BEFORE THE WASHINGTON UTILITIES AND TRANSPORTATION COMMISSION In the Matter of the Pricing Proceeding for Interconnection, Unbundled Elements, Transport and Termination, and Resale In the Matter of the Pricing Proceeding for Interconnection, Unbundled Elements, Transport and Termination, and Resale for U S West Communication, Inc. In the Matter of the Pricing Proceeding for Interconnection, Unbundled Elements, Transport and Termination, and Resale for GTE NORTHWEST INCORPORATED ) ) ) ) ) ) ) ) ) ) ) ) ) ) DOCKET NO. UT-960369 DOCKET NO. UT-960370 DOCKET NO. UT-960371 POST-HEARING BRIEF OF MCI RE AVOIDED COSTS ) INTRODUCTION It has long been the policy of this state to promote competition in all telecommunications markets. See, e.g., RCW 80.36.300(5). With the passage of the Federal Telecommunications Act of 1996 (FTA), national policy has caught up with Washington. Moving from that theoretical policy, however, to competitive reality will not be easy. In the two years since the Commission wrote that it faced "many difficult issues as it attempts to facilitate the transition of the telecommunications industry from a monopoly market structure to a competitive market", Fourth Supplemental Order, Docket UT-941464 (Interconnect case), page 6, Washington Utilities and Transportation Commission v. U S West Communications, Inc., et al., Consolidated Docket UT-941464 (WUTC October 31, 1995).competition is still just emerging. U S West and GTE still dominatethe landscape. Competitive LECs (CLECs) are operating, but their market share remains minuscule. The decisions theCommission makes in both phases of this docket will determine whether local competition ever evolves from its current nascent stage to the vigorous effective competition Congress envisioned. MCI has proposed an avoided cost study that will ensure that competition has a reasonable chance to evolve in Washington. The Commission should reject any arguments that MCI's model fails to preserve the current return of the incumbent LECs (ILECs). Such arguments are based on the false premise that it is possible and appropriate to protect the ILEC's current revenue streams when setting wholesale prices. It is not. More importantly, it is essential for the broader public interest that the Commission rejects these arguments to ensure that competition has a chance to develop. If competition works, the public will benefit in the long term by an actively competitive market. If, on the other hand, the Commission errs on the side of trying to protect the monopoly ILECs from erosion of their current revenue streams, the state may never see the benefits of vibrant competition. While no course available to the Commission is without risk, the benefits of promoting competition far outweigh the minuscule risks associated with adopting costing methods and prices for UNEs and wholesale services that follow both the letter and the pro-competitive spirit of the FTA. DISCUSSION As to all issues other than avoided costs, MCI and AT&T have filed a joint brief. This brief addresses only avoided cost issues. V. Avoided Costs: a. Definitions and Assumptions. A threshold issue that may be raised in this proceeding is what is meant by "avoided" costs. MCImetro’s avoided cost study identifies retail costs that an efficient company would avoid in making a sale at the wholesale level. The Commission should reject any argument that "avoided" costs are only those actually avoided. In fact, no party has or reasonably can demonstrate "actual" avoided costs because all parties are basing their studies on historic cost data measured prior to the inception of resale competition. The FTA is designed to facilitate economically efficient entry of new competitors into the local exchange market. Thus, the relevant inquiry for determining an appropriate wholesale discount rate is to determine which retail costs are avoidable by an economically efficient competitor selling at wholesale, and not which costs an ILEC actually expects to avoid. This approach thus encourages incumbents to avoid costs that are avoidable, and thereby operate efficiently. With the advent of competition, such increased efficiency will translate into lower prices to consumers. The “actually avoided” standard that has been advocated by some is inimical to the goal of economically efficient pricing, as it would subsidize, if not encourage, the incumbent’s inefficiencies and create a disincentive to cost avoidance. i. Services to which discounts will apply. The FTA recognizes that merely removing the legal barriers to entry is insufficient to allow competition to develop. The enormous costs of duplicating the networks the ILECs have built over the last century constitutes a significant economic barrier to facilities based entry. Consequently, it is essential for competition to develop quickly that CLECs be permitted to resell all retail telecommunications services at wholesale rates in order for local competition to develop. Exh. 164 (DiTirro) at 3-4. The FTA makes it clear that in order to further the goal of development of local exchange competition, ILECs must offer all retail telecommunication services for resale and wholesale rates. Specifically, ILECs are required: To offer for resale at wholesale rates any telecommunication service that the carrier provides at retail to our customers who are not telecommunications carriers. . . . 