ORDER NO. 98-444

ENTERED NOV 13 1998

This is an electronic copy. It does NOT contain the Appendices or Footnotes. The hard copy of this document contains over 200 footnotes.  This version also does not contain Commissioner Smith's concurrence.

OF OREGON

UT 138

UT 139

In the Matter of the Investigation into )

Compliance Tariffs filed by U S WEST )

Communications, Inc., Advice Nos. 1661, )

1683, 1685, and 1690. )

ORDER

In the Matter of the Investigation into )

Compliance Tariffs filed by GTE Northwest )

Incorporated, Advice Nos. 589, 599, and 6ll. )

I. EXECUTIVE SUMMARY *

II. PROCEDURAL HISTORY *

III. ARE INCUMBENT LOCAL EXCHANGE CARRIERS REQUIRED TO COMBINE BUILDING BLOCKS? *

A. Telecommunications Act of 1996. *

B. Party Positions. *

C. Commission Decision. *

IV. BUILDING BLOCK ACCESS PROPOSALS *

A. USWC and GTE Proposals. *

B. Response to GTE and USWC Access Proposals. *

1. Cost Concerns. *

2. Space Availability. *

3. Loss of Service while Collocation Arrangements Are Established. *

4. Loss of Service During Cut-Over of Facilities. *

5. Additional Connections/Potential Points of Failure. *

6. Maintenance and Provisioning Difficulties. *

7. Security Concerns. *

8. Alternative Building Block Access Arrangements. *

C. Commission Decision -- Building Block Access Proposals. *

1. Background. *

2. Mandatory Collocation. *

3. USWC’s SPOT Frame Proposal. *

4. Direct Access. *

V. BUILDING BLOCK ACCESS ISSUES *

A. Under What Circumstances May Carriers Purchase Building Blocks From Tariff? (ISSUE II) *

1. Is an Interconnection Agreement Required Before a Carrier May Purchase Building Blocks? *

2. May a Carrier Purchase Building Blocks Under the Tariffs if it has Executed an Interconnection Agreement With the ILEC? *

B. May a Carrier Purchase ILEC Retail Services and Combine Those Services With Building Blocks? (Issue IV (D)) *

C. May Carriers Use Building Blocks to Provide Switched Access Service? (Issue IV(B)) *

D. Virtual Collocation. *

E. Common (Shared) Transport Building Block. *

F. Elimination of Building Blocks. *

G. USWC Cageless Collocation Proposal. *

H. Integrated Digital Loop Carrier (IDLC). *

VI. NONRECURRING COSTS *

A. Recurring vs. Nonrecurring Costs. *

B. Nonrecurring Costs vs. Competition Onset Costs. *

C. TSLRIC Methodology. *

D. Cost Principle No. 2 -- Choice of Technology. *

E. Nonrecurring Cost Studies. *

1. USWC Studies. *

2. GTE Study. *

3. Staff Study. *

4. AT&T/MCI Study. *

VII. NONRECURRING COST STUDIES -- DISPUTED ISSUES *

A. Service Order Processing Costs. *

1. Flow through and Fallout. *

2. AT&T/MCI *

3. USWC and GTE. *

4. Staff. *

5. Commission Decision -- Service Order Processing Costs. *

B. Provisioning Processes. *

1. POTS vs. Design Services. *

2. USWC. *

3. Joint Intervenors. *

4. Commission Decision -- Provisioning Processes. *

C. Integrated Digital Loop Carrier (IDLC). *

1. Party Positions. *

2. Commission Decision -- IDLC. *

D. Labor Times and Probabilities. *

1. USWC. *

2. AT&T, MCI and GTE. *

3. Staff. *

4. Commission Decision -- Labor Times and Probabilities. *

E. Jumper Activity Times. *

1. Party Positions. *

2. Commission Decision -- Jumper Activity Times. *

F. COSMIC Frames. *

1. Party Positions. *

2. Commission Decision -- COSMIC Frames. *

G. Disconnection Activities. *

1. Party Positions. *

2. Commission Decision -- Disconnection Activities. *

H. Dedicated Inside Plant (DIP) and Dedicated Outside Plant (DOP). *

1. Party Positions. *

2. Commission Decision -- DIP/DOP. *

I. Number of Work Activities per Visit. *

1. Party Positions. *

2. Commission Decision -- Work Activities per Visit. *

J. Loop Unloading Activities. *

1. Party Positions. *

2. Commission Decision -- Loop Unloading. *

K. GTE Outside Facilities Connection Charge. *

L. Testing Activities. *

1. USWC and GTE Proposed Tariffs. *

2. Staff. *

3. USWC. *

4. GTE. *

5. Joint Intervenors. *

6. Commission Decision -- Testing Activities. *

M. Staff Adjustments for Circuit Provisioning Functions. *

1. Party Positions. *

2. Commission Decision -- Circuit Provisioning Functions. *

N. Nonrecurring Cost Markup. *

1. Party Positions. *

2. Commission Decision -- Nonrecurring Cost Markup. *

VIII. TARIFF TERMS AND CONDITONS *

A. Customer Letters of Authorization. (Issue V.G) *

1. Party Positions. *

2. Commission Decision -- Letters of Authorization. *

B. Special Construction Charges. (Issue VII) *

1. Party Positions. *

2. Commission Decision -- Special Construction Charges. *

C. Tariff Terminology. (Issue VIII.A) *

D. NAC Capabilities and Provisioning. (Issue VIII.B) *

E. NAC Conditioning. (ISSUE VIII.D) *

F. Network Interface Device (NID). (Issue VIII.E) *

G. NACC Definition. (Issue VIII.H) *

H. LSR Limitations. *

I. Bona Fide Request Process. *

J. Recovery of Nonrecurring Costs. *

K. Other GTE Tariff Issues. *

1. Section 2.4 Paragraph 1. *

2. Section 2.4 Paragraph 2. *

3. Section 2.9. *

4. Section 2.10. *

L. Other USWC Tariff Issues. *

1. Section 2.2 *

2. Section 2.7.C *

3. Section 4.4 *

4. Section 6.1.B *

5. Section 6.2.A and 6.2.C *

6. Section 6.2.A.3 *

7. Miscellaneous USWC Tariff Sections. *

In 1990, the Commission initiated a proceeding to unbundle and reprice services provided by incumbent telecommunications utilities, including U S WEST Communications, Inc. (USWC) and GTE Northwest Incorporated (GTE). At the time of that decision, interexchange carriers and access providers were beginning to make competitive inroads and capture customers from the utilities. We determined that unbundling, uniform pricing, and nondiscriminatory availability of building blocks were necessary to implement a balanced program of regulation and competition as envisioned by the Oregon Legislature. Subsequently, Congress passed the Telecommunications Act of 1996, which, among other things, also requires incumbent telecommunications utilities to unbundle their services and provide competing carriers with nondiscriminatory access to network elements.

In this order the Commission examines tariffs filed by USWC and GTE to govern how building blocks will be supplied to competing telecommunications carriers. As a practical matter, the terms and conditions in these tariffs will determine whether local exchange competition will ever exist in this state. If the conditions under which competing carriers obtain access to building blocks are too burdensome, local competition will languish and the goals articulated by the 1996 Act and the Oregon Legislature will go unfulfilled. In this complex and detailed undertaking, our task has been to examine proposed building block access arrangements and specify tariff terms and conditions that eliminate roadblocks to entry without hindering the ability of the incumbent carriers to meet the challenges posed by a competitive local exchange market.

Consistent with recent decisions of the Eighth Circuit Court of Appeals, we reluctantly agree that USWC and GTE should not be required to combine unbundled elements for competing carriers. Although we believe the Court’s decision will produce building block access arrangements that are less efficient than those contemplated by this Commission, we are persuaded that abiding by the Court’s ruling is the only way to allow competition to move forward while these issues are under consideration by the Supreme Court of the United States.

At the same time, we reject USWC and GTE proposals which would force competitors to collocate and utilize unnecessary facilities to buy building blocks from the incumbent utilities. To the extent that USWC and GTE choose not to combine building blocks on behalf of competing carriers, we find that the competitors must have direct access to the incumbent carrier’s facilities in order to combine building blocks themselves. We find that direct access is necessary to provide competing carriers with nondiscriminatory access to building blocks and to allow local exchange competition to succeed.

In addition to building block access arrangements, we also examine a variety of upfront nonrecurring charges that USWC and GTE propose to charge competing carriers at the time building blocks are purchased. We conclude that many of the assumptions underlying the USWC and GTE nonrecurring cost studies are unreasonable and inconsistent with cost principles adopted in prior unbundling dockets. We also find that the resulting upfront charges place new telecommunications providers at a significant competitive disadvantage vis a vis the incumbent carriers. To rectify these problems, we instruct USWC and GTE to resubmit building block tariffs incorporating more reasonable cost assumptions.

This order also examines a variety of other issues raised by the USWC and GTE tariffs, including carrier eligibility to purchase building blocks, services that may be provided using building blocks, the proposed elimination of building blocks, and circumstances requiring the modification or construction of facilities to provide reasonable access to building blocks.

 II. PROCEDURAL HISTORY

On July 19, 1996, the Public Utility Commission of Oregon (Commission or OPUC) issued Order No. 96-188 in docket UM 351, unbundling the telecommunications services offered by U S WEST Communications, Inc. (USWC) and GTE Northwest Incorporated (GTE) into building blocks and establishing cost-based prices for those unbundled elements.

The Federal Communications Commission (FCC) subsequently issued Order 96-325 promulgating regulations to implement the interconnection and pricing provisions of §§251 and 252 of the Telecommunications Act of 1996 (the Act).

On November 1, 1996, the Commission issued Order No. 96-283 adopting additional building blocks consistent with the minimum list of unbundled network elements approved by the FCC. USWC and GTE were ordered to file tariffs establishing the approved building blocks and building block rates, together with applicable terms and conditions and any proposed nonrecurring charges.

On December 16, 1996, GTE and USWC filed tariffs in accordance with Order No. 96-283. GTE’s filing was designated Advice No. 589. USWC’s filing was designated Advice 1661.

At its January 21, 1997, public meeting, the Commission granted USWC and GTE additional time to make revisions to Advice Nos. 1661 and 589. USWC and GTE were ordered to refile their building block compliance tariffs on March 14, 1997. On that date, GTE and USWC made supplemental filings that completely replaced the earlier filings.

At its April 1, 1997, public meeting, the Commission approved a Commission Staff (Staff) recommendation allowing Advice Nos. 1661 and 589 to take effect subject to refund pursuant to ORS 759.185. The Commission also initiated dockets UT 138 and UT 139 to investigate the reasonableness of Advice Nos. 1661 and 589, respectively.

On June 20, 1997, USWC filed Advice 1677 modifying the tariff sheets in Advice No. 1661. Advice 1677 was designated docket UT 140 and was suspended by the Commission at its July 8, 1997, public meeting.

On June 25, 1997, the Commission issued Order No. 97-239 in docket UM 844, revising the building block rates approved in Order No. 96-283. The new building block rates were based upon revised cost study results approved by the Commission.

On July 8, 1997, USWC filed Advice 1683, replacing Advice No. 1677 and removing certain price increases. At a special public meeting held July 15, 1997, the Commission permanently suspended Advice No. 1677 and allowed Advice No. 1683 to take effect subject to refund. The investigation of Advice 1683 was added to the pending investigation in docket UT 138, and docket UT 140 was closed.

On August 20, 1997, GTE filed Advice No. 599 implementing revised building block prices in accordance with the new prices approved in Order No. 97-239 in docket UM 844. USWC made a similar filing -- designated Advice No. 1685 -- on August 29, 1997. At the September 9, 1997, public meeting the Commission approved a Staff recommendation allowing Advice Nos. 599 and 1685 to take effect subject to refund. The investigation of the nonrecurring charges included in Advice Nos. 599 and 1685 was added to the issues under consideration in dockets UT 138 and UT 139.

On October 14, 1997, USWC filed Advice No. 1690, specifying additional revisions to its building block tariff. At the November 4, 1997, public meeting, the Commission approved a Staff recommendation allowing Advice No. 1690 to take effect subject to refund. The investigation of Advice No. 1690 was also added to the issues under consideration in docket UT 138.

Dockets UT 138 and UT 139 were consolidated for hearing and disposition. The first hearing in this matter was held December 1-4, 1997. The hearing focused on the nonrecurring charges proposed by GTE and USWC, as well as various non-price terms and conditions in the companies’ tariffs. At the hearing, it became apparent that additional evidence would be necessary to address issues arising from the decision of the Eighth Circuit Court of Appeals rendered in October 1997. Additional testimony was filed regarding these issues in January and February 1998.

On January 15, 1998, GTE filed Advice No. 611, revising its tariff, PUC OR No. 15, to incorporate changes proposed in its testimony. The proposed tariff has been voluntarily suspended by GTE pending the outcome of this docket.

Further hearings were held February 24-26, 1998. Post-hearing briefs were submitted on March 17 and March 31, 1998. The parties to this proceeding are listed on Appendix A of this order.

