Though the Commission recognized that the combination of television stations within the same market could yield efficiencies that benefit consumers, the Commission also recognized that station combinations only have an overall public "welfare-enhancing" effect when the consolidation does not create a "new largest" entity. Order ¶ 194 (citing, inter alia, R. Preston McAfee & Michael Williams, Horizontal Mergers and Antitrust Policy, 40 J. Indus. Econ. 181-87 (June 1992)). Thus, the Commission determined that it had to limit allowable station combinations to those that would not create excessive market power in a "new largest" entity. Finding that a significant "cushion" of audience share percentage points generally separates top-four stations from the fifth-ranked stations, the Commission decided that a top-four restriction would ensure that station consolidations did not lead to excessive market power. Id. ¶ 195-96. The Commission also recognized that efficiencies are less prevalent when financially strong stations merge with each other. Id. ¶ 197. For example, top-four stations are already more likely to be originating local news46 and to have made the transition to digital television. Id. ¶¶ 198-99.

The Deregulatory Petitioners assert that the top-four restriction prevents small-market stations from yielding any of the benefits of consolidation. They argue that by effectively precluding any consolidation in markets with fewer than five stations, the Commission deprives the benefit of consolidation from stations — those in small markets — who need it most. But the Commission's local television rule is protective of small-market stations. The numerical limits already allowed, in effect, extra47 concentration to ensure that small-market stations would realize the efficiency benefits of consolidation. As for the smallest markets with fewer than five stations — where the top-four restriction operates to preclude any consolidation — it was not unreasonable for the Commission to conclude, as it did, that the detriment of concentrated market power — e.g., reduced incentive to improve programming of mass appeal — outweighed the efficiency benefits. Id. ¶¶ 197, 200. And, while we recognize that the Commission "cannot save an irrational rule by tacking on a waiver procedure," ALLTEL Corp. v. FCC, 838 F.2d 551, 561 (D.C.Cir.1988), we note that the modified rule allows the Commission to waive the top-four restriction in small markets where those consolidations would be beneficial overall.48

The Deregulatory Petitioners also object to the top-four restriction because, they argue, it unjustifiably treats all top-four ranked stations the same. The essence of their objection is the Commission's finding of a "general separation" between the top-four stations and other stations. They point out that in many markets, especially small-and medium-sized ones,49 the third and fourth-ranked stations could combine without exceeding the audience share of the first or second-ranked stations. Thus, they contend, there is a more substantial "separation" between the third and fourth-ranked stations than between the fourth and fifth-ranked stations.50

But we must uphold an agency's line-drawing decision when it is supported by the evidence in the record. Sinclair, 284 F.3d at 162; AT&T Corp., 220 F.3d at 627. Here there is ample evidence in the record to support the Commission's restriction on combinations among the top-four stations as opposed to the top-three or some other number. The Commission found a "cushion" of audience share percentage points between the fourth and fifth-ranked stations in most markets. Order ¶ 195. Networks' national audience statistics, which are generally reflected in local market rankings of affiliated stations, also show a substantial 60% drop in audience share between the fourth and fifth-ranked networks. In the ten largest markets, the top-four stations combined control at least 69% (and an average of 83%) of the local commercial share in their respective markets, and in all of the ten largest markets a combination between the third and fourth-ranked stations would produce a new largest station. Local Television Ownership and Market Concentration Study, in UCC Reply Comments at attachment 5.51 Furthermore, the Commission found that permitting mergers among top-four ranked stations generally leads to large HHI increases.

Thus we conclude that the Commission's decision to retain the top-four restriction is supported by record evidence. Accordingly, we extend deference. See Sinclair, 284 F.3d at 162.

D. We remand the specific numerical limits for the Commission's further consideration.

The Commission decided to construct its numerical limits to ensure that most markets would have six equal-sized competitors because the HHI score of a six-member market — 1667 (6 × (100 ÷ 6)²) — is below the Department of Justice and Federal Trade Commission's 1800 threshold for highly concentrated markets for antitrust purposes. Order ¶¶ 192-93 (citing Merger Guidelines § 1.51(c)). Thus the Commission decided to allow triopolies in markets of 18 stations or more (18 ÷ 3 = 6 equal-sized competitors) and duopolies in markets of 17 or fewer (both limits subject to the restriction on a combination of top-four stations). Id. ¶ 193.

The Commission's assumption of equal market shares received flak from both ends of the objecting spectrum. The Citizen Petitioners point out that television stations' market shares vary widely and argue that it is arbitrary for the Commission to base its numerical limits on a rudimentary station "head count" of outlets. The Commission's rationale for its triopoly rule requires that we accept a combination of the first, fifth, and sixth-ranked stations as the competitive equal of a combination of the 16th, 17th, and 18th-ranked stations, just because each combination consists of the same number of stations. While the Citizen Petitioners demonstrate the Commission's flawed rationale with examples of what the modified rule allows, the Deregulatory Petitioners demonstrate the same flaw by pointing out what the modified rule forbids. There is no logical reason, they argue, why it should be impermissible to have five duopolies and one triopoly (a total of six competitors)52 in a market with 13 stations when it is possible that the triopoly could have a lower combined market share than any or all of the duopolies.

The Commission defends its equal market shares approach with the suggestion that market share, which varies with each season's new programs, is too "fluid" to be the basis for its regulations. Id. ¶ 193. But elsewhere in the local television ownership rule the Commission found that the market share was stable enough to rely on for support of its top-four restriction.53 Id. ¶ 195. And not only is the Commission's "market share is too fluid" rationale inconsistent with other aspects of the rule, it is unsupported. The Order cites no evidence to support its assumption that market share fluctuates more in television broadcasting than in other industries. Nor does it refute the Citizen Petitioners' suggestion that this is unlikely to be the case because, unlike most other industries, television station owners face a barrier to market entry (requirement of a license) and the number of market participants (television station owners) is in decline.

The Citizen Petitioners also object to the numerical limits because they allow markets that are already highly concentrated to become even more concentrated. For example, Philadelphia is a market with more than 18 stations, so the modified rule would allow the duopolist Viacom to acquire a third station and potentially increase its audience share from 25% to 34%. Philadelphia's HHI score is already 2037, well above the Commission's 1800 target, and a Viacom triopoly would raise the score to 2487. We acknowledge that the Commission never intended the numerical limits to represent a "mechanical application of the DOJ/FTC Merger Guidelines." Id. ¶ 197. But it expressly chose its six equal-sized competitor benchmark to ensure that markets would not exceed the Merger Guidelines' 1800 threshold for highly competitive markets.54 Id. ¶ 192. After justifying 1800 as the target, the Commission relaxed the local television ownership rule to allow more concentration in markets that already exceed the target-which is not just some markets, but most. UCC Reply Comments attachment 5 (28 out of 33 markets studied had HHIs that exceeded 1800, based on 2001 data).55

The deference with which we review the Commission's line-drawing decisions extends only so far as the line-drawing is consistent with the evidence or is not "patently unreasonable." Sinclair, 284 F.3d at 162. The Commission's numerical limits are neither. No evidence supports the Commission's equal market share assumption, and no reasonable explanation underlies its decision to disregard actual market share. The modified rule is similarly unreasonable in allowing levels of concentration to exceed further its own benchmark for competition (1800)-a glaring inconsistency between rationale and result. We remand the numerical limits for the Commission to support and harmonize its rationale.

E. We remand the Commission's repeal of the Failed Station Solicitation Rule.

The Citizen Petitioners also challenge the Commission's repeal of the Failed Station Solicitation Rule ("FSSR"), 47 C.F.R. 73.3555 n.7, which required a waiver applicant to provide notice of the sale to potential out-of-market buyers before it could sell the failed, failing, or unbuilt television station to an in-market buyer.56 Order ¶ 225.

The Commission promulgated the FSSR in its review of the local television rule that Congress had required under § 202(c) of the 1996 Act. To alleviate concerns that its decision to allow duopolies would undermine television station ownership by minorities, the Commission created the FSSR to ensure that qualified minority broadcasters had a fair chance to learn that certain financially troubled — and consequently more affordable — stations were for sale. 1999 Television Rule Review, 14 F.C.C.R. 12,903, ¶¶ 13-14, 74. In the current Order, however, the Commission does not explain that preserving minority ownership was the purpose of the FSSR, nor does it argue that the FSSR was harmful or ineffective toward this purpose. The only support for its decision to eliminate the FSSR is its prediction that "the efficiencies associated with operation of two same-market stations, absent unusual circumstances, will always result in the buyer being the owner of another station in that market."57 Order ¶ 225.

By failing to mention anything about the effect this change would have on potential minority station owners, the Commission has not provided "a reasoned analysis indicating that prior policies and standards are being deliberately changed, not casually ignored." Greater Boston TV Corp. v. FCC, 444 F.2d 841, 852 (D.C.Cir.1970). Furthermore, while the Commission had promised in 1999 to "expand opportunities for minorities and women to enter the broadcast industry," 1999 Television Rule Review, 14 F.C.C.R. 12,903, ¶ 14, the FSSR remained its only policy specifically aimed at fostering minority television station ownership. In repealing the FSSR without any discussion of the effect of its decision on minority television station ownership (and without ever acknowledging the decline in minority station ownership notwithstanding the FSSR), the Commission "entirely failed to consider an important aspect of the problem," and this amounts to arbitrary and capricious rulemaking.58 State Farm, 463 U.S. at 43, 103 S.Ct. 2856; see also Copps Dissenting, 18 F.C.C.R. at 13,970-71 (chastising the Commission for "fail[ing] to conduct rigorous analysis of today's rules on minorities and women"); Adelstein Dissent, 18 F.C.C.R. at 13,997 (same). For correction of this omission, we remand.59

Petitioners challenge the Commission's decision to modify its local radio ownership rule, which limits the number of commercial radio stations that a party may own in local markets of different sizes, 47 C.F.R. § 73.3555(a), by, among other changes, adopting a new method for determining the size of local markets. They also argue that the Commission failed to justify its decision to retain the rule's specific numerical limits. We affirm the Commission's decision to modify the rule (including modifying its method for determining local market size), but we agree that its decision to retain the numerical limits was arbitrary and capricious, and hence remand for the Commission's further consideration.

A. Regulatory Background and the 2002 Biennial Review

In 1992 the Commission abandoned its one-to-a-market limit on radio station ownership and implemented tiered numerical limits that allowed for common ownership of as many as six (three AM and three FM) stations — but no more than a combined 25% of the market's total audience share — in the largest markets of 40 or more commercial stations.60 In 1996 Congress directed the Commission to relax the tiered limits even more, to allow for as many as eight radio stations (but no more than five in the same AM or FM service) to be commonly owned in the largest markets of 45 or more commercial stations.61 Additionally, Congress required that the Commission review these limits biennially and repeal or modify them as necessary in the public interest. Id. § 202(h), 110 Stat. at 111-12.

In its 1998 and 2000 biennial reviews, the Commission decided to retain the same numerical limits enacted by Congress after finding that they continued to be necessary in the public interest. 1998 Biennial Regulatory Review, 15 F.C.C.R. 11,058, ¶ 59, 2000 WL 791562 (2000); 2000 Biennial Regulatory Review, 16 F.C.C.R. 1207, ¶ 32, 2001 WL 38123 (2001). But separately the Commission initiated a proceeding to consider modifying the way in which it "determine[s] the dimensions of radio markets and counts the number of stations in them." Definition of Radio Markets, 15 F.C.C.R. 25,077, ¶ 1, 15 F.C.C.R. 25077 (2001). That rulemaking was later consolidated with the 2002 biennial review. Notice, 17 F.C.C.R. 18,503, ¶ 7.

In its 2002 biennial review, the Commission again decided that the numerical limits set by Congress in 1996 remained necessary in the public interest. Order ¶ 239. But the Commission changed its method for determining the size of a radio station market. Since the Commission first promulgated tiered numerical limits in 1992, it had defined the markets as the overlapping area of radio stations' coverage. 47 C.F.R. § 73.3555(a)(3)(ii); 57 Fed.Reg. 18,089 (Apr. 29, 1992). For each proposed combination of stations, the local market was defined by "the number of ... stations whose principal community contours overlap, in whole or in part, with the principal community contours of the stations" that were proposed to be commonly owned. 47 C.F.R. § 73.3555(a)(3)(ii).62

Following the 2002 biennial review, the Commission decided to replace this so-called "contour-overlap methodology" with the geography-based market definition used by Arbitron, a private entity that measures local radio station audiences for its customer stations. Order ¶¶ 275-76. Within these so-called "Arbitron Metro markets," the number of total radio stations would include both commercial and noncommercial stations that are licensed to a community within the market, as well as stations located outside the market that attract a minimum level of listenership within, see id. ¶¶ 279-81, 295. Under the 1996 Act, only commercial stations had counted in determining the size of the market.

Though the Commission decided to grandfather any existing radio station combinations rendered noncompliant by the switch to Arbitron Metro markets, it restricted transfer of those combinations, subject to a limited exception for sales to eligible small businesses. Id. ¶ 489.

In an additional modification to the local radio ownership rule, stations under Joint Sales Agreements — as noted earlier, when a licensee authorizes a broker station to sell advertising time on the licensee's station in return for a fee — would be "attributed" to the brokering entity for the purpose of determining the brokering entity's compliance with the applicable limit. Id. ¶ 317.

B. We uphold the Commission's new definition of local markets.

Both the Deregulatory Petitioners and the Citizen Petitioners object to the Commission's changes to the way in which it defines a local radio market. Predictably, the Deregulatory Petitioners object to the modification that decreases the size of local markets (using Arbitron Metro markets instead of the contour-overlap methodology) — which potentially lowers the applicable numerical limit in a given market — while the Citizen Petitioners object to the modification that increases the size of local markets (inclusion of noncommercial stations) ___ which potentially raises the applicable numerical limit in a given market. But we conclude that the Commission has justified these changes under § 202(h) and they are not arbitrary and capricious under the APA.

1. The Commission justified using Arbitron Metro markets.

We reject out of hand the Deregulatory Petitioners' argument that, by changing the market definition methodology, the Commission effectively increased the numerical limits in violation of § 202(h)'s presumption in favor of deregulation. As discussed in Part II above, § 202(h) is not a one-way ratchet. The Commission is free to regulate or deregulate as long as its regulations are in the public interest and are supported by a reasoned analysis. The change to Arbitron Metro markets is both.

Under the existing rule, to determine whether an entity may acquire a radio station under the local radio rule, the Commission first must know how many radio stations are in that station's local market (called the "denominator" figure). The size of the market determines which numerical limit applies. Second, the Commission must determine how many radio stations in that market would be owned by the same entity if the entity acquired the stations it proposes (called the "numerator" figure). If this figure is within the numerical limit, the transaction may proceed.

Under the contour-overlap methodology, the Commission calculates the numerator by counting the acquiring entity's radio stations that all have overlapping signal contours. A station whose signal contour overlaps with some but not all of the other stations' contours will not be counted in the numerator. The Commission calculates the denominator by counting all of the stations whose contours intersect with at least one (not all) of the contours of another station in the numerator.

The Commission found that the contour-overlap methodology results in inconsistency between the numerator and denominator figures for a given transaction. Some radio stations owned by the acquiring entity may be counted in the denominator (because their contours intersect with at least one of the numerator-station's contours) but not the numerator (because their contours do not intersect with all of the other numerator-station's contours). An acquiring entity might actually own stations that do not count toward determining how many more it may own within the numerical limit, but do count toward determining the size of the market, where a bigger number correlates to a higher applicable limit. Order ¶ 253. The numerator/denominator inconsistency gives rise not only to an artificially low numerator/denominator ratio, which favors combinations but also the potential that combination owners could strategically acquire stations — those whose signals overlap with some, but not all of their already-owned stations' signals — to end-run the numerical limits altogether. Id. ¶ 254. Furthermore, a "perverse incentive" problem arises, as the contour-overlap methodology actually encourages consolidation of powerful radio stations because the stations with larger signals are more likely to create larger markets, which makes it more likely that their owner would be able to acquire even more radio stations in the market. Id. ¶ 257.

The Commission determined that it could not fix these problems with the contour-overlap methodology "merely by excluding commonly owned stations from the denominator or including those stations in the numerator." Id. ¶ 255. Excluding those stations from the denominator would mean the Commission was determining the size of the market based on who owns the stations, "a distinction that would be both unprincipled and unprecedented in the history of competition analysis." Id. On the other hand, including all commonly owned stations that are represented in the denominator could overly inflate ownership levels by including "outlying stations far from the area of concentration." Id.

The Commission also found that its transaction-specific market definitions frustrated its ability to benchmark and compute the level of competition in a given area. Thus it determined that "a local radio market that is objectively determined, i.e., that is independent of the radio stations involved in a particular acquisition, presents the most rational basis for the defining radio markets." Id. ¶ 273. For these reasons, the Commission decided to jettison the contour-based methodology and instead employ Arbitron's existing geographical definitions of local markets, finding they are the "established industry standard" that "represent a reasonable geographic market delineation within which radio stations compete." Id. ¶ 276.63

As the Deregulatory Petitioners point out, the Commission's reliance on Arbitron Metro markets is not without flaw. One obvious problem is that this method of measure does not cover the entire country. In fact, about 70% of counties in the United States are not within one of the 287 Arbitron Metro markets. But people tend to be clustered in specific population centers, as 78% of the United States' population over the age of 12 lives in Arbitron Metro markets. Nor has the Commission ignored the minority of the public who does not live in Arbitron Metro markets. It initiated a new rulemaking proceeding to develop market definitions for these areas, and in the interim decided to use a modified contour-overlap methodology that "minimize[s] the more problematic aspects of that system." Id. ¶ 285.

The Deregulatory Petitioners argue that the Commission failed to demonstrate that the contour-overlap methodology was actually harming competition. One petitioner called the Commission's explanation of the numerator/denominator inconsistency an "exercise in graduate school theoretical economics." Tr. at 394. But the Commission did find evidence of potential competitive harm in a working group study suggesting that station consolidation under the contour-overlap regime has resulted in an increase in advertising prices. Order ¶ 261 n. 548. And, as described above, the Commission observed that the contour-overlap methodology created a "perverse incentive" problem, as the most powerful stations whose contours intersect more stations (and thus have a larger denominator) are more able than smaller stations to acquire more stations. Id. ¶ 257. The Commission also found that the contour-overlap methodology impairs its ability accurately to measure and compare competition in consistently defined markets. Id. ¶ 259. The Commission's findings provide ample justification for the conclusion that the contour-overlap methodology's resulting inconsistencies are more than just abstract theory.

