Full opinion text
Statement for the Court filed PER CURIAM. Opinion for the Court filed by Circuit Judge GINSBURG. Opinion for the Court filed by Circuit Judge RANDOLPH. Opinion for the Court filed by Circuit Judge ROGERS. Opinion dissenting in part filed by Circuit Judge RANDOLPH. PER CURIAM. In these consolidated cases, various cable companies and municipalities petition for review of several orders of the Federal Communications Commission implementing the Cable Television Consumer Protection And Competition Act of 1992, Pub.L. No. 102-385, 106 Stat. 1460 (codified in scattered sections of 47 U.S.C.). We are issuing three separate opinions, each addressing a distinct category of issues. In the opinion for the court authored by Judge Ginsburg, we address what the parties call the “rate issues,” various challenges brought under the 1992 Cable Act and the Administrative Procedure Act to certain FCC decisions concerning the rates that regulated cable companies may charge. In the opinion for the court authored by Judge Randolph, we consider the claim of various cable companies that the FCC, in implementing the Cable Act, violated the First Amendment to the United States Constitution. Finally, in the opinion for the court authored by Judge Rogers we review what the parties call the “rules issues,” various claims made by cable companies and a group of cities concerning the scope of the FCC’s cable regulations and the role of local governments in regulating cable. GINSBURG, Circuit Judge: In addressing the “rate issues,” we consider challenges made by a group of cable companies, by a group of cities, and by Blade Communications, Inc., an individual cable company. Put simply, the cable petitioners argue that the FCC’s new ratemaking regime results in rates that are too low and that it should be set aside both because it violates the 1992 Cable Act and because it is arbitrary and capricious in violation of the Administrative Procedure Act. Blade Communications argues more specifically that the Commission’s rules improperly penalize it for having had rates that were lower than those charged by most cable systems prior to the imposition of controls. The cities argue that other aspects of the ratemaking regime run afoul both of the 1992 Cable Act and of the APA; generally, they ask us to set aside those parts of the FCC’s rules that they claim permit cable companies to charge unlawfully high rates. We hold that, with one exception, the Commission struck an appropriate balance between the competing interests of the cable companies and their subscribers, in violation neither of the 1992 Cable Act nor of the APA. The one exception is the Commission’s treatment of so-called gap-period external costs; on that issue, we grant the cable companies’ petition and vacate the rule. I. BACKGROUND Under the Cable Act of 1992, any cable system that does not face “effective competition,” as defined in the Act, is subject to rate regulation. 47 U.S.C. § 543(a)(2). The definition of effective competition includes three types of situations, to wit: (A) fewer than 30 percent of the households in the franchise area subscribe to the cable service of a cable system; (B) the franchise area is— (i) served by at least two unaffiliated multichannel video programming distributors each of which offers comparable video programming to at least 50 percent of the households in the franchise area; and (ii) the number of households subscribing to programming services offered by multichannel video programming distributors other than the largest multichannel video programming distributor exceeds 15 percent of the households in the franchise area; or (C) a multichannel video programming distributor operated by the franchising authority for that franchise area offers video programming to at least 50 percent of the households in that franchise area. 47 U.S.C. § 543(i)(l). Only a cable system that finds itself in one of those three situations, which the Commission calls respectively a “low penetration system,” an “overbuild,” and a “municipal system,” is exempt from rate regulation. 47 U.S.C. § 543(a)(2). The Act divides the cable services of a system that is subject to rate regulation into three categories: (1) the basic service tier; (2) cable programming service; and (3) video programming offered on a per channel or per program basis, which alone is not subject to rate regulation. 47 U.S.C. §§ 543(a)(1), (¿X2). The basic service tier includes local broadcast channels; those non-commercial public, educational, and government-access channels that the cable system is required by its franchise to carry; and such additional channels as the cable operator may in its discretion include in this tier. 47 U.S.C. § 543(b)(7). The Act provides that a cable subscriber must purchase the basic service tier in order to gain access to other service tiers, 47 U.S.C. § 543(b)(7), and instructs the Commission to establish regulations that “ensure that the rates for the basic service tier are reasonable[,]” and are “designed to achieve the goal of protecting subscribers ... from rates ... that exceed the rates that would be charged for the basic service tier if such cable system were subject to effective competition.” 47 U.S.C. § 543(b)(1). Each local franchising authority that has been certified by the FCC may enforce the FCC’s basic service tier rate regulations within its franchise area. 47 U.S.C. §§ 543(a)(2)-(6). Cable programming service includes all cable channels that are neither part of the cable system’s basic tier offering nor offered on a per channel or per program basis. 47 U.S.C. § 543(Z)(2). The Act charges the Commission (rather than local franchising authorities) with enforcement of the rate regulations for cable programming service, 47 U.S.C. § 543(a)(2)(B); the Commission must establish criteria to identify and create procedures for lowering any “unreasonable” rate for cable programming service. 47 U.S.C. § 543(c)(1). FCC review of rates for cable programming service is triggered on a case-by-case basis when a subscriber, franchising authority, or other relevant State or local governmental entity files a complaint. .47 U.S.C. §§ 543(c)(1)(B), (c)(3). In implementing the Act, the FCC promulgated rules for determining the highest rate that each regulated cable system could charge initially, as well as rules for calculating allowable rates on a “going-forward” basis. The Commission further decided to apply both sets of rules in a “tier-neutral” fashion, meaning that the same methodology is used to set rates for the basic service tier and for cable programming service. Implementation of Sections of the Cable Television Consumer Protection and Competition Act of 1992: Rate Regulation, Report and Order and Further Notice of Proposed Rulemak-ing, 8 F.C.C.R. 5631, 5759-60, 5881-82 (1993) (hereinafter “Rate Order”); see also Implementation of Sections of the Cable Television Consumer Protection and Competition Act of 1992: Rate Regulation, First Order on Reconsideration, Second Report and Order, and Third Notice of Proposed Rulemaking, 9 F.C.C.R. 1164, 1182-85 (1993) (hereinafter “First Reconsideration”). The tier-neutral approach is designed to ensure that a cable system has no incentive to move programming between the basic service and cable programming service tiers: the incremental charge it can make for a particular channel will be the same regardless whether the channel is placed in the basic service tier or in a