 III. ARE INCUMBENT LOCAL EXCHANGE CARRIERS REQUIRED TO COMBINE BUILDING BLOCKS?

A. Telecommunications Act of 1996.

Section 251(c)(3) of the Act states that incumbent local exchange telecommunications carriers (ILECs) have:

. . . the duty to provide, to any requesting telecommunications carrier for the provision of a telecommunications service, nondiscriminatory access to network elements on an unbundled basis at any technically feasible point on rates, terms and conditions that are just, reasonable, and nondiscriminatory in accordance with the terms and conditions of the agreement and the requirements of this section and section 252. An incumbent local exchange carrier shall provide such unbundled network elements in a manner that allows requesting carriers to combine such elements in order to provide such telecommunications service.

In its First Report and Order implementing the local competition provisions of the Act, the FCC promulgated rules which, among other things, required ILECs to combine unbundled network elements (UNEs) for requesting telecommunications carriers. ILECs were also required to provide access to preassembled UNEs and refrain from disassembling already-combined UNEs.

Several of the FCC rules were appealed to the United States Court of Appeals for the Eighth Circuit. In a decision entered July 18, 1997, the Court vacated §51.315(c)-(f) of the FCC rules regarding access to unbundled elements. On October 14, 1997, the Court entered an order on rehearing that also vacated §51.315(b) of the FCC rules. In its rehearing order, the Court decided to strike that portion of its initial decision regarding the combination of network elements and substitute the following:

We also believe that the FCC’s rule requiring incumbent LECs, rather than the requesting carriers, to combine network elements that are purchased by the requesting carriers on an unbundled basis, 47 C.F.R. §51.315(c)-(f), cannot be squared with the terms of subsection 251(c)(3). The last sentence of subsection 251(c)(3) reads, "An incumbent local exchange carrier shall provide such unbundled network elements in a manner that allows requesting carriers to combine such elements in order to provide such telecommunications service." 47 U.S.C.A. §251(c)(3) (emphasis added). This sentence unambiguously indicates that requesting carriers will combine the unbundled elements themselves. While the Act requires incumbent LECs to provide elements in a manner that enables the competing carriers to combine them, unlike the Commission, we do not believe that this language can be read to levy a duty on the incumbent LEC to do the actual combining of the elements. The FCC and its supporting intervenors argue that because the incumbent LECs maintain control over their networks, it is necessary to force them to combine the network elements, and they believe that the incumbent LECs would prefer to do the combining themselves to prevent the competing carriers from interfering with their networks. Despite the Commission's arguments, the plain meaning of the Act indicates that the requesting carriers will combine the unbundled elements themselves; the Act does not require the incumbent LECs to do all the work. Moreover, the fact that the incumbent LECs object to this rule indicates to us that they would rather allow entrants access to their networks than have to rebundle the unbundled elements for them.

Section 251(c)(3) requires an incumbent LEC to provide access to the elements of its network only on an unbundled (as opposed to a combined) basis. Stated another way, §251(c)(3) does not permit a new entrant to purchase the incumbent LECs assembled platform(s) of combined network elements (or any lesser existing combination of two or more elements) in order to offer competitive telecommunications services. To permit such an acquisition of already combined elements at cost based rates for unbundled access would obliterate the careful distinctions Congress has drawn in subsections 251(c)(3) and (4) between access to unbundled network elements on the one hand and the purchase at wholesale rates of an incumbent LEC's retail services for resale on the other. Accordingly, the Commission's rule 47 C.F.R. §51.315(b), which prohibits an incumbent LEC from separating network elements that it may currently combine, is contrary to §251(c)(3) because the rule would permit the new entrant access to the incumbent LECs network elements on a bundled rather than an unbundled basis.

Consequently, we vacate rule §51. 315(b)-(f) as well as the affiliated discussion sections.

B. Party Positions.

USWC and GTE contend that the Eighth Circuit’s decision unambiguously holds that incumbent LECs are not required to combine network elements, and that state regulatory agencies are likewise precluded from requiring ILECs to combine such elements. USWC and GTE maintain that any state-imposed "rebundling requirement" necessarily conflicts with §251(c)(3) of the Act and is therefore preempted under the Supremacy Clause of the United States Constitution as well as numerous decisions of the Supreme Court.

In support of their argument, USWC and GTE emphasize that several sections of the Act prohibit state commissions from taking actions that conflict with §251. For example, §261(b) allows state commissions to enforce or promulgate regulations only "if such regulations are not inconsistent with [sections 251 to 261 of the Act]." Section 261(c) similarly provides that states may impose additional requirements on carriers only if the State’s requirements are not inconsistent with §§251 to 261. In addition, §251(d)(3) of the Act preserves state access and interconnection regulations from FCC preemption only where they are consistent with the requirements of §251. Finally, while §251(e)(3) permits state commissions to establish or enforce state law requirements in an interconnection agreement, the agreement must nevertheless comply with the requirements of section 251.

USWC and GTE emphasize that conflict preemption occurs "where the state law stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress." They maintain that any OPUC decision requiring ILECs to combine elements is preempted because it would substantially prevent the implementation of §251 by rendering the Act’s resale requirement a nullity and placing incumbent carriers at an unfair disadvantage in the telecommunications marketplace. USWC notes:

One of Congress’ primary objectives was to provide alternative means of access for competing carriers to enter the local telephone services market. Under one alternative, a competing carrier pays wholesale rates for finished services that it can immediately resell to its customers. Under the other alternative, carriers purchase unbundled elements of the incumbent LEC’s network on a dedicated basis through an up-front investment. In doing so, the purchaser of the unbundled elements assumes certain business risks and costs that the purchaser of finished services for resale does not incur. In its order on rehearing, the Eighth Circuit emphasized that any rebundling obligation would "obliterate" the goals and objectives of the Act by eliminating the distinction between these two alternatives . . .

GTE contends that forcing ILECs to combine building blocks would also undermine the Act’s goal of competitive neutrality:

Such a rule would allow new entrants to purchase complete, bundled access to GTE’s entire network under the guise of purchasing "unbundled" elements. As the Eighth Circuit recognized, however, purchasing a full platform of combined UNEs for providing local service is no different than purchasing the wholesale service GTE offers for resale -- purchasing under either label effectively allows a new entrant to resell local service provided exclusively over GTE’s own network. If this Commission were to impose such a rule, that would provide GTE’s competitors a second avenue for obtaining wholesale service, but at a different price -- indeed, a price potentially well below the wholesale rate specified by section 252(d)(3) of the federal Act. By giving entrants the choice of two different prices for obtaining effectively the same thing, the Commission would simply create an opportunity for rampant arbitrage. As the Eighth Circuit concluded, this result would "obliterate the careful distinctions Congress has drawn between UNEs and wholesale services."

Aside from the preemption issue, GTE and USWC also contend that, as a matter of sound public policy, new entrants rather than ILECs should bear the cost and responsibilities of combining building blocks. If new entrants can purchase preassembled building block platforms that replicate ILEC retail services without incurring any additional cost, they will simply opt for the least expensive method of market entry. They will not add any value to the retail service or have any incentive to invest in new network technologies. GTE also stresses that there is no assurance that CLECs will share with customers any profits earned by arbitraging the difference between the combined UNE rate and the wholesale rate for ILEC services.

AT&T, MCI, and TCG (Joint Intervenors) argue that the Eighth Circuit decision does not restrict state authority to order ILECs to combine unbundled elements. These parties assert that the Court’s ruling was limited to the FCC’s regulations and did not address other possible sources of the ILEC duty to combine elements. They maintain that Oregon law authorizes the Commission to prevent ILECs from "dismembering their networks purely for the sake of raising barriers to entry into the local service market." The Joint Intervenors urge the Commission to require USWC and GTE to combine building blocks irrespective of whether federal law allows ILECs to discontinue such combinations.

The Joint Intervenors contend that numerous provisions of the Act make clear that federal law supplements, rather than supplants, state law in this area. They note that the Eighth Circuit reaffirmed the importance of §2(b) of the 1934 Act, which limits FCC jurisdiction and preserves the traditional boundary between interstate/foreign services which are a federal concern, and intrastate service, which is predominately a matter of state law.

The Joint Intervenors also argue that the Act does not expressly or implicitly preempt state law or Commission authority with respect to the provisioning of building blocks. They observe that the Act expressly provides that state regulations, orders, and policies establishing access and interconnection obligations of LECs shall be given effect unless they are inconsistent with the requirements of §251 or substantially prevent implementation of the purposes of the Act. In particular, §251(d)(3) provides:

Nothing in this section [251] shall affect the ability of a State to impose, on a competitively neutral basis . . . requirements necessary to . . . protect the public safety and welfare, ensure the continued quality of telecommunications services, and safeguard the rights of consumers.

In this context, the Joint Intervenors point out that the Eighth Circuit stressed that §251(d)(3):

does not require all state Commission orders to be consistent with all of the FCC’s regulations under section 251. . . . It is entirely possible for a state interconnection or access regulation, order, or policy to vary from a specific FCC regulation and yet be consistent with the overarching terms of section 251, and not substantially impair the implementation of section 251 or Part II. In this circumstance, subsection 251(d)(3) would prevent the FCC from preempting such a state rule, even though it differed from an FCC regulation.

Given that the Act expressly acknowledges state authority to supplement the minimum requirements of federal law, the Joint Intervenors contend that no congressional intent to preempt state law may reasonably be inferred. They note that preemption is inferred only:

where Congress has legislated comprehensively to occupy an entire field of regulation, leaving no room for the states to supplement federal law, or where the state law at issue conflicts with federal law, either because it is impossible to comply with both, or because the state law stands as an obstacle to the accomplishment and execution of congressional objectives.

Moreover, "[t]o justify preemption, an asserted conflict must be particularly

sharp . . . where Congress legislates in a field which the States have traditionally occupied." Joint Intervenors maintain that, because states have traditionally regulated intrastate telephone service, and because the provisions of the Act regarding local exchange service are interstitial, not comprehensive, there is no basis to conclude that a Commission decision requiring ILECs to combine network elements is in conflict with the Act.

Joint Intervenors also deny the ILECs claim that a Commission order requiring building block combinations will "stand as an obstacle to the accomplishment and execution" of the Act’s objectives. To the contrary, Joint Intervenors assert that a rebundling requirement is the most efficient, least discriminatory means of enabling competing carriers to combine building blocks and will only hasten the achievement of the Act’s central objective: introducing competition into local exchange markets and "erod[ing] the monopolistic nature of the local telephone service industry."

Finally, Joint Intervenors dispute USWC/GTE’s claim that mandatory ILEC building block combinations will obliterate the distinction between resale and building blocks. They assert:

[D]ifferences between resale and combined building blocks abound. Total service resale limits the CLEC to use of the ILEC’s total services. Use of building blocks enable CLECs to combine their own facilities, including operator services/directory assistance platforms, with building blocks. In this manner, use of combined building blocks will promote CLEC deployment of their own facilities because it will enable CLECs efficiently to shed individual building blocks as competition develops and market opportunities permit. A requirement under state law that ILECs combine building blocks will not obliterate those distinctions and, therefore, is not preempted by federal law.

Staff also argues that the Eighth Circuit decision was limited to a review of the FCC’s authority under the Act and did not preempt states from requiring ILECs to combine building blocks. Staff points out that the Act does not prohibit states from adopting regulations relating to provisioning building blocks and strongly favors state actions to preserve telecommunications competition. It notes that the Court reaffirmed state jurisdiction over intrastate telecommunications. Staff maintains that requiring GTE and USWC to combine building blocks is consistent with the Act and promotes local exchange competition because it is nondiscriminatory, efficient, and cost effective.

Staff argues that a CLEC "combines" building blocks when it designs its system and orders the specific building blocks necessary to provision the desired service. This process includes activities such as establishing points of interconnection and access arrangements, establishing interconnection to emergency and operator services, engineering the network, integrating facilities and establishing billing systems. Once the CLEC has performed this combination function, the ILEC bears responsibility for "installing" the selected building blocks by making the functionality of the building blocks available to the CLEC. Thus, the ILEC would physically connect -- or install -- building blocks ordered by the CLEC.

OGI agrees with Joint Intervenors and Staff that the Commission has the authority to order ILECs to combine building blocks. At a minimum, the Commission should require USWC and GTE to combine unbundled dedicated interoffice transport building blocks with network access channel building blocks, as well as any additional elements that may be required to connect customer premises to CLEC switching centers. To the extent the Commission declines to order ILEC combinations, OGI requests that the Commission grandfather all existing combinations used to serve OGI customers.

C. Commission Decision.

We agree with USWC and GTE that the Eighth Circuit has interpreted §251(c)(3) of the Act to hold that incumbent LECs cannot be required to combine unbundled network elements. The Court’s decision extends not only to so-called "UNE-Platforms" but also to "any combination of two or more elements." ILECs may voluntarily combine elements if they decide it is in their best interest, but are not required to do so.