Next we reject the Deregulatory Petitioners' argument that the Commission improperly departed from its past precedent, namely, its 1992 and 1994 decisions to retain the contour-overlap methodology. See Revision of Radio Rules and Policies, 7 F.C.C.R. 6387, ¶¶ 37-40, 1992 WL 690638 (1992); Revision of Radio Rules and Policies, 9 F.C.C.R. 7183, ¶ 35, 1994 WL 617591 (1994). In State Farm, the Supreme Court said that when an agency reverses its "former views" it is "obligated to supply a reasoned analysis for the change beyond that which may be required when an agency does not act in the first instance." 463 U.S. at 41-42, 103 S.Ct. 2856; see also Rust v. Sullivan, 500 U.S. 173, 186, 111 S.Ct. 1759, 114 L.Ed.2d 233 (1991). Here the Commission supplied a reasoned analysis for the change — that "[i]t was only after the ownership limits were substantially raised in the 1996 Act that the scope of the market distorting effects of that system became manifest." Order ¶ 262.

The Deregulatory Petitioners argue that because radio is a signal-based enterprise, any market definition that ignores the actual reach of station signals and the scope of their listenership does not accurately measure competition. But we believe that it was reasonable for the Commission to conclude that fixed, geography-based market definitions more readily enable accurate measurement and comparison of competition than a transaction-specific, contour-based definition. Id. ¶ 259.

Finally, the Deregulatory Petitioners point out that Arbitron makes money from its station customers, and it has financial and other obligations to those stations and its shareholders. Therefore, they argue, Arbitron's economic incentives render its market definitions subject to manipulation. But the Commission recognized these potential problems and adequately established specific safeguards to deter potential manipulation, including a two-year buffer period before any party can receive the benefit of either a change in Arbitron Metro market boundaries or the addition of more radio stations to the market. Id. ¶ 278. The Deregulatory Petitioners' suggested conjectural flaws do not persuade us that the Commission's adoption of Arbitron Metro markets was arbitrary and capricious. We also reject their contention that the Commission's use of Arbitron markets delegates governmental power to a private entity. Arbitron will only provide a mechanism for measuring concentration. Because the Commission remains the sole arbiter of whether a proposed radio station combination serves the public interest, no improper delegation will occur.

For these reasons, we conclude that the Commission's decision to replace contour-overlap methodology with Arbitron Metro markets was "in the public interest" within the meaning of § 202(h) and that it was a rational exercise of rulemaking authority.

2. The Commission justified including noncommercial stations.

The Citizen Petitioners object to the Commission's decision to count both noncommercial and commercial stations in determining the size of an Arbitron Metro market.64 They argue that the inclusion of noncommercial stations increases the number of stations in a given market and thereby allows more consolidation under the numerical limits. It is impermissibly inconsistent, they argue, to increase effectively the numerical limits in this way while at the same time defending the existing numerical limits from arguments that they should be relaxed. But we reject this argument's factual premise that counting noncommercial stations will increase significantly the number of stations in a given market. Even counting noncommercial stations in the definition of markets, the change to Arbitron Metro markets operates as a net decrease in most markets' size.65 This undercuts the Citizen Petitioners' argument that it was inconsistent for the Commission to increase the sizes of local markets while justifying the existing numerical limits. It was not arbitrary and capricious, therefore, for the Commission to take account of noncommercial stations in determining the ownership limits.66

C. We uphold the Commission's transfer restriction.

In applying its local ownership rules, the Commission decided to grandfather existing radio, television, and radio/television combinations, and did "not require entities to divest their current interests in stations in order to come into compliance with the new ownership rules." Order ¶ 484.67 Under past Commission rules, radio stations created in compliance with the rules were freely transferable under the Commission's policy that maintaining the status quo did not create new harm to public interest goals. See Revision of Radio Rules and Policies, 7 F.C.C.R. 6387, ¶ 48. Switching course, in this Order the Commission refused to permit the transfer or sale of grandfathered combinations that violate its local ownership limits except to certain "eligible entities" that qualify as small businesses.68 Order ¶ 488. We reject the Deregulatory Petitioners' arguments that this modification to the local radio ownership rule contravenes § 202(h), constitutes arbitrary and capricious rulemaking, and violates the United States Constitution.

1. Transfer restriction is "in the public interest."

The Commission successfully demonstrates that the transfer restriction is "in the public interest" as required under § 202(h).69 The Commission notes that forcing divestitures at the time of transfer "could afford new entrants the opportunity to enter the media marketplace" and "could give smaller station owners already in the market the opportunity to acquire more stations," which would promote the public interest in having competitive and diverse radio markets. Id. ¶ 487. Additionally, the Commission found a strong public interest in curtailing the perpetuation of "combinations [that] were created pursuant to a market definition that we conclude fails to adequately reflect competitive conditions." Id. Unlike its decision not to require noncompliant combinations to divest, the Commission decided to restrict transfers because it found no "countervailing considerations" (such as the expectancy interests of existing station owners) to outweigh the public interest. Id. This rationale satisfies the public interest requirement under § 202(h).

2. Transfer restriction is reasoned decisionmaking.

According to the Deregulatory Petitioners, the transfer restriction is arbitrary and capricious because the Commission "failed to consider an important aspect of the problem" — as required under State Farm, see 463 U.S. at 43, 103 S.Ct. 2856 — by ignoring the fact that combination-owners' investments are diminished by the rule. Requiring station combinations to spin off noncompliant stations depresses the price of the station combinations as a whole, which deprives combination-owners (as sellers) of their expected investments. But an agency is not restricted in its rulemaking by the expectations of the regulated. Borough of Columbia v. Surface Transp. Bd., 342 F.3d 222, 236 (3d Cir.2003) (recognizing that parties who enter the market knowing that the industry is heavily regulated "cannot be said to have been surprised that the [agency's] decisions might thwart its goals"); see also Sinclair, 284 F.3d at 166 (upholding a modification to the local television ownership rule "notwithstanding that it may upset some expectations").

The Deregulatory Petitioners also suggest that the Order arbitrarily and capriciously exempted sales to "eligible entities" from the transfer restriction because the Commission did not purport to analyze whether small businesses ($6 million or less in annual revenue) would actually be able to obtain financing necessary to acquire a noncompliant radio station combination. But the rationality of the transfer restriction does not depend on the "eligible entities" exception. It is enough for the Commission to decide, as it did, that when it comes to small business the relative diversity advantages to a market in which an eligible-entity transfer occurred would outweigh the disadvantages to the market's competition interests from the perpetuation of a noncompliant combination. The number of times the exception is invoked is irrelevant to the reasonableness of the transfer restriction as a whole.70

3. Transfer restriction is constitutional.

We reject also the Deregulatory Petitioners' suggestion that the Commission's restriction on the transferability of pre-existing noncompliant consolidations "raises serious constitutional questions"71 under the Due Process and Takings Clauses of the Constitution's Fifth Amendment.72

In order to succeed in a due process or takings case under the Fifth Amendment, the plaintiff must first show that a legally cognizable property interest is affected by the Government's action in question. See Cleveland Bd. of Educ. v. Loudermill, 470 U.S. 532, 538, 105 S.Ct. 1487, 84 L.Ed.2d 494 (1985) (requiring protected property interest for due process violation); Webb's Fabulous Pharmacies, Inc. v. Beckwith, 449 U.S. 155, 160-61, 101 S.Ct. 446, 66 L.Ed.2d 358 (1980) (requiring protected property interest for taking). But broadcast licenses, which are the subject of the Commission's restriction on transferability, are not protected property interests under the Fifth Amendment. See FCC v. Sanders Bros. Radio Station, 309 U.S. 470, 475, 60 S.Ct. 693, 84 L.Ed. 869 (1940) ("The policy of the [1936 Communications] Act is clear that no person is to have anything in the nature of a property right as a result of the granting of a license."); see also CBS, Inc. v. FCC, 453 U.S. 367, 395, 101 S.Ct. 2813, 69 L.Ed.2d 706 (1981) (noting that the broadcast spectrum is part of the public domain and that a broadcaster merely has "use of a limited and valuable part of the public domain; when he accepts that franchise it is burdened by enforceable public obligations" (quoting Office of Communication of the United Church of Christ v. FCC, 359 F.2d 994, 1003 (D.C.Cir.1966))); 47 U.S.C. § 301 (broadcast licenses provide for the "use ... but not the ownership" of channels of communication). Broadcast licensees accept their licenses subject to the Commission's power to regulate their transfer pursuant to "the public interest, convenience, and necessity." 47 U.S.C. § 310(d). Because a legally cognizable property interest is not implicated by the Commission's decision to restrict the transfer of pre-existing, noncompliant station combinations, a Fifth Amendment violation will not occur.

D. We affirm the attribution of Joint Sales Agreements.

Under a Joint Sales Agreement ("JSA"), a radio station authorizes a broker to sell advertising time on the station in exchange for a fee. In the Order, the Commission decided to count stations brokered under a JSA toward the brokering station's permissible ownership totals (that is, to "attribute" JSAs), as long as (1) the brokering entity owns or has an attributable interest in one or more stations in the local market, and (2) the joint advertising sales amount to more than 15% of the brokered station's advertising time per week. Order ¶ 317. The Commission justified this departure from its prior policy of nonattribution with its finding that in-market JSAs above the 15% threshold convey to the brokering entity a degree of "influence or control" sufficient to warrant limitation under the ownership rules — i.e., a "realistic potential to affect [the brokered station's] programming and other core operating decisions."73 Id. ¶ 318. Furthermore, the brokering entity has the ability and the incentive to exercise market power because it controls the advertising revenue of the brokered stations and assumes the financial risks and rewards of advertising. Id. ¶ 320. The Commission gave parties to JSAs two years to end their agreements or otherwise come into compliance with the local ownership rules. Id. ¶ 325. We reject the Deregulatory Petitioners' arguments that this modification to the local radio ownership rule contravenes § 202(h), constitutes arbitrary and capricious rulemaking, and violates the Constitution.

1. Attribution of JSAs is "necessary in the public interest."

In the Order the Commission explained that, due to their potential to convey influence to brokering entities, JSAs "may undermine [the Commission's] continuing interest in broadcast competition sufficiently to warrant limitation under the multiple ownership rule." Id. ¶ 318. The Commission's decision to modify its attribution policy to "reflect accurately the competitive conditions of today's local radio markets," id. ¶ 321, and thus prevent its local radio rule from being undermined, is a modification "in the public interest" under § 202(h).74

2. Attribution of JSAs is reasoned decisionmaking.

An agency that departs from its "former views" is "obligated to supply a reasoned analysis for the change beyond that which may be required in the first instance" in order to survive judicial scrutiny for compliance with the APA. State Farm, 463 U.S. at 41-42, 103 S.Ct. 2856. Here the Commission acknowledged that its decision to attribute JSAs was a departure from its prior policy of nonattribution, promulgated in 1999. Order ¶ 320 n. 698. The Commission explained that, "upon reexamination of the issue," it found that JSAs convey the potential for brokering entities to influence the brokered stations; thus a policy of attributing JSAs is necessary to "reflect accurately competitive conditions of today's local radio markets." Id. ¶¶ 320, 321.

The Deregulatory Petitioners argue that this proffered rationale is insufficient because the Commission does not explain what, if anything, transpired since 1999 that accounts for its change of position. But because we interpret the Commission's modification as a correction of its prior policy's "inaccuracy," the Commission's failure to cite a particular intervening change is not fatal to its reasoned analysis. In this context we accept that the Commission's determination — upon "reexamination of the issue" that the JSAs convey (and always have conveyed) a potential for influence — sufficiently rationalizes its decision to jettison its prior nonattribution policy and replace it with one that more accurately reflects the conditions of local markets.

3. Attribution of JSAs is constitutional.

We also disagree with the Deregulatory Petitioners' suggestion that the attribution of JSAs "raises serious constitutional concerns"75 under the Fifth Amendment's Takings Clause. Contracts such as JSAs are protected property interests under the Fifth Amendment, see United States Trust Co. of N.Y. v. New Jersey, 431 U.S. 1, 19 n. 16, 97 S.Ct. 1505, 52 L.Ed.2d 92 (1977), but here the Commission has not invalidated or interfered with any contracts. In deciding to attribute JSAs, it is has simply decided that stations subject to JSAs should, in certain circumstances, count toward the regulatory limit in determining how many stations the brokering entity may own in a market. Moreover, station owners have no vested right in the continuation of any particular regulatory scheme. Folden v. United States, 56 Fed. Cl. 43, 61 (2003) (parties "in a highly regulated field such as FCC licensing can have no distinct investment-backed expectations that include a reliance upon a legislative and regulatory status quo"); cf. Connolly v. Pension Benefit Guar. Corp., 475 U.S. 211, 222-27, 106 S.Ct. 1018, 89 L.Ed.2d 166 (1986) (no regulatory taking occurred because government had not appropriated property for its own use, did not impose a severe economic impact, and did not interfere with reasonable expectations). Thus we reject the Deregulatory Petitioners' suggestion that the attribution of JSAs will result in a regulatory taking.

E. We remand the numerical limits to the Commission for further justification.

In the Order, the Commission announced its decision to retain the existing numerical limits on radio ownership that Congress established in § 202(b) of the 1996 Act. The existing limits preclude the common ownership of radio stations with overlapping principal community contours, except that: (1) in a radio market with 45 or more commercial radio stations, a party may own up to eight commercial radio stations, not more than five of which are in the same service (that is, AM or FM); (2) in a radio market with between 30 and 44 commercial radio stations, a party may own up to seven commercial radio stations, not more than four of which are in the same service; (3) in a radio market with between 15 and 29 commercial radio stations, a party may own up to six commercial stations, not more than four of which are in the same service; and (4) in a market with 14 or fewer commercial radio stations, a party may own up to five commercial radio stations, not more than three of which are in the same service, except that a party may not own more than 50 percent of the stations in that market. 1996 Act, § 202(b)(1), 110 Stat. at 110; 47 C.F.R. § 73.3555(a)(1).

First, the Commission defended its use of numerical limits from those commenters who argued that there should be no limits at all on local radio station ownership. The Commission explained that radio is a closed-entry market, i.e., all available radio frequency spectrum has been licensed. So the only way to develop market power is to buy existing stations. Thus, the Commission says, numerical limits are a reasonable way of preventing available capacity from being "locked up" in the hands of a few owners. Order ¶ 288.

The Commission then explained that the specific numerical limits that it currently employs are effective to this end. Citing economic literature, the Commission suggested that a market with five equal-sized competitors is a sufficiently competitive market. The existing limits allow for roughly five equal-sized firms in each market. (For example, the eight-station limit for markets of 45 or more stations allows for five eight-station combinations). Id. ¶ 289. The Commission concluded that the existing limits ensure sufficiently competitive local markets.76 The Commission also concluded that the existing limits are not overly restrictive based on data showing that the top-four stations in each metro market are thriving in terms of revenue and audience share. Id. ¶ 290.

As explained in Part II.B above, § 202(h) operates to extend the "reasoned analysis" requirement (which ordinarily applies only to an agency's decision to promulgate new regulations or modify existing ones) to apply also to the Commission's decision to retain existing regulations. Although we accept the Commission's rationale for employing numerical limits (as opposed to other regulatory approaches such as a case-by-case analysis), we conclude that the Commission's decision to retain the specific numerical limits is not supported by a reasoned analysis.

1. The Commission's numerical limits approach is rational and in the public interest.

The Commission's decision to retain a numerical limits approach to radio station ownership regulation is "in the public interest." Without numerical limits, radio markets risk becoming "locked up" in the hands of a few owners (or even one owner) because all of the available radio frequency spectrum has been licensed — a high barrier to new market entrants. Order ¶ 288. Based on record evidence, the Commission justifiably concluded that numerical limits are necessary "to guard against consolidation ... and to ensure a market structure that fosters opportunities for new entry into radio broadcasting." Id. ¶ 291. For example, a MOWG study found that, since the existing limits were imposed in 1996, the number of radio station owners declined by 34% even though the number of stations increased by 5.4%. George Williams & Scott Roberts, Radio Industry Review 2002: Trends in Ownership, Format, and Finance (MOWG Study No. 11) at 3 (Sept. 2002). Additionally, the record shows that today 10 parent companies — the largest of which, Clear Channel Communications, owns 1200 stations nationwide, or 10% — dominate the radio industry and control about two-thirds of both listeners and radio revenues nationwide. Id. at 4. In contrast, prior to the 1996 Act's deregulation, the largest nationwide radio station combinations had fewer than 65 stations each. Id.

Furthermore, the record shows how increased consolidation has increased station prices, which limits opportunities for new market entrants and as a result limits diversity in station ownership and output. UCC Comments at 18. Consolidation has also reduced the amount of locally produced radio content, as large group-owners often broadcast remotely from national offices instead of having local employees produce programming. Comments of Future of Music Coalition, MB Docket 02-277 at 13-14 (Nov. 20, 2002). The record contains examples of consolidated stations that eliminated local news production. UCC Comments at 37-38. The evidence thus supports the Commission's conclusion that, by continuing to limit the consolidation of radio stations, numerical limits are "in the public interest" as required under § 202(h).

2. The Commission did not support its decision to retain the existing numerical limits with reasoned analysis.

Both the Citizen Petitioners and the Deregulatory Petitioners argue that the Commission's decision to retain the existing numerical limits was arbitrary and capricious. Predictably, the Citizen Petitioners argue that the Commission should have tightened the existing limits, and the Deregulatory Petitioners argue that the Commission should have relaxed them. But both sides' predominant argument is essentially the same: the numerical limits are not supported by the Commission's theory that they ensure five equal-sized competitors in most markets. While, as discussed above, substantial evidence supports the Commission's decision to retain the numerical limits structure of its local radio ownership rule, we also agree with the Petitioners that the Order lacks a reasoned analysis for retaining these specific numerical limits. We thus remand for the Commission's additional justification.

a. The Commission did not sufficiently justified "five equal-sized competitors" as the right benchmark.

The Commission relied on game theory to support its premise that five equal-sized competitors ensure that local markets are fragmented and structurally competitive. Order ¶ 289 n. 609 (citing Louis Phillips, Competition Policy: A Game Theory Perspective Ch. 2 (1995); Timothy F. Bresnahan & Peter C. Reiss, Entry and Competition in Concentrated Markets, 99 J. Pol. Econ. 997 (1991); Reinhard Selten, A Simple Model of Imperfect Competition Where Four Are Few and Six Are Many, 2 Int'l J. Game Theory 141 (1973)). The Citizen Petitioners and the Deregulatory Petitioners both dispute that these articles support the Commission's selection of five equalsized competitors as the appropriate benchmark. The Deregulatory Petitioners argue that the articles fall short because they do not rule out market structures other than equal-sized competitors (such as one large firm and many small ones) as equally competitive markets. This argument, unanswered by the Commission, warrants a response because (as discussed below) the record evidence supports neither actual nor potential existence of equal-sized competitors.

The Citizen Petitioners argue that the three articles the Commission cites are superseded and contradicted by the Department of Justice and Federal Trade Commission's most recent re-write of the Merger Guidelines, under which a market with five equal-sized competitors is considered "highly concentrated." See Merger Guidelines § 1.51(c) (a highly concentrated market has an HHI score higher than 1800). The Commission's decision to rely on theory that conflicts with the Merger Guidelines is suspect because, in the same Order, the Commission relied on the Merger Guidelines to derive its new limits for local television station ownership. See Order ¶¶ 192-93. The Commission did not address this discrepancy in its Respondent's Brief,77 and should do so on remand.

b. The Commission did not sufficiently justified that the existing numerical limits actually ensure that markets will have five equal-sized competitors.

