Full opinion text
Petition for review denied and judgment affirmed by published opinion. Judge MICHAEL wrote the opinion, in which Judge WIDENER and Judge NIEMEYER joined. MICHAEL, Circuit Judge. Direct broadcast satellite (DBS) service has recently joined cable and broadcast television as a major force in the market for delivering television programming to consumers. In these consolidated cases, representatives of the satellite industry raise various constitutional challenges to Congress’s efforts to regulate competition in that market through the Satellite Home Viewer Improvement Act of 1999 (SHVIA). Pub.L. No. 106-113, 113 Stat. 1501A-523. In addition, petitioners from the broadcast industry argue that one provision (the “a la carte rule”) of the FCC’s order implementing SHVIA must be struck down as an unreasonable interpretation of the statute. By enacting SHVIA, Congress sought to promote competition between the satellite and cable industries by creating a statutory copyright license that allows satellite carriers to carry the signals of local broadcast television stations without obtaining authorization from the holders of copyrights in the individual programs aired by those stations. The Act also imposes a “carry one, carry all” rule, which was designed to “preserve free television for those not served by satellite or cable and to promote widespread dissemination of information from a multiplicity of sources.” H.R. Conf. Rep. No. 106-464, at 101 (1999) (SHVIA Conference Report). The rule, which is scheduled to take effect on January 1, 2002, will require satellite carriers that choose to take advantage of the statutory copyright license by carrying one broadcast station in a local market to carry all requesting stations within that market. We hold, as did the district court, that the carry one, carry all rule does not violate either the First Amendment or the other constitutional provisions cited by the satellite carriers. We also hold that the FCC’s a la carte rule, which allows satellite carriers to offer local broadcast stations to their subscribers either individually or as part of a single package, is not arbitrary, capricious, or contrary to law. See 5 U.S.C. § 706(2)(A). I. A. Nearly all consumers receive their television programming through one of three delivery systems: broadcast television, cable, or satellite. Broadcast television stations transmit electromagnetic signals over the air, and these signals can be captured by any receiving television antenna within range. Twenty percent of American television households rely exclusively on broadcast stations for their television programming. Viewers pay no fee to receive broadcast signals. Instead, broadcast stations are supported by advertisers who pay for air time at rates determined by the audience sizes for particular programs. The most popular broadcast stations are affiliated with one of the four major television networks (ABC, CBS, NBC, and Fox). The major network affiliates compete for viewers and advertisers with various independent broadcasters, including independent commercial stations, noncommercial stations, and affiliates of emerging networks (UPN, WB, and PAX). The broadcasters’ principal competitors in the television programming delivery market are the cable and satellite industries. Cable and satellite companies now serve around 80 percent of America’s television households. Unlike broadcasters, their primary source of revenue is subscription fees. Cable television distributes its signals to subscribers over a local network of wires. It has for many years been the leading provider of television programming to American homes. Roughly 67 percent of television households currently subscribe to cable. Although cable subscribers must pay for the right to receive cable signals, they receive better picture quality and a wider variety of programming options than do television viewers who rely on antennas. Today, 84 percent of cable systems offer their subscribers at least 30 channels, including national non-broadcast channels (such as ESPN, MTV, CNN, and The Weather Channel) and regional non-broadcast channels (such as the New England Sports Channel). Cable operators also retransmit the signals of local broadcast stations to their subscribers. Providers of DBS (direct broadcast satellite) service deliver television programming by uplinking signals to satellites orbiting in space and then beaming those signals to receiving dishes connected to subscribers’ television sets. In the 1980s satellite dishes were 6 to 10 feet in diameter, and satellite carriers primarily served customers in rural areas. During the 1990s satellite carriers such as EchoStar and DirecTV developed much smaller dishes and began to compete with cable for subscribers in urban and suburban areas. Today, satellite carriers provide service in each of the nation’s 210 television markets and serve about 13 percent of television households. Whereas cable systems deliver their signals to subscribers over local wire networks, satellite is primarily a national service. The satellites currently used by DBS providers occupy one of three positions in the Earth’s orbit (called full CO-NUS slots) that allow the satellites to transmit a single beam covering the entire continental United States. The beam from a full CONUS satellite contains multiple frequencies, and compression technology enables multiple television channels to be carried on each frequency. The FCC licenses the use of 32 frequencies at each orbital slot; thus, there are 96 total frequencies that satellite carriers can use to reach satellite subscribers across the United States. Currently, 50 of these frequencies are licensed to EchoStar and 46 to DirecTV. Using their current compression ratios, EchoStar and DirecTV each have the ability to carry between 450 and 500 channels via full CONUS satellites. Every channel carried on these satellites is beamed to the homes of all subscribers; however, channels that individual subscribers do not pay to receive are blocked by the use of software in the subscriber’s home satellite equipment. Like cable, satellite service is financed by subscription fees, and it offers better picture quality and more viewing options than broadcast television. Satellite carriers can provide their customers with more national and regional non-broadcast channels than most cable systems; yet before SBVIA was passed, satellite carriers had difficulty competing with cable for urban and suburban customers. The root cause of this difficulty was plain. Cable systems, but not satellite carriers, provided their customers with access to the signals of local broadcast stations. This competitive advantage was rooted not only in technology but also in federal copyright law. Unlike satellite carriers, cable operators have never been required to obtain copyright clearances for the broadcast programming they retransmit to their subscribers. To explain the origins of cable’s competitive advantage over satellite, we must briefly review the history that produced the current legal regime governing the relationships among cable, broadcast television, and satellite. In 1965, long before cable became a major force in the television programming delivery market, the FCC imposed “must-carry” rules on cable systems requiring them to retransmit the signal of any requesting broadcast station that was “significantly viewed” in its local market. See Quincy Cable TV, Inc. v. FCC, 768 F.2d 1434, 1438-43 (D.C.Cir.1985) (discussing the early history of cable regulation). The FCC feared that cable might undermine free, local broadcasting unless “local broadcasters were assured access to the whole of their allocated audience.” Id. at 1441. Specifically, the must-carry rules were designed to aid newer UHF stations, which lacked the signal quality enjoyed by more established VHF stations. The FCC was concerned that in the absence of carriage obligations cable systems would carry only VHF stations and that non-carried UHF stations would be placed at a competitive disadvantage in the race for advertising dollars. See id. Copyright law had little effect on cable until Congress passed the Copyright Act of 1976, Pub.L. No. 94-553, 90 Stat. 2541, which provides that owners of copyrights in audiovisual works such as television programs have the exclusive right to authorize public performances of those works. 