cable programming service tier. See First Reconsideration, 9 F.C.C.R. at 1183. II. Analysis The cable petitioners challenge the FCC regulation with respect to: (1) the methods the Commission used to arrive at the initial rates that a cable system may charge; (2) the rules it prescribed concerning the amount a system may charge for equipment; (3) its decision to regulate the basic service tier and cable programming service on a “tier-neutral” basis; and (4) its treatment of costs that the cable companies incurred during the “gap period” between passage of the Act and the date upon which each, cable operator became subject to rate regulation under the Act. The cities in turn challenge various aspects of: (1) the rules for setting initial rates; and (2) the “going-forward” rules. Insofar as the various challenges involve the FCC’s interpretation of the 1992 Cable Act, we apply the rule of Chevron U.S.A. Inc. v. NRDC, 467 U.S. 837, 842-43, 104 S.Ct. 2778, 2781-82, 81 L.Ed.2d 694 (1984): If the statute is clear, then that is the end of the matter, for we “must give effect to the unambiguously expressed intent of Congress.” If, however, the statute is silent or ambiguous with regard to a specific issue, then we must accept the interpretation of the agency so long as it is reasonable. Id. at 843, 104 S.Ct. at 2781. As discussed in the opinion for the court addressing the First Amendment aspects of this case, no “grave constitutional question” is implicated by the FCC’s interpretation of the statute, and hence we need not consider whether Chevron deference would apply if such a question were to arise. Insofar as the various challenges amount to a claim that the Commission behaved arbitrarily and capriciously, we review the record with an eye to whether the agency has “examine[d] the relevant data and artieulate[d] a rational connection between the facts found and the choice made.” Motor Vehicle Manufacturer’s Ass’n of the United States v. State Farm Mutual Automobile Ins. Co., 463 U.S. 29, 43, 103 S.Ct. 2856, 2866, 77 L.Ed.2d 443 (1983). How taut is our “surveillance of the rationality of agency decisionmaking, however, depends upon the nature of the task assigned to the agency.” National Cable Television Ass’n v. Copyright Royalty Tribunal, 724 F.2d 176, 181 (D.C.Cir.1983). Because agency ratemaking is far from an exact science and involves “policy determinations in which the agency is acknowledged to have expertise,” our review thereof is particularly deferential. United States v. FCC, 707 F.2d 610, 618 (D.C.Cir.1983); see also Northern States Power Co. v. FERC, 30 F.3d 177, 180 (D.C.Cir.1994). We shall first address both groups of petitioners’ challenges to the rules for setting initial rates, then take up both groups’ objections to the going-forward rules. We shall turn last to the other issues raised by the cable petitioners. A. The Rules For Setting Initial Rates After considering a number of alternative ways of determining whether basic service tier rates are reasonable, the Commission decided to rely primarily upon the yardstick provided by systems that face effective competition. Rate Order, 8 F.C.C.R. at 5761-67. To this end, the Commission gathered rate (and other) information from 141 cable systems that satisfy the statutory standard for effective competition, as well as from a random sample of approximately 300 cable systems that do not. Rate Order, 8 F.C.C.R. at 5761. Using multiple regression analysis, the FCC originally isolated three factors other than competition that affect a system’s rates: the number of channels it offers, the number of subscribers it has, and the number of signals it receives by satellite. Id. at 5768-69, 6143-44. Controlling for those factors, the Commission compared the rates charged by the non-competitive and the competitive systems and found that there was a 10 percent “competitive differential,” meaning that on average, non-competitive systems charged about 10 percent more than similar systems that faced effective competition. Id. at 5766, 6145. Using the same data, the Commission also developed a benchmark formula for calculating the per channel rate that a cable system subject to effective competition would charge, taking into account the number of channels, of subscribers, and of satellite-delivered signals provided by that system. Id. at 5770-71. Originally the FCC decided that systems not facing effective competition would be required to use the benchmark formula to set their initial rates, except that no system would be required to reduce its existing rates by more than 10 percent. Id. at 5771-72. Any system that was at or below its benchmark rate would not have to reduce its rates at all, id., and any system could opt out of the regime entirely by requesting that its rates be set by means of conventional cost-of-service regulation. Id. at 5797-5800. Although the Commission, upon reconsideration, first refused to alter these rules, First Reconsideration, 9 F.C.C.R. at 1173-77, it later agreed to make substantial changes. See Implementation of Sections of the Cable Television Consumer Protection and Competition Act of 1992: Rate Regulation, Second Order on Reconsideration, Fourth Report and Order, and Fifth Notice of Proposed Rulemaking, 9 F.C.C.R. 4119 (1994) (hereinafter “Second Reconsideration”). The Commission re-analyzed the data concerning the aggregate difference between the rates charged by systems that do and systems that do not face effective competition, and determined that the competitive differential was actually 17 percent rather than 10 percent. Id. Most of the increase traces to two refinements in the Commission’s methodology. First, in a further multiple regression analysis, the FCC isolated and controlled for certain additional factors that affect a system’s rates. Id. at 4155-59. More significant, however, is the change in the way the Commission weighted the rate data. In the Rate Order, 8 F.C.C.R. at 5644, the Commission had arrived at the 10 percent competitive differential by aggregating the data for systems facing effective competition (as defined) and comparing the result to the rate data for systems that do not face effective competition. Because 79 of the systems in the group facing effective competition group were low penetration systems (compared to 46 overbuilds and 16 municipals), this approach gave the greatest weight to the data for low penetration systems and the least weight to the data for municipal systems. In the Second Reconsideration, 9 F.C.C.R. at 4153-55, the Commission disaggregated the data for the three types of systems facing effective competition and calculated a competitive differential for each group. This approach yielded competitive differentials of approximately one percent for low penetration systems, 13 percent for overbuilds, and 37 percent for municipal systems. Id. at 4160, 4296. For various reasons discussed below, the FCC decided that the overbuild sample provides the best indicator of the effects of competition upon rates. Id. at 4160-66. The FCC adjusted the overbuild figure from 13 percent to 16 percent, however, by factoring in the percentage of each system in that group that is actually overbuilt by a competing system. Id. at 4162. The Commission then used the competitive differentials for low penetration and municipal systems only as factors in deciding to raise the overall differential further to 17 percent, id. at 4165-66, of which also more below. Having