Regardless whether five equal-sized competitors is the right benchmark for competition, the Commission did not sufficiently justified that five equal-sized competitors would emerge or actually have emerged under the numerical limits. It defies logic to assume that a combination of top-ranked stations is the competitive equal to a combination of low-ranked stations just because the two combinations have the same number of stations. The Commission itself acknowledges that "radio station groups with similar numbers of radio stations [can] have vastly different levels of market power." Order ¶ 290.

Furthermore, evidence shows that the existing numerical limits do not ensure five equal-sized competitors. According to the record, most markets are dominated by one or two large station owners.78 And the top-four station owners together control the lion's share of the market.79 Even if these four station-owners were "equal-sized" (they are not), the HHI score of such a market would be 2500 (4 × (100 ÷ 4)²), well above the Merger Guidelines' 1800 threshold for highly concentrated markets.

The Commission does not explain why it could not take actual market share into account when deriving the numerical limits.80 Had it proffered the "market share is too fluid" rationale, we have already rejected that explanation in the context of the local television ownership rule and the Cross-Media Limits. We also note that the Commission has in the past extolled the value of audience share data for measuring diversity and competition in local radio markets.81 So the Commission's reliance on the fiction of equal-sized competitors, as opposed to measuring their actual competitive power, is even more suspect in the context of the local radio rule.

For these reasons, the Commission's numerical limits cannot rationally be derived from a "five equal-sized competitor" premise. We thus remand for the Commission to develop numerical limits that are supported by a rational analysis.

c. The Commission did not support its decision to retain the AM/FM subcaps.

The Deregulatory Petitioners challenge the Commission's decision to retain the AM/FM subcaps, which the Commission justified on the grounds that FM stations have technological and economic advantages over AM stations. Order ¶ 294. But the Deregulatory Petitioners point out, and we agree, that this does not explain why it is necessary to impose an AM subcap at all. The Commission does not respond in its brief to this particular criticism. Thus it should do so, or modify its approach, on remand.

Though we affirm much of the Commission's Order, we have identified several provisions in which the Commission falls short of its obligation to justify its decisions to retain, repeal, or modify its media ownership regulations with reasoned analysis. The Commission's derivation of new Cross-Media Limits, and its modification of the numerical limits on both television and radio station ownership in local markets, all have the same essential flaw: an unjustified assumption that media outlets of the same type make an equal contribution to diversity and competition in local markets. We thus remand for the Commission to justify or modify its approach to setting numerical limits. We also remand for the Commission to reconsider or better explain its decision to repeal the FSSR.82 The stay currently in effect will continue pending our review of the Commission's action on remand, over which this panel retains jurisdiction.

SCIRICA, Chief Judge, dissenting in part, concurring in part.

Although I concur in some parts of the Court's comprehensive analysis of this complex agency order, including its rejection of the constitutional challenges, I respectfully dissent from its decision to vacate and remand. In my view, the Court's decision has upended the usual way the judiciary reviews agency rulemaking. Whether the standard is "arbitrary or capricious," "reasonableness," or some variant of a "deregulatory presumption," the Court has applied a threshold that supplants the well-known principles of deference accorded to agency decision-making. In so doing, the Court has substituted its own policy judgment for that of the Federal Communications Commission and upset the ongoing review of broadcast media regulation mandated by Congress in the Telecommunications Act of 1996.

I would lift the stay and allow the Commission's media ownership rules to go into effect. It is not the role of the judiciary to second-guess the reasoned policy judgments of an administrative agency acting within the scope of its delegated authority. Allowing the biennial (now quadrennial) review process to run its course will give the Commission and Congress the opportunity to monitor and evaluate the effect of the proposed rules on the media marketplace. More importantly, it will ensure that accountability for these crucial policy decisions rests with the political branches of our government.

In 1934, Congress delegated broad authority to the Federal Communication Commission to regulate "interstate and foreign commerce in communication by wire and radio," and to grant station broadcast licenses that served "public convenience, interest, and necessity." Communications Act of 1934, 47 U.S.C. §§ 151, 309(a). This statutory authority to regulate broadcast media is girded by an obligation "to preserve an uninhibited marketplace of ideas in which truth will ultimately prevail." Red Lion Broad. Co. v. FCC, 395 U.S. 367, 390, 89 S.Ct. 1794, 23 L.Ed.2d 371 (1969). The FCC's delegated responsibility to foster a robust forum for national debate is unique in administrative law and essential to the vibrancy of our deliberative democracy. See Buckley v. Valeo, 424 U.S. 1, 49, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976) ("Democracy depends on a well-informed electorate.").

Preserving the "marketplace of ideas" does not easily lend itself to mathematical certitude. While other independent federal agencies may act with greater measurable precision in reducing pollution emissions, defining safety standards or even establishing interest rates, the FCC operates in the less scientific arena of speech and debate.83 In this realm, the Commission's mandate to maintain viewpoint diversity in the national broadcast media is complicated both by the "elusive" concept of diversity,84 and by the inherent uncertainty regarding the prospective effects of structural rules.85 Even the direct regulation of specific broadcast content — e.g., content guidelines for news and public affairs programming — does not necessarily assure viewpoint diversity will be achieved.

The Commission's duty to regulate broadcast media in the public interest is further complicated by the unpredictable impact of emerging technologies on the media marketplace. When the Commission was first formed, AM or "standard" radio was the only broadcast medium of practical concern. See 2002 Biennial Regulatory Review — Review of the Commission's Broadcast Ownership Rules and Other Rules Adopted Pursuant to Section 202 of the Telecommunications Act of 1996, 18 F.C.C.R. 13,620, 13,649-651 ¶¶ 90-94, 2003 WL 21511828 (2003) ("Order") (detailing history of the modern media marketplace). Today, the modern media marketplace includes literally thousands of radio and broadcast television stations, hundreds of national, regional and local non-broadcast television networks delivering a vast range of content over cable and direct broadcast satellite systems, and perhaps most significantly, the Internet and a host of digital technology-enabled interactive services. Id. ¶¶ 95-128. In designing media ownership rules for this dynamic technological landscape, the Commission must balance the potentially negative effects of license ownership concentration on programming diversity against the economic viability of new or struggling media outlets.

Over the past seventy years, the Commission has actively adjusted its license ownership rules on an ongoing basis to foster the growth of new media outlets while maintaining a focus on programming diversity. For much of its early history, the Commission operated with the assumption that diversification of ownership best served the public interest in promoting programming diversity.86 For example, in 1938 the Commission denied Genesee Radio Corporation's application for a second AM station license after concluding it was "not in the public interest to grant the facilities for an additional broadcast station to interests already in control of the operation of a station of the same class and in the same community." Genesee Radio Corp., 5 F.C.C. 183, 186 (1938). For the next forty years, media regulation continued primarily "on the theory that diversification of mass media ownership serves the public interest by promoting diversity of program and service viewpoints, as well as by preventing undue concentration of economic power."87 NCCB, 436 U.S. at 780, 98 S.Ct. 2096.

The Commission has departed from this baseline presumption when strict adherence to a policy of ownership diversity threatened to drive some media outlets from the market. For example, in the early 1970s, the Commission relaxed its "one-to-a-market" rule to permit co-ownership of AM and FM licenses in the same market after observing that most AM-FM combinations were "economically and/or technically interdependent," and that most free-standing FM stations were not economically viable without a co-owned AM license revenue. See Amendment of Sections 73.35, 73.240 and 73.636 of the Commission's Rules Relating to Multiple Ownership Rules of Standard, FM and Television Broadcast Stations, 28 F.C.C.2d 662, 671 ¶ 33, 1971 WL 23029 (1971). The Commission similarly permitted UHF-radio combinations on a case-by-case basis to encourage the development of the UHF broadcasting medium. Id. at 674. Likewise, in repealing national ownership caps for television and radio, the FCC concluded the potential economic efficiency gains from "group ownership actually further[], rather than frustrate[], the foremost First Amendment goal of augmenting popular discussion of important public issues." Amendment of Section 73.3555 of the Commission's Rules Relating to Multiple Ownership of AM, FM, and Television Broadcast Stations, 100 F.C.C.2d 17, 20, 1984 WL 251222 (1984).

The dynamic evolution of the media ownership rules — with the Commission intermittently encouraging and discouraging economic concentration — demonstrates the virtual impossibility of drafting a single, static regulatory structure that consistently serves the public interest for an extended time period. The Commission's statutory mandate to regulate broadcast media "in the public convenience, interest, and necessity" has been interpreted to require this sort of iterative, if imprecise, rulemaking. Nat'l Broad. Co. v. United States, 319 U.S. 190, 225, 63 S.Ct. 997, 87 L.Ed. 1344 (1943) ("If time and changing circumstances reveal that the `public interest' is not served by application of the Regulations, it must be assumed that the Commission will act in accordance with its statutory obligations.").

The Telecommunications Act of 1996, Pub. L. No. 104-104, 110 Stat. 56, codified this ongoing review process by requiring the Commission to conduct a biennial review of its media ownership rules and "repeal or modify" any rule that no longer serves the public interest. Id. at § 202(h), 110 Stat. at 111-112.88 On the cusp of an unprecedented revolution in communication technologies, Congress set in motion this statutorily-prescribed process of media deregulation based on the conviction that increased competition in the media marketplace would best serve the public interest.89 The significance of the mandatory review mechanism should not be ignored. This particular set of ownership rules — like any of those passed by the Commission — is not cast in iron. No single set of proposed rules can perfectly capture the dynamic nature of the media marketplace. These rules serve as an important point of reference in an ongoing review process. Nevertheless, the latest rules represent a reasoned step towards the pro-competitive, deregulatory media ownership framework contemplated by Congress.90

Short-circuiting the statutory review process deprives both the Commission and Congress the valuable opportunity to evaluate the new rules and the effects of deregulation on the media marketplace. Under the original implementation schedule, the Commission already would have initiated its assessment of the proposed rules in preparation for the 2004 Biennial Review. See Telecommunications Act of 1996 § 202(h). Vacating and remanding the proposed rules to the Commission will preserve the existing rules in place for months or even years,91 and the resulting delay will likely leave the public worse off than if these rules were allowed to take effect.92

Given the dynamic nature of the industry, the task of crafting a regulatory structure that reflects the realities of the media marketplace requires the Commission to make predictive judgments about the future. Courts have consistently recognized the Commission's authority and unique expertise in making such estimations. See Cellco P'ship v. FCC, 357 F.3d 88, 98 (D.C.Cir.2004) ("Nothing ... suggest[s] that under § 11's biennial review mandate the Commission could no longer rely on its predictive judgment or properly-supported inferences in determining to retain a regulation."). Courts also have correctly acknowledged it is virtually impossible for an agency to compile an unchallengeable factual record in support of forward-looking rules designed to anticipate the future development of the marketplace. As the Supreme Court noted, "[i]n such circumstances complete factual support in the record for the Commission's judgment or prediction is not possible or required; `a forecast of the direction in which future public interest lies necessarily involves deductions based on the expert knowledge of the agency.'" NCCB, 436 U.S. at 814, 98 S.Ct. 2096 (quoting FPC v. Transcontinental Gas Pipe Line Corp., 365 U.S. 1, 29, 81 S.Ct. 435, 5 L.Ed.2d 377 (1961)).

Foremost among the future developments facing the Commission is the explosive growth of the Internet as a media source. The unique qualities of the Internet provide an unimaginable breadth of accessible information as well as a forum for individual expression. The Supreme Court recognized the unique nature of this media source when it wrote:

The Internet ... offer[s] a forum for a true diversity of political discourse, unique opportunities for cultural development, and myriad avenues for intellectual activity." 47 U.S.C. § 230(a)(3) (1994 ed., Supp. V). While "surfing" the World Wide Web, the primary method of remote information retrieval on the Internet today, individuals can access material about topics ranging from aardvarks to Zoroastrianism. One can use the Web to read thousands of newspapers published around the globe, purchase tickets for a matinee at the neighborhood movie theater, or follow the progress of any Major League Baseball team on a pitch-by-pitch basis.

Ashcroft v. ACLU, 535 U.S. 564, 566, 122 S.Ct. 1700, 152 L.Ed.2d 771 (2002) (citations omitted).

By requiring the Commission to craft a set of media ownership rules that perfectly account for the effects of concentration, economic efficiency, diversity and future technological progress, the majority constructs a high bar and impedes the review process established by Congress. Instead of asking the Commission to start from zero, the prudent step would be to allow these reasoned rules to go into effect, monitor the resulting impact on the media marketplace, and allow the Commission to refine or modify its approach in its next quadrennial review. In the interim, if the Commission goes too far or ignores the scope of its statutory mandate, Congress itself may act, as it already has with respect to the national television ownership rules, see Consolidated Appropriations Act of 2004, 118 Stat. 3, to modify or repeal any rule it deems to be no longer in the public interest.

There are, of course, alternative approaches the Commission might have taken in crafting the proposed media ownership rules. But the Commission acted within its delegated authority by adopting a set of rules which it properly determined would serve the public interest. Our review of those rules is necessarily cabined within our traditional constitutional role. The courts "must guard against the danger of sliding unconsciously from the narrow confines of law into the more spacious domain of policy." Phelps Dodge Corp. v. NLRB, 313 U.S. 177, 194, 61 S.Ct. 845, 85 L.Ed. 1271 (1941). Whether or not we agree with this particular set of rules, it is not our role to overturn the Commission's reasoned policy judgments. See Chevron, U.S.A., Inc. v. Natural Res. Def. Council, 467 U.S. 837, 866, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984) ("The responsibilities for assessing the wisdom of such policy choices and resolving the struggle between competing views of the public interest are not judicial ones: `Our Constitution vests such responsibilities in the political branches.'") (quoting TVA v. Hill, 437 U.S. 153, 195, 98 S.Ct. 2279, 57 L.Ed.2d 117 (1978)). Questions of media diversity and ownership command strongly divergent views. But those questions should be answered by officials within our legislative and executive branches of government.

This is an appeal of an agency decision under the Communications Act of 1934, 47 U.S.C. § 151 et seq. This Court's jurisdiction is based on 47 U.S.C. § 402(a), and 28 U.S.C. § 2342(1). Although there are some similarities, I differ from the majority on the applicable standard of review. Moreover, I believe the majority's subsequent analysis oversteps the appropriate standard. In doing so, the majority substitutes its own judgment for policy decisions meant to be resolved by the Agency.

As noted, Petitioners bring a litany of challenges to the Commission's Order. Different standards of review apply to the various challenges. In reviewing a contention that an agency rule is arbitrary and capricious, our standard of review is governed first by the Administrative Procedure Act, 5 U.S.C. § 706. Under the APA, this Court is instructed to "hold unlawful and set aside agency action, findings, and conclusions" that are found to be "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law." § 706(2)(A); New Jersey Coal. for Fair Broad. v. FCC, 574 F.2d 1119, 1125 (3d Cir.1978).

The scope of review under the "arbitrary and capricious" standard is "narrow, and a court is not to substitute its judgment for that of the agency." Motor Vehicle Mfrs. Ass'n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 77 L.Ed.2d 443 (1983). Nevertheless, in reaching its decision, the agency must examine the relevant data and articulate a satisfactory explanation for its action including a "rational connection between the facts found and the choice made." Id. (quoting Burlington Truck Lines, Inc. v. United States, 371 U.S. 156, 168, 83 S.Ct. 239, 9 L.Ed.2d 207 (1962)). Normally, an agency rule is arbitrary and capricious where:

the agency has relied on factors which Congress has not intended it to consider, entirely failed to consider an important aspect of the problem, offered an explanation for its decision that runs counter to the evidence before the agency, or is so implausible that it could not be ascribed to a difference in view or the product of agency expertise. The reviewing court should not attempt itself to make up for such deficiencies; we may not supply a reasoned basis for the agency's action that the agency itself has not given.

Id.; see also Robert Wood Johnson Univ. Hosp. v. Thompson, 297 F.3d 273, 280 (3d Cir.2002). We will, however, "uphold a decision of less than ideal clarity if the agency's path may reasonably be discerned." Motor Vehicle Mfrs. Ass'n, 463 U.S. at 43, 103 S.Ct. 2856 (quoting Bowman Transp., Inc. v. Arkansas-Best Freight Sys., Inc., 419 U.S. 281, 286, 95 S.Ct. 438, 42 L.Ed.2d 447 (1974)).

The scope of our review is necessarily influenced by the authority allocated to the FCC in its enabling act. The 1934 Act grants "broad discretion" to the Commission to allocate broadcast licenses in the "public interest, convenience and necessity." FCC v. WNCN Listeners' Guild, 450 U.S. 582, 594, 101 S.Ct. 1266, 67 L.Ed.2d 521 (1981). The Act's public interest standard is a "supple instrument for the exercise of discretion by the expert body which Congress has charged to carry out its legislative policy." Id. at 593, 101 S.Ct. 1266 (quoting FCC v. Pottsville Broad. Co., 309 U.S. 134, 138, 60 S.Ct. 437, 84 L.Ed. 656 (1940)). The Commission's broad authority suggests that "the Commission's judgment regarding how the public interest is best served is entitled to substantial judicial deference," and "is not to be set aside" as long as its implementation of the public interest standard is "based on a rational weighing of competing policies." Id. at 596, 101 S.Ct. 1266.

Finally, the standard of review is even more deferential "where the issues involve `elusive' and `not easily defined' areas such as programming diversity in broadcasting." Sinclair Broad. Group v. FCC, 284 F.3d 148, 159 (D.C.Cir.2002). Because many aspects of this administrative order involve policy determinations on such elusive goals, a "rationality" standard is often appropriate. See NCCB, 436 U.S. at 796-97, 98 S.Ct. 2096 (finding the Commission acted rationally in determining diversification of ownership would enhance the likelihood of achieving diversity of viewpoints). Additionally, where issues involve line-drawing determinations, our review is necessarily deferential to agency expertise, see AT&T Corp. v. FCC, 220 F.3d 607, 627 (D.C.Cir.2000), so long as these decisions do not run counter to the evidence before the agency. Sinclair, 284 F.3d at 162.

Several Deregulatory Petitioners challenge the Order as violating § 202(h) of the 1996 Act. This section reads:

The Commission shall review its rules adopted pursuant to this section and all of its ownership rules biennially as part of its regulatory reform review under section 11 of the Communications Act of 1934 and shall determine whether any of such rules are necessary in the public interest as the result of competition. The Commission shall repeal or modify any regulation it determines to be no longer in the public interest.

Because the Order was promulgated as part of the biennial review required by this section, our review is necessarily informed by Congress' direction in the statute that in addition to reviewing its ownership rules biennially to "determine whether any of such rules are necessary in the public interest as the result of competition[,] [t]he Commission shall repeal or modify any regulation it determines to be no longer in the public interest." § 202(h).