47 U.S.C. § 251(c)(4) (emphasis added). A "retail" service is a service provided to subscribers "who are not telecommunications carriers." Id. Thus, for example, residential services are clearly retail services that must be made available for resale at a discount. Volume discounted services should also be available for resale at wholesale rates. First, because ILECs sells volume discounted services to noncarrier (i.e., "retail") customers, the services are clearly subject to the resale requirement of section 251(c)(4)(A). Moreover, simply because a service is provided at a volume discount does not mean that there are no savings for the ILECs when the service is sold at wholesale. For example, a consumer may purchase a single roll of paper towels for a certain price or a dozen rolls of paper towels for a smaller price per unit. That the consumer receives a volume discount on the bulk purchase does not transform the sale from retail to wholesale. The same holds true with telecommunications services. The retailing costs on the sale are avoided when the same services are sold at wholesale regardless of whether the sale is a single unit or a volume sale. ILECs will realize substantial reductions of retailing costs on all their wholesale services. For example, in the case of Centrex-type services, the ILECs currently bear costs for the manual ordering process that their retail customers use. These costs will be eliminated in the wholesale environment by the transition to electronic interfaces and electronic ordering. See, e.g., Exh. 165 (DiTirro) at 13 - 14. The Commission should also require ILECs to sell certain services, even if the Commission does not agree they are telecommunications services to CLECs--at at least the retail rate. Examples of such services are voice mail, inside wire, and inside wire maintenance. Although MCI believes these services are telecommunications services, the Commission, in previous orders, has disagreed. See, e.g., Commission Order Approving Interconnection Agreement, In the Matter of the Petition for Arbitration of an Interconnection Agreement Between MCImetro Access Transmission Services, Inc. and U S West Communications, Inc. (WUTC Docket No. UT-960310, August 18, 1997). [Hrt][Hrt][Hrt][Hrt][Hrt][Hrt] These are services that many of the ILEC's customers find to be valuable. If the CLEC cannot continue to provide the end user with their familiar services, then this will constitute a barrier to the CLEC winning that customer. The Telecommunications Act not only requires that telecommunications services be made available for resale, the Act also requires ILECs to: Not to prohibit, and not to impose unreasonable or discriminatory conditions or limitations on, the resale of such telecommunications service . . . . 47 U.S.C. § 251(C)(4)(B) (emphasis added). ILEC's refusal to resell services that--although they are not telecommunications services are related to the telecommunications services that must be resold, constitute an unreasonable and discriminatory limitation on the resale of the underlying telecommunications services. It is hard to comprehend how it could be considered reasonable for the ILECs to refuse to wholesale services to CLECs when the CLECs are willing to pay full retail rates. Since these services are not regulated, presumably the ILECs are making an acceptable profit on these services at the retail rates. Their only possible motivation for refusing to allow the services to be resold together with the telecommunications services over which they have a monopoly is that the ILECs recognize that if they give the CLECs access to their voice mail and inside wire services, the CLECs will find it easier to compete for the ILEC's customers. Any condition or limitation that makes it harder for CLECs to compete for the customers of the ILECs should be presumed to be unreasonable. ii. Separated or unseparated data. The purpose of using "separated" data is to recognize that the retail services subject to resale are largely intrastate in nature. MCI's avoided cost study addresses this fact by using ARMIS 43-04 data on state operations rather than the "subject to separations" data the FCC has used. Exh. 164 (DiTirro) at 15. Absent the use of state specific cost data, both the numerator and denominator of the discount calculation would include expenses allocated to services that will not be resold. Id. ARMIS Report 43-04 provides the necessary breakdown of the relevant costs on a state specific basis. Using this data, MCI's model results in a wholesale discount of 23.71% for U S West and 18.11% for GTE. Exh. 164 (DiTirro) at 17, Appendix I. Because ILECs will avoid some interstate costs when their services are resold, MCI's approach of using only intrastate costs actually works in favor