17 U.S.C. § 106(4). Although the Copyright Act of 1909 had granted copyright holders a similar right to control public perfor- manees of copyrighted works, cable operators’ secondary transmissions of broadcast programs were not regarded as additional performances of those programs under the 1909 Act and thus did not infringe upon the intellectual property rights of the program copyright holders. See Fortnightly Corp. v. United Artists Television, Inc., 392 U.S. 390, 400-01, 88 S.Ct. 2084, 20 L.Ed.2d 1176 (1968). The 1976 Act’s legislative history reveals that Congress intended to countermand the result in Fortnightly by designating cable’s secondary transmissions of broadcast programs as public performances of those programs: “a cable television system is performing when it retransmits the broadcast [of a television program] to its subscribers.” H.R.Rep. No. 94-1476, at 63 (1976), reprinted in 1976 U.S.C.C.A.N. 5659, 5677. As a result, copyright law now generally requires parties seeking to retransmit the signal of a broadcast television station to obtain authorization from those holding copyrights in each of the programs broadcast by that station. Cable operators, however, have never been subject to this general rule. Congress found “that it would be impractical and unduly burdensome to require every cable system to negotiate with every copyright owner whose work was retransmitted by a cable system.” Id. at 89, 1976 U.S.C.C.A.N. at 5704. As a result, the 1976 Copyright Act granted cable operators a statutory license that allows them to retransmit broadcast television signals without securing authorization from program copyright holders. See 17 U.S.C. § 111(c). As cable grew in popularity, cable operators and cable programmers began to chafe at the FCC’s must-carry rules. Eventually, cable interests challenged the rules as an unconstitutional burden on freedom of speech, and the rules were struck down by the D.C. Circuit in 1985. See Quincy Cable TV, 768 F.2d at 1463. The FCC moved quickly to adopt a modified set of carriage rules, but the D.C. Circuit again decided that the rules violated the First Amendment. See Century Communications Corp. v. FCC, 835 F.2d 292, 293 (D.C.Cir.1987). After a period of several years during which cable enjoyed the benefits of a statutory copyright license without the burdens of carriage obligations, Congress reimposed must-carry rules by passing the Cable Television and Consumer Protection and Competition Act of 1992 (Cable Act). Pub.L. No. 102-385, 106 Stat. 1460. Like the FCC’s original must-carry rules, the Cable Act’s must-carry rules were designed to protect broadcast stations that might be refused carriage in the absence of a must-carry requirement. The Cable Act must-carry rules were also challenged on First Amendment grounds, but were ultimately upheld by the Supreme Court in Turner Broadcasting System, Inc. v. FCC, 512 U.S. 622, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994) (Turner I), and Turner Broadcasting System, Inc. v. FCC, 520 U.S. 180, 117 S.Ct. 1174, 137 L.Ed.2d 369 (1997) (Turner II). In Turner I the Supreme Court held that the must-carry rules were content-neutral restrictions on speech, Turner I, 512 U.S. at 662, 114 S.Ct. 2445, reviewable under the intermediate First Amendment scrutiny standards established by United States v. OBrien, 391 U.S. 367, 88 S.Ct. 1673, 20 L.Ed.2d 672 (1968). The Court further recognized that the interests served by the must-carry rules — preserving local broadcast television, promoting the “widespread dissemination of information from a multiplicity of sources,” and promoting fair competition in the television programming market — were important “in the abstract.” Turner I, 512 U.S. at 662-63, 114 S.Ct. 2445. However, the Court remanded the case for further factual development on the question of whether these interests were genuinely advanced by the must-carry rules and whether the rules were a narrowly tailored means of promoting these interests. After the remand the Court held in Turner II that the cable must-carry rules fully satisfied the O’Brien test. As a result, cable operators today enjoy the privileges of a statutory copyright license but are also bound by carriage obligations. As the above history indicates, the cable industry developed within a context of extensive FCC regulations, but it has been largely unaffected by copyright considerations. Before 1976, cable retransmissions of broadcast programs were not regarded as additional performances of these programs and therefore did not infringe program copyrights. Since 1976, cable has enjoyed a statutory copyright license. In contrast, the satellite industry has until recently enjoyed relative freedom from FCC oversight, but its history has been shaped by a series of changes in federal copyright law. When the 1976 Copyright Act was passed, home satellite service did not exist, and the statutory copyright license created by 17 U.S.C. § 111(c) was granted only to cable systems. Although early satellite carriers lacked the channel capacity to retransmit the signals of local broadcast stations to the local audiences of those stations (called “local-into-loeal” service), the carriers did have the capacity and the desire to offer their subscribers some distant network signals. In other words, they had enough capacity to carry the signals of network affiliates in a major city (usually New York) throughout the entire country, but not enough capacity to carry local network affiliates in smaller cities. It was unclear, however, whether satellite carriers’ secondary transmissions of the signals of network affiliates infringed the copyrights in the various programs broadcast by those stations. Satellite carriers argued that they might be eligible to use the § 111 copyright license as “cable systems,” 17 U.S.C. § 111(c), or that their secondary transmissions of broadcast signals were noninfringing under the exemption for passive carriers in 17 U.S.C. § 111(a)(3). Still, they asked Congress to resolve the uncertainties about their copyright liability by passing new legislation. Congress responded with the Satellite Home Viewer Act of 1988. Pub.L. No. 100-667, 102 Stat. 3949. Recognizing that the high transaction costs involved in clearing the rights to network signals might impede the growth of the fledgling satellite industry, Congress gave carriers a limited copyright license to retransmit the signals of distant network broadcast stations to unserved households that were unable to receive an adequate over-the-air signal through a conventional rooftop antenna. 