BEFORE THE WASHINGTON UTILITIES AND TRANSPORTATION COMMISSION In the Matter of the Pricing Proceeding ) for Interconnection, Unbundled Elements, ) DOCKET No. UT-960369 Transport and Termination, and Resale ) ) In the Matter of the Pricing Proceeding ) for Interconnection, Unbundled Elements, ) DOCKET NO. UT-960370 Transport and Termination, and Resale for ) ) US WEST COMMUNICATIONS, INC. ) ) In the Matter of the Pricing Proceeding ) for Interconnection, Unbundled Elements, ) DOCKET NO. UT-960371 Transport and Termination, and Resale for ) ) GTE NORTHWEST INCORPORATED ) GTE’S PRE-HEARING BRIEF GTE Northwest Incorporated (“GTE”), by counsel, pursuant to the Commission’s directive at the prehearing conference on September 21, 1998, and the Commission’s Sixteenth Supplemental Order on Prehearing Conference, hereby files its Pre-Hearing Brief. This Brief will focus primarily on the over-riding issues before the Commission in this proceeding – the statutory and constitutional right of ILECs to recover all of their actual costs of providing unbundled network elements (“UNEs”), and the proper pricing policies and mechanisms to ensure such full recovery. The Brief will also summarize GTE’s positions with respect to the remaining issues in this proceeding. I.Full Cost Recovery Issues Part I of this Brief will address Issues V.1 through V.10 of the Commission’s July 10, 1998 Issues List. A. Overview The Telecommunications Act of 1996 (the “Act”) requires that, in selling (or, more accurately, leasing) its UNEs, GTE have the opportunity to recover all of its actual costs, which are the costs GTE currently incurs to provide telephone service using GTE’s existing local exchange network. This is the only plausible interpretation of the plain language of section 252(d)(1), which requires that UNE prices shall “be based on the cost . . . of providing” UNEs and “may include a reasonable profit.” It is also the only way to reconcile the Act’s UNE pricing provision with its pricing provision for resale. On an annual basis, GTE’s total actual costs are equal to its regulated revenues – approximately $470.1 million (provided by the underlying UNEs) in Washington. This formula follows from the history and purpose of regulation in this state. GTE has been and continues to be a regulated company. A primary purpose of regulation is to ensure that the regulated company is given the opportunity to recover its prudently-incurred investment and operating costs plus a reasonable profit – nothing more and nothing less. Through its regulation of GTE, the Commission provides GTE with that opportunity. Accordingly, the revenues produced through GTE’s tariffed rates reflect GTE’s actual costs plus a reasonable profit. To conclude otherwise would mean that this Commission has failed to perform it statutory duties. See Wash. Rev. Code § 80.36.080 (1998). There is no evidence of such a failure here. Accordingly, GTE’s pricing proposal for UNEs is designed to provide GTE the opportunity to recover its actual costs, as measured by its current revenues, including a reasonable profit. The Commission has already estimated the total element long-run incremental cost (“TELRIC”) of the key elements in GTE’s network. GTE continues to believe that the Commission’s TELRICs are too low. GTE therefore respectfully reserves the right to continue to object to the Commission’s decisions in this regard. However, these TELRICs do not capture GTE’s common costs, which are those forward-looking costs that cannot be assigned to particular UNEs. Thus, GTE proposes that its forward-looking common costs be recovered through the use of a fixed allocator. GTE has calculated its forward-looking wholesale related common costs in the state of Washington to be $136.8 million, $18.2 million of which is currently recovered through services that are not provided via UNEs. This leaves $118.6 million in common costs to be recovered in UNE prices. That number is equal to 55 percent of GTE’s total direct costs of $214.3 million, which GTE has calculated based on the TELRICs previously developed by this Commission. Increasing each previously established TELRIC by 55 percent will produce UNE prices that reflect GTE’s Commission established TELRICs plus the company’s wholesale related common costs ($214.3 million + $118.6 million = $332.9 million). This $332.9 million is the total revenue GTE would receive if it were to become a 100% wholesaler of UNEs, assuming that its wholesale UNE prices were based on Commission approved TELRICs plus 55% to recover the company’s common costs. This amount, however, cannot be reconciled with the company’s current revenues of $470.1 million (provided by the underlying UNEs, which, when adjusted to net out avoided retailing expenses, drops to $442.7 million. The $442.7 million figure is derived by using the wholesale discount of 10.1%, applying it to retail services subject to resale, and subtracting retailing expenses that the company would avoid if it were to become a 100% wholesaler of finished services. As Staff witness Glen Blackmon noted in his Phase I prefiled testimony: The TELRIC estimates should be used as a check on prices, to ensure that no prices are set below cost. Once that test is met the objective should be to establish rough parity between wholesale or unbundled network element rates and corresponding resale rates for finished services. Finished services and wholesale elements are generally close substitutes for each other, since the latter are piece parts of the former. Pricing one below the other sends the market incorrect signals that distort the choices of both consumers and competitors and it could constitute undue discrimination. See Testimony of Glenn Blackmon at 9 (March 28, 1997). GTE agrees with Mr. Blackmon. Stated another way, the wholesale prices for finished services and for their component parts (i.e., UNEs) should be comparable, since elements are “piece parts” of services. Several truths emerge from this analysis. First, either the Commission-approved TELRICs are way too low, or GTE’s current revenues less avoided expenses are vastly in excess of what they should be, and there is no evidence of that. Second, the comparatively small common cost mark-ups proposed by other parties (see, e.g., MCI’s proposed 7.14% mark-up) are all grossly insufficient to cover GTE’s substantial common costs. For instance, using MCI’s mark-up, GTE’s total revenues would be only $229.6 million ($214.3 + ($214.3 x 7.14%)). Not only is this $100 million less than the total revenues yielded by GTE’s 55% allocator, it is a whopping $213.1 less than $442.7 million. Third, in order to afford GTE the opportunity to recover all of its actual costs of providing universal service, an additional adjustment is necessary to mitigate arbitrage opportunities that result from the disorientation in GTE’s retail rates. In particular, GTE must be permitted to assess a surcharge that will ensure “rough parity” between GTE’s wholesale and retail rate structures and protect GTE’s continuing obligation to provide universal service. As explained below, absent such consistency, CLECs will be granted significant arbitrage opportunities that will deny GTE the opportunity to recover its actual costs, let alone a reasonable profit. Such a denial would clearly be contrary to the Act’s central cost recovery and universal service mandates. This interim surcharge will be necessary until the Commission establishes a permanent and sufficient USF mechanism. Finally, GTE proposes a Competitive Transition Charge (“CTC”) to recover any stranded costs GTE incurs. Stranded costs are defined as the current dollar value of prudent investments (or ratebase) no longer recoverable because of a change in policy, e.g., the opening of markets into which entry previously was prevented by franchise limitation and/or ordered rates that do not allow the company to recover its actual costs. The change in policy wrought by the Act, and the resulting competitive entry by CLECs while ILECs remain regulated, will strand some of GTE’s costs by causing loss of market share and the lowering of prices. The Commission can substantially mitigate these adverse effects by adopting the other elements of GTE’s pricing proposal, by rebalancing rates, and by establishing a competitively neutral USF mechanism. Nonetheless, to the extent that some stranded costs remain, GTE is entitled to recover them. It proposes to do so through the CTC. B. The Act Requires That GTE Be Given the Opportunity to Recover All of Its Actual Costs. Section 252(d)(1) of the Act commands that prices for UNEs “shall be based on the cost . . . of providing” the UNE. That directive requires that UNE prices be based on all the actual costs that GTE has incurred in constructing and operating the network that it must now make available to CLECs in the form of UNEs. The cost “of providing” a particular network element is necessarily GTE’ s cost, not the cost of another carrier or a hypothetical one. As the incumbent carrier it is GTE, and only GTE, that is doing the “providing” of actual UNEs to new entrants. And the UNEs GTE will be providing are those contained in GTE’s actual network, not those of a “yet unbuilt superior one.” See Iowa Utilities Board v. FCC, 120 F. 3d 753, 813 (8th Cir. 1997). Thus, only GTE’s costs are relevant under the terms of the Act. Section 252(d)(1) goes on to state that prices for UNEs “may include a reasonable profit.” This language also establishes that GTE must fully recover its actual costs because a firm can never earn a profit until it first recovers its actual costs. Even the FCC has recognized this reality. See In re Implementation of the Local Competition Provisions of the Telecommunications Act of 1996, First Report and Order, CC Docket No. 96-98, FCC 96-325 (rel. August 8, 1996) (“First Report and Order”) at ¶ 699 (there can be no “profit” unless a firm earns “the total revenue required to recover all the costs of [the] firm”) (emphasis added). An examination of the Act as a whole also confirms that UNE prices must reflect GTE’s actual costs. Under the Act, CLECs can take advantage of an ILEC’ s network by (1) reselling an ILEC’s services and (2) leasing