This statutory language has been interpreted to correspond with the deregulatory process codified in the 1996 Act. See Fox I, 280 F.3d at 1033 (noting that in § 202(h) Congress instructed the Commission to "continue the process of deregulation"); Sinclair, 284 F.3d at 159 ("This sentence appears as the last sentence of Congress's instruction that the review of each of its ownership rules every two years `which the court characterized as designed to continue the process of deregulation.'") (quoting Fox I, 280 F.3d at 1033).

The FCC recognized its heightened burden for maintaining regulations in its Order, noting that "Section 202(h) appears to upend the traditional administrative law principle requiring an affirmative justification for the modification or elimination of a rule."93 Order ¶ 11. The FCC further recognized the direction from Fox and Sinclair that "Section 202 carries with it a presumption in favor of repealing or modifying the ownership rules." Order ¶ 11; Fox, 280 F.3d at 1048; Sinclair, 284 F.3d at 159.94 The FCC must demonstrate that retention of the rules remains in the public interest. See Fox, 280 F.3d at 1044 (faulting the Commission for providing no analysis of the state of the television broadcast market in maintaining the National Television Ownership Rule). As noted at the outset, this requirement does not preclude the FCC from relying on predictive judgments and expertise in crafting its rules. See id. at 1051 (noting that the "court should ordinarily defer to the Commission's predictive judgments"). The Commission's authority to rely on predictive judgments and empirical assumptions was recently affirmed by the Court of Appeals for the D.C. Circuit. See Am. Family Ass'n, Inc. v. FCC, 365 F.3d 1156, 1166 (D.C.Cir.2004) (noting "necessarily wide latitude to make policy based on predictive judgments deriving from [the Commission's] general expertise").

At issue on the § 202(h) standard of review is whether and to what extent the presumption to repeal or modify ownership rules turns on the definition of the term "necessary." The FCC interpreted "necessary" in § 202(h) to mean "useful," "convenient" or "appropriate" rather than "required" or "indispensable." 18 F.C.C.R. 4726, 4730-36 (March 14, 2003); Order ¶ 11 n. 15 (citing same). The FCC's interpretation of "necessary" is correct.95 The term "necessary" in statutes has been found repeatedly to mean "useful" or "appropriate" rather than "indispensable." 18 F.C.C.R. at 4731 n. 24 (citing, inter alia, Morgan v. Commonwealth of Va., 328 U.S. 373, 377-78, 66 S.Ct. 1050, 90 L.Ed. 1317 (1946)); see also Cellco, 357 F.3d at 99.

The FCC's interpretation of "necessary" does not neutralize the deregulatory flavor of the 1996 Act. Under the FCC's interpretation, a regulation must be useful (albeit not indispensable) in promoting the public interest in order to survive the biennial — now quadrennial — review process. The "presumption," therefore, is that a regulation will be vacated or modified if it does not continue to be in the public interest. This is different from the traditional approach to rule retention, which would counsel for retention of a rule unless there were reasons to change it. In this sense, Congress' mandate and the FCC's approach is "deregulatory" because it would repeal or modify rules no longer found to be in the public interest.

The FCC looked to the structure of the 1996 Act in reaching its interpretation. The FCC recognized that "necessary" appears in other sections of the Communications Act that discuss the FCC's rulemaking powers, and in those contexts "necessary" has taken on the "useful" meaning. See, e.g., 47 U.S.C. § 201(b). Holding the review process to a more stringent standard would lead to the absurd result of the FCC issuing a rule as merely useful, but then having to revoke it at the next review because it is not indispensable. Furthermore, if the rulemaking and review definitions of "necessary" were different, the FCC could simply sidestep the more stringent review standard by repealing and then reissuing any useful rules that were not indispensable.

The rulemaking provisions are permissive, while the biennial review is mandatory. Compare, e.g., id. (Commission "may prescribe such rules and regulations as may be necessary in the public interest") with § 202(h) ("The Commission shall review its rules ... biennially ... and shall determine whether any of such rules are necessary in the public interest as the result of competition. The Commission shall repeal or modify any regulation it determines to be no longer in the public interest.") (emphasis added). That the FCC is permitted to make rules but required to review them, however, does not shed light on the meaning of the term "necessary." The mandatory nature of the review is not undermined by the "useful" interpretation. The FCC must repeal or modify a rule that is no longer useful in promoting the public interest.

Recently in Cellco, 357 F.3d 88, the D.C. Circuit upheld the Agency's interpretation of "necessary" in a similar review provision of the 1996 Act.96 The FCC's "useful" interpretation of the statutory language is a reasonable one, and should be upheld.97

The 1996 Act's legislative history bolsters the FCC's interpretation. Section 202(h) falls within the ambit of a larger review provision, Section 11 in the 1996 Act. The conference report regarding Section 11, H.R. Conf. Rep. No. 104-458, at 185 (1996), interprets "no longer in the public interest" as "no longer meaningful." This phrase, of course, is essentially synonymous with the phrase "no longer useful" or "no longer appropriate," and much less exacting than "no longer indispensable." Thus, the Conference Report's definition of "no longer in the public interest" supports the FCC's interpretation of "necessary."

Despite § 202(h)'s admittedly deregulatory tenor, the statute does not foreclose the possibility of increased regulation under the biennial review if the Commission finds such action in the public interest. Nothing in the 1996 Act overcomes the broad rulemaking authority of the FCC. See 47 U.S.C. § 303(r). Any limitation must be clearly stated by Congress. See, e.g., Am. Hosp. Ass'n v. NLRB, 499 U.S. 606, 613, 111 S.Ct. 1539, 113 L.Ed.2d 675 (1991) (stating that "if Congress had intended to curtail in a particular area the broad rulemaking authority [it has] granted ... we would have suspected it to do so in language expressly describing an exception [to the authority]").98

Finally, though the Order must be consistent with the remand directives from Fox and Sinclair, and those reasoned decisions are instructive, they are not the law of the case in the current appeal. This case involves petitions for review of the FCC's comprehensive reexamination of a larger set of its broadcast ownership rules, in which a different set of parties participated, a different record was compiled, and a different result reached. See, e.g., Hamilton v. Leavy, 322 F.3d 776, 786 (3d Cir.2003) (Law of the case doctrine applies only to the "same litigation.").99

But Sinclair is instructive for another reason relating to the proper application of our standard of review. In Sinclair, the D.C. Circuit found the Commission failed to demonstrate its exclusion of non-broadcast media from its definition of "voices" in its local television ownership rule was not arbitrary and capricious. 284 F.3d at 165. The Commission's failure in the order at issue in Sinclair is notably different in both kind and degree from the purported "flaws" the majority believes require vacating the FCC's Order here.

In sum, the standard of review is governed foremost by the APA's requirement that the FCC's rules not be arbitrary and capricious. Beyond this, we bear in mind the requisite flexibility afforded the Commission's rulemaking under the 1934 Act, and the particular deference granted to the Commission's decisions on diversity and other difficult to define goals. Finally, § 202(h) overlays a deregulatory tenor on our review. While I would not term this standard a "deregulatory presumption," the FCC is required to demonstrate that its rules remain useful in the public interest.

III. Rules at Issue in this Appeal and Agency Rationale for Rules

The majority partially vacates and remands the Order for the Commission to re-evaluate certain purported flaws. With regard to the Diversity Index, the majority finds the weight assigned to the Internet, as well as the assumption of equal shares for outlets within the same media type lacks record support. The majority also finds the Commission inconsistently derived its Cross Media Limits from the Diversity Index results. In regard to the local ownership rules, the majority finds the television and radio rules' reference to a particular market structure — one with equal-sized firms — as the Commission's underpinning for adopting specific ownership limits, is not supported by the record. In my view, none of these purported flaws reach the level of arbitrary and capricious decision-making or violate § 202(h).100

The majority's conclusions cannot be evaluated without examining the underlying rules and the FCC's rationale for the changes. The evolution of the challenged rules puts the Agency's mammoth task in perspective. This includes discussion of the FCC's goals and objectives in its biennial review. In my view, the Agency rationale supports the rule changes and demonstrates that the Commission's decisions were neither arbitrary nor capricious.

In accordance with its 2002 biennial review obligations, the Commission initiated a comprehensive review of six media ownership rules. Order ¶ 1. As part of its 2002 Biennial Review, the Agency sought comment on four broadcast ownership rules: the Local Television Multiple Ownership Rule on remand from Sinclair; the Radio-Television Cross-Ownership Rule; the National Television Multiple Ownership Rule on remand from Fox;101 and the Dual Network Rule. 2002 Biennial Regulatory Review, 17 F.C.C.R. 18,503, 2002 WL 31108252 (2002). The FCC also incorporated pending proceedings on local radio ownership, Rules and Policies Concerning Multiple Ownership of Radio Broadcast Stations in Local Markets; Definition of Radio Markets, 16 F.C.C.R. 19,861 2001 WL 1402453 (2001), Definition of Radio Markets, 15 F.C.C.R. 25,077 (2000), and proceedings on newspaper/broadcast cross-ownership. Cross-Ownership of Broadcast Stations and Newspapers, 16 F.C.C.R. 17,283, 2001 WL 1097041 (2001).

The Order at issue here commenced with a notice of proposed rulemaking released in September 2002. 17 F.C.C.R. 18,503, 2002 WL 31108252 (2002) ("NPRM"). To help guide its analysis, the Commission established a Media Ownership Working Group ("MOWG"), which commissioned twelve studies ranging from consumer surveys to economic analyses of media markets. Interested parties filed thousands of pages of comments, consisting of legal, social, and economic analyses, empirical and anecdotal evidence, and industry and consumer data to support their positions.

The result was a 256-page order in which the Commission analyzed the record and determined, by a three-to-two vote, to modify its ownership rules to provide a "new, comprehensive framework for broadcast ownership regulation." Order ¶ 3. The FCC adopted the Order on June 2, 2003, and released it a month later, on July 2, 2003. A summary of the Order was published in the Federal Register on August 5, 2003. 68 Fed. Reg. 46,286 (Aug. 5, 2003).

In its Order, the Commission identified three longstanding policy goals that would continue to guide its ownership rules: diversity, competition, and localism. The Commission found five types of diversity relevant to ownership policy: viewpoint diversity, program diversity, outlet diversity, source diversity, and minority and female ownership diversity. Order ¶ 18.

Viewpoint diversity applies to the availability of media content reflecting a variety of perspectives. Id. ¶ 19. This goal endeavors to ensure a robust marketplace of ideas. The Commission adhered to its "longstanding determination that the policy of limiting common ownership of multiple media outlets is the most reliable means of promoting viewpoint diversity." Id. ¶ 26.

Programming diversity refers to a variety of programming formats and content. Id. ¶ 36. The Commission found that programming diversity was generally best achieved by reliance on competition between delivery systems. Id. ¶ 37. According to the Commission, outlet diversity means that in a given market there are multiple independently owned firms. Id. ¶ 38. The Commission found that regulating the ownership of outlets to achieve outlet diversity was preferable to attempting to engineer outcomes directly, because ownership regulation reduces the need for the Commission to make subjective judgments about program content. Id. ¶ 39.

Source diversity refers to the availability of media content from a variety of content producers. Id. ¶ 42. In light of dramatic changes in television markets, including the increase in the number of channels available to most households, the Commission determined that source diversity need not be an objective of its broadcast ownership policies. Id. ¶ 43. Minority and female diversity refers to policies which encourage minority and female ownership of media sources, and the Commission reaffirmed this policy goal in the Order. Id. ¶ 46.

As set forth in the preamble to the 1996 Act, Congress believed greater competition and reduced regulation would inure to the public benefit. 1996 Act, preamble, 110 Stat. 56. The Order reaffirmed the Commission's "longstanding commitment to promoting competition by ensuring pro-competitive market structures," Order ¶ 57, recognizing "[c]onsumers receive more choice, lower prices, and more innovative services in competitive markets than they do in markets where one or more firms exercises market power." Id. In satisfying its competition goal, the Commission also noted competitive markets help contribute to the related goal of viewpoint diversity. Id. ¶ 58. The Commission pointed out that competition also furthers the goal of product innovation. Id. ¶ 69.

Federal regulation of broadcasting historically has placed significant emphasis on localism. The policy goal of localism addresses whether broadcast stations are responsive to the needs and interests of their local communities. Id. ¶ 74. Localism is contained in congressional directives to the Commission and has been judicially reaffirmed as a valid regulatory objective. Id. ¶ 73; see NBC v. United States, 319 U.S. 190, 203, 63 S.Ct. 997, 87 L.Ed. 1344 (1943) ("Local program service is a vital part of community life. A station should be ready, able, and willing to serve the needs of the local community...."). The Commission sought to promote localism through market structures that take advantage of media companies' incentives to serve local communities. To measure localism, the Commission focused on the selection of programming responsive to local needs and interests, and local news quantity and quality. Order ¶ 78-79.

The Commission considered the regulatory framework that would best achieve its goals. It recognized that bright line rules and case-by-case analysis offer different advantages. The Commission concluded that bright line rules provide greater "certainty to outcomes, conserve resources, reduce administrative delays, lower transaction costs, increase transparency of [the] process, and ensure consistency in decisions." Id. ¶ 82. The Commission determined that a case-by-case process accounts for the particular circumstances of each transaction but is fraught with regulatory problems, id. ¶ 84, such as high administrative costs and a lack of planning and investment predictability for media owners.102 Based on the efficiencies and predictability of bright line rules, this regulatory model supports most of the Agency's rules. Id. ¶ 85. The Agency reiterated the bright line rules were adopted based on a comprehensive review of the media marketplace and its assessment of what rules were necessary to promote its goals.

2. New Media Landscape and the Growth of the Internet

In its Order, the Commission took substantial notice of the expanding media market and significant technological advances that influence its rules. The Commission commented: "This Order confronts that challenge by determining the appropriate regulatory framework ... in a world characterized not by information scarcity, but by media abundance." Id. ¶ 89.

Significant technological advances in the media marketplace over the last twenty-five years required new regulatory responses from the Commission. Among these advances in the distribution of information were the Internet, satellite distribution systems, cable television and video-cassette recorders. Today there are over 308 satellite delivered non-broadcast networks available for carriage over cable, direct broadcast satellite systems ("DBS"), and other multi-channel viewed programming distribution ("MVPD") systems. Within more traditional sources, the number of broadcast television and radio outlets have also grown significantly in the last 40 years. Id. ¶ 121.

This explosion of new media sources is best illustrated by the Internet. The Commission commented:

The Internet, as an entirely new medium, composed of an amalgam of all the technologies that preceded it, completely transformed the way in which we communicate in unimaginable ways. These advances not only enabled the provision of vast amounts of content they also put more control in the hands of the public, allowing them to control what, when and how they receive information.

Order ¶ 111. The Internet provides a forum for unlimited numbers of independently administered voices. Additionally, content on the Internet is multimedia; it can be read, heard, and viewed simultaneously. Id. ¶ 118. The Commission also pointed out that experiencing web content is a highly individualized activity where people can access the Internet 24 hours a day anywhere they have access to a web browser. Id.

The Internet has spawned a new way of looking at media. The first graphical interface for the Internet was introduced in 1989. In 1992 there were only 50 websites in the world. By 1994, there were as many as 3,000 web sites in use. By year-end 2000, there were more than 30 million web sites. Id. ¶ 117. Approximately 42.5 million American households subscribed to an Internet access provider in 2000. Id. ¶ 120. According to submitted comments, 62.6 million households in 2001 were online, and that total is expected to rise to 86.3 million, or 76% of total U.S. households by 2006. Comments of Tribune Co., at 10.

According to the Commission, virtually every major media company has a corresponding web site, and any person with access to a web-hosting file server can create a web site, which the public may access. Internet users can view sites of their own choosing or can use a search engine, such as Google News, which presents information gathered from approximately 4,500 news sources worldwide.103 Id. ¶ 119. In conclusion, the Commission determined that:

there are far more types of media available today, far more outlets per-type of media today, and far more news and public interest programming options available to the public today than ever before. Although many of these new outlets are subscription-based... the competitive pressure placed upon free, over-the-air media has led to better quality and in some cases, an increase in the quantity of some types of content. In the next five to ten years, we expect more free, over-the-air content to become available as new technologies (i.e., digital transmission) are applied to these traditional media (i.e., broadcast television).

Id. ¶ 128. This dynamic, rapidly changing media landscape provided the backdrop for the Commission's ownership rules.

a. History of Cross-Ownership Regulation

For many years the Commission maintained rules governing cross-ownership of media in local markets. From 1975 through the present Order, the Newspaper/Broadcast Cross-Ownership Rule prohibited a single company from owning a full-service radio or television broadcast station and a daily newspaper serving the same community. See 47 C.F.R. § 73.3555(d) (2002). The rule was intended to promote media competition and diversity. 50 F.C.C.2d at 1074. The Commission examined the newspaper/broadcast cross-ownership prohibition in 1998 and concluded that the rule continued to serve the public interest because it furthered diversity. 15 F.C.C.R. at 11,105-08 ¶¶ 89-93. In that review, the Commission noted the rule should be continually reviewed to ensure it remained in the public interest. Specifically, the Commission noted "[t]here may be instances, for example, in which, given the size of the market and the size and type of the newspaper and broadcast outlet involved, sufficient diversity and competition would remain if a newspaper/broadcast combination were allowed." Id. at 11,109 ¶ 95.

In 1970 the Commission issued a "one-to-a-market" Radio/Television Cross-Ownership Rule, which proscribed common ownership of more than one full-time broadcast station (radio or television) in the same market. Amendment of Sections 73.35, 73.240, and 73.636 of the Commission Rules Relating to Multiple Ownership of Standard, FM and Television Broadcast Stations, 22 F.C.C.2d 306, 307-08 ¶¶ 5-8, 1970 WL 18044 (1970). Like the newspaper/broadcast rule, the purpose of the rule was twofold: (1) to promote competition in broadcasting; and (2) to promote diversification of programming sources and viewpoints. Id. at 307 ¶ 3. In 1989, the Commission relaxed its "one-to-a-market" rule by implementing a lenient waiver policy for applications involving radio and television combinations in the top 25 markets. Second Ownership Order, 4 F.C.C.R. 1741 ¶ 1, 1989 WL 510875 (1989). The 1996 Act required the Commission to extend the radio-television cross-ownership presumptive waiver policy to the top 50 television markets "consistent with the public interest, convenience, and necessity." § 202(d). Although the Commission retained the rule after the 1996 Act, it stated that it would continue to monitor the impact of the rule on the radio and television industries and would consider further modification of the rule in future biennial reviews. 14 F.C.C.R. at 12,949 ¶ 106.