We also agree with the ILECs that the Eighth Circuit has interpreted "combining" to mean the process of physically combining or connecting unbundled elements. While we believe there is merit to the interpretation offered by Staff -- indeed, it is the very process that the Commission contemplated would occur in an unbundled environment -- the language used by the Court clearly indicates that requesting telecommunications carriers are required to physically combine network elements themselves.

The preemption issue is more problematic. The issue presented is whether the Eighth Circuit decision merely delineates the scope of the FCC’s jurisdiction as alleged by Staff and Joint Intervenors, or whether it prescribes the requirements of the Act as they relate to the obligation to combine network elements, as USWC and GTE contend. In making this inquiry, the court must determine whether a state decision requiring incumbent LECs to combine elements is inconsistent with, or necessarily conflicts with, the purposes of the Act. In addition, the court must consider whether such a requirement would make it impossible to comply with, or pose an obstacle to, accomplishing the objectives of the Act.

Staff and Joint Intervenors corrrectly observe that the Eighth Circuit did not rule that states are preempted from requiring ILECs to combine elements. We also observe that the Act provides states with considerable latitude to enact regulations that go beyond the minimum regulations promulgated by the FCC. In particular, states are vested with broad authority to protect the public safety and welfare, ensure the continued quality of telecommunications services, and safeguard the rights of consumers. Indeed, the Eighth Circuit was careful to emphasize that states retain a significant role in implementing the requirements of the Act.

Nevertheless, the issue before us essentially involves the physical connection and arrangement of network facilities. If the Act contemplates that ILECs cannot be compelled to combine elements, is it possible for states to prescribe mandatory combination requirements without being "inconsistent" with the Act? Unlike other cases where jurisdictional conflicts have been resolved by allocating costs between federal and state jurisdictions, it is difficult to envision how building blocks could be provisioned one way for intrastate purposes and another way for interstate purposes. In evaluating such a situation, the appellate court would have to ask, "Is it possible for a federal "no combine" policy to coexist with a state "combine" policy, or will there be unavoidable conflicts that prevent substantial implementation of the federal policy?" Based on the record before us, we cannot conclude that such conflicts can be avoided.

Furthermore, the Eighth Circuit’s decision strongly suggests that any regulation which requires incumbent LECs to combine elements will pose an obstacle to the congressional objectives underlying the Act. As noted above, the Court held that permitting new entrants to acquire any existing combination of two or more elements at cost based rates "would obliterate the careful distinctions Congress has drawn in subsections 251(c)(3) and (4) between access to unbundled network elements on the one hand and the purchase at wholesale rates of an incumbent LEC’s retail services for resale on the other." In view of this statement, it is possible that a state-mandated combination requirement may be construed to obstruct the purposes of the Act.

At the same time, we agree with Staff and Joint Intervenors that the Eighth Circuit’s interpretation of the Act will produce UNE access arrangements that are more costly, more time-consuming to provision, and less efficient from a network engineering standpoint than if unbundled elements are combined by the ILECs. As we discuss in greater detail below, the evidence amply demonstrates that ILEC networks were never designed to be "decombined" and "recombined" in the manner contemplated by the Court’s decision. It is even more unfortunate that this result has come to pass essentially because of a pricing dispute over the potential for requesting carriers to arbitrage the difference between the wholesale rate and the combined UNE rate. Our extensive experience with unbundling of telecommunications services has shown that the pro-competitive goals of the Act and this Commission will not be achieved unless regulatory policy choices correspond with efficient engineering and network management decisions.

The Commission thus faces a dilemma. On the one hand, it is likely that an OPUC-imposed combination requirement would be appealed immediately to federal court. It is also likely the court would grant a stay of the combination requirement until the Supreme Court decides whether the Eighth Circuit’s interpretation of §251(c)(3) is correct. While it is difficult to say how long it will take for the Supreme Court to rule on this matter, it could easily take a year or more. If the Supreme Court ultimately decides in favor of the Eighth Circuit on the combination issue, it would not end the matter because the preemption issue is not currently before the Supreme Court on appeal. That could mean an additional delay of several more years while that issue works its way through the federal appellate process.

On the other hand, requiring CLECs to combine elements will almost certainly reduce network efficiency, impair service quality and delay the emergence of effective local exchange competition. The numerous problems associated with requiring CLECs to combine elements are discussed at length in this order.

The Commission does not believe that it is preempted by the Eighth Circuit's holding that ILECs cannot be required under the Act to combine network elements. Nevertheless, to avoid conflict with that holding while it is being reviewed by the United States Supreme Court, the Commission reluctantly adopts UNE access arrangements that are consistent with the Eighth Circuit's holding. Although this will produce access arrangements that are less satisfactory than those which result from requiring ILECs to combine elements, we believe that the options made available to requesting carriers in this order will be adequate and nondiscriminatory. More importantly, these arrangements will allow local exchange telecommunications competition to move forward -- albeit at a slower pace than would otherwise be the case -- while the courts grapple with the difficult and contentious process of interpreting the Act.

If local exchange telecommunications competition is to succeed, it is imperative that the Commission avoid the protracted litigation and delay that would result from a judicially-imposed stay of an OPUC-ordered combination requirement. We would rather adopt building block access arrangements that are less optimal but still have a reasonable chance of stimulating competition, than choose a course of action that will hamstring competitive entry for years to come. If the judicial process ultimately produces a resolution that permits more efficient ILEC combinations, we can always revisit the issue at a later date. In the interim, the Commission is persuaded that the access arrangements set forth in this order offer a workable solution to provisioning building blocks in Oregon.

A. USWC and GTE Proposals.

USWC and GTE propose to provide access to unbundled elements by requiring competing carriers to establish collocation arrangements at ILEC central offices. These proposals contemplate that the ILECs will deliver unbundled building blocks to either a collocation space assigned to an individual carrier or to a shared collocation area. Within the collocation space, a carrier would combine the building blocks necessary to provision service to its customers.

USWC and GTE do not propose to restrict the types of telecommunications services that requesting carriers can offer through unbundled elements. Carriers will be permitted to combine building blocks with technically compatible equipment owned by the carrier, except for certain types of equipment (e.g., switching equipment) that may not be collocated. USWC and GTE will also provide requesting carriers with the same building block functions, features, and technical capabilities that the ILECs provide to themselves.

GTE and USWC state that they will combine building blocks to the extent necessary to create the unbundled network elements mandated by the FCC in 47 C.F.R. §51.319. For network security reasons, GTE also agrees to provide the NAC (loop) building block and the Network Interface Device (NID) in either a separate or combined form. USWC and GTE will not combine any other building blocks for requesting carriers.

USWC states that it will not combine unbundled elements with finished services (retail services) or with interexchange services. GTE, on the other hand, agrees to provide a combination of building blocks and retail services to requesting carriers.

GTE’s proposal would permit CLECs to establish physical collocation arrangements, virtual collocation arrangements, or common collocation arrangements. These proposals are described below. If one of these types of collocation is not possible because of space limitations or technical reasons, GTE will negotiate with the CLEC to implement a mutually acceptable alternative. GTE proposes to provide physical collocation in Oregon under the prices, terms, and conditions of its federal physical collocation tariff. It will file an intrastate version of that tariff if requested by the Commission.

Under GTE’s physical collocation alternative, the CLEC would establish a segregated space or "cage" in a secured area within a GTE central office. The CLEC would place its own equipment in the cage and would have either escorted access or direct access to install, maintain, and repair the collocated equipment within the cage. The CLEC would then combine building blocks within the collocation space.

For example, to combine a NAC (loop) building block with a NACC building block (switch port), the CLEC would supply a jumper NAC (tie cable) from its collocation space to the GTE main distribution frame (MDF) for each building block. GTE would connect the tie cables to terminal locations on the MDF. GTE would then cross-connect the tie cable terminal locations to locations on the MDF where the NAC and NACC terminate by using jumper wires (short jumpers) for each building block. Within the collocation space, the CLEC would use another short jumper to connect the tie cable from the NAC to the tie cable from the NACC to combine the two building blocks. This arrangement is illustrated on Appendix B of this order. Appendix B also illustrates how a carrier that purchases the NAC building block and self-provisions its own switching would combine those elements under GTE’s proposal.

GTE points out that CLECs and competitive access providers (CAPs) currently use physical collocation to gain access to various components of GTE’s network. According to GTE, CLECs currently use tie cables to bring unbundled loops to their physical collocation cages and attach these tie cables to fiber optic termination equipment that provide access to the CLEC’s interoffice transmission facilities. GTE also states that CAPs have combined GTE network components within physical collocation spaces in the manner described above for approximately five years in hundreds of central offices throughout the country. GTE states that physical collocation has become so standard in the local exchange industry that CLECs can easily obtain information and expertise necessary to gain access to and recombine building blocks.

The terms and conditions for virtual collocation are set forth in GTE’s intrastate tariff. Carriers that choose virtual collocation may combine building blocks without establishing a separate collocation space within the central office. In this situation, GTE installs, repairs, and maintains the equipment necessary to terminate a carrier’s cable facilities at the GTE central office and connects those facilities to GTE’s network. Although the equipment is dedicated to the use of the carrier, the carrier does not have access to it. According to Mr. Hartshorn, GTE will make virtual collocation available only if there is insufficient space for physical collocation.

GTE states that carriers can combine building blocks using virtual collocation in a manner similar to the physical collocation option. NAC, NACC, or transport building blocks would each be cross-connected to tie cables that would be connected to the virtually collocated transmission device dedicated to the CLEC. The tie cables connected to the virtually collocated device would be cross-connected to separate transport cables for each building block. The transmission cables for each building block would then run from GTE’s central office to the CLEC’s facility. Within its facility, the CLEC can either connect the transmission cables for the various building blocks to its own network or combine the building blocks with each other. GTE contends that its virtual collocation proposal allows CLECs to perform the same building block combinations that are possible in a physical collocation arrangement.

GTE’s common collocation option is designed to allow limited common access to requesting carriers for the purpose of combining building blocks. Upon request, GTE will construct a 5’ X 20’ area within its central office. Within this caged area, GTE can erect up to 24 vertical uprights. Each vertical will be sold as a whole to a requesting CLEC, and is capable of accommodating 1,000 DS0 connections, or 500 NACs and 500 NACCs. A requesting carrier may purchase as many verticals as are available. The verticals would also accommodate DSX patch panels necessary for access to higher bandwidth dedicated transport facilities.

Once a CLEC purchases a common collocation vertical frame, it would connect the terminal blocks on the vertical frame to tie cables that extend to cross-connects (terminal blocks) on the GTE MDF, similar to the way a CLEC using physical collocation will provide tie cables between its collocation cage and the GTE MDF. The CLEC can then combine NACs, NACCs, and transport facilities by making the necessary connections between points on the vertical frame. If space is available, requesting carriers may also collocate a limited amount of their own equipment in the common collocation area to facilitate termination and rebundling of building blocks. Space for collocating CLEC equipment will be made available on a first-come, first-served basis.

GTE notes that common collocation differs from physical and virtual collocation in three main respects. First, it may significantly lower transaction costs of combining GTE building blocks because the common collocation space could be established within a shorter time and at a lower cost than with the other collocation offerings. Second, the common collocation area, while secured, would be common to all CLECs using frames within the common collocation space. Thus, while CLECs would not have exclusive access as with physical collocation, they would have direct access to the network connections unlike the case with virtual collocation. Finally, CLECs desiring to collocate their own equipment in the common collocation space would be more limited than they would be in a physical collocation arrangement.

GTE is still in the process of developing prices, terms, and conditions for its common collocation offering. Although it prefers to negotiate common collocation under contracts with requesting carriers, GTE will file an intrastate tariff if requested by the Commission.

GTE maintains that its collocation options will allow requesting carriers sufficient flexibility to implement a range of different business plans. It asserts that these arrangements provide a reasonable means of accommodating different carriers, maintaining network integrity, and advancing the competitive goals in the Act. GTE also agrees to provide requesting carriers with the information they need to select, establish, and use collocation arrangements to access and recombine building blocks. If GTE cannot provide any of the three collocation options, it will seek to negotiate alternative arrangements for building block access. Such arrangements may include, for example, mutually agreed meet points located outside the central office. If alternative arrangements cannot be negotiated or are unavailable, requesting carriers may have to purchase resold GTE service.

USWC also proposes to allow requesting carriers access to building blocks through physical collocation and virtual collocation arrangements. Although USWC has not finalized all of the details of those proposals, it submitted an illustrative tariff that includes many of the terms and conditions under which the physical and virtual collocation options would be offered.

The main difference between GTE’s physical and virtual collocation proposals and those recommended by USWC is that USWC proposes to insert an additional, intermediate distribution frame -- known as the Single Point of Termination, or SPOT frame -- between the USWC MDF and the CLEC’s collocation space. SPOT frames would be required in physical collocation and virtual collocation arrangements, as well as in the "cageless" collocation arrangements described below.

