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 U S WEST COMMUNICATIONS, INC. ) ) ) In the Matter of the Pricing Proceeding ) DOCKET NO. UT-960371 for Interconnection, Unbundled Elements ) Transport and Termination, and Resale ) for GTE NORTHWEST INCORPORATED ) ) RESPONSE TESTIMONY OF REX KNOWLES NEXTLINK WASHINGTON, INC. August 20, 1998 Response Testimony of Rex Knowles Page 1 Q. PLEASE STATE YOUR NAME, EMPLOYER, AND BUSINESS ADDRESS. A. My name is Rex Knowles. I am a Director Regulatory and External Affairs for NEXTLINK Management Services, Inc., 111 East Broadway, Suite 1000, Salt Lake City, Utah 84111. I. BACKGROUND Q. PLEASE IDENTIFY AND DESCRIBE THE PARTY ON WHOSE BEHALF YOU ARE TESTIFYING. A. I am testifying on behalf of NEXTLINK Washington, Inc. ("NEXTLINK"), an affiliate of NEXTLINK Management Services, Inc. NEXTLINK is a competitive local exchange company ("CLEC") that provides facilities-based local and long distance telecommunications services in Washington in competition with U S WEST Communications, Inc. ("U S WEST") and anticipates competing with GTE Northwest Incorporated ("GTE"). Q. WHAT ARE YOUR RESPONSIBILITIES WITH NEXTLINK MANAGEMENT SERVICES, INC.? A. I am responsible for all regulatory, legislative, municipal, and incumbent local exchange carrier ("ILEC") initiatives in several states in the U S WEST region, as well as in Nevada. Q. WHAT IS YOUR BUSINESS AND EDUCATION BACKGROUND? A. I graduated from Portland State University in Portland, Oregon, with a degree in Business Administration/Finance Law in 1989. I was employed by United Telephone of the Northwest from 1989 to 1993 as a regulatory staff assistant and product manager responsible for incremental cost studies and creation and implementation of extended area service ("EAS") and 911. From 1993 to 1996, I was employed by Central Telephone of Nevada as manager of revenue planning and research and was responsible for supervising cost study preparation and developing and implementing regulatory reform, including opening the local exchange market to competition and alternative forms of regulation for ILECs. I joined NEXTLINK Management Services, Inc., in my current position in the Spring of 1996. Q. HAVE YOU PREVIOUSLY TESTIFIED IN OTHER REGULATORY PROCEEDINGS? A. Yes, I have provided testimony on costing and pricing issues in a proceeding before the Utah Public Service Commission determining rates for access to, and interconnection with, U S WEST's network as required by the Telecommunications Act of 1996 ("Act") and Utah state law. Q. WHAT IS THE PURPOSE OF YOUR TESTIMONY? A. The purpose of my testimony is to address NEXTLINK's position and proposals on three issues raised in the direct testimony filed in Phase II of this proceeding: (1) Pricing of unbundled network elements, specifically the need to adopt rates for unbundled loops based on geographically deaveraged loop costs; (2) Cost recovery for interim local number portability that complies with the FCC's local number portability orders; and (3) The impact of GTE's proposed "Interim Universal Service Charge" and "Competition Transition Charge," which would eliminate any realistic opportunity for a CLEC to provide local exchange service in competition with GTE. I also discuss the issue of terms and conditions included in the compliance tariffs filed or to be filed by U S WEST and GTE. II. GEOGRAPHIC DEAVERAGING Q. WHAT IS NEXTLINK'S POSITION ON HOW UNBUNDLED ELEMENTS SHOULD BE PRICED IN THIS PROCEEDING? A. NEXTLINK agrees with the testimony filed by William Page Montgomery on behalf of AT&T Local Services on how the prices for unbundled network elements should be developed in this proceeding. NEXTLINK also agrees with Mr. Montgomery and other witnesses that costs calculated based on averages are problematic. Rather than focusing on averaging cost model outputs, however, NEXTLINK's primary concern with averaging is the Commission's decision not to geographically deaverage costs, particularly loop costs. In paragraph 274 of its Eighth Supplemental Order, the Commission chose "not to deaverage UNE and interconnection rates at this time," agreeing with "parties who argue that it is more appropriate to consider this issue in the context of universal service reform, deaveraged retail prices, and the extent of competitive activity in Washington State." This decision precludes