In 1999, the Radio/Television Cross-Ownership Rule was modified to its current form. The Commission found the growth of media outlets, the efficiencies of joint ownership, and the public service benefits obtained from joint operations all supported its decision to allow common ownership of radio and television stations. Id. at 12,948 ¶ 102. Currently, the Radio/Television Cross-Ownership Rule allows a party to own up to two television stations and up to six radio stations in a market where at least 20 independently owned media voices would remain post-merger. Where parties may own a combination of two television stations and six radio stations, the rule allows a party alternatively to own one television station and seven radio stations. A party may own up to two television stations (as permitted under the local television rule) and up to four radio stations (as permitted under the local radio rule) in markets where, post-merger, at least ten independently owned media voices would remain. A combination of one television station and one radio station is allowed regardless of the number of voices in the market. 47 C.F.R. § 73.3555(c) (2002).

In light of the 1999 relaxation of the rule, the Commission decided to proceed cautiously and monitor the impact of the new rules on diversity and competition. See 2000 Biennial Review, 16 F.C.C.R. 1207, 1218 ¶ 32, 2001 WL 38123 (2001).

In the Order at issue, the Commission repealed both cross-ownership rules and replaced them with more flexible Cross Media Limits, which established more finely-tuned cross-ownership limits in small and medium-sized markets, and placed no restrictions on cross-ownership in large markets beyond the intra-service local caps.

To provide historical perspective, and to place the repeal of the Newspaper/Broadcast Cross-Ownership Rule in proper context, it is important to note that the ban was adopted nearly three decades ago, in a media environment that has been dramatically and irreversibly transformed in the intervening years. In the nearly thirty years since the newspaper/broadcast cross-ownership prohibition was enacted, the media landscape has changed dramatically, as the Commission extensively documents in its Order. Among these changes are the growth of cable, VCRs, satellite video and audio services, and the Internet. Order ¶¶ 106-19. As noted, media outlets per market within broadcast television and radio have grown during this period as well, id. ¶ 121, providing viewers with a multitude of choices not available twenty or even ten years ago. With other technological advancements likely in the near future, the Commission recognized that its rigid rule had become outdated. The FCC concluded that "the question confronting media companies today is not whether they will be able to dominate the distribution of news and information in any market, but whether they will be able to be heard at all among the cacophony of voices vying for the attention of Americans." Id. ¶ 367.

b. Repeal of Newspaper Broadcast Cross-Ownership Prohibition

The FCC looked to its policy objectives of promoting competition, localism and diversity, and concluded the Newspaper/Broadcast Cross-Ownership Rule (NBCO): 1) was neutral with respect to competition; 2) actually undermined localism by preventing newspaper-broadcast mergers that would provide more and higher-quality local news to their markets; and 3) was not necessary to preserve viewpoint diversity in most markets. Id. ¶ 330.

In terms of competition, the Commission concluded "the local newspaper market is distinct from the local broadcasting market," and therefore "a newspaper/broadcast combination ... cannot adversely affect competition in any product market." Id. ¶ 332. Synergies and cost-reductions of cross-ownership could also translate into increased competition. Id. ¶ 337. As discussed, the FCC measured the promotion of localism through two metrics: "the selection of programming responsive to local needs and interests, and local news quantity and quality." Id. ¶ 78. The FCC determined cross-ownership promotes localism because it allows newspaper stations and broadcast stations to pool their resources, allowing the whole of their coverage to exceed the sum of their parts. "Specifically, MOWG Study No. 7 found that while non-network owned but network-affiliated stations provide, on average, 14.9 hours per week of local news and public affairs programming, newspaper-owned affiliated stations provide almost 50% more such programming, averaging 21.9 hours per week." Id. ¶ 344. In addition, the study found the average number of hours of local news and public affairs programming provided by same-market, commercially owned television-newspaper combinations was 25.6 hours/week, compared to 16.3 hours/week for the sample of television stations owned by out-of-market newspapers. Id. The FCC also found corresponding advantages in quality of local coverage provided by newspaper-owned stations, as shown by ratings (measuring consumer approval) and industry awards (measuring critical approval). Id. ¶ 345. Substantial evidence supports the Commission's decision that cross-ownership translates into pro-competitive public interest benefits.

The FCC concluded the negative effects of the NBCO Rule on localism hurt diversity as well: if local stations do not have the resources to provide a high quantity and quality of news, the number of strong voices in their market will be diminished. The FCC cited record evidence both supporting and refuting the notion that common ownership leads to common viewpoint among media properties. From this conflicting evidence, the FCC inferred that "although there is evidence to suggest that ownership influences viewpoint, the degree to which it does so cannot be established with any certitude." Id. ¶ 364.

The Commission also examined the growth in media outlets, including the addition of cable and the Internet as viable news sources. It concluded the large number of new voices rendered the NBCO unnecessary for purposes of protecting diversity, because any loss in independent voices resulting from cross-ownership would, in all but the most concentrated markets, be counterbalanced by the plethora of alternative voices. The Order concluded "competing media outlets abound in markets of all sizes...." Id. ¶ 365. Specifically, the Commission pointed to the expanding role of cable in providing varied content, including local news and public affairs programming. Id. The Order continued, noting:

The Internet, too, is becoming a commonly-used source for news, commentary, community affairs, and national/international information. Seventy-two percent of Americans are now online and spend an average of nine hours weekly on the Internet. MOWG Study No. 3 suggests that consumers generally view Internet news sources as a substitute for daily newspapers and broadcast news ... the Internet does play an important role in the available media mix.

Id. ¶ 365. It is also worth noting that none of the 30 million web sites existing in the year 2000 existed when the NBCO Rule was first adopted.

The record supports: (1) the growth in television and radio outlets; and (2) that cable and the Internet were properly cited in the growth of media outlets. Additionally, the benefits to localism from cross-ownership provide a rational basis for repealing the blanket prohibition. Moreover, the Agency's decision on the mixed record is particularly reasonable in light of the deregulatory tenor of § 202(h).

c. Cross-Media Limits

The record and the Commission's experience led it to conclude that more narrowly focused limits were necessary in certain specific situations to guard against "an elevated risk of harm to the range and breadth of viewpoints that may be available to the public," resulting from cross-ownership of media properties. Order ¶ 442. Having examined the record evidence, the Commission reasonably concluded that new limits were necessary "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." Id. ¶ 432.

In that vein, the FCC adopted a methodology termed the Diversity Index to provide its new media cross-ownership framework with an empirical footing. The Diversity Index takes account of certain media outlets available to consumers, the relative importance of those media as sources of local news, and ownership concentration across these media. Id. ¶ 391.

Based on detailed analysis of the record, informed by the Diversity Index, the FCC adopted the following Cross Media Limits:

----------------------------------------------------------------------------------------------
Number of Television Stations in Market     Cross-Ownership Limits
----------------------------------------------------------------------------------------------
1-3 television stations                     No cross-ownership of newspapers, radio stations,
                                              or television stations.
----------------------------------------------------------------------------------------------
4-8 television stations                     Single entity may own a daily newspaper and
                                              either: (1) a television station and up to half
                                              the maximum number of radio stations allowed
                                              by the local radio rules; or (2) no television
                                              station and as many radio stations as allowed
                                              by the cross-ownership rule. Additionally, a
                                              television owner who owns a duopoly (two
                                              television stations) may not own a newspaper;
                                              likewise, a newspaper owner may only own one
                                              television station.
----------------------------------------------------------------------------------------------
Over 9 television stations                  No cross-ownership limits.
----------------------------------------------------------------------------------------------

As noted, in crafting its Cross Media Limits, the FCC drew upon the Diversity Index, which it modeled after the Herfindahl-Hirschmann Index ("HHI"). The HHI was originally constructed to measure market concentration in the context of economic competition; if a given market's HHI was too high, it signaled the market was at risk of being monopolized. By analogy, the FCC used its Diversity Index to determine which media markets presented the highest risk of ownership concentration. Underlying this objective is the premise that media ownership influences viewpoint, so that highly concentrated media ownership could negatively impact the diversity of viewpoints disseminated within a market.

The FCC chose the HHI as its market concentration metric over a simple firm count because of the former's greater sensitivity for concentration. Id. ¶ 396. The HHI sums the squares of firms' market shares in a given market. The higher a market's HHI, the more concentrated it is. For example, an economic market with only one firm (a monopoly) would have an HHI of 10000 (100 squared), while a market shared equally among ten firms ("Market A") would have an HHI of 1000 (10 times 10 squared). A firm count system would treat Market A as evenly concentrated with a ten-market firm having a market share breakdown of 30-30-5-5-5-5-5-5-5-5 ("Market B") because each market contains ten firms. In contrast, the HHI supports the intuition that Market B is actually vastly more concentrated, with an HHI of 2000. Furthermore, unlike the firm count system, the HHI recognizes that a merger between two large firms creates a more concentrated market than a merger between two smaller firms. If the two 30% firms merged in Market B, its HHI would rise to 3800, while a merger of two 5% firms would increase the HHI to 2050. The firm count system would undiscerningly treat both mergers the same, however, by noting that both markets would now have 9 firms instead of 10.

The FCC adapted the HHI to viewpoint diversity by ascertaining different media's shares of consumers' local news preferences. MOWG Study No. 8, a nationwide survey, asked respondents to identify sources used in the past seven days for local news and current affairs, offering the following answer choices: "television, newspaper, radio, Internet, magazines, friends/family, other, none, don't know, and refuse." Id. ¶ 402. The Study then asked respondents to identify the sources they had used in the past 7 days for national news. According to Study No. 8, the five leading sources of local news were television, newspapers, radio, the Internet and magazines. In apportioning media shares, however, the FCC decided to discount magazines entirely because they only accounted for 6.8% of respondents' answers, and only .6% of respondents identified magazines as their "primary" source of local or national news. Furthermore, the Pew Research Center conducted a study that showed only 4.2% of their responses cited news magazines as a source of local or national news. The FCC concluded such small figures for local or national news presaged even lower figures for local news only, because, with isolated exceptions, most magazines are national in focus. Accordingly, the FCC reasoned that magazines would have a negligible weight in the Diversity Index and should properly be excluded, at least until the next biennial review. After excluding magazines, and adjusting the remaining market shares accordingly, the user shares of the remaining four media were 33.8% for television, 28.8% for newspapers, 24.9% for radio and 12.5% for the Internet.

The FCC then explored whether any of the media could be broken down into submedia (i.e., television into broadcast and cable, newspapers into daily and weekly). The FCC decided to remove cable/satellite stations/providers from the voice pool because it was unclear whether the Study No. 8 respondents who said they received local news from cable television were in reality watching broadcast channels on their cable platforms. Id. ¶ 414. Additionally, local cable stations are only available to approximately one third of all cable subscribers. Id. Thus, the FCC counted only broadcast stations in its television percentage. The Order also notes the exclusion of cable from the Diversity Index will be reviewed in subsequent reviews. Such reviews will include follow-up MOWG questions concerning non-broadcast media. Id.

The FCC broke down newspapers into daily and weekly subcategories, with the survey data indicating that 70.3% of newspaper respondents read dailies and 29.7% read weeklies. The FCC used these weights to divide the total newspaper share (28.8%) among daily (20.2%) and weekly (8.6%) newspapers. Finally, the FCC separated the Internet into cable modem users and "other users." The Commission's studies demonstrated the appropriate break-down was 18.3% cable users and 81.7% dial-up or DSL users. Therefore, of the 12.5% of the media market allocated to the Internet, that percentage was broken down into 2.3% for cable users, and 10.2% for dial-up and DSL users.

Having apportioned shares among the various media based on actual use figures, the FCC turned to weighing different outlets within the same media. Here, the Commission moved away from actual use and attributed each outlet an equal share of its media type.104 In other words, in a market with ten television stations, the FCC did not determine what percentage of viewers actually received their local news from each of the ten stations; it simply apportioned each 10% of the television share, meaning a 3.4% share of the overall local news "market." The FCC added commonly owned shares together to divide shares of the market by owner. For example, a market with ten television stations, two of which are commonly owned, would have share figures of 6.8 and eight 3.4's rather than ten shares of 3.4. Similarly, the FCC added cross-owned shares from different media. For example, a market with ten television stations and ten radio stations with one cross-ownership combination would have share figures of 5.9, nine 3.4's and nine 2.5's, rather than ten 3.4's and ten 2.5's. Once the FCC determined the breakdown of shares amongst various independent owners, it squared those shares and added them together to get the Diversity Index for the market in question. See Order App. C (calculating Diversity Indices for various sample markets).

The FCC then sought to determine the Diversity Indices for various sample markets as a function of their number of Television stations. Specifically, the FCC calculated the Diversity Index for every market in the U.S. with 1-5, 15 or 20 television stations. The FCC then calculated the Diversity Indices for ten randomly selected markets with 6 television stations, 7 television stations, 8 television stations, 9 television stations and 10 television stations; the FCC chose ten-market samples for each television station grouping in the 6-10 station range because so many media markets have 6-10 television stations. Using this data, the FCC determined the average Diversity Index for a market as a function of the number of television stations within that market. Order App. D. The FCC also calculated the average increase in HHI in each tested market that would result from various cross-ownership merger scenarios. Id.

The FCC noted that in antitrust analysis, a market is not considered moderately concentrated until its HHI exceeds 1000. The FCC's calculations revealed that markets with three or fewer television stations had average Diversity Indices of over 1000, and that any cross-ownership mergers in such small markets would lead to significant Diversity Index increases; consequently, the FCC decided to prohibit any cross-ownership in markets with three or fewer television stations. In markets with four to eight television stations, which had base Diversity Indices below but near 1000,105 the FCC determined that dangerously large Diversity Index increases would result from daily newspaper-television duopoly mergers, as well as mergers between a daily newspaper, a television station and more than half the radio stations allowed by the local radio rule. Finally, the FCC concluded that large markets, with low indices and modest Diversity Index effects from potential mergers, did not need any cross-ownership regulation at all.

The Diversity Index was not promulgated as an ownership rule. Instead, it played a supporting role in the development of the Cross Media Limits. In adopting the Diversity Index, the FCC noted: "While the Diversity Index is not perfect, nor absolutely precise, it is certainly a useful tool to inform our judgment and decision-making. It provides us with guidance, informing us about the marketplace and giving us a sense of relative weights of different media." Order ¶ 391. As the Commission stated, the "cross-media limits are based on a set of assumptions drawn directly from the record evidence in this proceeding ... the CML [Cross Media Limits]... ultimately rest[ ] on our 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." Id. ¶ 435. In the end, the Diversity Index provided the Commission with a rough picture of the amount of media diversity concentration in markets, and how changes as a result of cross-media combinations could affect the level of diversity in markets.

The Commission's use of the Diversity Index constitutes a rational effort to lend a factual foundation to a thorny regulatory question. The development of the Diversity Index represents a bona fide approach that employs empirical evidence to measure and inform policy choices, which do not easily lend themselves to calculation. I would affirm the Commission's effort to lend a modicum of empirical support to a regulatory issue that had not been fully revisited in nearly thirty years. Perhaps other measures might have been employed, but the development and use of the Diversity Index was rational and reasonable, and clearly neither arbitrary nor capricious. The FCC sufficiently demonstrated the new rule was "necessary in the public interest."

Adopted in 1964, the Commission's "duopoly" rule prohibited ownership or control of television stations with overlapping Grade B signal contours. 47 C.F.R. § 73.3555(b) (1998). Section 202(c)(2) of the 1996 Act instructed the Commission to conduct a rulemaking "to retain, modify, or eliminate" its local television ownership limitations. 110 Stat. 111. Under the 1996 Act, the Commission adopted rules permitting common ownership of two television stations with overlapping contours, so long as (1) one of the two stations is not ranked among the top four, and (2) at least eight independently owned full-power commercial and noncommercial stations remain in the market after the merger. 47 C.F.R. § 73.3555(b) (2002).

This rule was remanded to the Commission by the Court of Appeals for the D. C. Circuit in Sinclair, 284 F.3d 148. In that case, the court affirmed that "the local ownership rule furthers diversity at the local level and is necessary in the `public interest' under § 202(h)," id. at 160, but held that the FCC had not justified different definitions of "voices" in the local television and radio-television cross-ownership rules. Id. at 165. The court held that the Commission had "failed to demonstrate" why it excluded non-broadcast media from the ownership rules' eight-voices test. Id.

On remand, the agency modified the Local Television Ownership Rule. The Commission determined the old local television rule was not necessary to further public interest in competition and viewpoint diversity. Order ¶ 133. But the Agency determined that some ownership limits remained necessary to promote competition. Id. The modified rule permits common ownership of up to two commercial stations in markets that have 17 or fewer full-power commercial and non-commercial stations, and up to three commercial stations in markets that have 18 or more stations. Id. ¶ 186. The Agency also prohibited combinations which would result in a single entity acquiring more than one station ranked in the top four in the market based on audience share. Id. Therefore, the rule prohibits ownership of more than one television station in any market with fewer than five television stations.106 Id. The new local television limits are reproduced in the following table:

---------------------------------------------------------------------------------------------
Number of stations in a Market              Number of combinations allowed
---------------------------------------------------------------------------------------------
                                            No combinations (per operation of the top-four
four or fewer                                restriction).
---------------------------------------------------------------------------------------------
                                            Two stations may be combined, as long as both of
five to seventeen                            the combining stations are not in the top four.
---------------------------------------------------------------------------------------------
                                            Three stations may be combined, as long as two
                                             of the combining stations are not in the top
eighteen or more                             four.
---------------------------------------------------------------------------------------------

The Commission found the common ownership of television stations in local markets can "result in consumer welfare enhancing efficiencies," by eliminating redundant expenses and increasing opportunities for "cross-promotion and counter-programming." Id. ¶ 147. These efficiencies would enable local broadcasters to better compete with cable and digital providers and would also "spur the transition to digital television," which the FCC identified as a goal in the public interest. Id. ¶ 148.

Such common ownership was not found to have a detrimental effect on audience ratings. Id. ¶ 150. Importantly, the Commission concluded that "owners/operators of same-market combinations have the ability and incentive to offer more programming responsive to the needs and interests of their communities and that in many cases, that is what they do." Id. ¶ 164. In fact, "[a]udience share data ... reveals that common ownership of two broadcast television stations has generally improved audience ratings." Id. ¶ 150.

According to the Commission, the new rule does not compromise its diversity goals because the modified caps ensure at least six firms in markets with twelve or more stations. Id. ¶ 207. The Commission explained that it would permit additional concentration in markets with less than twelve television stations because the economies of local stations justified an increase in market concentration as markets get smaller. Id. ¶ 201. Indeed, the Commission noted that its local television rule — designed to protect competition — necessarily also protects diversity. Id. ¶¶ 178, 180. Additionally, the Commission pointed out that television broadcast stations are not the only media contributing to viewpoint diversity in local markets. Numerous other sources for news are available, even within the delivered video market. See, e.g., 17 F.C.C.R. 26,901, 26,909-52 ¶¶ 15-111.

In regard to the top-four restriction, the Commission explained, "in local markets, there is a general separation between the audience shares of the top four-ranked stations and audience shares of the other stations in the market." Order ¶ 195. Therefore, these mergers would often lead to significant increases in market consolidation. In addition, the Commission found that combinations of the top-four stations were less likely to yield public interest benefits in terms of local news or improved technology. Id. ¶ 199.