17 U.S.C. § 119(a)(2)(B) (1994), amended by § 119(a)(2)(B) (Supp. V 1999). “Unserved households” were defined in part as those located outside a station’s “Grade B contour,” as defined by FCC regulations. Id. § 119(d)(10) (1994), amended by § 119(d)(10) (Supp. V 1999). Congress limited the copyright license to retransmissions to unserved households in order to respect the terms of network-affiliate contracts, which grant local affiliates the exclusive right to broadcast their networks’ programming in their local markets. Congress reasoned that local affiliates would lose the benefits of their bargained-for exclusivity rights if satellite carriers were allowed to import distant network signals into the affiliates’ local markets. The limited license created by the 1988 Act spurred the growth of satellite carriers and expanded the viewing options of unserved households, though it did little to help satellite effectively compete with cable in urban and suburban areas where viewers could already receive high quality broadcast signals. In the late 1990s two problems led Congress to revisit its work in the Satellite Home Viewer Act of 1988. First, cable continued to enjoy a virtual monopoly in subscription television services in metropolitan areas, and as a result cable subscribers were paying too much for those services. See S.Rep. No. 106-51, at 1-2 (1999). Second, the 1988 Act failed to deal adequately with the problems of satellite subscribers who lived within the Grade B contours of network television stations. Although these subscribers were ineligible to receive distant network signals under the 1988 Act’s limited copyright licence, many of them were unable to receive satisfactory over-the-air signals from their local network affiliates and thus remained without any meaningful access to network programming. See id. at 3-6. In addition, administrative difficulties in deciding which households were genuinely unserved led to increasingly bitter disputes between satellite carriers and broadcasters, leaving bewildered, angry consumers stuck in the middle. See, e.g., ABC, Inc. v. PrimeTime 24 17 F.Supp.2d 478 (M.D.N.C.1998), aff'd, 184 F.3d 348 (4th Cir.1999). Progress in satellite technology seemed to hold out a solution to both of these problems. Satellite carriers had developed the channel capacity to offer local-into-local service to their subscribers in some markets. Congress recognized that satellite carriers that could offer viewers access to their local broadcast stations could compete effectively with cable and drive down the price of subscription television services. See H.R.Rep. No. 106-79, pt. 1, at 14 (1999). In addition, Congress understood that the difficulties in deciding which households were sufficiently un-served to receive distant network signals would be irrelevant if satellite subscribers could watch their local network affiliates. See id. In 1997 and 1998 Congress conducted extensive hearings on the possibility of extending the satellite carriers’ statutory copyright license to include local-into-local retransmissions. Representatives from the various industries had ample opportunities to state their views. Satellite industry representatives argued that without a statutory copyright license, the high transaction costs of obtaining the rights to retransmit broadcast programs would effectively prevent satellite carriers from offering local-into-local service. The satellite representatives further explained that their inability to carry local broadcast stations placed them at a competitive disadvantage to cable because viewers want to be able to receive all of the televisions channels they watch from a single source: “most people who walk into a satellite dealer’s showroom turn around and walk out because they can’t get their local TV channels through DBS.” Reauthorization of the Satellite Home Viewer Act: Hearing Before the Subcommittee on Telecommunications, Trade, and Consumer Protection of the House Committee on Commerce, 106th Cong. 73 (1999) (1999 House Hearing) (statement of David Moskowitz, an EchoStar executive). The cable representatives stressed the need for regulatory parity, arguing that if satellite carriers were granted a statutory license comparable to that enjoyed by cable, they should also be subject to the must-carry rules and other regulatory constraints imposed on cable. See The Copyright Office Report on Compulsory Licensing of Broadcast Signals: Hearing Before the Senate Committee on the Judiciary, 105th Cong. 50 (1997) (statement of Decker Anstrom, President, National Cable Television Association). Finally, representatives of the broadcast industry emphasized that mandatory carriage rules were needed to ensure that satellite local-into-loeal service would not undermine Congress’s interest in preserving a vibrant system of free, local broadcast television. They explained that without carriage rules satellite carriers would choose to retransmit only the signals of major network affiliates in most markets and that competing stations in those markets would suffer from the loss of access to a significant (and growing) part of their potential audiences: No rational doubt may exist that a local station denied access to a portion of its inmarket audience is injured. Lack of carriage reduces potential audience and, therefore, actual audience. Reduced audiences translate to reduced revenue. Even where revenue reductions are less than fatal, they still affect a station’s ability to provide the best practicable service to the public. At best, a local station which a satellite carrier refuses to carry would be placed at a demonstrable disadvantage vis-a-vis competing broadcast television stations which are carried. Copyright Licensing Regimes Covering Retransmission of Broadcast Signals (Part II): Hearing Before the Subcommittee on Courts and Intellectual Property of the House Committee on the Judiciary, 105th Cong. 68 n. 38 (1998) (1998 House Hearing) (statement of James J. Popham, Vice President and General Counsel, Association of Local Television Stations). Congress sought to balance all of these concerns when it enacted SHVIA on November 29, 1999. The Conference Report explained that the legislation was designed to promote competition between satellite and cable while “preserving] free television for those not served by satellite or cable systems.” SHVIA Conf. Rep. at 101. Two interrelated provisions form the heart of SHVIA. The first provision, codified at 17 U.S.C. § 122, amended the Copyright Act to create a statutory copyright license for satellite carriers similar to that enjoyed by cable operators. The license enables satellite carriers to make secondary transmissions of a broadcast station’s signal into that station’s local market without obtaining the authorization of those holding copyrights in the individual programs broadcast by that station. 17 U.S.C. § 122(a). Satellite carriers that make use of this license pay no royalties to program copyright holders. The second provision, codified at 47 U.S.C. § 338, amended the Communications Act of 1934 by creating mandatory carriage rules, including the carry one, carry all rule that is the focus of this litigation. 