of the ILECs. Even though ILECs will avoid interstate costs, the rate elements covering these costs (subscriber line charge and the carrier common line charge) are not discounted, which creates a windfall for the ILECs on these interstate avoided costs. Nevertheless, MCI recommends the use of intrastate data, as MCI has done in its model, because of its simplicity and the ability to use publicly available ARMIS data. Exh. 164 (DiTirro) at 16. iii. TELRIC or embedded studies. Although the FCC found that the "bottom up" TELRIC methodology was appropriate for developing UNE costs, the FCC rejected TELRIC or TSLRIC for use in calculating an avoided cost discount. The appropriate methodology for determining the wholesale rate under the FTA is a "top down" approach. This requirement is derived from the FTA itself, which defines "wholesale rates" as follows: For the purposes of section 251(c)(4), a state's commission shall determine wholesale rates on the basis retail rates charged to subscribers for the telecommunication service requested, excluding the portion thereof attributable to any marketing, billing, collection, and other costs that will be avoided by the local exchange carrier. 47 U.S.C. § 252(d)(3). Because the FTA requires the state to set the wholesale rate using the retail rate as the starting point, TSLRIC is not an appropriate method for determining a wholesale rate unless rates have been set based on TSLRIC. See FCC Order, ¶ 915. U S West's rates in Washington have not been set based on TSLRIC. Rather, the rates have been set based on a traditional rate base, rate of return method and TSLRIC was only used for purposes of determining a pricing floor. Commission Decision And Order Rejecting Tariff Revision; Requiring Refiling, Washington Utilities and Transportation Commission v. U S West Communications, Inc., at 91 (Docket No. UT-950200, April 11, 1996). Even Ms. Gude, of U S West, recognizes that U S West's rates do not contain consistent markups over TSLRIC for the various U S West services. Transcript at 2787. Accordingly, it is completely inappropriate to use a TSLRIC approach in determining a wholesale discount. MCI's witness described the numerous flaws of U S West's alternate TSLRIC avoided cost study: USWC incorrectly uses forward-looking costs on a bottom-up basis. Consequently, USWC uses costs that are different than those costs and costing methodologies used to set the rates to which the discount will be applied, in direct violation of the Act and the FCC Order. USWC does not eliminate all the costs avoided when wholesaling services and USWC proposes to add back costs it allegedly incurs in wholesaling. USWC fails to properly identify avoided indirect costs. Finally and most importantly, USWC’s TSLRIC avoided cost study is not verifiable or adequately documented. Consequently, the Commission cannot rely on USWC’s TSLRIC avoided cost study to set the avoided costs discount for Washington. Exh. 165 (DiTirro) at 2 - 3. Even U S West has recognized the legal barriers to using TSLRIC. Although U S West has submitted a TSLRIC study in this docket, it has also prepared and submitted an embedded avoided cost study. Exh. 197 (Gude). No other party in this docket has submitted a TSLRIC avoided cost study. iv. ARMIS or proprietary data. It is desirable in all costing proceedings to use nonproprietary data, such as ARMIS. Only by using nonproprietary data can the Commission look past the "black box" and ensure that the costing methodology, whatever it is, is applied consistently using verifiable input data. The Commission should not adopt a methodology such as the two proposed by U S West which can "permit manipulation, prevent verification, and result in unreliable and anticompetitive avoided cost discounts" by using proprietary data. Exh. 165 (DiTirro) at 5. U S West's embedded avoided cost study is perhaps the most extreme example of how many opportunities for manipulation can be created by using proprietary data. It is not just a black box, it is a series of black boxes. U S West starts with proprietary embedded cost data. Then U S West uses its proprietary TSLRIC embedded cost study to derive the discount factors. In another proprietary formula, U S West applies the TSLRIC factors to the embedded costs. Transcript at 2824-30. This kind of study invites manipulation of the model to result, in U S West's and GTE's avoided cost studies, in unrealistically low discount percentages that will deter competitive entry. See, e.g., Exh. 165 (DiTirro) at 3, 6. The Commission should not adopt any avoided cost method that relies on proprietary data. v. Discounted percentage based on revenues or expenses in denominator. No one disputes that the FTA requires that the discount percentage developed by this Commission should be applied to existing ILEC retail rates. Each of the studies submitted to the Commission does this in the end. The FTA does not, however, require that the calculations to determine that percentage should be determined using current revenues. Using revenues in the demonimator to calculate the