the Commission from establishing rates for unbundled loops consistent with the Act's pricing standards. Q. WHY WOULD A SINGLE UNBUNDLED LOOP RATE BASED ON A STATEWIDE AVERAGE COST BE INCONSISTENT WITH THE ACT'S PRICING REQUIREMENTS? A. Section 252(d)(1) of the Act states that costs for interconnection and unbundled network elements "(A) shall be -- (i) based on the cost (determined without reference to a rate-of-return or other rate-based proceeding) of providing the interconnection or network element (whichever is applicable), and (ii) nondiscriminatory, and (B) may include a reasonable profit." A single unbundled loop rate based on a statewide averaged cost would not be cost-based, would be discriminatory, and would not include a reasonable profit. Q. WITH RESPECT TO THE FIRST PRICING REQUIREMENT, WHY WOULD A SINGLE UNBUNDLED LOOP RATE NOT BE BASED ON COST IF IT IS BASED ON A STATEWIDE AVERAGE COST? A. As far as I am aware, no party in this proceeding has disputed that loop costs vary by geographic area. Indeed, the cost models submitted in this proceeding and on which the Commission based its statewide average cost are designed to develop costs for discrete geographic areas, whether those areas are wire centers, census block groups, etc. The cost estimates developed by those models demonstrate a vast disparity between the cost of a loop in an urban area and the cost in a rural area. The Hatfield Model sponsored by AT&T and MCI, for example, segregates UNE costs into nine areas, and according to Mr. Klick's testimony in Phase I, calculates loop costs for U S WEST in Washington ranging from $5.59 in the most densely populated area to $68.29 in the least populous regions. A single, statewide loop "cost" based on these figures represents an average, which means that the statewide loop cost exceeds the cost for urban loops but is less than the cost for rural loops. Stated differently, consumers and CLECs who obtain loops in Seattle pay extra for those loops as a subsidy to those obtaining loops in sparsely populated areas who pay the same rate despite the higher cost. Accordingly, a CLEC obtaining loops in Seattle at a price based on a statewide averaged cost of $17.00 per loop is not paying a rate based on the $5.59 that it costs U S WEST to provide each of those loops. Q. BUT DOESN'T ESTIMATING COSTS FOR LOOPS AND OTHER TELECOMMUNICATIONS FACILITIES AND SERVICES NECESSARILY REQUIRE AVERAGING? A. Yes, but only to the extent necessary to obtain reasonably accurate data and not on a statewide basis. Optimally, a company would charge a customer only for the specific costs the company incurs to serve that particular customer, but calculating those costs would be impractical to say the least. Larger ILECs like U S WEST and GTE, therefore, have long used models that estimate costs based on defined geographic areas with common characteristics -- most prominently population density -- at a level of detail or "granularity" sufficient to produce reliable results. The Hatfield Model and the BCPM model estimate costs using a similar approach. Costs within each type of geographic area are averaged to establish a uniform cost for that area, but the common characteristics of each type of area limit the extent to which averaging is necessary. An office building in Bellingham, for example, is likely to be comparable in terms of service costs to an office building in Spokane, just as the costs to serve a large orchard outside Wenatchee and a timber farm near Forks may be roughly comparable. A statewide average, on the other hand, represents a uniform rate for geographic areas that have nothing in common except a location within Washington. The office building in Spokane is entirely different than the timber farm on the Olympic Peninsula, yet a statewide average assumes that the costs are the same. Stated arithmetically, a loop that costs $5.59 in an urban area can be averaged with $68.29 loop in a rural area, but the resulting averaged cost of $36.94 does not reflect the cost of providing the loop in either area. All else being equal, the less costs are averaged, the greater the accuracy of the resulting cost estimate. The record evidence would enable the Commission to establish geographically deaveraged loop costs, and those costs -- not the statewide average costs -- represent the "cost . . . of providing the . . . network element" as required in Section 252(d)(1)(A)(i). Q. WITH RESPECT TO THE SECOND PRICING REQUIREMENT IN SECTION 252(d)(1), WHY WOULD A SINGLE UNBUNDLED LOOP RATE BASED ON A STATEWIDE AVERAGE COST BE DISCRIMINATORY? A. The FCC's authority over pricing of unbundled network elements remains at issue before the Supreme Court, but the FCC has concluded that "nondiscriminatory," as that term is "used throughout section 251, applies to the terms and conditions an incumbent LEC imposes on third parties, as well as on itself." In re Implementation of the Local Competition Provisions in the Telecommunications Act of 1996, CC Docket No. 96-98, FCC 96-325, First Report and Order ¶ 218 (Aug. 8, 1996) ("Local Competition Order") (emphasis added). Section 251(c)(3), like Section 252(d)(1), requires that ILECs provide unbundled network elements at rates that are "nondiscriminatory." If U S WEST incurs a cost of $5.59 to provide a loop in Seattle, the nondiscrimination provisions of the Act require that U S WEST not impose any higher costs on a CLEC. Imposing costs of $17.00 on a CLEC for a loop that costs $5.59 would be discriminatory as that term has been interpreted by the FCC. Q. BUT DOESN'T U S WEST ALSO INCUR THE SAME STATEWIDE AVERAGED COST TO PROVIDE LOOPS ON WHICH THE COMMISSION'S SINGLE LOOP RATE WOULD BE BASED? A. No more than a large real estate developer with holdings throughout the state incurs the same average cost to obtain an acre of land in downtown Seattle as an acre of range land in Eastern Washington. One could average real estate prices on a statewide basis, but that figure would not reflect the cost of land in any one particular area, even though the developer obtains property in a variety of areas. Similarly, the average of U S WEST's loop costs throughout the entire state is not equivalent to U S WEST's costs to provide a loop in any one geographic area, any more than an average of U S WEST's costs to provide local service to its customers throughout the entire state is equivalent to U S WEST's costs to provide service in a particular geographic area. The entire basis of universal service funding is a recognition that local service costs vary by population density. Section 254 of the Act and Docket No. UT-980311 would be useless if Congress and the Commission did not have some confidence that costs can be measured on a sufficiently accurate basis to determine those geographic areas where the cost to provide basic local exchange service is sufficiently higher than costs in other areas to require that customers in those "high cost" areas need financial support to ensure affordable rates. Loop costs are part of these local service costs, and they can and should be disaggregated to form the basis of geographically deaveraged rates that reflect the costs of the loops provided, not an average of all loop costs in Washington. To the extent that the Commission finds that U S WEST or GTE need, and are entitled to, a subsidy to support provisioning loops in more remote areas of the state, that subsidy should be explicit support from a properly structured universal service fund, not an implicit subsidy contained in the rates the ILECs charge to competitors. Q. WHY ISN'T THE DIFFERENCE BETWEEN AVERAGED AND DEAVERAGED COST A "REASONABLE PROFIT," AS AUTHORIZED BY SECTION 252(d)(1)(B)? A. A mark-up of almost $11.50 over the cost of a $5.59 loop -- over 200% -- is not a "reasonable profit." Q. WHY SHOULDN'T THIS ISSUE BE ADDRESSED IN THE CONTEXT OF "UNIVERSAL SERVICE, DEAVERAGED RETAIL PRICES AND THE EXTENT OF COMPETITIVE ACTIVITY," AS STATED BY THE COMMISSION? A. Section 252(d)(1) requires that unbundled network elements be priced based on their cost with an option to include a reasonable profit without regard to any other issue, including those specified by the Commission. Section 254 establishes the principles for universal service reform, including the requirements in subsection (b)(4) that providers "make an equitable and nondiscriminatory contribution to the preservation and advancement of universal service," and in subsection (e) that universal service subsidies be explicit, rather than implicit in the rates charged to competitors and consumers. The Commission's decision not to deaverage loop prices impacts universal service, as I have explained in testimony filed in Docket No. UT-980311(a), but