established the new 17 percent competitive differential, the FCC further decided that all systems not facing effective competition must reduce their initial rates by 17 percent. Id. at 4166-68. The Commission did, however, provide some important qualifications. First, as in the old regime, a cable system can avoid the automatic (now 17 percent) reduction by opting for cost-of-service regulation. Id. at 4168. Second, if a 17 percent reduction would put a particular cable system’s rates below the rate allowable under the benchmark formula, then the system is required to reduce its rates only to the benchmark level until such time as the Commission has confirmed the accuracy of the 17 percent competitive differential by gathering and analyzing industry cost data. Id. at 4168-69. Similarly, “small operators,” defined by the Commission as those that have 15,000 or fewer subscribers and that are not affiliated with a larger operator, are not required to implement initial rate reductions at all until the Commission completes its analysis of industry cost data. Id. 1. The Cable Companies’ Petitions The cable petitioners advance a wide array of arguments challenging the methods by which the Commission arrived at the 17 percent competitive differential; they also challenge its decision to require all regulated systems (with the exceptions mentioned above) to institute the 17 percent rate reduction. Their arguments, though impressive in scope, are ultimately unconvincing. a. Low Penetration Data The cable petitioners first contend that the Commission violated the Act by failing to assign “proportionate weight” to the data for low penetration systems in establishing the competitive differential that forms the basis for the initial rate reductions the agency required. Recall that the Act provides that “regulations [to ensure the reasonableness of basic rates] shall be designed to achieve the goal of protecting subscribers ... from rates ... that exceed the rates that would be charged for the basic service tier if [the] cable system were subject to effective competition.” 47 U.S.C. § 543(b)(1). The Act further states that “in prescribing such regulations, the Commission ... shall take into account” seven listed factors, one of which is “the rates for cable systems, if any, that are subject to effective competition....” 47 U.S.C. § 543(b)(2). These provisions may well require the Commission, in establishing “reasonable” initial rates, to consider information about systems facing effective competition, which is, of course, precisely what the FCC has done. They do not, however, even suggest how the Commission should weigh the rate data from each subcategory of systems facing effective competition. Indeed, the text of the Act and its legislative history do not even provide the Commission with any guidance about how to weigh the seven factors that it is supposed to take into account, let alone how to weigh subeategories of data relevant to one particular factor. The only other direction that the Act gives the FCC for the establishment of rates for the basic tier is that it must “seek to reduce the administrative burdens on subscribers, cable operators, franchising authorities, and the Commission,” to which end it “may adopt formulas or other mechanisms and procedures.” 47 U.S.C. § 543(b)(2)(A), (B). The Congress thus refrained from micromanaging the Commission in the way that the cable petitioners now ask the court to do. In the absence of any statutory requirement that could be read to require the Commission to give “proportionate weight” to the rates charged by low penetration systems, however, we are not at liberty to oblige the petitioners. Perhaps anticipating the futility of the “proportionate weight” argument, the cable companies argue in the alternative that the Act requires the Commission to assign at least some weight to the low penetration data, and that the Commission failed to do so. This claim is factually incorrect. Although the 37 percent competitive differential for municipal systems caused the Commission to believe that the 16 percent differential for overbuilds was too low, the Commission “discounted [the municipal system data] somewhat ... on account of [its] consideration of low penetration systems, which had only a one percent competitive differential.” Second Reconsideration) 9 F.C.C.R. at 4166, 4195. Thus we see that the Commission did give some weight to the data for low penetration systems. The cable petitioners further contend that even if the Commission’s treatment of the low penetration data did not violate the Act, it was nonetheless arbitrary and capricious for the agency not to have given the data greater weight. The Commission, however, articulated a powerful economic rationale for according only minimal weight to those data. Under the statute, a cable system falls into the low penetration category if it serves less than 30 percent of the homes in its franchise area, regardless of its penetration rate for the subset of homes that it actually passes. 47 U.S.C. § 543(Z )(1)(A). The Commission noted that both its own data and an industry study suggest that a substantial number of the systems in the low penetration group serve more than 30 percent of the homes they pass. Second Reconsideration, 9 F.C.C.R. at 4162. The Commission reasonably concluded, therefore, that low penetration may reflect only the geographic limitations of the system rather than the presence of substitutes that restrain the cable operator from exercising market power; thus the low penetration group may well include systems that, due to market power, are able to charge rates substantially above the competitive equilibrium point. The FCC could not verify this possibility without collecting and analyzing extensive data concerning the low penetration group’s costs, but this was not realistically possible within the 180-day statutory deadline for the FCC to promulgate regulations. 47 U.S.C. § 543(b)(2); see National Ass’n of Regulatory Utility Comm’rs v. FCC, 737 F.2d 1095, 1124, 1138-42 (D.C.Cir.1984) (accepting agency ratemaking decision based upon agency’s expertise and best available information despite agency’s failure to amass additional useful data). Bearing in mind that one of the Congress’s declared purposes in enacting the Cable Act was to eliminate the effects of undue market power, see § 2(b)(5), we conclude that the FCC’s decision to give the data for low penetration systems only limited weight was reasonable. b. Overbuild Data The cable petitioners next challenge the FCC’s decisions to: (1) give the greatest weight to the overbuild data; (2) adjust the overbuild differential from 13 percent to 16 percent; and (3) adjust the final competitive differential further from 16 percent to 17 percent. The cable petitioners’ statutory challenge to the Commission’s decision to give the greatest weight to the overbuild data is based upon the same premise as its statutory challenge to the agency’s decision to give little weight to the low penetration data, and therefore fails for the reasons discussed in the previous section. Therefore it remains for us to decide only whether the Commission’s action was arbitrary and capricious. The FCC reasoned that because overbuilds face actual head-to-head competition, they provide the most accurate data for the purpose of simulating competitive cable rates, Second Reconsideration, 9 F.C.C.R. at 4612, a proposition that seems at first glance to