Under USWC’s proposal, building blocks would be delivered to the SPOT frame over tie cables that run between the USWC MDF to the horizontal side of the SPOT frame. Other tie cables will connect the vertical side of the SPOT frame to each carrier’s physical or virtual collocation space. CLEC technicians would have access to the SPOT frame to combine building blocks. This process requires the CLEC technician to install and connect jumper wires between the appropriate termination points on the SPOT frame. Appendix C shows how building blocks would be delivered when (a) a CLEC buys both an NAC and an NACC from USWC and (b) a CLEC buys an NAC from USWC but provisions its own switching.

If a CLEC wants to combine its facilities with USWC’s building blocks, it must collocate in some manner or perform the combinations at its own premises. On the other hand, if a CLEC does not self-provision any of its own facilities and buys all of the building blocks from USWC, the CLEC may make all of the necessary combinations at the SPOT frame without having to collocate.

USWC maintains that the SPOT frame architecture is reliable, economical, and allows combination of building blocks in the most efficient manner possible. It notes that the SPOT is currently installed in six Oregon central offices and in a number of other USWC territories. The SPOT frame architecture is also incorporated in existing collocation tariffs as well as in interconnection agreements arbitrated under the Act.

Whereas current procedures call for USWC technicians to make connections at the SPOT frame, USWC’s proposal in this case allows requesting carriers to make their own connections at the SPOT. For carriers that do not want to make connections at the SPOT frame, USWC agrees to cable any building block through the SPOT frame to the carrier’s collocation area.

Another difference between the USWC and GTE physical collocation and virtual collocation proposals involves the provisioning of tie cables. As noted above, GTE’s proposal contemplates that the CLEC will provision the tie cables between the GTE MDF and the carrier’s collocation space. USWC, on the other hand, requires its technicians to install the tie cables between the MDF and the spot frame and between the spot frame and the carrier’s collocation space. Under USWC’s proposal, therefore, the carrier would pay a recurring monthly charge for each tie cable (jumper NAC) building block that is used.

At the February hearing, USWC offered a third collocation option in addition to its physical and virtual collocation proposals. This option -- labeled "cageless" collocation -- is a form of physical collocation that would allow carriers to purchase collocation space in single frame increments. Requesting carriers would not have to erect secured cages and would be permitted to place their own equipment alongside USWC equipment. Carriers could install any equipment permitted in a physical or virtual collocation arrangement as long as adequate space was available. Tie cables would be used to connect carrier-supplied equipment to the SPOT frame where USWC would deliver the building blocks desired by the carrier. As with USWC’s other collocation options, requesting carriers would combine building blocks at the SPOT frame.

USWC developed cageless collocation to address concerns regarding inadequate central office space for physical collocation. It expects that cageless arrangements will also take less time to install than physical collocation because less construction is required. Although the terms and conditions of the cageless option are still being developed, USWC has implemented similar offerings in Arizona and Nebraska. It has also included cageless options in its interconnection contracts with TCG.

B. Response to GTE and USWC Access Proposals.

Staff, OGI, and the Joint Intervenors oppose the mandatory collocation proposals made by GTE and USWC. These parties maintain that the building block access arrangements offered by the ILECs are less efficient, more costly, and inferior in quality to the manner in which USWC and GTE provision services to themselves. They allege that the discriminatory nature of the ILEC proposals places new entrants at a competitive disadvantage and impedes the progress of local exchange competition.

1. Cost Concerns.

Staff, OGI, and the Joint Intervenors contend that the USWC and GTE building block access proposals will cause new entrants to incur unnecessary and substantially increased costs; first, to establish collocation arrangements, and second, to purchase additional building blocks.

OGI witness Jack Pestaner discussed the costs that OGI would incur to establish collocation arrangements under the USWC and GTE proposals. OGI is the only telecommunications carrier currently purchasing building blocks from the USWC and GTE tariffs. It provides combined local and toll services to customers in the Portland metropolitan area by supplying its own switching and purchasing high capacity NACs, interoffice transport, and multiplexing building blocks on a combined basis from USWC and GTE.

Mr. Pestaner testified that GTE’s physical collocation rates will cost OGI an average of $50,000 per central office to acquire space, power, and equipment, and an additional $1,000 per month per central office for rent and power. In order to serve its West Portland EAS customers, OGI will need to collocate in nine central offices, costing approximately $450,000 in up-front, one-time charges and $10,000 in monthly fees. Mr. Pestaner emphasized that the high cost of establishing these physical collocation arrangements will reduce the pace of competitive entry and will effectively prevent facilities-based carriers from serving low-density areas. As a result, many Oregon customers will be denied the benefits of local exchange competition.

OGI, TCG, and other parties acknowledged that USWC’s cageless collocation and, to a lesser extent, GTE’s common collocation proposal, may reduce the cost of establishing physical collocation arrangements. However, because these alternatives have not been finalized or approved by the Commission, there is no way for competing carriers to evaluate whether they will be viable from a cost standpoint.

In addition to the cost of establishing collocation arrangements, the USWC and GTE proposals contemplate that requesting carriers will have to purchase more building blocks than the ILECs must use to supply the same service to their customers. For example, AT&T witness Kenneth Wilson and TCG witness Frank Croan emphasize that a carrier purchasing both an ILEC NAC and an ILEC NACC to provision basic local service will have to buy at least two additional jumper NACs and two additional distribution frame terminations that USWC and GTE do not have to purchase to supply the same service. This means that requesting carriers must pay more each month to buy the extra building blocks.

2. Space Availability.

Staff, OGI, and the Joint Intervenors also question whether there is adequate space available in USWC and GTE central offices to permit the various forms of collocation contemplated by the ILEC proposals. These parties question whether USWC and GTE will have the ability to accommodate all carriers that might be required to establish physical collocation spaces. Also, because the specifics relating to common and cageless collocation have not been finalized, it is impossible to tell whether there will be adequate space in ILEC central offices to accommodate all of the carriers that seek to use those arrangements.

TCG witness Croan emphasizes that the 5’ X 20’ common collocation space proposed by GTE will limit all CLECs in a central office to a maximum of 12,000 customers, assuming a two-wire loop. If four-wire loops are deployed -- as required by virtually all of the new digital services -- the number of potential customers served would be cut in half. Given that a typical large central office may serve approximately 100,000 lines, Mr. Croan maintains that GTE’s proposal will stifle competition.

AT&T witness Wilson emphasizes that USWC has had problems providing adequate collocation space in other jurisdictions. In the state of Washington, for example, USWC has stated that there is no available collocation space in several central offices. Disputes over this issue have precipitated a number of complaints in that jurisdiction. Mr. Wilson notes that, if similar disputes arise in Oregon, the time and expense CLECs will incur to establish access arrangements will increase substantially.

AT&T/MCI witness Bonni Petti and MCI witness Michael Hydock also maintain that USWC and GTE do not adequately address issues relating to the provision of facilities to accommodate the recombination of building blocks by requesting carriers. Mr. Hydock observes that the USWC tariff makes no provision for dealing with situations where there is limited space to establish collocation arrangements. Nor does it indicate how USWC will allocate space among CLECs on the SPOT frame where space limitations exist. Furthermore, the USWC tariff does not specify the terms and conditions governing access to the SPOT frame. This omission is problematic because many wire centers are unmanned, thus requiring CLEC technicians to coordinate with ILEC personnel to gain access to those facilities. In addition to these concerns, Ms. Petti states:

[U]nder its proposed change to tariff revisions, US WEST must install (or expand) SPOT frames dedicated to the recombination function at numerous end offices throughout Oregon. US WEST does not specify when or how it proposes to do this. The illustrative [US WEST collocation tariff] does not state where in US WEST’s buildings the SPOT frames are (or will be) located. This leads to two concerns: The first is whether US WEST can make these frames available at the request of the CLECs in a reasonable amount of time in all the offices. Second, depending on the number of CLECs requesting common space, the frame could be sized for too few or too many CLECs. This uncertainty may lead to placing the SPOT in a location that is not in close proximity to the MDF or a collocation area. The placement [of the SPOT] is critical and could change the transmission characteristics of some or all the circuits if several floors are involved.

Although GTE and USWC acknowledge that there are locations where collocation is not possible, they do not agree that space availability will be a significant problem. The ILECs also emphasize that it is important for CLECs to forecast their demand for collocation so that space requirements may be reasonably accommodated.

3. Loss of Service while Collocation Arrangements Are Established.

Opponents of the USWC and GTE proposals argue that requesting carriers will experience substantial delays while mandatory collocation arrangements are established. They allege that significant delays in the process of implementing collocation will reduce the ability of new carriers to provide prompt service to their customers.

At the hearing, USWC indicated that it will endeavor to construct collocation facilities (including cageless arrangements) within 90 days. USWC’s illustrative collocation tariff, however, allows the company up to a total of 145 days to provision virtual or physical collocation after a carrier submits a request for collocation.

GTE also indicated that common collocation facilities could probably be constructed in a central office within approximately 90 days after a contract is negotiated and an order is placed by the requesting carrier. This contrasts with GTE’s interstate physical collocation tariff, which allows more than six months to establish collocation arrangements. GTE’s intrastate virtual collocation tariff does not specify a time frame for completing collocation arrangements.

OGI witness Pestaner contends that the ILEC proposals will subject OGI’s existing customers to lengthy service outages if building blocks are "decombined" by the ILECs and not "recombined" until collocation arrangements are put in place. He fears that it will take months to establish the necessary collocation arrangements and that the extended interruption in service will prove disastrous to OGI and its customers.

In addition to the time necessary to establish collocation arrangements, AT&T witness Wilson and MCI witness Michael Hydock observe that the USWC SPOT frame and GTE common collocation proposals create a serious risk that CLEC customers will be left without service during the cut-over from one carrier to another. Joint Intervenors and OGI contend that these circumstances will make it very difficult for new carriers to compete on an equal footing with the ILECs.

According to Mr. Wilson, there is no way to avoid some period of time in which the customer is out of service while facilities are disconnected and reconnected. Although the length of such outages could be reduced by permitting the CLECs to "pre-wire" most of the necessary cabling and cross-connections, that scenario assumes a level of ILEC cooperation that is unlikely to materialize under actual competitive conditions, and does not take into account human error and system flaws that could drastically increase out-of-service time.

Mr. Wilson testified that customer out-of-service time will be substantial if the CLEC and ILEC do not agree to pre-wire connections. Outages will occur because at least two separate disconnect/reconnect procedures (one at the MDF and one at the SPOT frame) would need to be completed. Even longer outages may occur if the prewiring is not done correctly due to misidentified terminal block assignments or cable and pair numbers, defective connections, or "assignments not spare." Mr. Wilson believes that problems are likely, especially given the difficulty of maintaining accurate and parallel ILEC/CLEC inventories of all block assignment and frame locations, together with the numerous points of failure on the collocation circuit.

Even if the CLEC and ILEC agree to prewire connections, appropriate methods and procedures must be developed to ensure that (a) no active call is on the line; (b) an efficient sequencing process is used for disconnecting and reconnecting each terminal; and (c) experienced technicians are available to handle cut-overs. Mr. Wilson emphasizes that, unless procedures to minimize service interruptions are implemented, customers will be inconvenienced, and possibly even endangered. In either case, new entrants are likely to lose business because of customer dissatisfaction. Messrs. Wilson and Hydock emphasize that the USWC and GTE tariffs do not include any provisions dealing with coordination of personnel or procedures to minimize out-of-service conditions.

5. Additional Connections/Potential Points of Failure.

As noted above, the building block access arrangements proposed by GTE and USWC require the use of a greater number of jumper wires and distribution frame terminations than is necessary if the ILECs combine building blocks. Staff, OGI, and the Joint Intervenors maintain that the added building blocks and connections create several additional potential points of failure that will adversely affect the quality of service received by carriers purchasing building blocks from USWC and GTE. In contrast to the circuit paths required under the USWC and GTE collocation proposals shown on Appendices B and C, Appendix D illustrates the circuit path that would exist if either the ILECs combined building blocks or requesting carriers were allowed access to the ILEC MDF to make the necessary combinations.

Several witnesses testified that the additional facilities and connections required by the SPOT/Common collocation proposals will reduce reliability and impair overall service quality. According to Mr. Wilson, a typical ILEC loop connection in a wire center has only two points of connection to a frame -- one on the terminal connecting to the loop, and the other on the terminal making the connection to the switch port. These connections must be considered points of failure because they are places where the loop connection is most likely to come apart. Under the USWC and GTE proposals, ILEC loops that are combined with ILEC switching will have a minimum of six points of failure. Thus, the ILEC proposals triple the possibility that CLEC loops will fail.

Mr. Wilson testified that, while employed for AT&T, he was responsible for a nationwide asset management project that involved removing a large number of unnecessary, intervening distribution frames from AT&T’s network. The decision to remove the frames resulted in a significant improvement in efficiency and customer satisfaction. Mr. Wilson states:

Efficient engineering principles require that the amount of equipment and the number of terminations in any circuit path should be minimized. Additional equipment is additional expense. Additional equipment also means additional connections in the circuit path, which means additional points of failure in the network. The old adage that "anything that can go wrong will" still holds true. In the case of the network, additional points of failure are sure to fail at some time. If a new piece of equipment is added in the circuit path, at least three new points of failure are added: the piece of equipment itself can fail, the connection into the new piece of equipment can fail, and the connection out of the new piece of equipment can fail. From a network engineering perspective, efficient network design instructs that additional equipment will result in extra expense and extra reliability problems and therefore additional equipment should be minimized.