the issue of whether or not to geographically deaverage loop costs need not and should not be made in the context of universal service reform. Unlike this proceeding, the focus in establishing revised universal service funding is on compliance with Section 254 and state universal service principles, not appropriate cost-based pricing of unbundled network elements under Section 252(d)(1). The second issue the Commission tied to loop cost deaveraging is geographically deaveraged retail service prices, which involves appropriate rate setting, not unbundled network element cost-based pricing. The ILECs and the Commission have established statewide averaged retail rates for local exchange services as a result of past rate cases, and geographic deaveraging of those rates is less an issue of underlying service costs than of policy and rate of return principles. Section 252(d)(1)(A)(i) states that the cost of unbundled network elements must be "determined without reference to a rate-of-return or other rate-based proceeding," precluding the Commission from conditioning proper unbundled network element prices on the outcome of a retail rate case. Geographically deaveraging loop prices, moreover, would have a negligible practical effect on ILEC rate issues, and certainly would not, in isolation, require geographically deaveraged retail rates. U S WEST and GTE provide no more than an handful of loops in Washington -- based on information U S WEST has provided in other states, I understand that U S WEST provides approximately 2,500 unbundled loops throughout its entire 14 state region, representing 0.0017% of the 14 million access lines U S WEST serves -- making the financial impact of geographic loop deaveraging de minimus on the ILECs. These numbers are likely to remain small in light of the cost and delays associated with access (i.e., collocation and related issues), ordering, and provisioning loops -- issues entirely unrelated to the recurring price. It is not worth the effort to overcome these other issues, however, if the loop price exceeds or approaches the ILEC's retail rates, as will a geographically averaged loop rate. An analysis of the "extent of competitive activity," the Commission's third issue, is similarly not a prerequisite to geographically deaveraging loop rates. Loop rates that substantially exceed their geographically deaveraged cost will have a significant dampening effect on competition in Washington. The Commission does not need a lengthy inquiry to come to the common sense conclusion that companies will not enter a market or expand their existing limited entry if the costs imposed on them exceed the revenues they can reasonably hope to generate. It should be obvious that competitors have a minimal share of the local exchange market in this state and that the Commission needs to do what it can to facilitate greater and more widespread competitive entry if it hopes to bring the benefits of effective competitive choice to Washington consumers. The proposal to tie geographically deaveraged loop prices to resolving universal service funding, deaveraging retail rates, and exploring the "extent of competitive activity" thus is nothing more than a red herring that serves only to delay the advent of effective competition. Indeed, I understand that the cost issues being resolved in Docket No. UT-980311(a) will result only in a report and proposal to the Legislature (as opposed to the establishment of an alternative universal service fund), that only the ILECs and the Commission can initiate a proceeding to rebalance retail rates, and that only the Commission can formally investigate the "extent of competitive activity" in Washington. I am not aware of any dockets that have been opened or any proceeding in which the Commission will be conducting an inquiry or otherwise considering the issues the Commission apparently believes are related to geographic deaveraging of loop costs, much less any proceeding other than this one in which unbundled loop prices will be established. The Commission's decision not to geographically deaverage loop costs "at this time," therefore, means that any unbundled loop rate based on a statewide averaged loop cost compels CLECs for the foreseeable future to vastly overcompensate U S WEST and GTE for the overwhelming majority, if not all, unbundled loops obtained in