be nearly self-evident. The cable petitioners point out, however, that the competitive differential for overbuilt systems falls over time. That trend, they argue, suggests that newly overbuilt cable systems engage in “price wars”: Either the incumbent system charges below-cost rates designed to drive the entering competitor out of the market or the new entrant charges below-cost rates in order to “greenmail” the incumbent into offering to buy it out upon favorable terms. Accordingly, the companies suggest that the overbuild sample reflects artificially low rates and, correspondingly, an artificially large differential from non-competitive systems. In the Second Reconsideration, 9 F.C.C.R. at 4163-64, the Commission offered its own hypothesis that the diminution in the competitive differential over time could be due to the emergence of “parallel or coordinated pricing.” Id. at 4163-64. According to that theory, competing cable companies learn over time how to collude in, or tacitly to coordinate, their pricing and therefore exercise greater market power than would be possible if they were truly competing. As the Commission suggested, this theory gains support from the fact that there are typically only two systems in any overbuilt area, which makes collusion or tacit coordination more plausible than it otherwise would be. Moreover, because information about rates is readily available and cable companies do not enter into long-term contracts with their subscribers, each duopolist would be able to detect and to retaliate against the other’s slightest departure from the rate upon which the two had expressly or impliedly agreed. Id.; see also William J. Baumol & Alan S. Binder, Economics: Principles and Policy 599 (5th ed. 1991) (ability to offer secret discounts undermines ability of oligopoly to cartelize). The FCC’s explanation therefore suggests that in light of the structure of the local cable market, it may be only in the early stages of direct head-to-head competition that overbuilt systems actually charge competitive rates. Although the Commission’s explanation cannot be proven without additional data, and although “[a] theory of ratemaking must be reasonable, explained, and supported,” it “is not subject to the same substantiation principle as the substantial evidence test applicable to factfinding.” National Ass’n of Greeting Card Publishers v. United States, 607 F.2d 392, 401 (D.C.Cir.1979) (quoting Continental Airlines Inc. v. CAB, 551 F.2d 1293, 1301 (D.C.Cir.1977)). The Commission’s theory is not so implausible that reliance upon it is unreasonable, especially when one considers that there is no evidence in the record, either anecdotal or analytical, to provide empirical support for the cable companies’ price war hypothesis. Therefore we reject the cable petitioners’ claim that the Commission’s decision to rely primarily upon the overbuild data is arbitrary and capricious. As mentioned above, the FCC adjusted the overbuild figure from 13 to 16 percent by factoring in the percentage overlap between each pair of overbuilt systems in that group. Second Reconsideration, 9 F.C.C.R. at 4162, 4284-85. The Commission hypothesized that not all overbuilds are equal because the intensity of the competition that an overbuilt system faces is likely to vary with the extent to which it actually overlaps with a competing system. Id. at 4284. That assumption seems completely reasonable, and the cable petitioners do not take issue with it as a theoretical matter. They argue, however, that the Commission unreasonably assumed that the intensity of competition is directly proportional to the percentage of overlap, and, where it lacked adequate data, unreasonably assumed that the percentage of overlap was the least possible given the percentage of the total franchise area covered by the respective systems. With regard to the first objection, we simply note that the Commission’s assumption of a linear relationship between the two variables certainly was reasonable, if only because it would have been virtually impossible to derive a more precise understanding of the relationship between the extent of overlap and the intensity of competition in the time available. See NARUC, 737 F.2d at 1124 (The “scope of agency expertise is often pragmatically circumscribed ... by the need to respond to ... regulatory problems ... within a reasonable period of time”). As for the second objection, while it would have been possible to determine the actual amount of overlap between each pair of competing systems for which it did not already have that datum, the Commission’s decision not to do so was a logical one; a new competitor, before attempting to compete head-to-head with the incumbent, typically will lay cable in areas of the franchise to which the existing cable system does not already provide service. Especially in light of the time constraint the Commission faced and the difficulty of gathering additional information, we see nothing unreasonable in its making that simplifying assumption. The cable companies’ challenge to the FCC’s adjusting the overbuild differential of 16 percent to the final competitive differential of 17 percent fares no better. They argue that this adjustment is improperly based upon the Commission’s assumption that parallel or coordinated pricing between overbuilt systems resulted in an artificially small overbuild differential. We have already held that the Commission’s reliance upon the parallel or coordinated pricing theory is not arbitrary and capricious. Moreover, the Commission did not rely solely upon that theory in reaching its decision to adopt the 17 percent figure; in the Second Reconsideration, 9 F.C.C.R. at 4166, the Commission made it clear that it also relied upon the data for municipal systems (which indicated a much larger differential) and the availability of the cost-of-service ratemaking alternative for any system the rates of which would otherwise be unduly reduced. Cumulatively, those factors adequately support the Commission’s decision to adjust the competitive differential upwards by the additional one percent. c. Large System Data The cable petitioners note that when the rate data gathered by the Commission are divided between large and small systems, the line of demarcation being set at 5,000 subscribers, the competitive differential for large systems is statistically insignificant. Based upon that observation, they conclude that large systems that do not face effective competition (as defined in the Act) do not exercise market power (i.e., do not charge supracompetitive rates), and thus the Commission’s decision to apply the 17 percent reduction of initial rates to large systems was arbitrary and capricious. When the cable petitioners made the same argument to the Commission they did not offer any explanation of why large systems that do not face effective competition either cannot or for some reason do not exercise market power to raise their rates. In their brief before this court, however, the cable companies suggest that large systems are generally located in areas that have more entertainment alternatives (e.g., broadcast television channels, video stores, cinemas, and live entertainment, including sporting events) and that these substitutes deprive the cable companies of market power. The FCC responds to the cable companies’ theory by noting that it has already analyzed one of those substitutes — additional broadcast television