The FCC further justified the specific numerical caps in its local television ownership rule by employing the Department of Justice and Federal Trade Commission's "standard approach" to evaluating competitive harms of an increase in horizontal market concentration: as noted, the Herfindahl-Hirschmann Index. In developing the new local ownership tiers, the FCC looked to the HHI used in merger analysis, selecting an HHI of 1800 as the benchmark for competitively suspect combinations. The DOJ/FTC merger guidelines recognize the HHI level of 1800 as the maximum level for "moderate" concentration. The FCC selected the upper bound for moderate concentration, instead of the lower bound of 1000, in recognition of the competitive pressures exerted by the cable networks. Id. ¶ 192. The ownership rules ensure that in markets with 12 or more stations, there will be at least six competing firms. Id. ¶ 207. Six equal-sized competitors in a market corresponds to an HHI of 1800. While the rule allows for fewer firms in smaller markets, the Commission explained that such increased levels of concentration were justified by the economics of smaller market stations. Id. ¶ 201.

The Commission justified its assumption of equal-share television stations. Television market shares vary widely. Id. ¶ 193. Due to shifts in programming popularity and product innovation, market shares in broadcasting are more fluid than in other industries. The Commission found that television shares can shift substantially over the life of a firm's investment in a particular station. Thus, in constructing its rules, the Commission focused on "a firm's `capacity' to deliver programming" — i.e., "the number of licenses that a firm controls in a market." Id. As the Commission explained, license caps — as opposed to limits on market share — are well-suited to serve the Commission's diversity goals. Id. ¶¶ 129, 178. The Commission's amendment of the local television rule is well-supported by the record and would provide a benchmark for future revisions.

For many years the Commission limited local radio ownership to no more than one AM and FM station with overlapping signal contours. See, e.g., Amendment of Section 73.3555 of the Commission's Rules, 4 F.C.C.R. 1723, 1723 ¶¶ 5-6, 1989 WL 511018 (1989). In 1992 the Commission relaxed its rules to permit common ownership of radio stations in line with the size of various local market tiers. Revision of Radio Rules and Policies, 7 F.C.C.R. 2755, 2776 ¶ 40, 1992 WL 689675 (1992), on reconsideration, 7 F.C.C.R. 6387, 1992 WL 690638 (1992). In its 1992 rulemaking, the Commission considered using the Arbitron rating service to define radio markets, and the number of stations in the market. Id. at 2778 ¶ 45. In the end, however, the Agency decided to define radio markets based on signal contour overlap.107

Section 202(b)(1) of the 1996 Act required the FCC to revise its limits on the number of stations an entity could own in a particular market by adjusting both the market tiers and the numerical limits. See Implementations of Sections 202(a) and 202(b)(1) of the Telecommunications Act of 1996 (Broadcast Radio Ownership), 11 F.C.C.R. 12,368, 1996 WL 740755 (1996). The 1996 Act expanded multiple radio station ownership rights by setting the following common ownership limits: eight stations in markets with forty-five or more stations, with five or fewer in the same service (AM or FM), § 202(b)(1)(A); seven stations in markets with thirty to forty-four stations, with four or fewer in the same service, § 202(b)(1)(B); six stations in markets with fifteen to twenty-nine stations, with four or fewer in the same service, § 202(b)(1)(C); and five stations in markets with fifteen to twenty-nine stations, with three or fewer in the same service, provided that no single entity could own more than fifty percent of the stations in such market, § 202(b)(1)(D). Congress authorized the FCC to further relax these limitations in order to permit greater common ownership if such ownership "will result in an increase in the number of radio broadcast stations in operation." § 202(b)(2). The limits are summarized below:

---------------------------------------------------------------------------------
                                                      Number of stations a party
                                                       may own which are of the
Number of commercial    Number of stations a party     same service (AM or
 stations in market:     may own:                      FM):
--------------------------------------------------------------------------------
45 or more              8                             5
--------------------------------------------------------------------------------
between 30 and 44       7                             4
--------------------------------------------------------------------------------
between 15 and 29       6                             4
--------------------------------------------------------------------------------
                        5 [with the exception that a
                          party cannot own more
                          than 50% of the stations
14 or fewer               in such market]             3
--------------------------------------------------------------------------------

The new limits imposed by Congress will be regularly reviewed as the 1996 Act required the FCC to review biennially (now quadrennially) its media ownership rules and modify or repeal those it found not "necessary in the public interest." 1996 Act, § 202(h).

At the time Congress instituted these changes, the FCC defined a radio market as the geographic area covered by the overlapping contours of the stations proposed for common ownership. See 47 C.F.R. § 73.3555(a)(1) (1995). Under that definition, local markets were determined with reference to stations' "principal community contour" — essentially, the area reached by a station's signal at a certain level of strength. See id. § 73.3555(a)(3). For each proposed combination of stations, the local market was defined by "the number of... stations whose principal community contours overlap, in whole or in part, with the principal community contours of the stations" that were proposed to be commonly owned. Id. Nothing in the 1996 Act addressed the methodology for defining radio markets or identifying the number of radio stations in a market for purposes of the ownership rules.

The FCC initiated a rulemaking to change the definition of radio markets. 1998 Biennial Reg'y Review — Review of the Commission's Broadcast Ownership Rules and Other Rules Adopted Pursuant to Section 202 of the Telecommunications Act of 1996, 15 F.C.C.R. 11,058, 11,091-94, 2000 WL 791562 (2000). In this review, the Commission expressed concern that its method of defining markets and identifying stations in the market led to "unrealistic results," and were both "illogical" and "contrary to Congress' intent." Id. 11093-94 ¶ 65, 67. Therefore, the Commission commenced a proceeding seeking comment on whether and how it should modify the way in which it "determine[s] the dimensions of radio markets and count[s] the number of stations in them." Definition of Radio Markets, 15 F.C.C.R. 25,077 ¶ 1, 2000 WL 1827671 (2001). The FCC eventually consolidated that rulemaking with another proceeding regarding the local radio ownership rules, Rules and Policies Concerning Multiple Ownership of Radio Broadcast Stations in Local Markets; Definition of Radio Market, 16 F.C.C.R. 19,861 ¶ 1, and then rolled the entire proceeding into the 2002 biennial review. See Omnibus NPRM, 17 F.C.C.R. at 18,506.

Since the passage of the 1996 Act, record evidence demonstrates dramatic transformation in the radio industry. In each of the last six years, about 20% of U.S. radio stations have changed hands in a frenzy of deal-making. From 1996 to 2002, the number of radio stations owners declined by 34% even though the number of commercial stations increased by 5.4%. Within individual markets, small numbers of large owners often dominate. The audience shares of the top four firms in smaller markets are even higher than in larger markets. The audience share of the top four firms in markets 51-100 and 101-289 is 92.5% and 93.9% respectively. This has led to high market concentrations, as measured by the HHI. When conducting its 2002 biennial review, the FCC was faced with significant consolidation in radio markets.

In the Order at issue, the Commission determined that it should retain the numerical limits in the local radio rule, but that it should revise the contour overlap method. The FCC proposed to replace the contour-overlap methodology with a geographic methodology, based on the Arbitron Metro Survey Areas. Arbitron, the principal radio rating service in the country, has defined radio markets (called Arbitron Metros) for most of the more populated urban areas of the country. Arbitron Metros are based on Metropolitan Areas established by the Office of Management and Budget.

The geography-based, Arbitron Market definition relies on the market definitions of Arbitron radio rating service, a private entity that measures local radio station audiences for its customer stations. See Order ¶¶ 275-76.108 Additionally, the Agency decided to include commercial and non-commercial stations that are licensed to a community in the relevant Arbitron Metro,109 as well as stations located outside the Metro that attract a minimum level of listenership in the Metro, see id. ¶¶ 279-81, 295, in calculating the relevant local radio market.

The Commission identified several flaws in the manner in which it presently defines radio markets, the contour-overlap market definition. Id. ¶ 239. One major flaw is known as the "Pine Buff" problem or "numerator-denominator" inconsistency. The FCC described the inconsistency this way:

[A] party is deemed to own only those stations that are represented in the numerator, i.e., stations that have mutually overlapping principal community contours. In calculating the denominator, however, any radio station whose principle community contour overlaps the principal community contour of at least one of the radio stations in the numerator is counted as being in the market, regardless of who owns the station. As a result, the denominator may include radio stations that are owned by the same party that owns the radio stations represented in the numerator. Because those stations are counted in the denominator, they are by definition "in" the market, but they would not count against the party's ownership limit in that market unless their principal community contours overlap the principal community contours of all the radio stations in the numerator.

Id. ¶ 253.

Numerator-denominator inconsistency causes two potential problems. First, because commonly owned stations in the denominator are not counted in the numerator, a party could use its own radio stations to increase the size of a market and "bump" itself into a higher ownership tier. The more common occurrence, according to the FCC, is that the inconsistency allows a party to own radio stations in the relevant market without having those stations count against the party's ownership limits. The contour-overlap methodology thus allows entities to circumvent the ownership limits, which are intended to promote competition and "to protect against excessive concentration levels in local radio markets." Id. ¶ 254.

The FCC also determined this irrationality in the contour-overlap methodology could not be fixed. If commonly owned stations are excluded from the denominator, then market size would be contingent upon who owned what station, "a distinction that would be both unprincipled and unprecedented in the history of competition analysis." Id. ¶ 255. If all commonly owned stations in the denominator are included in the numerator, "a party's ownership level in a particular market may be overly inflated by outlying stations far from the area of concentration." Id. Further, the FCC's interim adoption of a modified contour-overlap methodology for non-Arbitron Metro areas does not repudiate these findings. The FCC concluded temporary use of the old method was unavoidable during the pendency of the rulemaking process. Id. ¶ 283. And because this method was well-understood, it at least had the benefit of permitting orderly processing of radio station applications. The FCC noted, however, that the adjustments made could only "minimize the more problematic aspects of that system." Id. ¶ 285.

The contour-overlap approach is contrary to traditional antitrust principles, which employ "relevant geographic markets" for purposes of competition analysis. Though radio stations are signal-based products, the FCC effectively addressed this point by noting: "[r]adio stations serve people, not land," and "people in the United States tend to be clustered around specific population centers." Id. ¶ 273. This is demonstrated by the fact that the 287 Arbitron Metros cover only 30% of the counties in the United States, but cover 78% of the population above the age of twelve. Id. ¶ 282. It is also noteworthy that the Justice Department often treats Arbitron Metros as the relevant market for antitrust purposes. See United States v. CBS Corp. and Am. Radio Sys. Corp., Proposed Final Judgment and Competitive Impact Statement, 63 Fed. Reg. 18036, 18044-45 (Apr. 13, 1998). "The fact that radio signals are not congruent with geographic boundaries does not undermine the logic of relying on geographic areas to define radio markets." Order ¶ 273.

In addition, the "subjective market" problem has other drawbacks. Because stations with larger signal contours are more likely to create larger radio markets, the contour-overlap methodology disproportionately encourages the consolidation of powerful stations relative to smaller stations. Also, radio stations may overlap only a very small portion of a given "market." So while they are part of that market, these stations may be unable to serve effectively the listeners or advertisers affected by the proposed consolidation. Finally, the fact that every "market" is unique frustrates the FCC's ability to benchmark and compute the level of consolidation in a given area.

The FCC not only discussed the flaws in the contour-overlap methodology but also addressed concerns raised about the Arbitron Metro methodology. As the FCC conceded, "any methodology we develop may create anomalous situations in certain instances. But we cannot agree that our inability to achieve perfection in every instance justifies maintaining the current system." Id. ¶ 263. The FCC established safeguards to deter parties from attempting to manipulate Aribtron market definitions. Specifically, a radio station licensed by a community outside the Metro will not be considered "home" to a Metro for purposes of the ownership rules until that change has been in place for two years. Similarly, changes in Metro boundaries will be subject to a similar two year rule. The FCC believed these safeguards will ensure that changes made to Arbitron Metros will be made "to reflect actual market conditions and not to circumvent the local radio ownership rule." Id. ¶ 278.

The FCC cited several reasons why the Arbitron Metro methodology is a more rational way to define radio markets than the contour method. As stated, Arbitron Metros are consistently considered the relevant geographic market for antitrust purposes. Arbitron market definitions are also industry standards. It is reasonable for an agency to rely on industry defined standards. See AT&T Corp. v. FCC, 220 F.3d 607, 627 (D.C.Cir.2000) (stating that the FCC's reliance on "industry-approved metrics" in lieu of conducting its own competition study was not arbitrary and capricious).

Whether the contour-overlap methodology or the Arbitron Metro methodology more effectively promotes competition between radio stations cannot easily be quantified. Cf. NCCB, 436 U.S. at 796-97, 98 S.Ct. 2096 (rejecting argument that the rulemaking record did not demonstrate the new rule would in fact lead to increased diversity and stating that "diversity and its effects" are "elusive concepts" and "not easily defined"). "To restrict the Commission's action to cases in which tangible evidence appropriate for judicial determination is available would disregard a major reason for the creation of administrative agencies, better equipped as they are for weighing intangibles by specialization, by insight gained through experience, and by more flexible procedure." FCC v. RCA Communications, 346 U.S. 86, 96, 73 S.Ct. 998, 97 L.Ed. 1470 (1953) (internal quotations omitted). As discussed, the FCC provided several reasons why it believed the Arbitron Metro methodology is superior to the contour-overlap methodology. This type of "line-drawing" determination is entitled to "broad leeway" and deference. Sinclair, 284 F.3d at 159. An agency is "free to change its position" if it supplies "adequate data and a reasoned analysis to support the change." Natural Resources Def. Council, Inc. v. E.P.A., 790 F.2d 289, 298 (3d Cir.1986). The FCC met this burden with respect to its decision to switch from the contour-overlap market definition to Arbitron Metros.

Having settled on a revised local radio market definition, the Commission reaffirmed the ownership tiers established by Congress in the 1996 Act. The Commission emphasized that "[n]umerical limits help to keep the available capacity from becoming `locked-up' in the hands of one or a few owners, and thus help prevent the formation of market power" in competing for listeners. Order ¶ 288. The Commission observed that "[t]he current tiers ensure that, in markets with between 27 and 51 radio stations, there will be approximately five or six radio firms of roughly equal size," and that many of the top 100 metro markets fall within that range. Id. ¶ 289. Finding "that the concentration levels permitted by the current rule represent a reasonable and necessary balance for radio broadcasting that comports with general competition theory," the Commission "decline[d] to relax the rule to permit greater consolidation." Id. ¶ 290. The Commission also decided against more restrictive limits. The Commission recognized, however, "that greater levels of concentration may be needed to ensure the potential for viability of radio stations in smaller markets." Id. ¶ 292.

As with broadcast television, the Commission found that radio audience shares can change over time, and an appropriate evaluation of radio markets structure must take into account that licenses provide radio stations with the capacity to provide more popular programming. See id. ¶ 288. The Commission's assumption of equal shares when choosing its ownership limits takes account of these considerations. The Commission recognized that over the life of a firm's investment in a radio station, the market share of all stations in the market can shift substantially. Accordingly, the Commission chose an approach that counts all stations in the market as having similar capacity to compete for listeners. In evaluating the local radio ownership limits, the Commission sought "both to ensure a healthy, competitive radio market by enabling radio owners to achieve significant efficiencies though consolidation of broadcast facilities," while at the same time "ensur[ing] that such consolidation does not ... stifl[e] competitive incentives." Id. ¶ 293.

I agree with the Commission's determination that numerical limits on radio station ownership help preserve the limited number of station licenses from being monopolized by a few owners and help prevent the formation of market power. While interested parties may have struck the balance at a different point, or suggested rules that promoted different market structures, the line was the Commission's to draw. NCCB, 436 U.S. at 814-15, 98 S.Ct. 2096. The record supports its reasonable decision, which is neither arbitrary nor capricious.

The Commission also revised its rule to count noncommercial stations as part of the local radio market in order to account for the fact that such stations exert competitive pressures in the listening and program production markets. Order ¶ 239. The Agency found because "noncommercial stations... receive a significant listening share in their respective markets," their existence "exerts competitive pressure on all other radio stations in the market." Id. ¶ 295. Additionally, noncommercial stations count in determining the size of the local market. Id. The FCC also determined it would prohibit the transfer of station combinations that violate the local radio ownership rule at the time of sale, subject to a limited exception for sales to "eligible entities."110 Id. ¶ 489. The FCC found the disadvantages of allowing the transfer of combinations which do not comport with the ownership limits are not outweighed by countervailing considerations, such as owners' expectancy. Id. ¶ 487.

The FCC determined that Joint Sales Agreements111 should be attributable to the brokering licensee under the new rules.112 Id. ¶ 317. Where an entity owns or has an attributable interest in one or more stations in a local radio market, joint advertising sales of another station in that market for more than 15 percent of the brokered station's advertising time per week will result in counting the brokered station toward the brokering licensee's ownership caps. Id. The FCC justified this change in the attribution rule with its finding that in-market JSAs above the 15% threshold convey to the brokering entity a degree of "influence or control" sufficient to warrant attribution under the ownership rules. Id. ¶ 318. The FCC gave licensees two years from the effective date of its Order to terminate sales and programming agreements that do not comply with the modified local radio ownership rules, and declined to allow any transfers on non-compliant combinations that included the newly attributable sales and programming agreements. Id. ¶¶ 325, 491. I find the Commission's conclusions in the Local Radio Ownership Rule to be reasonable, and clearly not arbitrary and capricious.

For the reasons stated, I believe the Agency sufficiently supported its rule changes. The Commission's decisions were neither arbitrary nor capricious, were supported by the record and are in the public interest. The majority affirms some of the Commission's conclusions as well. I concur with the majority in its affirmance of certain of the Commission's conclusions. For the Cross-Ownership Rules, I agree the Commission's decision not to retain a ban on newspaper/broadcast cross-ownership is justified under § 202(h) and is supported by record evidence, and the Commission's decision to retain some limits on common ownership of different-type media outlets was constitutional and did not violate § 202(h). For the Local Television Ownership Rule, I agree the Commission's decision to retain the top-four restriction is justified and the threshold challenges to the Commission's regulatory approach fail. For the Local Radio Ownership Rule, I agree the Commission justified the switch to Arbitron Markets, that the Commission was warranted in including noncommercial stations, and the transfer restriction and attribution of Joint Sales Agreements should be upheld. I now turn to those areas where the majority finds the Commission's decisions unsupported.

A. Cross-Ownership Rule and the Diversity Index

The majority criticizes the weight given to the Internet in the Diversity Index, as well as the Commission's decision to attribute equal market shares to different outlets within the same media type. The majority also finds the Commission inconsistently derived the Cross Media Limits from the Diversity Index. Finally, the majority requires specific notice of the Diversity Index on remand.