AT&T/MCI witness Petti worked for several years as a regional manager for network operations, management, and engineering for Pacific Bell Telephone Company. She testified that reducing the number of facilities and connections generally results in improved service quality:

And I guess I have a different life experience in my 30 years in the operations world than [GTE witness Larry] Hartshorn, because any time you have an open lug available to have damage to it or be crossed or jumpers lifted, the more of those you have and the more human beings you have in the office working on that plant, the more chances are that something could happen to that customer, and that’s been my experience.

And when we, as Mr. Wilson explained, have designed out that kind of old equipment and eliminated those cross-connects, the report rate of trouble in the main frame, in my experience, and I had 82 buildings, we went from a high report rate of, oh, maybe three and a half reports per month per thousand lines to .10 on my main frames. That was a tremendous change, and a lot of that is because we used two things. One is we didn’t change the jumpers out, we used [dedicated inside plant] DIP and [dedicated outside plant] DOP, and the other is we designed out intermediate frames. We cabled everything we could.

TCG witness Frank Croan reiterated the importance of minimizing the number of facilities and points of interconnection in the network. His experience in the cable TV and telecommunications industries demonstrates that service calls can be substantially reduced, or even eliminated, by removing extra connections. Mr. Croan emphasizes:

When we talk about the proposals for a SPOT frame and for all of these additional connections, we’re not talking about simply one or two; we’re talking about multiples. We’re talking about eight, sometimes ten additional connections that are required to be made just to be able to take this circuit, this service, out of the MDF over to a place where a CLEC can cross-connect it through, and then come back to the MDF so that the ILEC could carry it on to another situation.

* * * * *

When you think about it, those connections are -- they’re creating a path. You think about a wire, a wire is an end-to-end communications path, and so long as it remains intact, it has its best transmission characteristics. As soon as you begin to cut this wire into pieces and put it back together, you are introducing an opportunity for something to occur at that juncture.

USWC and GTE deny that their proposed building block access arrangements will introduce points of failure or otherwise impair the ability of requesting carriers to offer service. GTE witness Larry Hartshorn testified that adding additional elements and connections inside the central office will have a negligible impact on end-to-end circuit reliability. USWC points out that the required connections do not involve outside plant, but rather will be made in a clean, climate-controlled central office environment. USWC reiterates that several LECs currently use the SPOT frame to combine elements. It also emphasizes that no studies were produced to support the claim that additional connections will increase the probability of circuit failure.

6. Maintenance and Provisioning Difficulties.

Joint Intervenors contend that the ILEC access proposals will produce a number of maintenance and provisioning problems that make it more difficult for new entrants to provide telecommunications services:

(a) The record indicates that the jumper wires that connect building blocks under USWC and GTE proposals are likely to be longer than those which now connect ILEC network elements. This is particularly true in large central offices such as USWC’s Portland Capitol office, where the MDF and SPOT frames are located several floors apart. Ms. Petti and Messrs. Croan and Hydock testified that longer jumper wires can cause transmission problems which prevent CLECs from using building blocks to provision service to customers. The ILECs, on the other hand, would not experience similar problems Mr. Croan explains:

If the combined distance of jumper NACs is 1,000 feet, the radius of the CLEC’s potential serving area is reduced by 1,000 feet when compared to the ILEC. For the DS1 provisioned using unbundled network elements, this is a 27 percent reduction in coverage area available to the CLEC when the SPOT frame is located 500 cable feet from the MDF. Keeping with this DS1 UNE example, if the distance between the MDF and SPOT frame is increased to 1,000 feet, the CLEC can complete over 51 percent of the area (using unbundled network elements) that the ILEC can reach using the exact same facilities. The problem is further illustrated if a CLEC is required to locate its facilities remotely, as suggested by GTE in those instances when central office space is unavailable.

Similarly, Mr. Hydock observes:

Under the U S WEST proposal, the effective length of the loop will be increased for CLECs. In the SPOT frame environment, the loop would need to be extended to the SPOT frame, and then back to the main distribution frame for the virtual connection to the port. The added length would include the distance from the MDF -- where the LEC currently joins the loop and the port -- to the SPOT frame, multiplied by two (the length of the cable for the loop, and the length of the cable for the port). The combined added distance could require added regeneration equipment within the central office. Essentially, the CLEC will be forced to obtain loops at terms and conditions that are inferior to those granted U S WEST’s retail operations.

(b) Ms. Petti and Mr. Wilson contend that the additional wiring caused by the ILEC proposals may create situations where jumper assignments on the distribution frame are not maintained properly. This is due to the fact that each CLEC occupying space on the SPOT frame or common collocation frame is responsible for maintaining a record of jumper connections. The record keeping process is made more difficult because the frame locations where the jumpers are terminated are not controlled by the CLECs, but are assigned by the ILEC. Thus, there is no assurance that the termination points for building blocks necessary to provision a service will be proximately located on the distribution frame.

Furthermore, the jumper termination points on the SPOT frame are encased in 4"X 6" plastic boxes that are situated side by side on the frame. Each box contains 100 jumper termination lugs. A sticker is attached to the cover of each box for purposes of identifying the jumper locations. Ms. Petti maintains that it will be difficult for CLECs to maintain adequate records of jumper assignments because it is physically difficult to enter notations on the box covers, and the covers are "typically lost over time." In contrast, the ILECs are able to maintain computer records of their jumper connections on the MDF.

(c) AT&T witness Wilson testified that the additional loop length required under the ILEC access proposals may also preclude CLECs from using the mechanized loop testing (MLT) capabilities of the ILEC switch. If an ILEC customer reports a service trouble, MLT allows ILEC repair personnel to electronically test, from a remote location, a customer’s facilities while the customer is still on the line. The addition of unnecessary tie cables and jumpers effectively increases the length of loop and changes the circuit parameters, making MLT unavailable. Thus, if a CLEC customer experiences trouble, an ILEC technician must be dispatched to manually test the loop. As a consequence, it may take several hours or days before the customer’s problem is diagnosed and resolved.

(d) Mr. Wilson also emphasizes that the SPOT and common collocation proposals limit the number of customers that can be provisioned with a combination of NAC and NACC elements in a given day. A primary source of delay is the manual work necessary to establish the cross-connections (i.e., install and connect the jumper wires) at the MDF and SPOT frame. Mr. Wilson cites a study prepared by Coopers and Lybrand for Bell Atlantic-New York, which estimated that a single large central office of a Bell Operating Company could provision a maximum of 143 loop/switch combination orders per day. Noting that an average central office in Oregon may serve over 50,000 lines, he contends that 143 lines per day will not support meaningful UNE-based competition. Furthermore, he contends that the Coopers and Lybrand estimate is overly optimistic because it: (a) assumes three shifts of two-person technician teams will work around the clock; (b) assumes that no ILEC frame work is necessary; (c) underestimates the number of ILEC job orders necessary for each loop-port combination; (d) does not account for additional time necessary to coordinate the cross-connect jobs technicians must perform; (e) underestimates new CLEC customers because it does not consider "customer churn;" and (f) does not take into account provisioning delays resulting from human error.

Mr. Wilson asserts that the manual processes required by the USWC and GTE building block access proposals will limit CLECs to a fraction of the customers that they would otherwise serve or that the ILECs are capable of serving. This will place significant restrictions on the ability of CLECs to market their services. CLECs will be unable to engage in mass marketing because such efforts are likely to produce demands for service far beyond that which can be provisioned at the central office. Instead, CLECs will be forced to market their services through controlled telemarketing or direct mail so that marketing could be shut down once capacity limits at individual central offices are met.

(e) Joint Intervenors also argue that the SPOT frame and common collocation proposals will subject CLECs to inferior service quality because both require unnecessary handling and removal of wires as customers change local service providers. Mr. Wilson observes that the jumper wires used to connect terminals on the MDF are very thin and frail and, in many cases, have been in place for several years. As significant competition develops and customers begin to churn, increased congestion on the distribution frames caused by installing new cross-connects and removing the old cross-connect wires will put unnecessary stress on jumper wires and result in broken connections. Frame jumpers are already subject to significant stress as technicians move jumpers within the frame and "mine out" old wires that are no longer being used. As this "pulling and tugging" increases with competition, CLEC service failures will also increase. Mr. Wilson argues that service failures will be borne disproportionately by the CLECs because recombinations at the SPOT/common collocation frames entail many more connections and points of failure than is required for ILEC loops.

(f) Mr. Wilson observes that the SPOT/common collocation proposals will require technicians to perform twice the amount of work for CLEC customers. In addition to directing a loop to the correct tie cable corresponding to the CLEC’s location on the SPOT/common collocation frame, technicians must also connect the CLEC’s return tie cable to the correct terminals on the MDF block that correspond to the correct switch port. Also, when there is circuit trouble, CLECs and ILECs will have to coordinate efforts to determine whether the source of failure is the SPOT/common collocation frame, the ILEC tie cables, the jumper wires, or the MDF. This process will become more difficult over time, as inevitable errors in recombination work cause incorrect disconnections and incorrect pairing of loop and switch ports. In contrast, no complicated effort is required when there is trouble on an ILEC customer’s line.

Ms. Petti describes the processes that would be necessary to transition a USWC customer to a CLEC or to change a CLEC customer over to another CLEC:

All of the activities must be coordinated for a working service, including checking the line prior to looping it through all of these extra blocks and cross-connects. The USWC technician and the CLEC technician must be at the office simultaneously, have the paperwork, prewire the SPOT frame, lay in the wire from the Jumper NAC blocks, one to the NAC (loop), and one to the NACC (port). The two USWC technicians, one at the loop and one at the port (unless they are in very close proximity such as a COSMIC frame in the same zone), must lift the old jumpers off and tie down the ones lying in readiness. Before lifting they must check to assure the line is idle (for special services/designed circuits this includes pre-checking to release the circuit) and must retest once it is tied down to be sure the cross-connect [jumper wire] has been placed on the right lugs. While USWC is performing this procedure the CLEC technician stands by in case the circuit does not work. If there is a problem, both (all) the technicians may put the customer back the old (USWC) way so they are not without dial tone.

Ms. Petti also disagrees with GTE’s attempt to compare the arrangements used by CAPs to connect loops and transport facilities with the building block access arrangements offered to CLECs in this case. She notes that, while the methods may be similar, there are significant differences in scope and scale that must be taken into account. CAPs generally prefer to serve large businesses where all of the loops from an entire building can be connected with fiber optics using SONET technology. Ms. Petti maintains that it will be very costly to serve the vast majority of POTS customers in this manner.

(g) The Joint Intervenors also contend that the additional connections required by the ILEC access proposals will cause available space on the MDF to be used up more rapidly than would be the case if the ILECs combined elements or the CLECs combined elements at the MDF themselves. For example, where a CLEC buys both the NAC and NACC, the USWC and GTE collocation proposals will require twice as many connections at the MDF. Ms. Petti asserts that MDF connections should be utilized more efficiently because increasing the capacity-- or "growing"-- an MDF requires an ILEC to perform "frame retermination projects" which are very difficult to accomplish.

(h) Integrated digital loop carrier (IDLC) technology allows customer NACs to be terminated directly into the ILEC switch without any physical connection to the MDF. Mr. Wilson argues that the concept of an IDLC NAC is incompatible with the proposed connection of the loop and switch through a SPOT/common collocation frame because IDLC loops cannot be manually disconnected from the switch on a customer-by-customer basis in the way that copper loops can. He alleges that separating IDLC loops from switching in a manner that complies with the SPOT/common collocation frame requirement would relegate CLECs to methods that are rarely available, impractical, and typically cause significant degradation of customer service. As a consequence, Mr. Wilson contends that ILEC customers served by IDLC will be unable to enjoy the benefits of competition.

The ILECs argue that IDLC technology will not make CLEC recombination of building blocks any more difficult. Where customers are served by IDLC, the ILECs will provide CLECs with NACs by attempting to use existing copper facilities that are routed through the MDF. If copper facilities are unavailable, USWC and GTE propose to install equipment in the central office that will enable them to extract the specific circuit supplying the customer. These loops would then be routed to the SPOT frame in the case of USWC or to either the common collocation frame or the carrier’s assigned collocation space in the case of GTE.

Ms. Petti responds that customers now served by IDLC facilities should not be placed on copper facilities previously abandoned by the ILEC because there is no assurance that the copper plant will be in good working condition. There is also the question of whether customers taken off IDLC and placed on copper will have access to the same capabilities as customers that continue to be served by ILEC IDLC. Finally, Ms. Petti expressed concern that the multiple conversions between optical and electrical signals contemplated by the ILEC proposal will diminish the signal quality that customers receive.