47 U.S.C. § 338(a)(1). Since SHVIA’s passage in late 1999 satellite carriers have enjoyed the benefits of the § 122 license without the burdens of the carry one, carry all rule. In order to phase in satellite carriage obligations, Congress decided that the rule would not become effective until January 1, 2002. Id. § 338(a)(3). As of that date the rule will require all satellite carriers who use the § 122 license to make local-into-local retransmissions in a local broadcast market to “carry upon request the signals of all television broadcast stations located within that local market.” Id. § 338(a)(1). In other words, the voluntary decision to carry one local station in a market under the statutory copyright license will trigger an obligation to carry all the requesting stations in that market. Like the cable must-carry rules, SHVIA’s carry one, carry all rule was designed to preserve a rich mix of broadcast outlets for consumers who do not (or cannot) pay for subscription television services. SHVIA adopts much of the machinery of the cable rules. For example, broadcast stations may seek carriage from satellite carriers under either § 338’s mandatory carriage requirements, 47 U.S.C. 325(b)(1)(C), or under the Act’s retransmission consent provision, 47 U.S.C. 325(b)(1)(A), which allows broadcast stations to receive compensation from satellite carriers in exchange for permission to retransmit the broadcast stations’ signals. In allowing broadcasters to choose between retransmission consent and mandatory carriage, both SHVIA’s carry one, carry all rule and the cable must-carry rules recognize the practical realities of the television marketplace. Major network affiliates have bargaining power and will be carried voluntarily in most instances, so they will normally elect to proceed under the retransmission consent provision. See Turner II, 520 U.S. at 191, 117 S.Ct. 1174 (explaining that the Cable Act’s retransmission consent provision reflects the “popularity and strength of some broadcasters”). In contrast, independent stations (affiliates of emerging networks, independent commercial stations, and noncommercial stations) may need the protection of must-carry rules and will normally elect mandatory carriage. The primary difference between § 338 and the cable must-carry rules is that § 338’s obligations are conditioned upon the satellite carrier’s voluntary choice to make use of the § 122 license in a particular television market, but the cable must-carry rules are mandatory in all markets. This difference reflects the technological dissimilarities between cable and satellite. Cable systems are local, and nearly all have enough channel capacity to carry all the broadcast stations in their local market and still provide an attractive mix of national and regional non-broadcast programming. The satellite carriers, in contrast, currently beam the same 450 to 500 channels throughout the continental United States and thus could not comply with a rule requiring them to retransmit the signals of each of the country’s roughly 1,600 local broadcast stations. See 1999 House Hearing at 95 (statement of David Mos-kowitz). While Congress recognized that satellite technology continues to improve and that eventually an unconditional must-carry rule might be possible, it still had to decide how to structure the satellite copyright license in the interim. Although minor variations were possible, Congress had two basic options. It could have decided to give satellite carriers a station-by-station copyright license that could be used free of carriage obligations. Under such a license, satellite carriers could “cherry pick” the stations they wanted in each local market. Congress predicted that in practical terms a station-by-station license would result in satellite carriers using their available local-into-local capacity to carry only the major network affiliates in as many markets as possible, beginning with the largest markets and working toward the smaller ones. The other choice was to impose the carry one, carry all rule in order to create a market-by-market copyright license. This kind of license forces satellite carriers to expand their local-into-local service one market at a time, beginning with the carriage of all requesting local stations in the largest markets and expanding from there. In sum, Congress either could have allowed satellite carriers to cherry pick by retransmitting some stations (the major network affiliates) in many markets, or it could have allowed satellite carriers to retransmit all of the stations in some markets. While satellite carriers would have preferred (and now argue for) the first result, Congress chose the second because it feared that cherry picking of major network affiliates within local markets would make it more difficult for non-carried stations in those markets to reach their audiences: Although the conferees expect that subscribers who receive no broadcast signals at all from their satellite service may install antennas or subscribe to cable service in addition to satellite service, the Conference Committee is less sanguine that subscribers who receive network signals and hundreds of other programming choices from their satellite carrier will undertake such trouble and expense to obtain over-the-air signals from independent broadcast stations. SHVIA Conf. Rep. at 102. • Non-carried stations in cherry-picked markets would “face the same loss of viewership Congress previously found with respect to cable non-carriage.” Id. at 101. Congress therefore concluded that the carry one, carry all rule would protect the ability of all local broadcasters to reach their audiences and thereby “preserve free television for those not served by satellite or cable systems and ... promote widespread dissemination of information from a multiplicity of sources.” Id. B. As directed by Congress, see 47 U.S.C. 338(g), the FCC issued a report and order implementing § 338’s carriage requirements on November 29, 2000. In the Matter of: Implementation of the Satellite Home Viewer Improvement Act of 1999: Broadcast Signal Carriage Issues, Retransmission Consent Issues, CS Docket Nos. 00-96 and 99-363 (2000) (SHVIA Order). The only aspect of the SHVIA Order at issue here is the a la carte rule adopted in paragraph 99 of the order. The FCC concluded in that paragraph that § 338(d) does not require satellite carriers to sell all local television stations in a given market as one package. Instead, the Commission ruled that carriers could offer local stations “as a package or a la carte, at comparable rates.” SHVIA Order at ¶ 99. This language would arguably permit satellite carriers to offer their subscribers a package including only the major network affiliates while offering independent broadcast stations only on an a la carte basis. However, in a subsequent order issued while these cases were under consideration, the FCC clarified its a la carte rule by stating that satellite carriers may not offer their subscribers a package including some subset of the local stations in a market (for example, the major network affiliates) while offering other local stations on an individual basis. See In the Matter of: Implementation of the Satellite Home Viewer Improvement Act of 1999: Broadcast Signal Carriage Issues, Order on Reconsideration, CS Docket No. 00-96, at ¶ 48 (2001) (Reconsideration Order). In other words, the Commission has now ruled that satellite carriers offering local-into-local service may offer their subscribers only two options: subscribers may buy the local stations of their choice a la carte at comparable prices, or they may buy a package of all the local stations for a price less than