channels — and found no evidence that it limits the market power of cable companies. Although it is, of course, possible that the other proffered substitutes do limit cable’s market power, the finding concerning broadcast television — which intuitively seems to be the closest substitute for cable television — takes much of the wind out of the cable petitioners’ sails. More important, however, the FCC’s response when originally faced with the issue in the course of rulemaking demonstrates that its decision not to treat large systems differently was not arbitrary and capricious. Concerned that the cable companies’ approach was “statistically risky” because it involved subdividing the already small sample of systems facing effective competition into still-smaller sub-samples, Second Reconsideration, 9 F.C.C.R. at 4160, the Commission analyzed the data for systems of all sizes in an effort to discern the relationship (if any) between system size and rates. Id. at 4159-4160,4301-03. The results lent no support to the cable companies’ contention that large systems without the constraint imposed by “effective competition” charge rates nearer to the competitive level than do small systems. Id. Moreover, the separate statistical analysis that the FCC did to take account of the percentage overlap between overbuilt systems — which, as we have seen, the Commission reasonably believes gives a more accurate measure of the competitiveness of a cable market — showed that the effect of system size on the competitive differential is not significant. Id. at 4159-60. In light of both the Commission’s statistical analyses and the failure of the cable petitioners to provide any support (beyond the comparison of systems on either side of the 5,000 subscriber mark) for their new theory, we cannot conclude that the FCC was unjustified in applying the 17 percent competitive differential to both large and small systems. d. Application of the 17 Percent Initial Rate Reduction to All Systems Not Facing Effective Competition The cable petitioners (including Blade Communications) attack the 17 percent initial rate reduction as, in effect, too blunt an instrument. Specifically, they argue that it is arbitrary and capricious because it falls equally upon all regulated systems, without regard to whether and by how much a particular system’s past rates exceeded the amount it would have charged had it been subject to effective competition. To impose an across-the-board 17 percent rate reduction upon all regulated systems might indeed force an historically low-priced system to lower its rates below the competitive level, but that is not what the Commission has done here. Faced with the statutory command to avoid placing an undue administrative burden upon franchising authorities, cable operators, subscribers, and itself, 47 U.S.C. § 543(b)(2)(A), and armed with express statutory permission to adopt formulas in order to meet this requirement, 47 U.S.C. § 543(b)(2)(B), the Commission established not only the general 17 percent rule but also a number of important exceptions thereto. Specifically because it recognized the possibility that some low-priced systems may not have exercised significant market power to raise past rates, the Commission accorded those systems (as well as unaffiliated systems with 15,000 or fewer subscribers) “transition relief’ from the 17 percent rule. Second Reconsideration, 9 F.C.C.R. at 4167-69, 4172-82. Low-priced systems — defined by the Commission as systems the rates of which would be below their revised benchmark rates if the full 17 percent reduction were imposed — are required to reduce their rates only to their revised benchmark level until the Commission completes an analysis (still on-going) of whether such systems face “unusual demand, cost or other influences” that would render the 17 percent reduction excessive. Id. at 4168-69, 4176-79. Presumably any system that kept rates near the competitive level, notwithstanding the absence of “effective competition,” did so not out of charity but because it faced an “unusual [elasticity of] demand,” meaning that consumers in its area were more inclined than consumers in the average market to forego subscribing to cable if rates were higher. We fully expect the Commission’s current study to address the plight in which Blade claims to find itself. Recognizing that the 17 percent reduction could be excessive for still other systems— those facing high costs rather than high elasticity of demand — the Commission also adopted the cost-of-service “safety valve,” whereby any system for which the 17 percent reduction would result in unreasonably low rates can instead opt to have its rates set on the basis of its individual costs. Second Reconsideration, 9 F.C.C.R. at 4195-97; Rate Order, 8 F.C.C.R. at 5797-5800. By thus establishing an easily applied general rule along with well-targeted exceptions, the FCC effectively balanced its twin responsibilities of ensuring reasonable rates and reducing administrative burdens. In sum, Blade’s concern with the plight of low-priced systems, though not fanciful, has been adequately addressed by the Commission. So too has the Commission addressed, via the cost-of-service safety-valve, the concerns of systems that face unusually high input costs. 2. The Cities’ Petition The cities attack the Commission’s initial-rate rules from a different perspective, arguing that the rates allowed by the Commission are too high to accomplish the purposes of the Act. They challenge both the 17 percent figure itself and several decisions that the Commission made in applying it. Their direct challenge to the 17 percent figure need not detain us long in light of our extensive discussion of that figure above and the cursory nature of the cities’ argument. Citing studies that apparently suggest that cable rates exceed competitive levels by more than 17 percent, the cities contend in a single sentence in their main brief that “[t]he FCC had ample evidence before it suggesting that, if anything, the differential [the FCC] identified was still far too low.” The slightly different question before us, however, is whether the Commission had adequate evidence for the conclusions it reached. It relied upon its own analysis of the cable industry rather than the studies of others, but that in itself hardly renders its decision improper. The Commission gathered extensive industry rate data both from systems that do and from systems that do not face effective competition and did a variety of regression analyses designed to control for variables other than competition that could affect rates. The agency brought its expertise and experience to bear in deciding how to weigh the various data sets (i.e., overbuild, low penetration, and municipal systems), ultimately arriving as we have seen at the 17 percent differential. The cities’ eonelusory assertion gives us no reason to believe that the FCC’s approach was any less reliable than the outside studies to which the cities point without elaborating upon the methodologies, assumptions, or data used in them. In sum, the Commission has supported its 17 percent rule with substantial evidence and has articulated a rational connection between the facts that it found upon the basis of that evidence and its ultimate decision to adopt the 17 percent initial rate reduction; meanwhile the cities have failed to provide convincing evidence that the Commission erred. The cities next offer two