UNEs. When a CLEC resells ILEC services, section 252(d)(3) of the Act requires that it purchase those services at wholesale rates equal to the ILEC’s current tariffed rates minus the ILEC’s avoided retailing expenses. Therefore, if the ILEC is a 100 percent reseller, it will receive revenues equal to its current revenues minus its avoided costs. This pricing rule is predicated on the fact that an ILEC’s current tariffed rates and resulting current revenues – based as they are on successful regulation by state Commissions – are “just and reasonable” and reflect the ILEC’s actual costs plus a reasonable profit. When a CLEC purchases UNEs, section 252(d)(1) requires that it pay the ILEC’s “cost,” which may include a reasonable profit. If one assumes that an ILEC is a 100 percent wholesaler of UNEs (i.e., that it leases all of its UNEs), then the total revenues it receives should, as in the case of resale, correspond to the ILEC’s actual cost as reflected by its current revenues (less its avoided retailing expenses). Otherwise, the ILEC would receive less revenue as a pure UNE wholesaler under the Act than it would as a pure reseller, even though the physical network and the costs of operating that network are essentially the same in both environments. It makes no sense to suppose that Congress, having pegged resale prices to the ILEC’s actual costs (as reflected in existing Commission approved retail rates), would disregard the ILEC’s actual costs in the UNE wholesale context. The Commission Staff appears largely to have endorsed this reasoning. In his Phase I testimony filed on March 28, 1997, Staff witness Glenn Blackmon stated: [T]he objective should be to establish rough parity between wholesale or unbundled network element rates and corresponding resale rates for finished services. Finished services and wholesale elements are generally close substitutes for each other, since the latter are piece parts of the former. Pricing one below the other sends the market incorrect signals that distort choices of both consumers and competitors, and it could constitute undue discrimination. See Testimony of Glenn Blackmon at 9. Since resale rates are pegged to the ILEC’s actual costs, so too should wholesale UNE rates. Finally, prices that reflect actual ILEC costs promote the pro-competitive goals of the Act. If CLECs can build and operate a more efficient network than GTE’s, they should be encouraged to do so. Setting UNE prices based on the cost of GTE’s actual network provides CLECs that incentive; setting lower prices that reflect a network that the CLEC theoretically could build does not. Such prices encourage CLECs to create proxy models, not networks. Prices based on such models are not the result of competition; they are simply the output of cost models, the best of which are still only guesses. In sum, UNE prices must allow ILECs to recover their actual costs plus a reasonable profit. As described below, GTE’s pricing proposal is the proper method for accomplishing this statutorily-mandated objective. The Commission Should Increase UNE Prices to a Level 55 Percent Above TELRICs to Reflect GTE’s Common Costs. GTE’s common costs are costs the company incurs that cannot be directly attributed to individual network elements. These common costs include, but are not limited to, administrative costs and overhead. No party to this proceeding denies that GTE reasonably will incur common costs on an ongoing basis, or that GTE is entitled to recover such costs. The only dispute is regarding the amount of GTE’s common costs. As reflected in the testimony of GTE witness Norris, GTE’s forward-looking wholesale related common costs are $136.8 million. Because $18.2 million of this amount is recovered through services not associated with UNEs, the common costs that must be recovered through UNE prices are $118.6 million. See Direct Testimony of Michael R. Norris at 9, (July 9, 1998). GTE proposes to recover its common costs through the use of a fixed allocator. A fixed allocator distributes common costs proportionally over all UNEs by “marking up” the TELRIC for each UNE by the same percentage. That percentage should correspond to the amount needed to raise UNE prices to the level needed to capture, in a 100 percent wholesale UNE environment, total costs (i.e., direct costs (TELRICs) plus common costs). For example, consider an ILEC whose total direct costs are $200 million and whose total common costs are $100 million. In this hypothetical, the TELRICs established for UNEs should result in cost recovery of $200 million assuming that the ILEC operates as a 100 percent wholesaler of UNEs. In order for this ILEC to recover all of its forward-looking costs, though, the TELRICs must be marked up. In this example, a mark-up of 50 percent ($100 million divided by $200 million) would be needed because increasing $200 million by 50 percent yields $300 million. Most parties to this proceeding agree that common costs should be recovered through a fixed allocator. The FCC has also recognized the reasonableness of this approach. See First Report and Order ¶ 696. Its advantages are manifest. First, it is easy to implement and verify. Second, it is equitable. Under the fixed allocator approach, those elements for which the CLEC is likely to have fewer alternative sources (e.g., the loop) do not receive a higher percentage mark-up than other elements for which there are many alternatives (e.g., the switch). In paragraph 251 of the Eighth Supplemental Order, the Commission asked the parties to select and justify a common cost allocation method. GTE prefers using a fixed allocator in this context, principally because of its administrative convenience. There may be circumstances, however, when it would be appropriate to deviate from the fixed allocator approach. In this proceeding, the primary dispute with respect to common costs is over the magnitude of the fixed allocator. This dispute, in turn, boils down to a disagreement over the magnitude of GTE’s common costs. As discussed above, the correct methodology for deriving the allocator is simply to divide total common costs (or at least those that are not recoverable through the pricing of non-UNE related services) by total direct costs. Based on the Commission-approved TELRICs, GTE’s total direct costs in Washington are $214.3 million. The missing variable, then, is the amount of GTE’s total common costs. GTE quantified its common costs in Washington by analyzing 1996 ARMIS data. GTE witness Michael Norris extracted from that data those costs that cannot be assigned to particular UNEs and thus are not captured in the Commission’s TELRICs. The sum of these costs equals GTE’s common costs. That sum was $136.8 million, of which $18.2 million is currently recovered in the rates of services such as special access and billing and collection which, therefore, are not provided via UNEs. No party can credibly show that Mr. Norris’ analysis includes as common costs any money that is captured in the Commission’ s TELRICs. That leaves approximately $118.6 million in common costs to be recovered in UNE prices through the fixed allocator. Accordingly, under the formula described above, the appropriate fixed allocator is 55 percent ($118.6 million ÷ 214.3 million). AT&T proposes a fixed allocator of only 10.4 percent for common cost recovery. It derives this figure based on a comparison of corporate operations expense to revenue less corporate operations expenses. It purports to use AT&T data to make this comparison. AT&T’s approach is severely flawed. First, there is no justification for using AT&T, as opposed to GTE, data. It is GTE that will be providing UNEs, and it is GTE that must recover the costs associated with doing so. AT&T’s costs are irrelevant to that objective. Second, the 10.4 percent figure does not even reflect all of AT&T’s common costs. It includes only AT&T’s Corporate Operations Expenses. As GTE witness Norris has testified, however, there are other types of common costs, such as General Support and Plant Operating Expenses, that are not included in the Commission’s TELRICs but that AT&T has simply ignored. Tracer proposes an even smaller mark-up of 4.05 percent. It argues that this mark-up is sufficient for U S West and should therefore be sufficient for GTE. But the premise that GTE’s allocator should match U S West’s is incorrect – the Act plainly requires that prices for UNEs be based on the cost of the ILEC actually providing the UNEs, not the costs of another carrier, hypothetical or real. Tracer witness Dr. Zepp acknowledges that, for GTE, the ratio of Corporate Operations Expenses alone to total revenues is 15.6 percent. This implies an allocator of 18.5 percent (15.6/(100-15.6)=18.5) based on the corporate operations component of common costs alone. Adding in the rest of GTE’s common costs would produce a significantly higher figure. Moreover, Dr. Zepp has failed to establish that U S West’s allocator should be only 4.05 percent. U S West’s evidence shows that its mark-up should be on the order of 18 percent. MCI proposes an allocator of 7.14 percent. However, that allocator is derived by comparing common costs with total regulated revenues, not total direct costs. As explained above, only a comparison of total common costs with total direct costs will produce an allocator that will enable GTE to recover all of its forward-looking costs. Moreover, MCI repeats the mistake of AT&T and Tracer by including only Corporate Operations Expenses as common costs. AT&T and MCI would have the Commission simply ignore the balance of GTE’s substantial common costs. Costs are costs—just because a significant component of a firm’s total costs are common rather than direct costs does not mean the costs were not legitimately incurred, and should not be recovered as a component of the firm’s total costs. In sum, only GTE has derived a common cost allocator that includes all of GTE’ s common costs and utilizes the correct formula. No party has rebutted GTE’s account-by-account computation of its common costs. Accordingly, the Commission should apply the 55 percent fixed allocator to GTE’s TELRICs. C. Sections 252(d)(1) and 254(f) of the