Before addressing these conclusions, it bears reiterating, as the Supreme Court recognized in FCC v. NCCB, that "[d]iversity and its effects are ... elusive concepts, not easily defined let alone measured without making qualitative judgments on both policy and First Amendment grounds." 436 U.S. at 796-97, 98 S.Ct. 2096 (internal quotations omitted). In reviewing the Agency's methodological choices and line-drawing in these areas, the Commission's determinations deserve deference. See Sinclair, 284 F.3d at 159. Furthermore, I agree with the Commission's explanation that the Diversity Index is not the final rule, nor the sole consideration relied upon, but rather a tool created to provide the Cross Media Limits with empirical footing. As such, the Agency's choice and formulation of a metric should be granted deference unless it can be shown to be unreasonable. The majority falls short of such a showing.

The Order concludes that diversity is best measured by outlets that provide local news. Order ¶ 391. The majority claims the Diversity Index grants too much weight to the Internet, understating the level of market concentration and overstating the level of diversity in a given market.113 Although conceding the Internet contributes to viewpoint diversity, the majority states, "we believe that the Commission gave too much weight to the Internet in deriving the Cross-Media Limits." But the Commission's line-drawing is not arbitrary or capricious. The majority has substituted its judgment for that of the FCC.

Although the majority finds the exclusion of cable from the Diversity Index to be justified, it holds the FCC's differential treatment of the Internet versus cable is unsupported. The FCC excluded cable because of serious doubts as to the extent cable provided independent local news as opposed to merely transmitting broadcast signals. The majority finds the Internet also largely transmits local news by providing another platform for local newspapers and broadcast stations. The majority holds the FCC must somehow discount responses that relied on websites that republished information reported by a broadcast or newspaper counterpart, or provide more persuasive evidence of significant independent local news on the Internet.

To provide a reference point, the FCC's distinction between cable and the Internet is distinguishable from the Commission's unsupported and conflicting definition of "voices" in Sinclair. See 284 F.3d at 162-65. In that case, the Court of Appeals for the D.C. Circuit noted the choice of eight competing voices and the definition of voices in the Commission's local television rule "are quintessentially matters of line drawing invoking the Commission's expertise in projecting market results." Id. at 162. But the court invalidated the Commission's definition of voices because the FCC did not adequately explain why it included only broadcast television stations as voices. Id. at 165. The court also found the Commission failed to explain its different definitions of "voices" in its local ownership rule and its cross ownership rule. Id. at 162.

In the Order at issue in Sinclair, the Commission provided two reasons for limiting its definition of voices. The first was a Roper Study, which "at no point ... compare[d] broadcast to non-broadcast television as sources of news, much less as sources of local news. Rather the Roper study simply indicated that nearly 70% of adults get most of their news from television...." Id. at 163. The second reason "involved `the unresolved questions about the extent to which [non-broadcast] alternatives are widely accessible and provide meaningful substitutes to broadcast stations.'" Id. at 164 (quoting Local Ownership Order P. 33, 64 Fed. Reg. 50,651). The Commission concluded that "in the absence of `definitive empirical studies quantifying the extent to which the various media are substitutable in local markets,' the `unresolved questions' on substitutability precluded further relaxation of local ownership restrictions." Id. The court determined that this "wait-and-see approach ... cannot be squared with [the Commission's] statutory mandate ... to `repeal or modify' any rule that is not `necessary in the public interest.'" Id. (quoting Fox Television Stations, 280 F.3d at 1042.).

In this Order, the Commission's methodologies, studies and analyses are significantly more comprehensive than in the Local Ownership Order reviewed in Sinclair. The Commission has presented valid reasons to exclude cable from the Diversity Index while including the Internet at a weight of 12.5%. Although not perfect, the MOWG studies are a significant improvement over the study in Sinclair, which did not compare broadcast television to cable television as a source of news. In the Order reviewed here, the Commission presents evidence on the substitutability of various media as news sources. See Order ¶ 409. The studies relied on by the Commission, the Diversity Index, and resulting Cross-Ownership Rule represent a reasonable approach to establish rational guidelines and are not arbitrary or capricious.

Contending the Internet provides predominantly national news, the majority disputes whether the Internet is a significant source of local news. The majority also notes that most people who go online for local news merely check the websites of their local newspaper or broadcast stations. The survey relied on by the Commission, MOWG Study No. 8, did not identify which websites respondents used as sources of local news.114 This omission can be fine-tuned on the next round of reviews, but, in my mind, because it does not approach the level of arbitrary or capricious action, it need not be remanded, further delaying the implementation of the Commission's rules. The Commission responds that the Internet provides online-only news sites such as Salon.com,115 online-only collections of news such as the Drudge Report,116 and otherwise provides a vast universe of information not contained in local newspapers or broadcasts.117 Order ¶ 427.

Although formal local news sites on the Internet are occasionally online platforms for newspapers and broadcast stations, individuals may nonetheless receive news from various local websites that directly transmit their own "news," rather than collect it as "news." Reply Comments of Media Gen., Inc., at 15-16. As the Commission noted, "[a]lthough many local newspaper and broadcast stations maintain websites with news content, that does not begin to plumb the extent of news sources on the Internet." Order ¶ 427. I agree. The majority contends the FCC should have discounted responses suggesting Internet use was constrained to online platforms for real world media. Even assuming the Commission had sufficient data to make such a calculation in this biennial review, it is not clear 12.5% was an inappropriate weight to attribute to the Internet. With future Internet growth, the Internet familiarity of the next generation of media users, and the fact that local information can be received outside of one's geographic locality, the Commission likely understated the Internet media share. These considerations only emphasize that determining the appropriate weight for each media source is as much art as science, and the choice of 12.5% was reasonable.

At several places in the Order, the Commission supported its chosen weight for the Internet in the Diversity Index. The FCC found, "[t]he web provides an unrestrained forum for the dissemination and consumption of ideas." Order ¶ 119. The Commission also found that the Internet "is becoming a commonly-used source for news, commentary, community affairs and national/international information." Id. ¶ 365. These findings are amply supported by record evidence. See Media Gen. Comments, App. 9; see also Media Gen. Reply Comments. Comments from Media General, as well as those from Hearst Corporation (referenced in the Order at ¶ 365) pointed to the existence of a significant number of local, independent websites on the Internet.118 Comments providing examples of websites with local content explicitly noted these examples represented a fraction of the websites with local content existing on the World Wide Web. See Media General Comments, 22-23 n. 50 ("Indeed, there is no way in which Media General's work could ever result in an exhaustive and complete list.... With more work, the totals could have easily trebled or quadrupled."). In its Order, the Commission noted 30 million websites existed in 2000. Order ¶ 117. The meteoric growth in the number of sites, as well as the sheer number of sites further belie the majority's contention that independent local content does not exist in substantial amount on the Internet. In formulating its Diversity Index, the Commission made clear it looked to potential sources of viewpoint. Id. ¶ 425. The sheer number of available outlets on the Internet supports the Internet's inclusion at 12.5%.

Moreover, the record contains a significant number of websites for local government bodies and civic organizations. See, e.g., Hearst Corp. Comments, MM Docket No. 01-235, at 11 & n. 36, Apps. A, C. As the Commission found, "[t]here 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." Order ¶ 427. The majority responds that "to accept [the Commission's] `universe of information' characterization of the Internet's viewpoint diversity, we would also have to disregard the Commission's professed intent to focus its consideration of viewpoint diversity on media outlets." The majority continues, "local governments are not, themselves, `media outlets' for viewpoint-diversity purposes. Like many entities, they just happen to use a particular media outlet — the Internet — to disseminate information."

I cannot agree with the majority's characterization. Government web sites are media outlets within the media type of the Internet. Particular websites — like stations within the television and radio mediums — are outlets. See Order ¶ 427 & n. 939 (referring to websites, stations, and publications as "information sources," not "types" of media). Also, the Commission has not excluded government and civic organization websites from its consideration in the Diversity Index. The Commission specifically avoided this sort of content analysis. See id. ¶ 424. These sites are relevant to diversity so long as they provide news and current affairs. The majority falls short of showing these sites fail to provide this service.119 Again, the choice not to exclude local government web sites from the diversity mix is a choice properly within the purview of the Agency. Their inclusion is neither arbitrary nor capricious.120

Beyond specific websites, other factors, discussed in the Order, support the weight the Commission attributed to the Internet. According to the Commission, the Internet lends itself to diverse viewpoints because of its combination of breadth and accessibility. As the Order noted, "accessing Web content is a highly individualized activity." Order ¶ 118. Simply typing a search into a search engine can produce numerous relevant websites. The breadth of the Internet allows access to specific issues at the click of a mouse. For example, a recent search revealed a number of websites relating to the extension of a bike path on the Schuylkill River in Center City Philadelphia. See e.g., http://freetheriverpark.typepad.com/ (newsletter of path access advocacy group) (last visited June 24, 2004); http:// www.centercityresidents.org/SRPaccess.htm (letter in support of park access from Center City Resident's Association) (last visited June 24, 2004).

Moreover, the Internet is unconstrained and provides a forum for a limitless number of voices and viewpoints. Recognizing the potential for viewpoint dissemination, political groups have taken their message to the Internet en masse. See e.g., Larry Fish, New Web Politics: Social Networking, Phila. Inquirer, April 27, 2004, at B1 (discussing political candidates' use of the Internet as a campaign tool). Media watchdog sites provide an alternative to a specific political view in significant numbers. See e.g., Jim Ruttenberg, New Internet Site Turns Critical Eyes and Ears to the Right, N.Y. Times, May 3, 2004, at A21 (describing Internet site, www.mediamatters.org, where liberal commentators monitor and counter conservative online and on-air commentators). Internet news sites have sprouted specifically to provide independent, local news. A particularly good example is the Independent Media Center of Philadelphia, see www.phillyimc.org, which over one million people have visited to access independent news since its founding in 2000. See www.phillyimc.org/about.shtml. Finally, interactive possibilities on the Internet such as message boards and chat rooms permit virtually unlimited viewpoint dissemination from a multitude of independent "sources." The Commission recognized the unique qualities of the Internet as they relate to diversity:

Whereas other forms of media allow for only a finite number of voices and editorially-controlled viewpoints, the Internet provides the forum for an unlimited number of voices, independently administered. Furthermore, content on the Web is multi-media; it can be read, viewed, and heard simultaneously. Since Web pages are stored on Web-hosting file servers, accessing Web content is a highly individualized activity, and any individual with access to a Web browser can access all available Web content 24-hours a day throughout the world.

Order ¶ 118.

After reviewing numerous websites referenced in the FCC's Order and supporting comments, I am convinced that the Commission acted reasonably by including the Internet in the Diversity Index at 12.5%. With over 70% of Americans now online,121 the FCC acted reasonably in counting the Internet as a significant addition to the media marketplace. Assigning the Internet a reasonable percentage of the overall market is a task within the range of Agency discretion. In any event, its assigned weights cannot be termed arbitrary and capricious. The Commission has committed to review the apportioning of media shares in its next review. Order ¶ 414; see also Am. Family Ass'n, 365 F.3d at 1166 (discussing the Commission's "duty to evaluate its policies over time to ascertain whether they work — that is, whether they actually produce the benefits the Commission originally predicted they would") (internal quotations omitted). Each review should bring the Diversity Index — or whatever metric the Commission chooses to measure diversity — closer to the true manner in which individuals actually get their information. Based on today's media marketplace, excluding the Internet from the Diversity Index, or limiting its importance, would have sent the Commission down an arbitrary path. The Cross Media Limits and the underlying Diversity Index constitute a worthy benchmark from which the Commission can continue to refine its Cross Ownership Rule. The Commission's rules are clearly not arbitrary and capricious.

As noted, I believe the Commission supported its treatment of the Internet in the Cross-Media Limits. Nonetheless, in the coming years, the Internet will present challenges for the Commission. In future quadrennial reviews, the FCC may want to reconsider how the Internet fits into the traditional concepts of measuring viewpoint diversity, especially the emphasis on local news. By nature, the Internet is uniform everywhere. Its content is not dependent on geographic or metropolitan boundaries. This fact should not undervalue this critical media as an important source for the dissemination of diverse information. In this respect, new modes to characterize diversity may be required. The Internet allows a dentist in Iraq to post a weblog with daily entries and photos from Baghdad for viewing anywhere in the world. See http:// healingiraq.blogspot.com (last visited May 23, 2004). Because this information may not qualify as "local news," it does not follow it should be excluded from the viewpoint diversity mix. Although independent news is all-important, the ability to replicate and distribute news on the Internet, long after its original published date, may cause the Commission to rethink its sole emphasis on "independent" sources when crafting future rules.

2. Equal Market Shares Within Media Type in the Diversity Index

The majority finds the Commission did not justify its decision to assign outlets of the same media type equal market shares. As noted, the FCC first apportioned shares based on actual use figures. But when weighing outlets within the same medium, it counted outlets equally.122 In light of its expertise and delegated authority to make these kinds of policy choices, as well as the underlying goal of viewpoint availability, I would affirm the Commission's decision to focus on media availability, rather than market share. Therefore, I would affirm the Commission's use of equal shares. Whether the assumption of equal market shares within media types is the perfect solution to a complex regulatory problem in every market is subject to legitimate debate. But I see no principled basis to invalidate the Commission's reasoned approach.

In constructing the Diversity Index, the FCC concluded that actual use data is reliable for differentiating among media types because a respondent who relies on television for local news will probably not switch to radio or newspaper because of inherent differences between the media. Order ¶ 409. When it came to weighing outlets within the same medium, however, the Commission decided to focus on an outlet's capacity to deliver information, rather than viewership or readership. This decision is supported by certain rationales. Id. ¶¶ 421-25.

Although it recognized other approaches were possible, the Commission adopted the "underlying assumption ... that all outlets have at least similar technical coverage characteristics." Id. ¶ 421. While this may be less true for radio stations, the Commission determined that Arbitron Metros truncate the service areas of larger, more powerful stations, equalizing service areas. Additionally, there are generally enough radio stations in a given market so that the per-station share is fairly small. Id.

The Commission found that substitutability within media types was much greater than substitutability between media types. Id. ¶ 422. So by breaking up the entire media market based on viewership, and each specific media type based on an equal share theory, the Commission chose a compromise that reflected individuals' media preferences and the relevant regulatory problem. The Commission provided the following common sense example to support its two complementary approaches:

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.

Id.

As the FCC explained: "media outlets can rapidly expand their distribution of content (including local news) at very low marginal cost." Id. ¶ 423. The FCC asserts that actual use data within a media type is not as reliable because a media outlet could conceivably increase or decrease its local news content. The majority cites the example of the Home Shopping Network switching to a news format as an example of the improbability of the Commission's position. That the Home Shopping Network will likely not switch formats does not negate the Commission's explanation that program shifts can and do occur. Radio stations do switch format.123 Furthermore, the Commission found local news can be added to television broadcasts at low marginal cost, especially in light of the efficiencies of cross-ownership. Id. Such a programming change could affect market share.

The Commission determined that current programming of a specific media outlet is not necessarily an accurate predictor of future programming. See Maurice E. Stucke and Allen P. Grunes, Antitrust and the Marketplace of Ideas, 69 Antitrust L.J. 249, 277 (2001) ("`[E]vidence of past production does not, as a matter of logic, necessarily give a proper picture of a company's future ability to compete.' ... Even if one could define markets and assign market shares in the marketplace of ideas, just how reliable would these historic market shares be under dynamic market conditions?") (quoting United States v. Gen. Dynamics Corp., 415 U.S. 486, 501, 94 S.Ct. 1186, 39 L.Ed.2d 530 (1974)). The majority finds this "does not jibe with the Commission's decision to assign relative weights to the different media types themselves...." This statement assumes the Commission must adopt a single strategy for apportioning shares both within and among outlets, and moreover, this strategy must perfectly account for the programming choices of newly acquired radio stations as well as the Home Shopping Network. In my view, the majority imposes an unreasonably high burden.

The Commission noted that adopting a usage measure to determine the most prominent sources for local news and public affairs would require difficult determinations about which content constitutes local news. For example, the Commission posed the question whether movie reviews should be considered. Order ¶ 424. Other gray areas would require similar determinations. Again, the Commission's goal is not to prescribe what content citizens access, but to ensure that a large range of viewpoints are available to the public. Id. ¶ 425. The equal shares approach adopted by the Commission comports with this goal. As one commentator noted:

In the marketplace of ideas ... what matters is the number of alternative information outlets available to consumers, not the current popularity ... of the ideas communicated by each outlet. Each source of ideas available to a given consumer is equally significant from a First Amendment perspective. The rational way to measure the "share" of each source of ideas available to a given set of consumers, therefore, is to give each source equal weight. It is availability and not usage of alternatives that should count, because it makes no sense to view the FCC's role as regulating the popularity, as opposed to the availability to consumers, of ideas and information. It is unpopular new ideas that may be of the greatest importance to the future. Such unpopular ideas are the essence of diversity in the marketplace of ideas.

Bruce M. Owen, Regulatory Reform: The Telecommunications Act of 1996 and the FCC Media Ownership Rules, 2003 Mich. St. DCL L. Rev. 671, 692 (2003).

The majority, apparently, would have chosen a different approach. But disagreement with the Commission's conclusions on such matters does not render those conclusions arbitrary and capricious. The Commission is entitled to "implement its view of the public-interest standard of the Act `so long as that view is based on consideration of permissible factors and is otherwise reasonable.'" WNCN Listeners' Guild, 450 U.S. at 594, 101 S.Ct. 1266 (quoting NCCB, 436 U.S. at 793, 98 S.Ct. 2096). In breaking up media types by market share and media outlets equally, in my view, the Commission incorporated public media preferences, its diversity goals, as well as industry characteristics. The Commission's decision to attribute shares within media equally was reasonable as it took account of permissible factors in light of the FCC's public interest goals.

3. Deriving the Cross Media Limits from the Diversity Index

The majority contends the Commission did not rationally derive its Cross Media Limits from the Diversity Index results. The majority correctly notes deference to the Commission's judgment reaches its zenith when assessing the rationality of the agency's line-drawing endeavors, see NCCB, 436 U.S. at 814-15, 98 S.Ct. 2096, but then takes issue with the "seemingly inconsistent" manner in which the line was drawn.

The majority overlooks the Commission's statement that the Diversity Index is a tool to help make judgments about cross-ownership, not the final rule. See Order ¶¶ 391, 435. As the Commission emphasized, the "cross-media limits are based on a set of assumptions drawn directly from the record evidence in this proceeding ... [and] ultimately rest[] on our 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." Id. ¶ 435 (emphasis added). In looking to the Diversity Index, the Commission did not abdicate its judgment for formula.

The Diversity Index — or the Cross Media Limits for that matter — will not provide the perfect limits in every local market. As the Commission stated:

[W]e 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 proscriptive rules themselves.

Id. ¶ 453.

The Commission decided to rely on bright-line rules in crafting its ownership rules. Id. ¶ 80. The FCC chose this approach because "bright line rules provide certainty to outcomes, conserve resources, reduce administrative delays, lower transaction costs, increase transparency of [the] process, and ensure consistency in decisions." Id. ¶ 82. As a result of the Commission's decision to adopt a bright-line rule rather than a case-by-case approach, the rules will not always perfectly address the diversity needs of a given market. This characteristic is inherent in nationwide, prescriptive rules, and in my view does not require remand.