challenges to the particular way in which the FCC implemented the 17 percent rule. First, they argue that the eost-of-serviee alternative that the Commission made available improperly favors cable operators: an operator may opt for cost-of-service regulation if it believes that its rates would otherwise be too low, but the local franchising authority may not subject the operator to that method of rate-setting if it believes that cable rates would otherwise be too high. This argument proceeds from a misunderstanding of the statutory mandate and the asymmetrical constitutional imperative that the FCC faces. The Commission concluded that the cost-of-service option should be made available to cable operators as a limited “safety-valve” for unusual cases in which the operator “would be harmed by applying the [17 percent rate reduction].” Second Reconsideration, 9 F.C.C.R. at 4166; accord Rate Order, 8 F.C.C.R. at 5755-56. That conclusion seems wise in light of the distinct possibility that an unexceptioned rate reduction could unconstitutionally yield confiscatory rates for cable systems that have not exercised market power significantly to raise rates in the past. See Southern Bell Tel. & Tel. Co. v. FCC, 781 F.2d 209, 214 (D.C.Cir.1986) (“rates obviously do not fall within a zone of reasonableness if they are so low as to be constitutionally confiscatory”). Moreover, because a cost-of-service regulatory proceeding is expensive for the cable operator, see Rate Order, 8 F.C.C.R. at 5755, the FCC can be confident that an operator will not lightly choose that option and it will indeed remain a limited exception to the general rule. By imposing a broad 17 percent reduction while allowing a limited eost-of-service safety-valve, the Commission effectively balanced the statutory goal of reducing administrative burdens and the constitutional necessity of avoiding confiscatory rates. This careful balance likely would be upset if local franchising authorities were allowed to impose cost-of-service regulation. In view of the comments filed by the cities during the proceedings under review, see Rate Order, 8 F.C.C.R. at 5754, it appears that a significant number of municipalities would impose cost-of-service ratemaking upon cable operators if authorized to do so, and that the statutory goal of administrative efficiency would be severely compromised thereby. Nor could such a loss of administrative efficiency be justified, as it would be in the current scheme, as necessary to prevent an unconstitutional outcome. Although we stop short of concluding that allowing franchising authorities to impose cost-of-service regulation would violate the Act (a question we need not decide here), we do conclude that the FCC’s decision not to give them that authority was reasonable in light of the Commission’s statutory mandate to reduce administrative burdens. The cities’ other challenge to the Commission’s implementation of the 17 percent rule goes to the transition relief it afforded small and low-priced systems, which they say frustrates the purpose of the Act. The cities argue that because the Commission’s transition relief rules allow a large percentage of all cable companies to avoid reducing their rates by the full 17 percent, the FCC has failed to meet its statutory mandate to ensure reasonable rates. As discussed above with regard to the cable petitioners’ challenges, the FCC promulgated the general 17 percent rule and the exceptions for transition as a package that was reasonably designed to meet the Act’s competing goals of administrative efficiency and reasonable rates. To that discussion, we add here only the observation that the transition relief is in fact transitory; the FCC proposes ultimately to apply the 17 percent rule to small and low-priced systems unless its further study produces evidence to support a different rate. Second Reconsideration, 9 F.C.C.R. at 4173, 4176-77. Although the cities raise the possibility that completion of that study may yet be far off, we need not address that concern now; no one argues that the time thus far elapsed is at all excessive in light of the complexity of the Commission’s task. B. The Rules for Setting “Going-Forward” Rates Having established the methods for setting the rate that a regulated cable system may charge initially (the 17 percent rate-reduction and the cost-of-service alternative), the Commission adopted a price cap regime (the so-called “going-forward” rules) for regulating future rates. See generally, Second Reconsideration, 9 F.C.C.R. at 4200-07; Rate Order, 8 F.C.C.R. at 5774-95. Under that regime, a cable system may adjust its initial per channel rate annually in order to reflect inflation and quarterly in order to reflect changes in certain of its costs that the Commission considers to be effectively beyond the cable operator’s control. Second Reconsideration, 9 F.C.C.R. at 4200-04; Rate Order, 8 F.C.C.R. at 5782-83. These so-called “external costs” are: (1) the retransmission consent fees cable operators pay to broadcasters; (2) programming costs; (3) taxes; and (4) franchise fees and the costs associated with other franchise requirements, including the provision of public, educational, and governmental-access programming. Second Reconsideration, 9 F.C.C.R. at 4201-02; Rate Order, 8 F.C.C.R. at 5783-90. The Commission decided to use a price cap regime because it would be less costly to administer than traditional cost-of-service regulation and would have the added advantage of providing operators with an incentive to be efficient — the lack of which is a notorious drawback of cost-of-service regulation— while still ensuring that rates remain reasonable. Rate Order, 8 F.C.C.R. at 5776-77; cf. National Rural Telecom Ass’n v. FCC, 988 F.2d 174, 177-78 (D.C.Cir.1993) (describing incentive to be efficient under price cap regulation). The cities advance an array of arguments that these rules are arbitrary and capricious and inconsistent with the 1992 Cable Act. The cable petitioners challenge only one specific provision as arbitrary and capricious. 1. The Cities’ Petition First. At the outset, the cities suggest that because the price cap formula is keyed to the rates that regulated systems are allowed to charge initially, it will only amplify the unreasonableness of those rates. Of course, that general contention depends upon the success of the cities’ attack upon the validity of the 17 percent rule for initial rates which, as we have already seen, failed. We turn therefore to the cities’ more specific points. Second. The cities claim that the FCC should not have treated franchise-related costs (including fees paid to franchising authorities) and programming costs as external because cable companies in fact have significant control over those costs. Although cable operators may indeed have some bargaining power vis-a-vis franchising authorities and programmers, there is no evidence in the record to support the cities’ contention that operators have substantial control over franchise-related and programming costs, ie., that franchising authorities and programmers are price takers. Moreover, the Commission’s decision to grant external treatment to such costs was in part meant to give effect to the specific provisions of the Act that require the Commission to take into account, in prescribing rate regulations for the basic service tier, both franchise fees and other costs associated with meeting franchise requirements. 