Act Require the Commission to Include a Surcharge Applicable to Certain UNEs and to Resale to Prevent CLECs from Pocketing USF Support. There can be no dispute that GTE’s retail rates are disoriented – that is, they contain implicit subsidies that support universal service in Washington. This disorientation in GTE’s retail rates provides significant arbitrage opportunities for CLECs. The testimony of GTE witness R. Kirk Lee presents examples of the arbitrage opportunities available to CLECs as a result of the implicit supports in GTE’s current retail structure. These examples use actual GTE revenue data from Washington. However, the problem can also be illustrated by a simpler example. Consider a hypothetical company, Washington Computer Associates (“WCA”), that manufacturers computers at $800 per unit and sells them at retail to both residential and business customers. Suppose that WCA incurs $100 in retailing expenses per computer and earns a competitive return (i.e., a reasonable profit) of $100 on each computer sold. Therefore, in order to recover to costs plus a reasonable profit, WCA charges $1,000 per computer to every residential and business customer. If WCA has ten residential and ten business customers, it produces 20 computers and receives $20,000 in revenues. It thereby covers its production and retailing costs and earns a reasonable profit. Now suppose that the government wants to encourage residential customers to own computers. To achieve this goal, it allows WCA to charge only $500 per computer to residential customers. To offset this implicit subsidy, the government allows WCA to charge $1,500 to every business customers. In this scenario, WCA will still produce twenty computers and still recover $20,000. ($5,000 from the sale of ten computers to residential customers at $500 plus $15,000 from the sale of ten computers to business customers at $1,500). Suppose, however, that the government then changes the rules and requires WCA to sell computers (or computer components) to its competitors at wholesale prices and that these wholesale prices are not consistent with WCA’s retail prices (because, principally, WCA remains obliged to sell its computers to residential customers at retail prices that are well below the company’s production costs). In that event, WCA will no longer be able to recover even its production costs of $16,000 ($800 x 20 computers). For example, if WCA is required to sell computers to Radio Shack at the cost-based wholesale price of $900 (production cost plus the $100 that formerly provided a profit), Radio Shack will simply target all of WCA’s business customers and sell them computers at a price less than the $1,500 WCA must charge to offset the effect of selling at below cost to residential customers. Eventually, WCA will be forced to sell computers to business customers at $900 in order to compete with Radio Shack. At this point, WCA will realize only $14,000 for its 20 computers ($5,000 from the 10 it sells to residential customers at $500 plus $9,000 from the 10 it sells to business customers at $900). Unable to cover even its production costs of $16,000, WCA will be forced out of business. This will occur not because WCA was less efficient than its competitors, but because government regulation created an imbalance between WCA’ s retail and wholesale prices, and its competitors naturally exploited the resulting arbitrage opportunities. Like WCA in our example, GTE’s existing retail rates contain significant implicit support; they do not reflect the true cost of providing service to Washington’s residential customers, especially those in rural markets. GTE is able to sustain these low rates only by virtue of the implicit subsidy supports provided by higher-priced services. But these implicit subsidies are not sustainable in a competitive environment because, as the computer example shows, competitors will be able to “cream skim” the customers who provide the support needed to maintain the subsidies. As the FCC has recognized: In a competitive market, a carrier that attempts to charge rates significantly above costs to a class of customers will lose many of those customers to a competitor. This incentive to entry by competitors in the lowest cost, highest profit market segments means that today’s pillars of implicit subsidies – high access charges, high prices for business services, and the averaging of rates over broad geographic areas – will be under attack. New competitors can target service to more profitable customers without having to build into their rates the types of cross subsidies that have been required of existing carriers who service all customers. Universal Service Report and Order, CC Docket No. 96-45, May 8, 1997, ¶ 17 (emphasis added). Accordingly, in the absence of retail rate rebalancing and the establishment of an explicit permanent universal service funding mechanism that removes all implicit subsidies in intrastate and interstate rates, the Commission should establish an interim universal service surcharge. The surcharge should be collected from CLECs that purchase GTE’s