Washington. Q. WHAT IMPACT DOES THE COMMISSION'S PRELIMINARY DECISION NOT TO GEOGRAPHICALLY DEAVERAGE LOOP COSTS HAVE ON THE ABILITY OF CLECS TO PROVIDE SERVICE TO WASHINGTON CONSUMERS? A. Most, if not all, loops the CLECs obtain from ILECs, at least initially, are likely to be in the most densely populated geographic areas where the loop costs are the lowest. Indeed, Hatfield Model cost estimates in six of the nine geographic areas in Washington are significantly lower than the $17.00 statewide averaged loop cost found by the Commission in its Eighth Supplemental Order. Even if the Commission established the loop rate for U S WEST at its $17.00 averaged cost, that price alone exceeds the rate for residential exchange service ($16.00, including End User Common Line) and is almost half the rate for basic business exchange service ($34.56, including EUCL), with comparable numbers applicable to GTE. Nor can the recurring cost of the loop be viewed in isolation. U S WEST and GTE have proposed high nonrecurring charges, including loop "conditioning" and testing. CLECs also must incur costs to collocate in the ILEC central office to access unbundled loops, including paying a recurring charge for Expanded Interconnection Channel Terminations ("EICTs") to connect each loop to the CLEC's collocated equipment. A $17.00 recurring charge when viewed in conjunction with these other charges makes use of ILEC unbundled loops to provide service in Washington very unattractive, resulting in a strong disincentive to invest limited resources in this state. Other states in which NEXTLINK affiliates operate have established much lower loop rates, generally as a result of geographic deaveraging. For example, the geographically deaveraged loop rates in Illinois are $3.72, $10.02, and $11.53, while in Ohio the rates are $9.19, $14.05, and $16.27, and in Pennsylvania the loop rates are $11.93 and $16.53. Not coincidentally, these states are also experiencing much higher usage of unbundled loops for entry into the local exchange market. NEXTLINK certainly would like to see Washington join these states in facilitating the development of effective local exchange competition, but such competition and its attendant consumer benefits will not develop in the foreseeable future if the sole economically viable means of entry is by using only facilities the CLEC has constructed itself. Q. WHAT PRICE DO YOU RECOMMEND THAT THE COMMISSION ESTABLISH FOR UNBUNDLED LOOPS? A. I recommend that the Commission establish at least three different rates for unbundled loops for each ILEC based on Commission-approved geographically deaveraged TELRIC estimates for each carrier. If the Commission continues to decline to geographically deaverage loop costs and prices in this proceeding, the Commission should establish interim loop rates at each ILEC's TELRIC subject to true-up to geographically deaverged loop prices established in a docket the Commission should establish to address geographic deaveraging of loop costs and prices. III. INTERIM LOCAL NUMBER PORTABILITY Q. WHAT IS NEXTLINK'S POSITION ON COST RECOVERY FOR INTERIM LOCAL NUMBER PORTABILITY? A. NEXTLINK agrees with Mr. Montgomery's testimony that the costs of interim local number portability should be recovered consistent with the way in which the costs of permanent local number portability will be recovered. This issue is of particular importance for NEXTLINK because it currently provides service in Spokane, which unlike the greater Seattle and Portland areas, is not among the 100 most populous Metropolitan Statistical Areas ("MSAs") and thus is not scheduled for deployment of permanent number portability until after the initial 100 MSAs have been converted. Accordingly, NEXTLINK and its customers in Eastern Washington will be dependent on interim local number portability for the foreseeable future, and the cost recovery mechanism the Commission establishes for interim local number portability will be in effect for some time in NEXTLINK's service area. Q. DO NEXTLINK'S INTERCONNECTION AGREEMENTS ADDRESS THIS ISSUE? A. NEXTLINK does not yet have an interconnection agreement with GTE, but NEXTLINK's agreement with U S WEST provides in Section IV.A.2., "The costs incurred by NEXTLINK and USWC of providing INP shall be recovered through a broad-based cost recovery mechanism, as described in the