Ms. Petti’s recommendation for serving IDLC customers is based upon more recent TR-303 IDLC technology rather than the TR-008 technology contemplated by ILEC proposal. This approach contemplates that electronic instructions would be delivered to the circuit pack located in the IDLC remote terminal, informing the circuit pack to direct certain calls to a "virtual" DS1 established by an operational support system known as OPS-INE. The virtual DS1 would then be routed to the CLEC’s collocation area in the ILEC central office for transport to the CLEC’s remotely located switch. At that point the CLEC would take responsibility for serving those customers.

USWC and GTE respond that IDLC is not a building block and there is no obligation to provision NACs using the specific technology recommended by Ms. Petti. The only requirement, they assert, is to provision NACs meeting the technical specifications established by the Commission. Both ILECs affirm that the building blocks provided to competing carriers will comply with all applicable technical requirements. In addition, neither USWC nor GTE currently has any TR-303 IDLC systems in its network. USWC also claims that Ms. Petti exaggerates the significance of this issue because less than three percent of its Oregon plant utilizes IDLC.

Under the USWC SPOT frame proposal, carriers will be collocated in a common area that is used by all new entrants. Likewise, the GTE common collocation arrangement also contemplates that carriers share access to the distribution frame for purposes of combining building blocks.

Ms. Petti states that customers requiring a high degree of security or extra protection will be reluctant to choose a CLEC as their provider if they know that their service would be exposed to additional potential points of failure or unknown individuals. Customers that use their telecommunications lines for government, scientific, or military information fall into this category, as well as customers concerned about trade secrets. While the ILECs have taken extra precautions to safeguard against tampering and inadvertent service interruptions, Ms. Petti observes that the CLECs would have no control over the common area.

Similarly, Mr. Hydock argues that allowing multiple carriers to access the SPOT frame will raise issues regarding security, network integrity, facilities management, and protection of proprietary and confidential business information. He observes that USWC’s tariff does not address any of these critical matters.

8. Alternative Building Block Access Arrangements.

AT&T witness Wilson maintains that there are better methods of providing access to unbundled elements than those recommended by USWC and GTE. While the most optimal engineering solution would be to require ILECs to provision existing building block combinations, Mr. Wilson recommends other access arrangements that allow recombination of network elements without the network inefficiency, service degradation, or increased expense associated with mandatory collocation. These methods would also provide the same or superior levels of network security associated with the ILEC building block access proposals.

The first alternative proposed by Mr. Wilson is to use the "recent change" capability of the switch to perform the required combinations electronically. This alternative would permit CLECs to combine elements in roughly the same manner as the ILECs do for themselves, but requires software development on the part of the ILECs and software vendors. In view of the limited progress the ILECs have made in developing software necessary to implement nondiscriminatory OSS access requirements, Mr. Wilson acknowledges that the recent change approach cannot not be considered a near-term solution.

The second alternative would allow CLEC technicians to have direct access to the MDF. Mr. Wilson suggests that the ILEC and CLECs agree upon a qualified third party vendor who would dispatch technicians to physically disconnect and reconnect building blocks at the MDF on behalf of each carrier. These technicians would be properly trained and qualified to perform the work and would be allowed unrestricted access to ILEC distribution frames. Under this approach, the technician would first verify that there were no active calls on the loop to be disconnected. Once that was done, the technician would "decombine" the elements by lifting one of the two wires from the terminal block and then "recombine" the elements by placing the wire back on the same terminal it was lifted from.

Although Mr. Wilson acknowledges that direct MDF access is unnecessary and illogical from a network engineering standpoint, he maintains that it is no less illogical than the problem it is intended to solve: that is, the ILECs voluntary decision to take apart existing combinations of network elements. Moreover, both Mr. Wilson and Mr. Croan emphasize that direct access is more efficient and less intrusive and time consuming than the collocation scheme proposed by the ILECs. Whereas the ILEC proposals require extensive coordination of ILEC and CLEC personnel to perform "double hot cuts," the direct access approach would require only a few minutes for a technician to remove and replace a wire.

Direct access is not without disadvantages, however. Removing and reattaching wires is a poor utilization of human resources and is obviously more costly than simply leaving connections in place. The manual nature of the activity also imposes some limits on the number of customers who can change their local provider in a given day.

The Commission decided that telecommunications services should be unbundled into building blocks, or network elements, in 1990, well before the advent of local exchange competition. At the time of our decision, interexchange carriers and access providers were beginning to make competitive inroads and capture customers from the ILECs. In Order No. 90-920, we concluded that unbundling, uniform pricing, and the nondiscriminatory availability of building blocks were necessary to implement the balanced program of regulation and competition specified by the Oregon Legislature in ORS 759.015.

Considerable time and effort was spent developing an incremental cost methodology and analyzing the many intricate details associated with the unbundling process. In July 1996, we issued Order No. 96-188 establishing building blocks and the building block prices applicable to USWC and GTE. The Telecommunications Act of 1996 had been enacted by that time, but the FCC had not yet issued its First Report and Order implementing the terms of the Act.

Like the Act, Order No. 96-188 does not preclude telecommunications carriers from purchasing all of the building blocks necessary to provision a finished retail service. Our thinking was that customers would benefit if carriers could purchase building blocks without restriction, because it would encourage the development of a variety of new service offerings. We also envisioned that building blocks would be used primarily by facilities-based carriers seeking to use ILEC loops and transport elements.

More importantly for purposes of this case, the Commission contemplated that building blocks would continue to be combined by the ILECs and provisioned in the manner described by Staff. That is, competitive carriers would submit orders for specific building blocks and the ILECs would install -- or physically connect -- the designated elements. As part of this process, the requesting carriers would pay nonrecurring charges to compensate the ILECs for reasonable costs incurred to process service orders and provision the requested building blocks. We contemplated this approach because it is the most efficient and reliable means of delivering building blocks to competitive providers from a network engineering standpoint and also ensures that customers will not be burdened with unnecessary costs. Indeed, we considered the benefits of this approach to be so transparent that we did not even address the matter.

As discussed above, the Eighth Circuit’s decision states that ILECs are not obligated to physically combine building blocks for requesting carriers. Requesting carriers may purchase all of the elements necessary to provision a finished service, but they are responsible for assembling those elements themselves. Unfortunately, the record in this case demonstrates quite clearly that the "decombine/recombine" process mandated by the Court will produce building block access arrangements that are less efficient and more complex from a network engineering and operations standpoint. Even more frustrating is that this process will generate unnecessary costs that will ultimately be borne by consumers of telecommunications services.

Our task in this case has been to select building block access arrangements that minimize the cost, complexity, and inefficiency associated with the decombine/recombine process, while ensuring that incumbent and non-incumbent carriers are treated in a reasonable and nondiscriminatory manner. Although the access arrangements adopted in this order are not the most optimal solution, we believe they offer the best opportunity for local competition to proceed in light of the Eighth Circuit’s decision. In the event the U.S. Supreme Court concludes that the Eighth Circuit was incorrect, we will revisit the access issue.

2. Mandatory Collocation.

The Commission rejects the USWC and GTE proposals to require collocation as a precondition to providing requesting carriers with access to building blocks. Although ILECs are obligated to offer physical and virtual collocation under the Act, there is nothing in the Act or the Eighth Circuit’s decision which states that collocation is the only means of obtaining access to unbundled elements. Moreover, the record in this case demonstrates that mandatory collocation will impede the progress of local exchange competition by subjecting requesting carriers to unnecessary costs, delays, risks, inefficiencies, and inferior service quality. For the reasons discussed below, we conclude that mandatory collocation would produce building block access arrangements that are discriminatory and anticompetitive in violation of the Act and Oregon law.

(a) The record indicates that requesting carriers will incur substantial expense in order to provide service under the mandatory collocation arrangements proposed by USWC and GTE. Carriers such as OGI will have to pay as much as $50,000 in up front costs and $1,000 in monthly expenses for each ILEC central office where collocation is required. At OGI’s current level of operations, this amounts to nearly one half million dollars in capital cost just to enable OGI to combine high capacity loops and transport elements. We agree with Mr. Pestaner that the magnitude of these costs, together with the monthly costs to purchase additional tie cables and frame terminations contemplated by the ILEC collocation proposals, will slow the pace of local exchange competition and discourage smaller carriers from serving low density areas. The only carriers that might be able to afford to establish more extensive collocation arrangements are larger, well-financed carriers whose primary interest is serving densely populated urban areas such as the Portland metropolitan area. Even then, the number of central offices in the Portland area would make it very costly to establish a ubiquitous presence. In all likelihood, the collocation of facilities would be limited to those central offices which serve major business districts rather than residential areas.

(b) The USWC and GTE mandatory collocation proposals do not adequately address whether there is sufficient central office space available to accommodate the likely demand for such arrangements. Section 251(c)(3) of the Act requires that requesting carriers must be given reasonable and nondiscriminatory access to unbundled elements. In order to show that mandatory collocation satisfies this requirement, the ILECs must show that they can adequately accommodate carrier requests for collocation in the foreseeable future.

USWC and GTE failed to make this showing. In fact, there is very little information in the record regarding space availability in ILEC central offices. GTE acknowledges that there is no room for physical collocation in its Tigard central office and states that information regarding its Portland area central offices is unavailable. Mr. Hartshorn also acknowledges that a substantial number of collocation requests would exhaust the space available in GTE’s Oregon central offices. USWC did not indicate whether there are space constraints at any of its Oregon central offices, but the record shows that the company has experienced space availability problems in Washington. Aside from this limited information, there is no basis upon which to judge the ability of USWC and GTE to accommodate the demand for collocation space.

USWC and GTE argue that it is extremely difficult to project the demand for collocation space unless competitive providers provide forecasts of their building block requirements. Intervenors acknowledge the importance of forecasting, but also point out that the growth of local exchange competition will depend on many factors, perhaps the most important being the ability to access building blocks on equitable terms. Given these uncertainties, we agree with Mr. Wilson and Ms. Petti that it is reasonable to assume that local markets characterized by effective competition will exhibit roughly the same amount of customer "churn" that now exists in long distance markets. On that basis, it is fair to assume that 40-50 percent of the customers now served by USWC and GTE may elect to receive local exchange service from competitive providers. USWC and GTE have not shown the level of customer demand for CLEC service that can be accommodated by their existing central office facilities. As a consequence, we have no way of knowing whether the ILECs’ ability to provide suitable collocation facilities will be overtaxed by their proposed building block access arrangements.

The ILECs could have shed more light on the space availability issue by conducting an analysis of available central office space under various levels of penetration by competing carriers. Such an analysis could have identified any offices where space limitations exist and discussed ILEC plans for allocating space among carriers or, alternatively, modifying or expanding central offices to accommodate collocation requirements. Without such information, we can only speculate regarding the amount of space available in the ILEC central offices and cannot conclude that USWC and GTE have sufficient facilities to enable them to provide reasonable building block access through mandatory collocation.

It might be argued that the lack of space to establish physical collocation arrangements is not fatal to the ILEC mandatory collocation proposals because the Act provides that carriers may use virtual collocation where there is insufficient space for physical arrangements. The record shows that virtual arrangements take up considerably less space than physical collocations. However, the virtual collocation arrangements offered by GTE and USWC are so complicated and fraught with potential problems that neither can reasonably be expected to be used by carriers to any significant extent to combine building blocks. As described by Mr. Hartshorn, GTE’s proposal contemplates that building blocks would be delivered over tie cables from the MDF to the carrier’s virtually collocated equipment and then delivered over transport facilities to the carrier’s remote facility where the carrier would combine the building blocks. USWC’s virtual collocation proposal requires that carriers combine building blocks at the SPOT frame. Both of these proposals are unnecessarily complex, require the purchase of additional building blocks, increase the potential for technical problems, and are likely to result in diminished service quality. Because the ILEC virtual collocation proposals are impractical and discriminatory in effect, they cannot be considered reasonable alternatives to physical collocation. As such, they do not mitigate our concerns about the lack of available space for physical collocation at USWC and GTE central offices.

(c) Aside from the problem of space availability is the question of the length of time necessary to establish collocation arrangements. USWC and GTE indicate that they will attempt to construct collocation areas within 90 days after arrangements with carriers have been finalized. While we have no reason to doubt that the ILECs will make good faith efforts to meet this objective, there are matters that may prolong the amount of time required to implement collocation arrangements. To begin with, the 90-day time period envisioned by the ILECs does not include any of the preliminary agreements that the carriers must enter into before the collocation area is constructed. Since both USWC’s illustrative tariff and GTE’s interstate physical collocation tariff allow approximately six months to finalize collocation arrangements, it is possible that several months could be consumed before construction is completed. USWC and GTE do not indicate the average amount of time it has taken to finalize the collocation agreements they have negotiated to date.