unbundled loops or ports, as well as from CLECs that enter using their own facilities exclusively. The surcharge must equal the average per line support provided by business customers. As shown in the testimony of Mr. Lee, the interim surcharge equals $37.17 per business line per month, based on GTE’s current retail rates. This interim universal service surcharge is required by the Act. As noted, section 252(d)(1) commands that GTE be allowed to recover, at a minimum, its actual costs. Absent the surcharge, GTE will be unable to do so. Moreover, section 254(f) of the Act requires that “every telecommunications carrier that provides intrastate telecommunications service shall contribute on an equitable and nondiscriminatory basis, in a manner determined by the State to the preservation and advancement of universal service in that State.” Until universal service is funded explicitly, only the inclusion of the interim surcharge in UNE prices will fulfill this statutory requirement in Washington. The necessity of such an interim surcharge, and its legitimacy under the Act, was recognized in Competitive Telecommunications Ass’n v. FCC, 117 F. 3d 753 (8th Cir. 1997). The issue there was whether the FCC had violated the Act’s cost-based pricing provisions by allowing ILECs to collect certain charges not related to the cost of UNEs in order to provide a temporary funding mechanism for universal service. Recognizing that universal service would be adversely affected in the absence of the charges in question, the Court concluded that the application of these charges on an interim basis was lawful. It stated: [T]he Act requires the reform of universal service subsidies and not, significantly, abolishment of universal service, even temporarily. Clearly, Congress did not intend that universal service should be adversely affected by the institution of cost-based rates. But the nine-month disparity between the deadline for implementation of cost-based service and the deadline for reform of universal service raises the threat of serious disruption in universal service for those nine months if cost-based service is required before universal service is funded by competitively neutral means . . . . If the FCC . . . had not instituted an interim access charge of some sort in order to subsidize universal service for the nine months before universal service reforms are complete, we think it apparent that universal service soon would be nothing more than a memory. 117 F. 3d at 1074 (emphasis added). Here, too, it is apparent that without the interim surcharge proposed by GTE, CLEC “cream skimming” will undermine universal service. To avoid that unlawful result, the interim surcharge must be added to UNE rates until permanent universal service reform is completed. In addition, a CLEC that chooses to self-provision intraLATA toll and instead pay GTE switched access charges will avoid paying rates specifically designed to support universal service. It is for this reason that an interim surcharge for resold services is also needed. GTE witness R. Kirk Lee calculated the interim universal service surcharge for resale at $0.17 per line per month for residential lines and $0.62 per line per month for business lines. This interim surcharge equals the amount of universal service support for intraLATA toll service currently generated by an average GTE customer line, net of the universal service support the same line would provide to GTE if it were collecting switched access rates instead. D. GTE Is Entitled to a Competitive Transition Charge to Recover Its Stranded Costs. If the Commission adopts the pricing proposals discussed above, GTE will be able to recover the vast majority of its actual costs. However, as shown in the testimony of GTE witness Michael Doane, some stranded investment would remain due, for example, to loss of market share. Because GTE is entitled to recover all of its actual cots, GTE proposes a competitive transition charge (“CTC”) to recover the remaining stranded investment. Stranded costs are the current dollar value of prudent investments no longer recoverable because of a change in regulatory policy. These previous expenditures are tied to prior regulatory requirements, including universal service obligations, requirements that certain rates be below-cost, carrier of last resort obligations, and service quality standards. Therefore, stranded costs represent investments that would have been uneconomic in the absence of a regulatory compact that protected the regulated firm’s opportunity to earn a fair return on invested capital. In this context, stranded costs can best be understood as a transition payment incurred in introducing competition into a previously regulated industry. GTE requests that the Commission implement this transition payment, the CTC, at the level necessary to enable GTE to recover whatever costs remain stranded given the Commission’s resolution of the other pricing issues before it. This payment is consistent not only with section 252(d)(1), for the reasons described previously, but also with sound economic policy.