or equal to the cost of subscribing to all the stations individually (for example, 12 stations at one dollar each or all 12 for $12 or less). See id. C. The present SHVIA litigation began on September 20, 2000, when SBCA, DirecTV, EchoStar, and DISH Network filed a complaint in the United States District Court for the Eastern District of Virginia alleging that SHVIA’s carry one, carry all rule, 47 U.S.C. § 338(a)(1), exceeds Congress’s power under the Copyright Clause and violates the First Amendment and the Due Process and Takings Clauses of the Fifth Amendment. NAB and PBS intervened on the side of the FCC in order to defend the statute. The satellite carriers moved for summary judgment, but the FCC and its supporting intervenors filed Rule 56(f) motions requesting discovery in order to respond to the satellite carriers’ motion. The district court granted their motions, and discovery began. The FCC and its supporting intervenors then filed a Rule 12(b)(6) motion to dismiss and (ultimately) a cross-motion for summary judgment. After the parties had conducted extensive discovery in support of their summary judgment motions, the district court granted the FCC and its intervenors’ motion to dismiss on June 19, 2001. See Satellite Broad. & Communications Ass’n v. FCC, 146 F.Supp.2d 803, 809 (E.D.Va. 2001). The district court treated the carry one, carry all rule as a content-neutral regulation of the satellite carriers’ speech and upheld the rule under intermediate First Amendment scrutiny. In addition, the court rejected the satellite carriers’ arguments alleging violations of the Copyright Clause and the Fifth Amendment. The satellite carriers appealed. That appeal has been consolidated with three petitions for review of the FCC’s SHVIA Order. SBCA filed a petition for review of the SHVIA Order in this circuit, and Ech-oStar filed a' similar petition in the Tenth Circuit. Both petitions advanced constitutional challenges to SHVIA similar to those advanced by the satellite carriers in the Eastern District of Virginia. NAB filed a petition for review in the D.C. Circuit challenging the FCC’s a la carte rule as arbitrary, capricious, and not in accordance with law. See 5 U.S.C. § 706(2)(A). Paxson Communications intervened in this last petition on the side of NAB. The three petitions for review were consolidated and assigned to this circuit pursuant to 28 U.S.C. § 2112(a)(3). II. A. The parties devote the lion’s share of their attention to the satellite carriers’ First Amendment challenge to the carry one, carry all rule enacted in 47 U.S.C. § 338. Satellite carriers, like cable operators, function primarily as conduits for the speech of others. They transmit programming from a variety of sources (national and regional non-broadcast channels, superstations, and local broadcast stations) to their subscribers “on a continuous and unedited basis.” Turner I, 512 U.S. at 629, 114 S.Ct. 2445. Yet both satellite carriers and cable operators engage in speech protected by the First Amendment when they exercise editorial discretion over the menu of channels they offer to their subscribers. See id. at 636-37, 114 S.Ct. 2445. Insofar as the carry one, carry all rule seeks to influence the exercise of that discretion, it is open to challenge under the First Amendment. It is important to be clear at the outset, however, that some of the speech interests present in the Turner cases are absent here. The cable must-carry rules burdened the speech of cable programmers who had to be dropped from some cable systems in order to make room for local broadcast stations. See id. at 675, 114 S.Ct. 2445 (O’Connor, J., dissenting). To the extent the First Amendment guarantees viewers and listeners the right to receive information, see Bd. of Educ. Island Trees Union Free School Dist. v. Pico, 457 U.S. 853, 867, 102 S.Ct. 2799, 73 L.Ed.2d 435 (1982) (plurality opinion); Red Lion Broad. Co. v. FCC, 395 U.S. 367, 390, 89 S.Ct. 1794, 23 L.Ed.2d 371 (1969), that right was also burdened by the cable must-carry rules because the voices of dropped cable programmers were effectively lost to entire communities. Here, however, the satellite carriers do not claim that the carry one, carry all rule will force them to drop national or regional non-broadcast programmers in order to carry more local broadcasters. They claim only that they will have to carry independent broadcast stations in larger markets at the expense of major network affiliates in smaller markets. See Brief for Petitioners SBCA and EchoStar at 26-27, 35 n. 13. While those major network affiliates would no doubt prefer to be carried on satellite services, the carry one, carry all rule does not deprive those affiliates of the ability to reach their audiences. In fact, they will be in exactly the same position under the carry one, carry all rule that they would have occupied if SHVIA had never been passed. Nor will any voices be completely lost to communities as a result of the carry one, carry all rule. The rule’s only negative effect on viewers will be that the major network affiliates in some markets will be available to local viewers through only two mediums (broadcast and cable) rather than three. In sum, the carry one, carry all rule burdens speech only to the extent that it affects satellite carriers’ decisions about how to allocate their capacity for offering local-into-local service by inducing them to carry a different set of local broadcasters than the carriers would have preferred. In considering the satellite carriers’ First Amendment challenge, our first task is to determine the appropriate standard of review. Turner I provides the starting point. In that case the Supreme Court held that the cable must-carry rules were content-neutral restrictions on speech and were therefore not subject to strict scrutiny. See Turner I, 512 U.S. at 662, 114 S.Ct. 2445. Here, the satellite carriers repeat many of the same arguments for strict scrutiny that were examined and rejected in Turner I. Although the satellite carriers labor mightily to distinguish that case, their effort ultimately fails. The Supreme Court has said that the “ ‘principal inquiry in determining content neutrality ... is whether the government has adopted a regulation of speech because of [agreement or] disagreement with the message it conveys.’ ” Id. at 642, 114 S.Ct. 2445 (quoting Ward v. Rock Against Racism, 491 U.S. 781, 791, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989)). This inquiry involves two steps. “First, we must examine the plain terms of the regulation to see whether, on its face, the regulation confers benefits or imposes burdens based upon the content of the speech it regulates.” Chesapeake and Potomac Tel. Co. of Va. v. United States, 42 F.3d 181, 193 (4th Cir.1994), vacated on other grounds, 516 U.S. 415, 116 S.Ct. 1036, 134 L.Ed.2d 46 (1996). If it does not, we then ask whether the regulation’s “ ‘manifest purpose is to regulate speech because of the message it conveys.’ ” Id. (quoting Turner I, 512 U.S. at 645, 114 S.Ct. 2445). The text of § 338(a)(1) provides: [E]ach satellite carrier providing, under section 122 of Title 17, secondary transmissions to subscribers located within the local market of a television broadcast station of a primary transmission made by that station shall carry upon request the signals of all television broadcast stations located within that local market. 