wholesale discount percentage is not in the public interest, and in fact, it is only desired by the ILECs to protect them from competition. Using revenues in the denominator results in a discount that is unreasonably low and that will deter, rather than encourage, competitive entry. The intent of the FTA is not to protect ILEC's rate of return. Use of expenses in the denominator and the numerator properly determines the relative portion of the ILEC's activities that are attributable to its retail operations compared to a wholesale operation. Mr. DiTirro explained why MCI uses expenses in the denominator rather than revenues or some other factor: MCI chose to use expenses as the most accurate measure of the portion of the ILEC’s operations attributable to retailing. Since MCI looks only at expenses for avoided costs, we believe only expenses should be included in the denominator for the determination of the appropriate avoided cost discount. It should be remembered that this process is not intended to set rates. The rates for services subject to resale have already been set and those rates already have a return element intended to recover capital costs. This process is intended to established a percentage which represents the portion of the ILECs retail operations which can be avoided when wholesaling services. Study methods often use a variety of factors for such determinations. For example, one could attempt to use headcount to measure retail operations, counting employees performing retail tasks. In such an example, it would be clearly inappropriate to interject irrelevant data such a the companies rate of return for the determination of the portion of the company attributable to retailing. For this same reason, if expenses are the medium for this calculation only expenses should be used for the determination of an accurate avoided cost discount. Exh. 165 (DiTirro) at 8 - 9. The reason that ILECs and others support use of revenues in the denominator is a misguided attempt to preserve the traditional rate of return of the ILECs. See, e.g., Transcript 2124-25 (Strain Cross). The Commission, however, should not approve of any method that tries to shelter ILECs from competition and preserves return on their embedded investment. There are a number of reasons why attempting to preserve the ILECs current rate of return in adopting a wholesale discount is not an appropriate goal. First, no one can predict what the effect of resale and competition will be on the ILECs overall rate of return. Exh. 165 (DiTirro) at 19. As staff witness Strain admitted, she did not know whether, in the long term, the rate base of an ILEC would remain the same (thus requiring a constant dollar return) if the ILEC began to provide a substantial portion of its services at wholesale. Transcript 2116-17. Second, it is inappropriate to protect rates of return in a competitive environment. For example, GTE has entered the interexchange market and U S West will in the future. Incumbent IXCs, such as AT&T and MCI, do not have their return on their investments protected. It is no more appropriate to try to protect the investments of U S West and GTE from the effects of competition than it would be to attempt to protect the investments of MCI and AT&T in the interexchange market from increased competition. Third, there is no reason to believe that it is necessary to attempt the difficult task of trying to preserve a particular return on existing ratebase. Staff witness Strain used an example of the effect on return of a company that was 50% wholesale and 50% retail. Exhibit 145, PMS-2. As she admitted, the difference in rate of return on a total company basis was half the difference on rate of return shown in her example. Transcript 2119. In a more realistic example of a company that becomes a 10% wholesaler and 90% retailer, the effect in Ms. Strain's example would be a reduction on the company's overall rate of return from 10% to 9.88% or only .12% reduction. Transcript 2122-23. This difference is not even sufficient to move a company that is earning at the mid point of its authorized rate of return range out of the authorized range. Transcript 2123. It may take years before the ILECs lose as much as 10% of their customer base to reseller competitors and become 10% wholesalers. In the meantime, many factors will change. To the extent the ILECs continue to retain monopoly power, there may be future rate cases. Rates may need to be adjusted upward or downward, depending on numerous factors beyond the Commission's control. To attempt to control just one factor, by using revenues in the denominator rather than costs, is futile and unnecessary. The "threat" to return on ratebase is simply too small and remote to justify the substantial reduction in the wholesale discount and the real and immediate barrier to competitive entry that will result from using revenues in the denominator. vi. Exclusion of OS/DA and NRCs from calculation. MCI excluded operator service and directory assistance (OS/DA) expenses from the numerator but not the denominator because those