Second, it is reasonable to assume that a mandatory collocation requirement would precipitate numerous requests for collocation. Not only would multiple carriers seek to collocate, but each of those carriers would likely want to collocate in more than one ILEC central office. USWC and GTE did not present evidence that would enable us to conclude that they have the ability to handle multiple collocation requests in a reasonable and timely fashion. There is little information, for example, concerning the number of ILEC personnel that would be devoted to this purpose or the availability of independent contractors in the event that additional resources are needed to handle a large number of CLEC collocation requests. Without a more detailed and comprehensive showing, there is simply no way for us to find that the ILECs have the ability to implement their mandatory collocation proposals in a manner that will allow local exchange competition to proceed without substantial delay.

(d) Another serious shortcoming of the USWC and GTE mandatory collocation proposals is the disruption of customer service that would occur after existing building block combinations have been taken apart by the ILECs and before new collocation arrangements can be established. As Mr. Pestaner emphasizes, OGI’s customers are faced with precisely this situation. Those customers would have their local service disconnected and have to obtain service from another carrier while OGI establishes the mandatory collocation arrangements required by USWC and GTE.

The inconvenience of having service disconnected is a significant concern for any customer, but the impact on OGI’s business customers would be especially severe. Most businesses rely heavily upon telecommunications service to maintain vital communications links with their customers. Having to involuntarily change providers -- and possibly telephone numbers -- can disrupt the flow of commerce and cause serious financial harm to any enterprise. We agree with Mr. Pestaner that the impact on OGI’s business operations would be nothing short of disastrous. It is very unlikely that any customer would be willing to resume service with OGI after being disconnected and forced to commence service with a different carrier. Surprisingly, there is no indication in the record that GTE or USWC acknowledge the severity of this problem. Nor do we find anything in the ILEC proposals to mitigate the adverse impacts on OGI’s customers.

(e) We agree with Joint Intervenors and Staff that the additional connections required by the ILEC collocation proposals increase the likelihood of circuit failure and diminish the quality of service that new entrants will be able to supply to their customers. This is particularly true in the case of the SPOT frame because even more jumper wires and connections are required.

GTE witness Hartshorn testified that the additional connections pose a negligible risk, but we are persuaded by the testimony of Ms. Petti and Messrs. Wilson, Croan, and Hydock to the contrary. The detailed explanations provided by these witnesses convince us that increasing the number of circuit connections increases the likelihood that something will go wrong. That risk is magnified by the extensive handling and manipulation of wires required by the ILEC proposals. It is also significant that all of these witnesses have experienced situations where reducing the number of circuit connections and intervening distribution frames lessened the number of failures and improved customer service quality. When questioned on this issue, even GTE witness Alfred Banzer acknowledged that additional circuit connections create the potential for additional problems and service breakdowns.

(f) Another troubling aspect of the ILEC access proposals is that they require more connections at the MDF and increase the rate at which the capacity of the MDF is exhausted. USWC witness Schmidt agreed with AT&T witness Petti that "growing" an MDF is extremely difficult. USWC and GTE did not indicate how they intend to address this problem as local competition expands beyond the minimal level that now exists.

There is insufficient evidence in the record to enable us to fully appreciate the operational and engineering implications of the MDF "retermination" projects described by Ms. Petti. We do not know, for example, the time or expense necessary to complete such projects or whether they would entail customer service outages. In addition, expanding the MDF to accommodate the additional connections required by the ILEC proposals would reduce the amount of space available for collocation unless the central office is also enlarged. As Mr. Croan points out, USWC’s SPOT frame requirement exacerbates any existing space limitation problem because increasing competition would force USWC to expand the SPOT frame as well as the MDF.

(g) The ILEC access proposals do not contain sufficient information regarding the technical and operational aspects associated with those proposals. There are a number of issues relating to facilities management, coordination of CLEC and ILEC personnel, and access to building blocks that are not adequately discussed. Without more complete information regarding the methods and procedures that will govern the provision of building blocks, we are left to speculate regarding numerous technical details and the overall adequacy of the USWC and GTE building block access proposals.

3. USWC’s SPOT Frame Proposal.

All of the collocation arrangements proposed by USWC require that building blocks must be combined at the SPOT frame. Although ILECs are required to provide collocation to requesting carriers under the Act, the Commission concludes that the addition of an intervening distribution frame such as the SPOT frame unnecessarily complicates access to building blocks and discriminates against competing carriers by forcing them to combine building blocks in an inferior, inefficient, and more costly manner. Aside from being operationally unnecessary, the SPOT frame places added strain on often already congested frames and delicate wiring, substantially increases the risk of human and mechanical failure, complicates central office maintenance, increases the cost of providing service, and seriously impairs the ability of new entrants to provide reliable local exchange service. Several of these deficiencies also apply to the GTE common collocation proposal.

(a) GTE’s physical collocation proposals demonstrate that there is no technological reason to require an intermediate distribution frame such as the SPOT frame as a precondition to obtaining access to building blocks. USWC’s proposal requires that tie cables run from the MDF to the SPOT frame and from the SPOT frame to the carrier’s collocation space. GTE’s proposal eliminates the SPOT frame entirely and simply runs tie cables from the carrier’s collocation space directly to the MDF.

USWC argues that the SPOT frame is necessary to establish a point of demarcation between ILEC and CLEC facilities. While the record indicates that this was the original reason why USWC developed the SPOT frame concept, it is not necessary to install an intervening distribution frame for that purpose. In GTE’s proposal, for example, the network demarcation point is where CLEC tie cables connect to the GTE MDF.

USWC maintains that the SPOT frame adds flexibility by allowing each carrier to work on its own network without interference, while simultaneously providing USWC with more space to make terminations on the MDF. These arguments are not persuasive. As discussed below, the record shows that the SPOT frame causes a number of operational problems for requesting carriers. It also requires many more connections on the MDF, thereby creating another set of potential problems.

(b) Because the SPOT frame proposal generally requires more jumper wires and connections, it increases the potential for circuit failure as detailed above. USWC argues that the possibility of failure is remote and emphasizes that the only method for combining elements on its network currently requires both a SPOT frame and collocation. It also emphasizes that no evidence has been produced showing an increased incidence of failures at the SPOT frames installed in USWC’s service territory.

USWC arguments are not compelling, however, because there is very little local exchange competition in Oregon and no reliable evidence concerning the extent of competition in other USWC territories. While there may be few problems with the SPOT frame architecture to date, we are persuaded by Mr. Wilson’s observation that:

[a]s significant competition develops and customers begin to churn, the continual activity and increased congestion on the frame caused by installing new cross-connects [jumper wires] and removing old cross-connects will put an unnecessary stress on the frames’ jumpers, at times causing a connection to inadvertently break. The impact of the increased strain on the frame and resultant service failures will be borne disproportionately by the CLECs, because recombination at the SPOT frame will double the number of cross-connects for CLEC loops compared to ILEC loops. Jumpers in a frame (especially the MDF) are already subject to significant pulling and tugging as technicians move other jumpers across or around the frame, or "mine" out old wires that are no longer being used. As this pulling and tugging increases with competitive activity, so too will the CLEC’s service failures.

In its post-hearing brief, USWC offers to cable any building block through the SPOT frame directly to a carrier’s collocation space. We presume this offer was made to alleviate carrier concerns about combining elements at the SPOT frame. While this approach may mitigate certain problems associated with the SPOT frame, it does not alter the fact it is simply unnecessary to require the insertion of an intervening distribution frame for requesting carriers to obtain access to building blocks.

(c) The SPOT frame creates a number of maintenance and provisioning problems that hamper the ability of new entrants to provide reliable service and compete with established incumbent providers. The problems identified by Ms. Petti and Messrs. Croan, Hydock, and Wilson include (a) restrictions on transmission capabilities and the inability to perform testing functions because of longer jumper wires; (b) record keeping difficulties because of manual inventory requirements and USWC’s refusal to pre-wire connections; (c) additional labor costs; and (d) limitations on the number of CLEC customers that can be provisioned on a daily basis. These obstacles unnecessarily complicate the process of combining building blocks and discriminate against new carriers by preventing them from providing service comparable to that provided by the ILECs. USWC did not adequately address these concerns.

In its post-hearing brief, USWC admits that using the SPOT frame to access building blocks is slower and more costly than obtaining ILEC-combined elements. It insists, however, that the unbundling process inherently involves disaggregating the network and that the SPOT frame is the most efficient way to combine building blocks. USWC further emphasizes that the Act only requires ILECs to treat all CLECs equally; not to provide CLECs with access that is "equal in quality," or identical to, the access that the ILEC provides to itself.

Although we agree that ILECs must treat all requesting carriers equally, the duty to provide nondiscriminatory access to unbundled elements is more comprehensive than that expressed by USWC. In our view, nondiscriminatory access requires that the building block access provided to competing carriers must be functionally equivalent to the arrangements the ILEC provides to itself. The record in this case shows that the ILEC mandatory collocation proposals -- and the SPOT frame proposal in particular -- are demonstrably inferior in several respects and will prevent requesting carriers from using building blocks in a manner that allows them to compete effectively with the ILECs in local exchange markets.

Furthermore, USWC’s statement that it is not obligated under the Act to provide "equal in quality" access to unbundled elements is incorrect. Pursuant to 47 C.F.R. §51.311(b) of the FCC regulations, ILECs must provide requesting carriers with access to network elements that is equal in quality to the access that ILECs provide to themselves. This rule was reviewed by the Eighth Circuit and left intact.

Even if the FCC had not mandated "equal in quality" access to unbundled elements, we would impose such a requirement to ensure the nondiscriminatory availability of building blocks. Section 253(b) of the Act provides that "[n]othing in this section shall affect the ability of a State to impose, on a competitively neutral basis . . . requirements necessary to . . . protect the public safety and welfare, ensure the continued quality of telecommunications services, and safeguard the rights of consumers." In addition, ORS 759.050(2)(c) authorizes the Commission to impose reasonable conditions on telecommunications providers to ensure fair competition.

(d) Because of the additional jumper wires and connections required by the SPOT frame and common collocation proposals, requesting carriers must purchase additional building blocks and incur additional monthly costs that ILECs do not incur to provision the same service. These extra costs make it more burdensome for new entrants to compete and compound the discriminatory effects of the ILEC proposals.

(e) USWC emphasizes that all of the interconnection agreements it has executed with competing carriers pursuant to the Act require that carrier facilities be routed through the SPOT frame. As we have noted, the record indicates that the SPOT frame was proposed originally by USWC for security reasons and as a point of network demarcation. It was not until later that USWC designated the SPOT frame as the location where requesting carriers must combine elements.

Since we find that the SPOT frame effectively precludes requesting carriers from obtaining nondiscriminatory access to building blocks as required by the Act and Oregon law, we conclude that it is unlawful for USWC to require any requesting carrier to combine elements at the SPOT frame. Carriers may elect to recombine elements at the SPOT frame if they choose, but they are not required to do so.

4. Direct Access.

The Commission is persuaded that the burdensome and discriminatory nature of the ILEC access proposals effectively precludes competing carriers from combining building blocks. To obviate the shortcomings of the ILEC proposals, we concur with the Joint Intervenors that requesting carriers should be allowed direct access to the MDF and other ILEC facilities for the purpose of combining building blocks. This process will enable carriers to connect the jumper wires between appropriate terminal connections directly on the MDF, rather than at the SPOT frame, common collocation frame, or individual carrier collocation area. As Mr. Croan explains, this approach has the very real and essential benefit of maintaining the exact same network configuration that the ILEC uses in its provision of service without involving the technically inferior access associated with the USWC and GTE proposals.

As we have stated, the Act contemplates that requesting carriers will have greater access to ILEC facilities than is contemplated by the USWC and GTE mandatory collocation proposals. Section 251(c)(3) of the Act provides that requesting carriers shall have "nondiscriminatory access to network elements on an unbundled basis at any technically feasible point. . ." Addressing this section, the Eighth Circuit held:

The FCC and its supporting intervenors argue that because the incumbent LECs maintain control over their networks it is necessary to force them to combine the network elements, and they believe that the incumbent LECs would prefer to do the combining themselves to prevent the competing carriers from interfering with their networks. Despite the Commission’s arguments, the plain meaning of the Act indicates that the requesting carriers will combine the unbundled elements themselves; the Act does not require the incumbent LECs to do all of the work. Moreover, the fact that the incumbent LECs object to this rule indicates to us that they would rather allow entrants access to their networks than have them rebundle the unbundled elements for them.

The Court’s decision makes clear that requesting carriers are entitled under the §251(c)(3) to enter ILEC premises for purposes of combining elements. We interpret this authorization to allow competing carriers to access to the MDF and any other ILEC facilities where building blocks may be combined.

USWC and GTE argue that allowing CLEC personnel to have access to the MDF will threaten the reliability, integrity, and security of the telephone network. Mr. Hartshorn observes that each MDF is connected to thousands of jumper wires and that every time these wires are handled there is a risk that they could be dropped, clipped, or connected to the wrong terminal, resulting in a service outage. He states that direct access will also require GTE and CLEC technicians to act in close coordination to ensure that connections are made properly and that record keeping problems are minimized. He also emphasizes that direct MDF access prevents GTE from ensuring the security of its network and will entail significant equipment, operational, and labor costs in order to guard against interference, to track connections, and to train technicians.