47 U.S.C. § 338(a)(1). On its face, the rule confers benefits on a content-neutral basis because satellite carriers must carry all requesting local broadcast stations regardless of the content of their programming. Cf. Turner I, 512 U.S. at 645, 114 S.Ct. 2445 (“The [must-carry] rules benefit all full power broadcasters who request carriage — be they commercial or noncommercial, independent or network affiliated, English or Spanish language, religious or secular.”). The satellite carriers argue, however, that the carry one, carry all rule is content-based on its face because its burdens are triggered by a satellite carrier’s decision to transmit certain content, namely, the signal of a local television station. We disagree. SHVIA’s carriage obligations are not triggered simply by the decision to carry a local broadcast station in a given market. Instead, they are triggered by the decision to carry that station by making use of the § 122 license. A satellite carrier that privately negotiates the required copyright clearances can retransmit the signal of a local broadcast station without incurring any carriage obligations, but a carrier that retransmits the same signal by means of the statutory copyright license must comply with the carry one, carry all rule. Thus, the burdens of the rule do not depend on a satellite carrier’s choice of content, but on its decision to transmit that content by using one set of economic arrangements rather than another. Accordingly, we hold that the carry one, carry all rule is content neutral on its face. Next, we must decide whether the carry one, carry all rule is content-based in its purpose. The satellite carriers claim that the rule has a content-based purpose because it seeks to promote the survival of independent broadcast stations, including affiliates of emerging networks, commercial independent stations, and public broadcasting stations. In support of this claim they cite language from the Conference Report indicating that Congress was concerned that satellite carriers would choose to carry only affiliates of the major networks and that other local broadcasters would be cut off from portions of their potential viewing audiences. See SHVIA Conf. Rep. at 102. In addition, the satellite carriers rely on legislative history suggesting that Congress sought to protect local broadcast stations because those stations provide valuable news and public affairs programming to their communities. See id. at 92. Again, we conclude that these points do not distinguish the present case from Turner I. Both Congress and the Supreme Court understood perfectly well that the cable must-carry rules primarily benefitted independent broadcast stations. See Turner I, 512 U.S. at 672-73, 114 S.Ct. 2445 (Stevens, J., concurring in part and concurring in the judgment); Turner II, 520 U.S. at 191-92, 117 S.Ct. 1174. Similarly, the statutory findings included in the Cable Act described broadcast television as “an important source of local news and public affairs programming, and other local broadcast services critical to an informed electorate,” Cable Act § 2(a)(ll), 106 Stat. at 1461, and stated that noncommercial television “provides educational and informational programming to the Nation’s citizens,” id. § 2(a)(8)(A), 106 Stat. at 1461. Nevertheless, the Court determined that neither must-carry’s benefits to independent broadcast stations nor Congress’s recognition of the distinctive value of local programming revealed a content-based purpose. Instead, the Court concluded, “Congress’ overriding objective in enacting must-carry was not to favor programming of a particular subject matter, viewpoint, or format, but rather to preserve access to free television programming for the 40 percent of Americans without cable.” Turner I, 512 U.S. at 646, 114 5.Ct. 2445. According to the Court, Congress’s findings did not show a content-based preference for broadcast television over cable television; they merely indicated that local broadcast television was intrinsically valuable and therefore worth preserving. Id. at 648, 114 S.Ct. 2445. In sum, Turner I held that the cable must-carry rules were content neutral in purpose even though the rules were meant to protect independent broadcast stations and even though Congress recognized that those stations provided valuable local programming. Here, the satellite carriers do no more than point to legislative history indicating that Congress had the same purpose in enacting SHVIA and that Congress continued to appreciate the distinctive value of local broadcast programming. This is not enough to establish that SHVIA has a content-based purpose. We conclude, then, that the carry one, carry all rule should not be subject to strict scrutiny. At most, it is a content-neutral measure that imposes incidental burdens on speech and is therefore subject to intermediate First Amendment scrutiny under United States v. O’Brien, 391 U.S. 367, 88 S.Ct. 1673, 20 L.Ed.2d 672 (1968). Because, as we explain below, the carry one, carry all rule passes constitutional muster under O’Brien, we need not address the FCC and its intervenors’ argument that the rule should be evaluated under a more lenient standard. B. O’Brien directs us to uphold a content-neutral regulation of speech if “it furthers an important or substantial governmental interest; if the governmental interest is unrelated to the suppression of free expression; and if the incidental restriction on alleged First Amendment freedoms is no greater than is essential to the furtherance of that interest.” O’Brien, 391 U.S. at 377, 88 S.Ct. 1673. Deciding whether an interest is important or substantial and whether that interest is unrelated to the suppression of free expression is usually simple enough, but this is only the beginning of the inquiry. As the Turner cases explain, the O’Brien standard gets its bite from a two-part analysis. First, we must determine whether the regulation materially advances an important or substantial interest by redressing past harms or preventing future ones. These harms must be “real, not merely conjectural,” and the regulation must “alleviate these harms in a direct and material way.” Turner J, 512 U.S. at 664, 114 S.Ct. 2445. If a regulation places even incidental burdens on speech without yielding some genuine benefit, it must be struck down. In making these evaluations, we must “accord substantial deference to the predictive judgments of Congress.” Id. at 666, 114 S.Ct. 2445. Our “sole obligation is to assure that, in formulating its judgments, Congress has drawn reasonable inferences based on substantial evidence.” Turner II, 520 U.S. at 195, 117 S.Ct. 1174 (internal quotation marks and citation omitted). If the regulation materially advances some important or substantial interest, we then proceed to the second part of the O’Brien analysis and ask whether the regulation is narrowly tailored to serve that interest. The O’Brien narrow tailoring requirement is less rigorous than that applied under strict scrutiny. The requirement is met “so long as the regulation promotes a substantial government interest that would be achieved less effectively absent the regulation.” Ward, 491 U.S. at 799, 109 S.Ct. 2746 (internal quotation marks and citation omitted). In other words, the regulation must not “burden substantially more speech than is necessary to further the government’s legitimate interests.” Id. In sum, the carry one, carry all rule satisfies the O’Brien standard if the rule materially advances at least one