costs will be avoided in a resale market. The FCC approved of MCI's treatment of the OS/DA. As Mr. Dunkle pointed out, U S West's operator and directory services are covered in part by rates from other services. Exh. 152 (Dunkel) at 69. In fact, as U S West witness Gude admitted, if U S West completely exited the operator service and directory assistance businesses altogether, U S West would avoid $24 million in costs and would lose only $12 million in revenues. Transcript at 2822. Although GTE contends that its operator service rates cover its costs in the aggregate, GTE will still realize a savings when resellers provide their own operator services. Transcript 2758-59. GTE witness Eachus agreed that GTE's operators receive a number of different types of calls that do not generate revenues. Id. GTE also receives nonrevenue directory assistance calls, because of the free calling allowance. Id. Thus, it is clear that GTE will also avoid operator service and directory assistance costs in a wholesale environment. MCI's avoided cost model appropriately excludes operator service and directory assistance costs from the numerator, because those costs will be avoided by the ILECs in a wholesale environment. vii. Separate discount for groups of services or one discount rate. MCI's model takes a "composite" approach rather than attempting to develop different discount rates for different services. There are several reasons that this method is preferable. First, the composite approach is much more easily understood. Second, and more importantly, the composite approach can be done using publicly available data, such as ARMIS reports. This is important because ILECs, using proprietary data, have attempted to understate the avoided costs. The composite approach has been broadly supported and adopted. In this docket, only U S West is advocating a service specific approach. GTE, which was advocating a service specific approach in its early arbitrations, is now proposing a single composite discount rate. Transcript at 2760. MCI witness DiTirro explained the difficulties and risks involved in attempting to develop service specific discounts: Service-by-service data are much harder to come by. Even if more detailed information were publicly available on a service-by-service basis, the consistency of the information would be questionable due to the numerous allocations and assumptions the ILEC would have to make to develop the service-specific information. Exh. 164 (DiTirro) at 11. Because of the lack of reliable, verifiable, service-specific data, as well as near consensus on a composite approach to determining a wholesale discount, the Commission should adopt a composite model, such as MCI's. b. Methodologies and Models. i. AT&T. Although MCI supports a different avoided cost model than AT&T, AT&T's model satisfies the criteria of the FTA. Importantly, AT&T's method is less prone to manipulation than U S West or GTE's models and is clearly more verifiable. Although AT&T uses revenues in the denominator, while MCI uses total expenses, AT&T properly accounts for the differences between revenues and expenses which U S West and GTE do not. Exh. 165 (DiTirro) 19-20. However, AT&T's model is more complicated. Consequently, MCI submits that its model should be adopted because it is simple, understandable, and verifiable. ii. GTE. GTE's avoided cost study suffers from five major flaws. First, GTE develops its costs using work centers in total company operations. Accordingly, it is impossible for the Commission to determine whether GTE's study relates, in any way, to GTE's costs in Washington. Exh. 165 (DiTirro) at 10. The second major flaw of GTE's avoided cost study is that it is based on the improper assumption that it will provide services to its wholesale customers in the same manner as it provides retail services. This grossly understates the costs that GTE can and will save in wholesaling services if it does so efficiently. Exh. 165 (DiTirro) at 10 - 11. The third major flaw of GTE's study is in its calculation of indirect avoided costs. GTE improperly uses total expenses in its denominator, rather than merely direct expenses. This approach inflates the denominator, resulting in an understated indirect avoided cost percentage. Exh. 165 (DiTirro) at 10 - 11. In a fourth error, GTE uses revenues in the denominator for determining the overall avoided cost percentage. As discussed above, this approach inflates the denominator and substantially understates the avoided cost discount. Expenses must be used in the denominator, as in MCI's model, in order to determine what portion of GTE's operations are retail versus wholesale. Revenues have no basis in determining the scope and nature of GTE's operations. Finally, GTE appears to suffer from an overall misconception of the avoided cost requirements of the FTA. GTE attempts to add back into wholesale prices its lost "contribution" from the services it assumes it will no longer sell to customers who migrate to resold services. Exh. 165 (DiTirro) at 11. This is contrary to Section 