FCC Number Portability Order. The costs shall be apportioned on the basis of active local numbers." This provision thus is consistent with the proposal advanced by TCG Seattle (which has the same agreement with U S WEST). My understanding is that this language was negotiated between U S WEST and TCG Seattle for inclusion in the agreement should the Commission conclude -- as it did -- that it would comply with the FCC Number Portability Order, and this provision was upheld by the federal district court against a challenge by U S WEST. To the extent that the Commission continues to comply with that order, it should adopt the TCG Seattle proposal. Q. DOESN'T THAT PROPOSAL REQUIRE THAT U S WEST AND GTE INCUR MOST OF THE COSTS OF INTERIM LOCAL NUMBER PORTABILITY? A. No. As I understand the proposal, each carrier would be responsible for recovery of its interim local number portability costs from its end-user customers, just as will be the case when permanent local number portability is deployed. U S WEST and GTE thus would be precluded from imposing all interim local number portability costs on CLECs, as the FCC has required, and subject to Commission approval, the ILECs can develop their own mechanism for recovering those costs from their retail customers. Q. WHAT ABOUT ACCESS CHARGES TO CALLS MADE TO PORTED NUMBERS? A. The FCC requires that the carriers share access charges in those circumstances, but the FCC did not establish a specific sharing mechanism. I am not aware that U S WEST has accounted for access charges it has collected (or imputed when acting as the toll provider) for calls made to numbers it ports to NEXTLINK, and U S WEST has yet to respond to a NEXTLINK data request asking for this information. In Utah, however, U S WEST responded to a similar data request by saying that it does not track interexchange minutes terminated to ported numbers and has not remitted any access charges to CLECs (although providing a proposal to do so). GTE responded to NEXTLINK's data request in this proceeding stating that it currently ports 968 numbers to CLECs, and although it does not track interexchange calls made to those numbers, GTE proposes to remit to the CLECs 85% of the access charges based on GTE's average monthly terminating access revenues per line. GTE's practice appears to be reasonable -- if "revenues" includes imputation of revenues when GTE is the toll provider -- given the relatively small amount of numbers ported and the short-term nature of interim local number portability. Q. HOW DOES NEXTLINK RECOMMEND THE COMMISSION ESTABLISH INTERIM LOCAL NUMBER PORTABILITY COST RECOVERY IN THIS PROCEEDING? A. NEXTLINK recommends that the Commission adopt the TCG Seattle proposal and that the Commission require both U S WEST and GTE to remit 85% of all access charges collected or imputed on toll calls made to ported numbers, based on average monthly terminating revenues (including imputed revenues) per line, if that traffic cannot be measured directly. In addition, I understand that U S WEST and GTE were authorized to charge CLECs for interim local number portability on an interim basis under tariffs they have filed, subject to true-up based on the outcome of further proceedings. Accordingly, U S WEST and GTE should be required to refund any charges -- both recurring and nonrecurring -- they have imposed on CLECs for interim local number portability except for the amounts for which the CLEC is responsible based on its total of active telephone numbers. U S WEST and GTE should also remit 85% of all access charge revenues they have collected or imputed on toll calls made to ported numbers since their interim tariffs went into effect according to the formula I have recommended. IV. GTE SURCHARGE PROPOSALS Q. WHAT IS NEXTLINK'S POSITION ON GTE'S PROPOSALS FOR A "COMPETITIVE TRANSITION CHARGE" AND "INTERIM UNIVERSAL SERVICE CHARGE"? A. NEXTLINK strongly opposes GTE's proposals for an "Interim Universal Service Charge" ("IUSF") and "Competitive Transition Charge" ("CTC"), which would be more appropriately called "Competition Termination Charges." Other witnesses will address these proposals from an economic perspective, but my testimony focuses on the impact these charges would have on the ability of a CLEC to provide local service in Washington. Q. WHAT WOULD BE THE IMPACT OF GTE'S PROPOSALS ON