substantial government interest in a narrowly tailored manner. The satellite carriers claim that the rule is simply a way to promote the interests of some speakers (independent broadcasters in large markets) at the expense of others (major network affiliates in small and medium-sized markets). Predictably, they contend that this interest is not even legitimate, let alone substantial or important. We agree that the carry one, carry all rule is meant to preserve the ability of independent broadcasters to reach their local audiences, but Congress did not view that goal as an end in itself. Instead, Congress recognized that protecting independent broadcasters from the harmful effects of satellite cherry picking would further two substantial government interests. The first is the government’s interest in preserving a multiplicity of local broadcast outlets for over-the-air viewers, those who do not subscribe to satellite or cable service. The second is the government’s interest in preventing its grant of a statutory copyright license to satellite carriers from undermining competition in local markets for broadcast television advertising. Though these two interests are closely related because both would be threatened in the same manner without the carry one, carry all rule, they are distinct because the first involves harms to over-the-air viewers while the second involves harms to local advertisers and to independent broadcasters themselves. As we explain next in the first part of our O’Brien analysis, we find that both interests are materially advanced by the carry one, carry all rule. We will now address each interest in turn. 1. Congress enacted the carry one, carry all rule to “preserve free television for those not served by satellite or cable systems and to promote widespread dissemination of information from a multiplicity of sources.” SHVIA Conf. Rep. at 101. These interests are clearly substantial and unrelated to the suppression of free expression. See Turner I, 512 U.S. at 662-63, 114 S.Ct. 2445 (stating that the government’s interests in “preserving the benefits of free, over-the-air local broadcast television” and “promoting the widespread dissemination of information from a multiplicity of sources” were both important and unrelated to the suppression of free expression). In deciding whether § 338 materially advances these two interests, it is useful to characterize them more precisely. We therefore adopt the characterization suggested in Turner II, where the Supreme Court treated the two interests as a unified whole: “Congress has an independent interest in preserving a multiplicity of broadcasters to ensure that all households have access to information and entertainment on an equal footing with those who subscribe to cable.” Turner II, 520 U.S. at 194, 117 S.Ct. 1174; see also id. at 193, 117 S.Ct. 1174 (stating that Congress’s interest in “preserving the existing structure of the broadcast industry discloses a purpose to prevent any significant reduction in the multiplicity of broadcast programming sources available to noncable households” (internal quotation marks and citation omitted)); id. at 226, 117 S.Ct. 1174 (Breyer, J., concurring in part) (describing the Cable Act’s purpose as “providing] over-the-air viewers who lack cable with a rich mix of over-the-air programming” and “assuring] the over-the-air public access to a multiplicity of information sources” (internal quotation marks and citation omitted)). In short, the government interests served by the carry one, carry all rule are best characterized as a single interest in preserving a multiplicity of broadcast outlets for over-the-air viewers. This characterization accurately captures the concerns expressed by Congress when it enacted § 338. In explaining that a market-by-market license promotes both the interest in preserving broadcast television for over-the-air viewers and the interest in promoting widespread dissemination of information from multiple sources, Congress expressed its concern that “absent must-carry obligations, satellite carriers would carry the major network affiliates and few other signals. Non-carried stations would face the same loss of viewership Congress previously found with respect to cable noncarriage.” SHVIA Conf. Rep. at 101. In other words, Congress understood that the threat to over-the-air viewers was not the loss of broadcasting as a medium, but the loss of the independent stations needed to provide those viewers with a rich mix of broadcast programming from multiple sources. The proper question, then, is whether the government’s interest in preserving a multiplicity of broadcast outlets for over-the-air viewers is materially advanced by the carry one, carry all rule. The satellite carriers appear to concede that the rule would effectively combat any real threat to this interest, but the carriers argue that they do not pose a real threat. Congress, however, concluded that satellite carriers would threaten this interest in the near future if it gave them a statutory copyright license without imposing carriage rules. See SHVIA Conf. Rep. at 101. Our review of that judgment has both factual and legal aspects. We must decide whether Congress’s factual predictions about the consequences of enacting a station-by-station copyright license were supported by substantial evidence in the legislative record. See Turner II, 520 U.S. at 195-96, 117 S.Ct. 1174. We may also look to evidence outside the legislative record in order to confirm the reasonableness of Congress’s predictions. See id. at 196, 204, 117 S.Ct. 1174. In addition, we must determine whether the predicted set of consequences is legally sufficient to count as a real threat to the government’s interest in preserving a multiplicity of broadcast outlets for over-the-air viewers. Congress made extensive statutory findings in enacting the Cable Act. See Cable Act § 2(a), 106 Stat. at 1460-63. SHVIA’s legislative record is less extensive, but this does not affect our duty to accord substantial deference to Congress’s predictive judgments. “Congress is not obligated, when enacting its statutes, to make a record of the type that an administrative agency or court does to accommodate judicial review.” Turner I, 512 U.S. at 666, 114 S.Ct. 2445; see also Sable Communications of Cal., Inc. v. FCC, 492 U.S. 115, 133, 109 S.Ct. 2829, 106 L.Ed.2d 93 (1989) (Scalia, J., concurring) (“Neither due process nor the First Amendment requires legislation to be supported by committee reports, floor debates, or even consideration, but only by a vote.”). We can fairly conclude from the legislative record that Congress made the following three factual predictions: (1) that satellite carriers would soon capture a. significant part of the television programming delivery market, (2) that satellite carriers would continue to deny carriage to significant numbers of independent stations in markets where they chose to offer local-intoloeal service, and (3) that non-carried stations in those markets would be harmed by losing access to parts of their potential audiences. In sum, Congress predicted that if satellite carriers enjoyed a statutory copyright license but were not bound by carriage rules, significant numbers of independent broadcast stations would be harmed by losing access to significant portions of their audiences. As we explain below, each of Congress’s predictions was supported by substantial evidence. The harder question is the