47 U.S.C. §§ 543(b)(2)(C)(v), (vi); First Reconsiderar tion, 9 F.C.C.R. at 1211,1217-20. The Act is also intended to promote the expansion and diversification of cable programming, see §§ 2(a)(6), 2(b)(1), 2(b)(3), which is more likely to come about if cable operators may recoup their costs from subscribers willing to pay for more expensive and therefore presumably better programs. Although the statute may not require the Commission to grant external treatment to franchise-related and programming costs, the cited provisions surely give the Commission the authority to do so, particularly in the absence of evidence indicating that cable operators do in fact have substantial control over such costs. Third. Although their claim is rather vague, the cities seem to suggest that the FCC improperly allowed cable operators to double count certain of their costs by according external cost treatment to some expenses that they could also take into account in establishing their initial rates. This suggestion, however, reflects a misinterpretation of the form that the Commission created for reporting external costs. Because the Commission resolved any ambiguity that the form may have created by expressly stating that only changes in costs will be recognized and accorded external treatment, Second Reconsideration, 9 F.C.C.R. at 4201-02, the cities’ concern is unwarranted. Fourth. The cities next challenge as one-sided the Commission’s decision to allow cable companies to adjust rates for external cost increases quarterly while requiring them to adjust rates for external cost decreases only once a year. Although the Commission recognized that allowing quarterly increases would add to the administrative burdens it faces, it decided to provide this opportunity lest cable operators be unduly burdened by having to absorb cost increases for up to a year. First Reconsideration, 9 F.C.C.R. at 1233-35. That concern seems well-founded, particularly in light of (1) the constitutional concern that arises if a cable company is required to absorb costs to the point that its allowed rates become confiscatory, see Southern Bell Tel. & Tel. Co., 781 F.2d at 214 & n. 5, and (2) the administrative costs of the alternative, ie., cost-of-service regulation. Because subscribers have no parallel constitutional right that would be protected by requiring quarterly adjustments to reflect cost decreases, the Commission was within its statutory mandate in deciding that such a requirement would not be worth the resulting administrative burden upon the operators and the agency itself. See 47 U.S.C. § 543(b)(2)(A) (requiring the Commission to “reduce administrative burdens”). Moreover, the Commission limited the ability of a cable operator to exploit the disparity in the cost increase and decrease reporting periods by requiring that any operator that files for a cost increase must set off any cost decreases against cost increases in the same quarterly report. Second Reconsideration, 9 F.C.C.R. at 4202. Fifth. In a similar vein, the cities argue that the FCC improperly failed to require cable companies to report increases in advertising revenues and to offset those increases against any external costs. Because advertising revenues, we are told, fluctuate greatly, it would be administratively burdensome to incorporate their ups and downs into the rate-making process. Hence, the Commission’s decision not to do so appears to be reasonable. We do not rule out the possibility, however, that this decision could become unreasonable over time if evidence comes before the Commission clearly demonstrating that advertising revenues have become a steady and significant source of increased revenue for cable operators, thereby calling in question the reasonableness of the rates they may charge under the price cap formula. Sixth. The cities finally contend that the Commission arbitrarily refused to adopt a “productivity offset” (ie., a factor to account for cost-reducing productivity gains in the cable industry) against the inflationary increases allowed to cable operators. They point to the Commission’s decision to include such an offset in its telephone price cap regime, see generally, Policy and Rules Concerning Rates for Dominant Carriers: Report and Order and Second Further Notice of Proposed Rulemaking, 4 F.C.C.R. 2873 (1989), and argue that by failing to do so here, the Commission improperly subordinated the interests of consumers to those of cable operators. The Commission adequately explained the apparent disparity: the record in the telephone price cap proceeding included extensive evidence demonstrating that increases in productivity in the telephone industry significantly exceed those in the general economy. Rate Order, 8 F.C.C.R. at 5781-82. There is no comparable evidence concerning the cable industry in the present record. Therefore, we are constrained to affirm the Commission’s decision not to impose a productivity offset at this time. As with its decision not to offset advertising revenues, however, the Commission’s decision not to establish a general productivity offset could ultimately prove to be unreasonable if the Commission is ever confronted with evidence indicating that the cable industry does in fact benefit from productivity increases that significantly outstrip those in the general economy. 2. The Cable Companies’ Petitions The cable petitioners question only whether the external-cost rules for going-forward rates should have been applied to the “gap period” between passage of the Act and the date upon which each cable operator actually became subject to rate regulation under the Act. Although the most obvious way to set cable operators’ initial rates might have been simply to reduce by 17 percent the rates that they were charging when the new regulations took effect, the Commission decided not do that lest it build into the permitted initial rates any unwarranted rate increases that cable operators took after passage of the 1992 Cable Act. Second Reconsideration, 9 F.C.C.R. at 4170. The Commission instead decided to discount the rates charged by operators on September 30, 1992 (the last month before passage of the Act) by 17 percent, and then to adjust that rate to allow for inflation during the gap period. Id. at 4170-71. At the same time, however, the Commission decided not to allow cable operators to adjust their rates to reflect external cost increases incurred during the gap period, which the cable petitioners say is arbitrary and capricious. The FCC acknowledged that allowing adjustments for changes in external costs during the gap period would make the initial rates more accurate, but it decided that doing so would place undue administrative burdens upon both cable operators and the Commission. First Reconsideration, 9 F.C.C.R. at 1233. That explanation is patently unconvincing for two reasons. First, the Commission’s concern with the administrative burden upon cable operators is unnecessary: if the cost of recovering its increased external costs exceeds the revenue to be gained, the cable operator can be counted upon to forego the pleasure. Second, the Commission’s concern with the administrative burdens that it would face also rings hollow because the type of documentation to be reviewed would be the same as the external-cost documentation that the Commission must review for all post-gap periods pursuant to the going-forward rules. The Commission also attempts to justify its decision by noting that the disadvantaged cable operator can always turn to the eost-of-service option. But that option, by the Commission’s own admission, is costly for cable operators and the Commission alike, and is intended to be a limited “safety-valve” exception. Second Reconsideration, 9 F.C.C.R. at 4196. In sum, the Commission offers no reason to doubt that cable operators incurred external costs during the gap period, yet under its regulations they would never be able to recoup those costs short of opting for cost-of-service regulation — which would be akin to shooting a fly with a blunderbuss. Because the Commission’s proffered justifications are completely unacceptable, we hold that its decision to preclude a rate adjustment designed to recover changes in external costs incurred during the gap period is arbitrary and capricious. C. The Cable Companies’ Other Challenges The cable petitioners make two other claims, both of which question the FCC’s interpretation of certain provisions of the Act. Neither challenge can surmount the hurdle set for it by Chevron. 