252(c)(3). Nothing in the FTA or in the fundamental nature of competition ensures any competitor, including an ILEC, of a continued level of contribution from services it formerly provided but that its competitor is now providing. Rather, the FTA requires that the retail rate be discounted by the portion of the rates attributable to the costs that are avoided in the wholesale market. See 47 U.S.C. § 252(c)(3). Mr. DiTirro provided a number of examples of the flaws in GTE's cost study. Exh. 165 (DiTirro) at 12-17. In one of the most glaring examples, GTE identifies only 43% of customer services expenses as avoided. Id. at 13. GTE based this on the assumption that it will not use automated OSS interfaces, which would allow resellers to process their orders directly into GTE similar to methods GTE makes available to itself for its own retail products. This results in the reseller paying these manual processing costs twice because GTE failed to provide it with equitable treatment. Id. Thus, GTE not only discriminates against CLECs in the provision of OSS interfaces, in effect it also proposes to charge the CLECs extra for its anticompetitive and discriminatory treatment. GTE's avoided cost study contains too many fundamental flaws to be adopted, even with adjustments. The Commission should reject GTE's model in favor of MCI's. iii. MCI. MCI submits that its avoided cost model is superior to the other models submitted in this docket. First, MCI's cost model complies with Section 252(d)(3) of the FTA. Exh. 164 (DiTirro) at 5. Second, the MCI model properly employs a top down study that is based on the same costs employed to set retail rates. Id. at 5 - 6. Third, MCI's avoided cost study eliminates all the retailing costs which reasonably can be avoided when the ILEC wholesales services. Id. This ensures that new entrants are not saddled with duplicative retailing costs and maximizes the potential for competitive entry. Fourth, MCI's study identifies both direct and indirect costs that will be avoided. Id. Finally, MCI's cost study is the most verifiable and understandable of the studies submitted. MCI's study is based on publicly available reports of ILECs under FCC, Part 32 Uniform Systems of Accounts (USOA). Id. The data MCI used is not only publicly available and verifiable, the data is specific to the state of Washington. Id. at 11. MCI's study is conservative in many respects. It does not treat all expenses in certain categories as avoidable, in order to account for the fact that some marketing and sales expenses will continue to be incurred even in the wholesale environment. See Id. at 9. Thus, for example, MCI treats only 90% of marketing expenses in USOA account 6610 as avoided. Id. MCI properly uses expenses in the denominator. This ensures that a mathematically correct ratio is obtained that compares the portions of the ILEC's operations that are devoted to retailing versus wholesaling. Exh. 165 (DiTirro) at 17 - 18. This also avoids the problem of building in a rate of return element or rate of return guarantee into wholesale rates, as other parties' models attempt to do. Such a guarantee of a rate of return is not necessary, and is not possible in any event, as the market moves toward effective competitiveness. iv. Public Counsel. Public Counsel witness Dunkel developed a model based on MCI's model. Dunkel[[[ at 53. However, Mr. Dunkel made certain inappropriate changes to the MCI model. The most significant error, in terms of impact on the final discount factor, is his use of revenues in the denominator. This accounts for nearly all the difference between Mr. Dunkel's final percentage discount and MCI's. The reason for this is that although Mr. Dunkel made a number of changes that affected the numerator, the end result was nearly the same. Rounded, both Mr. Dunkel and Mr. DiTirro determined the numerator for U S West of $191 million. Transcript at 2165. With the denominator, however, Mr. Dunkel makes the same mistake as the Commission Staff of attempting to ensure that the ILEC enjoys the same level of profits whether it wholesales or retails its services. This approach would actually give ILECs an increased profit margin on resold services. Exh. 165 (DiTirro) at 18. Further, this approach is ratemaking from a revenue requirement perspective and not an accurate determination of the avoided cost discount. Id. Mr. Dunkel also makes some errors in his testimony regarding OS/DA cost and revenues. For example, he contends that MCI errs in leaving OS/DA revenues in its avoided cost study. Exh. 152 (Dunkel) at 67-68. In fact, MCI does not use revenues at all in its cost study, only expenses in both the numerator and the denominator. Exh. 164 (DiTirro), App. I. However, Mr. Dunkel does correctly recognize that some OS/DA costs are recovered from other services. See Exh. 152 (Dunkel) at 69. While the Public Counsel avoided cost study certainly has some merit, it suffers from a fundamental flaw of using revenues in the denominator. Accordingly, the Commission should reject the Public Counsel model in favor of the MCI model,