essentially legal one of whether the consequences Congress predicted constitute a real threat to the government’s interest in preserving a multiplicity of local broadcast outlets for over-the air viewers. We begin with the three factual predictions, addressing them one by one. Congress’s prediction that satellite carriers would become a significant force in the television programming delivery market was supported by substantial evidence. While Congress did not explicitly find that satellite carriers would attain any specific level of market penetration, it obviously projected that satellite carriers would attract enough viewers to put downward pressure on cable rates. Congress created the § 122 copyright license to produce this result, see SHVIA Conf. Rep. at 91-92; S.Rep. No. 106-51, at 1-3 (1999), and it cannot be unreasonable for Congress to proceed on the premise that its own legislation will be effective. Moreover, the rapid growth in satellite service since SHVIA’s enactment suggests that Congress’s prediction will soon come true. The number of DBS subscribers grew by 28 percent between June 1999 and June 2000. See FCC Seventh Annual Report at ¶ 61. In comparison, cable subscriptions grew at an annual rate of less than 1.5 percent. See id. at ¶ 65. DBS providers have attributed much of their growth to the § 122 license. See id. at ¶ 68; see also id. at ¶ 69 (stating that SBCA reported a 43 percent increase in the number of new DBS subscribers per month in the 6 months following SHVIA’s enactment). In its comments for the FCC’s Seventh Annual Report, SBCA predicted that DBS service might have as many as 16 million subscribers (over 15 percent of American television households) by the end of 2000. See id. at ¶ 65. In sum, Congress’s first prediction was entirely reasonable. Congress’s second prediction was that without § 338, satellite carriers would refuse to carry significant numbers of independent broadcast stations in markets where they offered local-into-local service. See SHVIA Conf. Rep. at 101. This prediction was supported by substantial evidence in the record before Congress. See, e.g., 1999 Senate Hearing at 10 (statement of Charles Ergen). In addition, the satellite industry’s track record under SHVIA has proven the prediction to be accurate. As of November 5, 2001, DirecTV carried the four major network affiliates in 41 markets (a total of 164 stations), but carried a total of only 13 independent stations in those 41 markets. See www.skyre-port. com/sky report/local, htm. Similarly, EchoStar carried the four major network affiliates in 35 markets (a total of 140 stations), but carried only 13 independent stations in those markets. See id. Thus, during SHVIA’s phase-in period the satellite carriers have confined their local-into-local offerings almost exclusively to major network affiliates. Moreover, they have plainly announced their intention to continue cherry picking the major network affiliates so long as they may lawfully do so. This preference for cherry picking will not simply wither away because it is rooted in the national character of existing satellite technology. As we have said, satellite carriers currently beam the same signals throughout the nation from satellites orbiting in full CONUS slots. As a result, a carrier who chooses to retransmit the signal of an independent commercial station in Boston beams that station’s signal to subscribers nationwide, but blocks out the signal at the homes of all subscribers who do not live in the Boston area. As this example illustrates, local-into-local service is a relatively inefficient use of a carrier’s finite channel capacity. National programming uses that capacity more efficiently because it can be expected to draw more viewers across the country as a whole than any local channel, even if the national programming is less popular within individual local markets than the programming of the independent broadcast stations in those markets. As a result, satellite carriers have economic incentives to favor national non-broadcast programming over local broadcast programming. While improvements in satellite technology, such as the deployment of spot beam satellites, may reduce the inefficiencies that result from carrying local broadcast stations, these inefficiencies will continue to exist to some degree so long as the beams from individual satellites cover multiple television markets. Yet satellite carriers also recognize that their product is less attractive to subscribers if they fail to carry the local affiliates of major networks because most people watch these stations about “65 percent of the time” they spend watching television. 1999 Senate Hearing at 7 (statement of Charles Ergen). Under these circumstances, the economically rational choice for satellite carriers is to cherry pick the major network affiliates in markets where the carriers offer local-into-local service. We conclude, then, that both the satellite industry’s track record and its economic incentives provide substantial evidence supporting Congress’s predictive judgment that satellite carriers will deny carriage to significant numbers of independent broadcast stations if the carriers are not bound by the carry one, carry all rule. Congress’s third prediction was that lack of satellite carriage would harm independent broadcast stations. See SHVIA Conf. Rep. at 101 (stating that “[n]on-carried stations would face the same loss of viewership Congress previously found with respect to cable noncarriage”). This general prediction rested on the following series of more specific predictions about the consequences of non-carriage. First, a non-carried station will effectively lose access to satellite subscribers if satellite carriers cherry pick the major network affiliates in its market. Second, a station’s loss of access to those subscribers will result in lower ratings, which will in turn lead to lower advertising revenues. Third, lower revenues will lead to a decrease in programming quality, which will then further depress ratings and station revenues, all in a vicious cycle. The record before Congress contains substantial evidence to support these predictions. See ante at 19 (statement of James J. Popham); Copyright Compulsory License Improvement Act: Hearings on H.R. 768 Before the Subcommittee on Courts and Intellectual Property of the House Committee on the Judiciary, 106th Cong. 47 (1999) (statement of John Hutchinson, CEO, Local Television on Satellite) (stating that “[s]at-ellite could effectively, like cable, become the bottleneck which disenfranchises stations”). Further, the predictions are essentially the same as those Congress made in passing the Cable Act. See Turner II, 520 U.S. at 208-09,117 S.Ct. 1174 (discussing testimony from broadcasters that explained how cable non-carriage leads to declines in ratings, advertising revenues, and program quality). As the Supreme Court explained in Nixon v. Shrink Missouri Government PAC, 528 U.S. 377, 391, 120 S.Ct. 897, 145 L.Ed.2d 886 (2000), the “quantum of empirical evidence needed to satisfy heightened judicial scrutiny of legislative judgments will vary up or down with the novelty and plausibility of the justification raised.” Because the justifications here were both familiar and plausible, Congress could reasonably conclude, at the very least, that non-carried stations would suffer some degree of harm in the absence of the carry one, carry all rule. The extent of that harm wo