CLECS? A. Adoption of GTE's CTC and ISUF charges would prevent competition from developing in GTE's service territories in Washington. As I understand the CTC proposal explained on page 43 of the Direct Testimony of Michael J. Doane, GTE calculates the CTC as the difference between its costs and its average revenue per line (from all sources, including access charges, toll, vertical features, etc.). This charge apparently would be levied on all competitors, regardless of whether they use any GTE facilities to provide local service. Mr. Doane contends that "an equally efficient CLEC could profitably enter by matching GTE's current retail rates" and a "CLEC that was more efficient could undercut GTE's current retail rate and profitably enter." I fail to see how any CLEC could profitably enter under these circumstances or would even try. In Mr. Doane's example, a CLEC would pay GTE $22.86 per month for the unbundled network elements and a CTC of $45.83 for the opportunity to generate GTE's average revenues of $73.51 -- a difference of $4.82, which represents GTE's retailing costs. The equally efficient CLEC incurs the same $4.82 in retailing costs and so would spend $73.51 to generate revenues of $73.51. I do not know what definition of "profit" Mr. Doane was using, but R. Kirk Lee, also testifying on behalf of GTE, states on page 12 of his direct testimony, "A firm can never earn a profit unless it first recovers its actual costs." (Emphasis added.) I see no opportunity to make a "profit," as that term is commonly understood, in the situation of the equally efficient CLEC whose revenues are equal to its costs. As for the more efficient CLEC, the only margin available for "profit" is to reduce the $4.82 in retailing costs since all other costs are fixed (in fact, paid to GTE) -- which represents less than a 6.6% return, even if the CLEC could eliminate all retailing costs (which would be impossible), far lower than GTE's authorized rate of return. Perhaps as an academician, Mr. Doane considers the opportunity to be twice as efficient as GTE and earn a 3.3% return on investment "profitable entry." As a business person operating in the real world, I do not, and neither do I believe would GTE, if presented with the same opportunity. No rational investor would accept such a return on investment when it could earn almost 3% in a passbook savings account and over 5% in a certificate of deposit without any risk. A wholly facilities-based carrier could potentially increase the margin in GTE's scenario by providing its own network rather than paying GTE $22.86 to use unbundled network elements. However, only a CLEC that could incur significantly lower costs to build its network -- including indirect and common costs -- could even hope to enter the market in GTE's service territories, and then would have to be willing to pay GTE over 62% of the CLEC's revenues just for the privilege of providing service in GTE's territory. Mr. Doane also totally ignores GTE's IUSF proposal when concluding that imposition of GTE's CTC would allow "profitable entry." Mr. Lee on page 26 of his direct testimony on behalf of GTE proposes a monthly IUSF of $37.17 per line served by a CLEC, regardless of whether that line is provided using a GTE unbundled loop or the CLEC's own facilities exclusively. This figure, like the CTC, is calculated as the difference between GTE's average revenue per line and the costs of GTE unbundled network elements needed to provide local service. The only difference between the IUSF charge and the CTC appears to be that Mr. Lee and Mr. Doane assume different costs for the unbundled elements. GTE's proposed IUSF charge, therefore, would have the same impact on competitive entry as the CTC. Imposition of both the IUSF charge and the CTC would be like using a bulldozer to bury a hamster. The total of GTE's proposed CTC ($45.83) and IUSF charge ($37.17) is $83.00, which substantially exceeds GTE's average revenue per line figure of $73.51, before factoring in the CLEC's costs to actually provide the service. Even Mr. Doane must realize that a CLEC, regardless of its efficiency, cannot recover its costs -- much less make a profit -- if it must pay GTE more for the opportunity to provide service in its territory than the revenues that service will generate. The only opportunity GTE's CTC and IUSF provide is a guarantee t