The majority finds the Commission allowed combinations (newspaper/television station/50% radio limit) that resulted in higher Diversity Index increases, while disallowing combinations (newspaper/television duopoly), which resulted in lower increases. But the Cross Media Limits were not based solely on the Diversity Index. In fact, the Commission directly addressed the majority's purported "inconsistency:"

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 clear that these 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 ownership 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 in diversity. Although the loss of diversity that might result were that owner to add a significant radio presence in the market in the market warrants a further 50% limit in the number of radio properties that owner might hold, such is not the case if a combination does not include any television properties.

Id. ¶ 467 (emphasis added).

A newspaper/television duopoly combination realizes diminishing cross-fertilization benefits in comparison to the combination of a newspaper and the first station in television or radio. Therefore, it was reasonable for the Commission to disallow newspaper/duopoly combinations even when the Diversity Index increase might be lower than that for newspaper/television station/50% radio combinations. It was rational for the Commission to deviate from the Diversity Index in creating its Cross Media Limits. This example highlights that the Cross Media Limits primarily address the impact of cross-ownership on diversity, without ignoring other ownership goals like competition and efficiency.124

Furthermore, the first task of the Diversity Index was to identify at-risk markets, then identify at-risk cross-ownership combinations. As the Commission commented:

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.

Id. ¶ 442 (emphasis added). The "horizontal" lines the Commission drew in its Cross Media Limits, i.e., the lines between small and medium markets, and medium and large markets, are supported by the Diversity Index results. See id. App. D. I would not vacate and remand an effective methodology because its results do not correspond to the Commission's final rule for all combinations in all markets.

The majority focuses on the methodological choices behind the Diversity Index and overlooks the end result — the Cross Media Limits. The Agency recognized the Diversity Index to be "a blunt tool" that is neither "perfect nor absolutely precise." Id. ¶¶ 392, 398. I see no reason to invalidate any aspect of the Commission's Index. The Cross Media Limits were the first structural change to the Commission's Newspaper/Broadcast Cross-Ownership Rule in nearly thirty years. The Diversity Index lent transparency and empirical footing to this massive undertaking. In my view, it was reasonable for the Agency, both in designing the Diversity Index and in its final Cross Media Limits, to rely on a balanced, empirical approach that will, as a result of § 202(h), be subject to periodic reexamination.

On remand, the majority advises the Commission to cure its "questionable" notice of the Diversity Index. This admonition goes beyond the notice requirements of the Administrative Procedure Act. Under the APA, an agency must publish notice of either the terms or substance of the proposed rule or a description of the subjects and issues involved. 5 U.S.C. § 553(b)(3). The Commission provided notice and opportunity to comment on the newspaper/broadcast cross-ownership rule in two separate notices of proposed rulemaking. See Newspaper/Broadcast NPRM, 16 F.C.C.R. 17,283, 2001 WL 1097041 (2001); 2002 Biennial NPRM, 17 F.C.C.R. 18,503, 2002 WL 31108252 (2002). As the majority recognizes, the Commission gave notice that it was considering "creating a new metric" to "reformulate [its] mechanism for measuring diversity and competition in a market," and that it was contemplating "design[ing] a test that accords different weights to different outlet types." Notice, 17 F.C.C.R at 18,539-40 ¶¶ 113-115. In the Newspaper/Broadcast NPRM, the Commission discussed at length questions regarding diversity and competition in relation to the Newspaper/Broadcast Cross-Ownership Rule, posing questions for public comment such as: "Is it possible that the effect on diversity will be different depending on the size of the markets involved, or the predominance of newspapers and broadcast stations in a particular local market." 16 F.C.C.R. at 17,292 ¶ 18. These and other statements in the NPRMs adequately addressed the subject of the new Cross Ownership Rule.

As an analytical tool that informed the Commission's judgments about how to account for diversity in different markets, and the need for ownership limits in these markets, the Diversity Index itself was formulated as a response to comments. "Rulemaking proceedings would never end if an agency's response to comments must always be made the subject of additional comments." Cmty. Nutrition Inst. v. Block, 749 F.2d 50, 58 (D.C.Cir.1984).125 As the majority notes, formulas derived from comments do not need to be re-noticed unless prejudice will result. Id. ("The response may ... take the form of new scientific studies without entailing the procedural consequences appellants would impose, unless prejudice is shown."). The Diversity Index provided an empirical foundation for formulating the final rule. The choices made in creating this empirical model were within delegated agency authority and were supported by the record. I do not find the alleged lack of notice to be prejudicial. Every methodological choice will have proponents and detractors, but so long as the choices are not arbitrary and capricious, they are valid. I would not expand the settled notice and comment requirements of § 553.

The majority upholds the top-4 limit in the Local Television Ownership Rule but vacates and remands the chosen numerical limits "for the Commission to harmonize certain inconsistencies and better support its assumptions and rationale."126

The FCC justified the specific numerical caps in its intra-service rules (radio and television) by employing the Department of Justice and Federal Trade Commission's "standard approach" to evaluating competitive harms of an increase in horizontal market concentration: the Herfindalh-Hirshmann Index (HHI). The DOJ/FTC Merger Guidelines recognize the HHI level of 1800 as the maximum level for "moderate" concentration. So the FCC designed its local television rule to keep markets within the 1800 limit. Order ¶¶ 192-93. As noted, the FCC selected the upper bound for moderate concentration, 1800, instead of the lower bound, 1000, in recognition of the competitive pressures exerted by the cable networks. Id. ¶ 192.

An HHI score of 1800 corresponds to having six equal-sized competitors in a given market. This led the FCC, in part, to determine in markets of 18 stations or more, an entity could own three stations, while in markets with 5 to 17 stations, an entity could own two stations. Id. ¶ 193. The Commission allowed ownership of two stations in markets with less than 12 stations — though this undercuts its 6 equal-sized competitors goal — because the economics of local broadcast stations justify slight increases in market concentration as market size decreases. Id. ¶ 201. The FCC relied on data from the National Association of Broadcasters showing that small-market stations have a harder time competing for revenue than stations in larger markets. Thus, the local television rule represents a "graduated tradeoff between optimal competition in the delivered video market (six station owners) and recognition of the challenging nature of broadcast economics in small to mid-sized markets." Id. ¶ 202.

In developing the new local ownership tiers, the majority concludes the FCC failed to justify its chosen numerical limits. The majority contends that allowing triopolies in large markets does not ensure that six equal-sized competitors will emerge because the FCC does not take into account existing respective audience shares among stations in a given market. The majority also finds the FCC cannot cite "fluid" market share as a basis for regulation yet, elsewhere in the same Order, use market share differentials as justification for retaining the top-four restriction. The FCC found, for example, "a general separation between the audience shares of the top four-ranked [local] stations and the audience shares of other stations in the market." Order ¶ 195.127 Furthermore, the majority finds the FCC did not cite empirical support for its "equal-sized firm" assumption, and that its assumption of equal market shares, id. ¶ 134, to derive its limits ignores reality. The majority finds that by assuming equal market shares the FCC allowed combinations based on unrealistic market percentages. It maintains the Agency did not rely on market share, but on a rudimentary "head count" of outlets. According to the majority, the inconsistent treatment of market share both within and between rules renders them arbitrary and capricious.128

In my view, the Commission justified its decision to assume equal market shares. The majority states the Commission sought to "ensure that markets would not exceed the Merger Guidelines' 1800 threshold for highly competitive markets." This statement portrays a heavier reliance on the Merger Guidelines than the Commission avowed. In discussing the relationship between its local television ownership limits and the Merger Guidelines, the Commission said: "a strict, overly simplistic application of the DOJ/FTC Merger Guidelines would potentially prohibit some welfare enhancing mergers and allow some anticompetitive mergers...." Order ¶ 193. The Commission also stated:

The DOJ/FTC Merger Guidelines ... are written not for a specific industry, but rather as guidelines intended for application across all industries. Our rules are formulated for a specific market — the delivery of video programming — and are based on an extensive record on the extent of competition in this market and the effect of our current local TV ownership rule. This record allows us to craft a more finely-tuned rule for this industry.

Id. ¶ 192. So, in fact, the Commission did not seek to "ensure" that no market would exceed the Merger Guidelines' 1800 threshold for highly competitive markets. Instead the Merger Guidelines provided a "starting point" from which the public interest dictated the final rule. Id.

The majority finds that no evidence supports the Commission's equal share assumption. I disagree. The FCC found that a station's market share is too fluid to use as a basis for regulation, so its decision to regulate instead a firm's "capacity" to deliver programming is justified. Id. ¶ 193. Because product innovation and program choice vary with each season, a firm's market share is more fluid than in other industries. Thus, the FCC focused on a firm's capacity to deliver programming. Id. Over the life of a firm's investment in a station, and the duration of its license, the market breakdown can shift substantially. For example, recently launched broadcast stations now vie with the "big four" for viewers. See id. ¶ 110. As recently as 1995, two new broadcast networks UPN and WB were launched and now both compete-often successfully-for viewers with the "big four" networks. See e.g., http:// www.viacom.com/prodbyunitl.tin?ixBusUnit=30 (discussing growing popularity of UPN network) (last visited May 25, 2004); see also Comments of Paxson Communications Corp., at 20 (noting retention of the UHF discount is "likely to encourage the emergence of a larger number of competitive broadcast networks to join the existing seven."). The Merger Guidelines, which the Commission looked to in setting its local license caps, contemplate the possibility that in instances where current revenue market shares are misleading indicators of competitive performance, equal shares can be imputed to each competitor. DOJ/FTC Merger Guidelines § 1.41 n. 15 ("Where all firms have, on a forward-looking basis, an equal likelihood of securing sales, the Agency will assign firms equal shares.").

Moreover, the majority discounts the Commission's assertion that local limits contribute to the Agency's diversity goals. "Ensuring that several competitors remain within each of the radio and television services, we also ensure that a number of independent outlets for viewpoint will remain in every local market, thereby ensuring that our diversity goal will be promoted." Order ¶ 129; see also id. ¶ 178. Commentators have recognized that outlets, or sources, should be counted equally in the marketplace of ideas:

I conclude that independently-owned outlets for, or sources of, ideas and information generally should each be counted equally as separate sellers in the marketplace of ideas, with response to the consumers whom they can reach (or the consumers who can reach them), without regard to the classification or popularity of their current content.

Bruce M. Owen, Regulatory Reform, 2003 Mich. St. DCL L. Rev. at 696.

Though not flawless in all markets, the Commission soundly justified its assumption of equal market shares as underpinning for its choice of license limits in its local television rule. In reviewing assumptions made by the Commission in a different context, the D.C. Circuit recently wrote:

These assumptions appear rational on the current record. We have no obvious way of verifying the FCC's assertion regarding the general characteristics ... We must defer to the Commission's expert judgment in the absence of record evidence indicating that the Commission's assumption is a clear error of judgment, or a showing that the empirical assumption is facially implausible or inconsistent.

Am. Family Ass'n, 365 F.3d at 1165-66. In my view, the majority has not shown error approaching this standard. As agency line-drawing deserves deference, and the Commission's choice of equal market shares is neither arbitrary and capricious, I see no basis for overturning this decision.

The majority supports the Commission's use of numerical limits in the local radio ownership rule but finds the Commission did not support its decision to retain the existing numerical limits established by § 202(b). The majority also vacates and remands the "five equal-sized firms" starting point, which undergirds the local radio rule, claiming the Commission's decision is not supported by "substantial evidence." In light of § 202(h), I agree that the Commission must explain its decision to maintain the existing limits. But I find the Commission to have justified its choices.

First, in reaching its conclusion, the majority employs too high a standard. The Commission's conclusions do not need to be supported by "substantial evidence." Rather, the Commission's line-drawing is generally not overturned "unless a petitioner can demonstrate that the lines drawn ... are patently unreasonable, having no relationship to the underlying regulatory problem." Sinclair, 284 F.3d at 162 (internal quotations omitted). The Commission's actions did not "run counter to the evidence before it," id. (quoting Motor Vehicle, 463 U.S. at 43, 103 S.Ct. 2856), and the Commission "provide [d] a reasoned explanation for its action." Id.

As discussed, the FCC maintained the radio tiers established by Congress in the 1996 Act. The FCC explained that retaining the existing tiers would result in five roughly equal-sized firms in each market. Order ¶ 289. The Commission explained its ownership rules ensure there will be roughly five or six owners in markets with between 27 and 51 radio stations, which should allow competitive market performance. Id.129 Citing economic literature, the FCC found that five equal-sized firms ensure fragmented, competitive markets. See id. ¶ 289 n. 609. The FCC also concluded the current limits are not overly-restrictive based on data showing the top four stations in each metro market are thriving in terms of revenue and audience share. Id. ¶ 290.

The majority asserts the economic studies on which the FCC relied do not support its premise that a market with five equal-sized competitors is comparable in terms of market performance to a fragmented, structurally competitive market. See id. ¶ 289 n. 609 (citing R. Selter, A Simple Model of Imperfect Competition Where Four Are Few and Six Are Many, 2 Int'l J. Game Theory 141 (1973); Louis Phillips, Competition Policy: A Game Theory Perspective Ch. 2 (Cambridge U. Press 1995); Timothy F. Bresnahan & Peter C. Reiss, Entry and Competition in Concentrated Markets, 99 J. Pol. Econ. 977-1009 (1991)). The majority contends the Commission failed to respond to petitions urging market structures other than equal-sized competitors, which could provide equally competitive markets. From the other direction, the majority picks up the argument of Citizen Petitioners that the game-theory articles should not be followed because they conflict with the most recent rewriting of the Merger Guidelines, which suggest a HHI score above 1800 corresponds with a highly concentrated market. Merger Guidelines § 151(c).

The Commission may favor a certain market structure if it reasonably supports its decision. The Commission is the appropriate body to choose between various equally competitive market structures when formulating its rules. See Am. Family Ass'n, 365 F.3d at 1163 ("Given that both options are rational, the FCC's choice of means is well within its discretion."). In structuring its rules, the Commission chose to take into account that radio audience shares change over time,130 and an evaluation of radio market structure must also take into account that licenses provide radio stations with the capacity, and likely the incentive, to provide more popular programming. See Order ¶ 288. It would be irrational if a radio station did not provide more popular programming because it would surpass its allotted market share.

In determining appropriate ownership tiers, the Commission's use of antitrust theory tailored to the radio industry is reasonable. The Commission is not required to strictly follow the Merger Guidelines in creating its limits. As the majority recognizes, antitrust regulation and FCC ownership regulation play different roles.131 While the majority may have struck the balance at a different point, the line was the Commission's to draw. Sinclair, 284 F.3d at 162.

The majority is not convinced by the Commission's assertion that five equal-sized competitors "would emerge or actually have emerged under numerical limits." Certainly, "radio station groups with similar numbers of radio stations [can] have vastly different levels of market power." Order ¶ 290. This assertion, however, does not require the Commission to adopt a market structure that ignores its other finding — that station shares can be fluid. Additionally, as discussed in relation to the Local Television Ownership Rule, an assumption of equal shares makes sense in terms of diversity, as each station has the capability of disseminating information. As the local ownership rules support the Commission's diversity objectives as well, equal shares are reasonable, and clearly not arbitrary and capricious. See id. ¶ 306 ("Our competition-based limits on local radio ownership thus promote viewpoint diversity, not only ensuring a sufficient number of independent radio voices, but also by preserving a market structure that facilitates and encourages entry in the local media market by new and under-represented parties.")

The majority finds the Commission "does not explain why it could not take actual market share into account when deriving the numerical limits." But in its Order, the Commission wrote:

We also reject arguments that we incorporate a market share analysis into the local radio ownership rule or that we continue to "flag" applications that propose radio station combinations above a certain market share.... Market share, however, must be considered in conjunction with the overall structure of the industry in determining whether market power is present. In radio, the availability of a sufficient number of radio channels is of particular importance in ensuring that competition can flourish in local radio markets. The numerical caps and the AM/FM service limits are designed to address that interest, and in our judgment, establishing a inflexible market share limit in our bright-line rule would add little, if any, benefit. We do not seek to discourage radio firms from earning market share through investment in quality programming that listeners prefer; our objective is to prevent firms from gaining market dominance through the consolidation of a significant number of key broadcast facilities. We do not believe that developing a market share limit would significantly advance that objective.

Order ¶ 300 (emphasis added).

The majority also points out the Commission has previously looked to market share for measuring diversity and competition in local radio markets. Again, the Commission expressly addressed its change in policy when it wrote:

We recognize that our conclusion differs from the Commission's view in 1992 that an audience share cap was necessary "to prevent consolidation of the top stations in a particular local market." But the audience share cap was never intended to be more than a "backstop" to the new numerical limits the Commission had established, which for the first time allowed a party to own multiple radio stations in a local market. The audience share cap was eliminated as a result of the revisions to the local radio ownership rule that Congress mandated in the 1996 Act, which left only the numerical caps in place. But because of the problems associated with the contour-overlap market definition and counting methodologies, we could not rely with confidence on those numerical limits to protect against undue concentration in local markets. As a result, we began looking at revenue share in our "flagging" process and the interim policy that we established in the Local Radio Ownership NPRM. Now that we have established a rational system for defining radio markets and counting market participants, we believe that the numerical limits will be better able to protect against harmful concentration levels in local radio markets that might otherwise threaten the public interest.

Order ¶ 301. As noted, the Commission based its switch to the Arbitron Metro market definition — which the majority affirms — in part, on the difficulty of guarding against undue consolidation of market share under the old contour-overlap method. See Order ¶ 257 ("[T]he contour-overlap system actually encourages consolidation of powerful radio stations because stations with larger signal contours are more likely to create larger radio markets, which make it more likely that a party would be able to acquire additional radio stations in that market."). With a more rational market definition, which, according to the Commission, will make consolidation less likely, it was not arbitrary or capricious for the Commission to de-emphasize market share in crafting its ownership limits.

In order to warrant judicial deference, an agency's line-drawing decision must be justified by a reasonable explanation and cannot run counter to the evidence. Sinclair, 284 F.3d at 162; see also NCCB, 436 U.S. at 814-15, 98 S.Ct. 2096. No record evidence invalidates the policy choices made by the Commission. Its assumptions and chosen limits correspond to a reasonable market structure for radio programming markets.132

The Supreme Court has upheld FCC policy based on factual determinations "that were primarily of judgmental or predictive nature." NCCB, 436 U.S. at 814, 98 S.Ct. 2096 (noting "complete factual support is not possible or required; `a forecast of the direction in which future public interest lies necessarily involves deductions based on the expert knowledge of the agency'") (quoting FPC v. Transcontinental Gas Pipe Line Corp., 365 U.S. at 29, 81 S.Ct. 435). I do not find § 202(h), or any other directive to the Agency, to require the Commission to ignore its well-informed predictions about which types of markets best foster its public interest goals, and consequently to create rules with these markets in mind.

For the foregoing reasons, I would affirm the Commission's rules.