1. Tier Neutrality As discussed in Part I, the Commission decided to apply its ratemaking rules in a “tier-neutral” fashion, meaning that the same methodologies and standards are used to establish allowable rates for both the basic service tier and the cable programming service tier(s). First Reconsideration, 9 F.C.C.R. at 1182-85; Rate Order, 8 F.C.C.R. at 5759-60, 5881-82. It did so in order to avoid creating any incentive for cable operators to move programming between the basic service and cable programming service tiers by making any such move revenue neutral. See First Reconsideration, 9 F.C.C.R. at 1183; Rate Order, 8 F.C.C.R. at 5759-5760. The FCC further explained that because it is simpler, the tier-neutral approach also serves to reduce the administrative burden upon all concerned. Id. The cable petitioners contend that the language, structure, and legislative history of the 1992 Cable Act simply do not permit the Commission to apply the same regulatory standard to the basic service tier and to cable programming service. That contention, however, is premised upon a significant misunderstanding of the Act. First, the cable petitioners misconstrue the Congress’s findings. Focusing solely upon § 2(a)(1) of the Act, they suggest that the Congress found that only basic cable rates were excessive. Although that section begins with a reference to then-recent increases in the rates for basic cable service, it ends with the broader conclusion that “[t]he average monthly cable rate has increased almost 3 times as much as the Consumer Price Index since rate deregulation.” While it is possible that the Congress meant, as the cable petitioners suggest, only that average monthly rates for basic cable service had increased by thrice the CPI, that is not what the provision says. Moreover, after reviewing the rest of the Congress’s findings it becomes clear that, regardless of the proper construction of § 2(a)(1), the Congress was concerned with what it perceived to be the excessiveness of cable rates in general, not the rates for a particular type of service. Indeed, the Congress immediately followed the finding referenced by the cable operators with a general finding (in § 2(a)(2)) that cable operators serving most cable subscribers do not face effective competition and consequently exercise undue market power. That finding admits of no distinction between basic tier and cable programming service. Similarly, the Congress made findings about increasing concentration and vertical integration in the cable industry generally. See §§ 2(a)(4), (5). The statute is even clearer when it broadly states that: It is the policy of the Congress in this Act to ... (4) where cable television systems are not subject to effective competition, ensure that consumer interests are protected in receipt of cable service; and (5) ensure that cable television operators do not have undue market power vis-a-vis video programmers and consumers. §§ 2(b)(4), (5). Although the cable petitioners would like to limit that policy to the basic service tier, the stubborn fact remains that the Congress directed it to the cable industry in general. Put simply, the legislature’s generalized approach to formulating the problem and to enumerating the objectives of the statute simply do not support the cable petitioners’ position that the Act is concerned only (or even concerned more) with rates for the basic service tier than with rates for cable programming service; if anything, the Congress’s non-tier-specific findings and policy statement support the FCC’s view that tier-neutral regulation is appropriate. The cable petitioners also mistake the significance of certain substantive differences between the statutory sections that govern respectively the basic service tier and cable programming service. Although the Act requires the Commission to establish regulations to ensure “reasonable” rates for the basic service tier, 47 U.S.C. § 543(b)(1), with respect to cable programming service it authorizes the Commission to correct “unreasonable” rates. 47 U.S.C. § 543(c)(1)(A). The cable petitioners perceive a middle ground between the “reasonable” and the “unreasonable,” suggesting that in light of those differing terms the Commission’s regulation of cable programming service must be more lenient than its regulation of the basic service tier. That suggestion is at the least counterintuitive; if the Congress intended to invoke different levels of regulatory stringency, it seems most unlikely that they would have used those cognate terms to describe the two regimes. Moreover, the Commission’s explanation— that the terminological difference reflects a procedural rather than a substantive distinction in the two regulatory schemes — is a good deal more persuasive. See Rate Order, 8 F.C.C.R. at 5875. Although the Act requires loeal franchising authorities actively to regulate rates for the basic service tier in accordance with established FCC standards, it precludes the Commission from reviewing a system’s rates for cable programming service unless and until it receives a complaint from a subscriber, the franchising authority, or some other relevant state or local governmental entity. Compare 47 U.S.C. § 543(b) with 47 U.S.C. § 543(c). Consequently, rates for the basic service tier will always be reviewed ex ante while rates for cable programming service will only be reviewed ex post. It therefore makes sense that the Congress would formulate the question respecting the basic service tier as whether, ex ante, a proposed rate would be reasonable, and yet formulate the question respecting cable programming service as whether, ex post, an existing rate is unreasonable. Because those key terms are strikingly similar and the slight difference between them is easily explained as a product of the different procedural postures in which they will arise, we conclude that this text actually supports the Commission’s tier-neutral approach rather than the contentions of the cable companies. The cable petitioners also point to other differences between the regulatory regimes for the basic service tier and for cable programming service. They note, for example, that the Act provides that “in establishing the criteria for determining in individual cases whether rates for cable programming services are unreasonable” the Commission shall “consider” six factors, 47 U.S.C. § 543(c)(2), se