It is apparent from the record that USWC and GTE propose to insulate themselves from the same types of risk that requesting carriers are exposed to under the ILEC access proposals. We do not understand how providing CLECs with access to the central office MDF compromises network security and integrity, but having CLECs share a SPOT or common collocation frame does not. Essentially, the ILECs claim that it is more important to protect the integrity of their circuits than it is to protect the integrity of CLEC circuits. We disagree. ILEC circuits are not entitled to any greater level of reliability or security than CLEC circuits. Nondiscriminatory access requires that all carriers should be placed on an equal footing whenever it is reasonable to do so.

In concluding that MDF access is necessary to ensure nondiscriminatory access to building blocks under the Act, we do not underestimate the concerns expressed by USWC and GTE regarding network integrity and security. We believe, however, that these concerns can be minimized by developing appropriate methods and procedures to govern access and by ensuring that only properly trained and qualified technicians are given access to the MDF and other ILEC facilities where building blocks are combined. Operational and labor costs can also be mitigated by adopting processes that require fewer technicians and eliminate the need for coordinated cut-overs. An example of the latter might be Mr. Wilson’s suggestion that a third party vendor be used to disconnect and reconnect building blocks.

Although direct access is far less burdensome than the access proposals forwarded by the ILECs, it is not as efficient or economical as having the ILECs combine building blocks themselves. For example, where a carrier elects to buy all of the building blocks necessary to provision a retail service, direct access will still require the manual removal and replacement of jumper wires, an activity that would be unnecessary if existing building block combinations are simply left in place. Also, various circumstances may preclude the ILECs and CLECs from using third party vendors to perform the required decombinations and recombinations. In that event, ILEC and CLEC technicians will have to engage in the more complicated process of performing coordinated cut-overs of customer service. These are not insignificant concerns because they will make the decombine/recombine process more cumbersome and increase the overall cost of providing service. Nevertheless, these problems pale in comparison to the complicated processes and unnecessary costs imposed by the ILEC access proposals.

As a final matter, it must be kept in mind that the ILECs are not compelled to dismantle existing combinations of building blocks. In fact, both USWC and GTE are careful to emphasize that they may continue to combine unbundled elements if they choose to do so. That being the case, the ILECs and CLECs may always enter into voluntary agreements allowing the ILECs to continue combining building blocks for a fee instead of having the CLEC perform the combinations themselves. Of course, such agreements would have to be implemented on a nondiscriminatory basis.

USWC and GTE argue that requesting carriers should be required to execute an interconnection agreement before being allowed to purchase building blocks. Although an interconnection agreement may be the prudent course of action in many cases, the Commission agrees with the non-ILEC parties that telecommunications carriers should be permitted to purchase building blocks under the tariffs authorized in this proceeding without having to negotiate separate contracts. Indeed, one of the primary reasons for having a building block tariff is to permit carriers to obtain access to unbundled elements without incurring the time and expense of negotiating an interconnection agreement. OGI currently purchases building blocks from GTE without having an interconnection contract in place.

The position taken by USWC and GTE also appears to be inconsistent with that taken by the FCC. In Application of Bellsouth, supra, that FCC stated:

[a Statement of Generally Available Terms (SGAT) under which a Bell Operating Company provides access to unbundled elements] can provide the proper vehicle for CLECs to use to enter the local market quickly without having to negotiate an interconnection agreement with an incumbent LEC. The Statement may be particularly useful to smaller carriers that wish to do business with the incumbent LEC without becoming involved with formal negotiations.

Although §271 requirements discussed in the Bellsouth case apply only to regional Bell operating companies such as USWC, it is clear that the FCC does not require carriers to enter into an interconnection agreement as a precondition to purchasing building blocks. Like an SGAT, the tariffs filed by GTE and USWC will detail the specific terms and conditions under which building blocks may be purchased by requesting carriers. We see no reason why a separate interconnection contract must be executed.

GTE argues that carriers with an executed interconnection agreement should not be able to pick and choose between tariff provisions and the terms of the interconnection agreement. It maintains that the agreement should be the exclusive means of obtaining facilities or services.

USWC contends that a building block tariff will preclude it from negotiating and entering into interconnection agreements under the Act. It maintains that there is a "direct and inescapable conflict" between the Act’s reliance on negotiated interconnection agreements and allowing carriers bound by such agreements to purchase under a tariff. USWC also claims that the tariff will (a) eviscerate any flexibility it may have in future negotiations under the Act because CLECs will have no incentive to negotiate interconnection agreements; and (b) prevent USWC from negotiating single agreements to govern interconnection in more than one state.

Joint Intervenors, OGI, and UNICOM argue that carriers with executed interconnection agreements should also be free to purchase building blocks under the tariff. They assert that such a requirement is necessary to ensure nondiscriminatory treatment among carriers. On the other hand, these parties maintain that tariff terms, conditions, and prices should not be imposed on any carrier that has executed an interconnection agreement or seeks to enter into such an agreement. They assert that parties will have no incentive to enter into bona fide negotiations if the terms of the tariff apply to all agreements arbitrated under the Act.

Staff states that the terms of an approved interconnection agreement should control over a tariff provision addressing the same issue. Conversely, the tariff should control where the interconnection agreement is silent on a given issue. In addition, staff emphasizes that the Commission retains authority to impose conditions on all carriers whenever necessary to implement telecommunications policy goals (e.g., "slamming" regulations).

The Commission does not agree with USWC’s claim that building block tariffs necessarily conflict with the negotiation/arbitration process set forth in the Act. Nor do we agree that building block tariffs necessarily impair ILEC flexibility in negotiating interconnection agreements. As explained above, carriers are not obligated to enter into interconnection contracts to purchase building blocks. They should not be precluded from using building blocks simply because they do not have the resources or expertise to engage in lengthy negotiations and contentious arbitration proceedings. The Commission has independent authority under Oregon law to establish reasonable building block access arrangements and to ensure that carriers who choose not to proceed under the Act also have the opportunity to compete with incumbent providers.

At the same time, we agree with Staff and the ILECs that carriers bound by interconnection agreements should not be permitted to use the building block tariff to supplement or supplant any terms, conditions, or prices that are covered in an interconnection agreement. A carrier should not be able to opt for more favorable tariff provisions if it agreed upon different terms in the contract negotiation process. Nor should a carrier be able to choose from the tariff if different terms were determined through an arbitration proceeding. This approach is necessary if there is to be any certainty to interconnection agreements. The argument advanced by OGI and Joint Intervenors that carriers must be able to freely select between tariff and contract provisions to ensure nondiscriminatory access to building blocks is not persuasive.

We also agree with Staff that carriers should be able to utilize the tariff if an issue is not covered under an interconnection agreement. ILECs will not be disadvantaged in this situation because there is no opportunity for carriers to pick and choose between the contract and the tariff. Moreover, this approach allows contracting carriers to purchase building blocks that may not have been authorized at the time the interconnection agreement was executed.

B. May a Carrier Purchase ILEC Retail Services and Combine Those Services With Building Blocks? (Issue IV (D))

USWC states that it will not allow requesting carriers to combine building blocks with retail services. GTE does not oppose allowing such combinations.

The Commission sees no reason why building blocks and retail services cannot be combined where the requesting carrier needs services in addition to building block functionalities. For example, OGI currently combines OC3 fiber optic lines purchased out of GTE’s tariff with DS1 transport facilities purchased out of the GTE building block tariff. The reason for this is that OC3 fiber facilities have not yet been denominated as a building block. Precluding OGI from purchasing this combination of fiber optic and transport facilities would penalize OGI’s customers while OGI attempted to self-provision its own fiber optic lines or develop another method of supplying service. We see no basis for such a result and, accordingly, reject USWC’s proposal.

C. May Carriers Use Building Blocks to Provide Switched Access Service? (Issue IV(B))

USWC and GTE argue that requesting carriers should not be allowed to use unbundled building blocks to provide switched access service. They allege that access charges provide a significant subsidy to basic services that will be undermined if carriers are permitted to substitute lower priced building blocks for switched access service. USWC and GTE contend that this issue should be addressed only after the Commission has rebalanced basic exchange rates and has adopted an explicit universal support service mechanism to offset existing subsidies included in switched access rates.

USWC further maintains that it is not required to provide the combination of building blocks necessary to replicate switched access pursuant to the decision in Iowa Utilities Bd. It also states that the FCC is currently reviewing this issue as part of its Further Notice of Proposed Rulemaking attached to the Third Order on Reconsideration.

Prior Commission decisions hold that requesting carriers may purchase the building blocks necessary to provide switched access service. This position is consistent with the regulations promulgated by the FCC.

In addition, we do not find anything in the Iowa Utilities Bd. decision which prohibits requesting carriers from purchasing all of the unbundled elements necessary to provide switched access service. On the contrary, the Eighth Circuit specifically held that CLECs may purchase all of the elements necessary to provision a finished service.

D. Virtual Collocation.

USWC witness Don Mason testified that USWC intends to revise its current virtual collocation tariff and may withdraw its virtual collocation offering altogether. Mr. Mason explained that virtual collocation may be inconsistent with the Eighth Circuit’s decision that ILECs do not have to combine network elements.

The Eighth Circuit decision does not relieve USWC from the obligation to provide virtual collocation. The Court’s decision regarding ILEC combinations focused on §251(c)(3) of the Act. It does not address §251(c)(6), which requires ILECs to offer physical collocation to requesting carriers, and if that is unavailable, virtual collocation.

In addition, the FCC has recently held that "new entrants must be able to choose virtual collocation as a method of combining network elements, regardless of whether physical collocation is practically available." The FCC’s position is based on 47 C.F.R. §51.321(b), which was not vacated by the Eighth Circuit.

For these reasons, USWC shall maintain a virtual collocation tariff. The company must offer that form of collocation in a manner consistent with FCC and OPUC requirements.

E. Common (Shared) Transport Building Block.

In Order No. 96-188, the Commission determined that common or "shared" transport should be unbundled as a building block. The FCC subsequently concluded in its Third Order on Reconsideration that shared transport should be categorized as an unbundled network element. The shared transport building block is comprised of interoffice transport transmission facilities that are shared between the ILEC and one or more requesting carriers. It includes ILEC transmission facilities between end offices, between tandems, and between tandems and end offices.

Unlike GTE, USWC refuses to provide the shared transport building block to requesting carriers. It argues that shared transport is actually a combination of unbundled elements rather than a single element, because it requires the ILEC to use multiple transport facilities together with the routing tables contained in ILEC switches in order to transport carrier calls. Accordingly, USWC contends that shared transport is an impermissible combination of network elements that ILECs are not required to offer under the Eighth Circuit’s ruling.

USWC’s position is contrary to the Act. On August 10, 1998, the Eighth Circuit upheld the FCC’s finding that shared transport is a network element and must be supplied by ILECs to requesting carriers pursuant to §251(c)(3) of the Act.

USWC’s position regarding shared transport also conflicts with the legislative mandate in ORS 759.015 to provide a balanced program of regulation and competition. USWC’s proposal obligates requesting carriers to purchase dedicated transport between each end office and between each end office and the tandem switch. In effect, this requires each CLEC to construct a parallel transport network. As AT&T witness Wilson and others emphasize, such an approach is extremely inefficient from a network engineering standpoint, poorly utilizes ILEC interoffice transport facilities, and requires unnecessary deployment of additional transport facilities by requesting carriers.

 The record shows that failure to provide shared transport would dramatically exacerbate problems CLECs have experienced obtaining an adequate number of trunk terminations on USWC switches. If carriers are required to install dedicated transport facilities as USWC proposes, there must be a sufficient number of trunk terminations available to accommodate demand. USWC did not disclose how many of its end office and tandem switches face such limitations or what measures would be necessary to resolve the problem.

Another problem with USWC’s position is that older switches may not have a sufficient number of routing indices to route all of the dedicated trunks that would be required. According to Mr. Wilson, a new routing index would be required for each new dedicated trunk. A large increase in the number of trunks could tax the routing capacity of vintage switches such as the 1AESS switch. As with the trunk termination issue, USWC did not address these concerns.

In addition, USWC’s proposal would delay the widespread growth of local exchange competition. If new entrants must install dedicated transport facilities between all ILEC end office and tandem facilities, they will probably concentrate their efforts in urban population centers. Carriers are unlikely to build transport facilities to serve less densely populated areas until their competitive position is firmly established. Access to shared transport, on the other hand, permits new entrants to extend their offerings to a greater number of customers without incurring the responsibility of building parallel transport networks.

In summary, there is no merit to USWC’s position regarding shared transport. Requiring new entrants to install dedicated transport facilities would necessitate the installation of duplicative facilities at substantial expense and without any assurance that USWC’s switches will be capable of terminating all of the additional trunks that would be required. The net effect would discriminate against competing carriers, delay the emergence of competition, and needlessly subject telecommunications consumers to additional costs. The Eighth Circuit has sustained the FCC’s determination that shared transport is a network element that is critical to the development of local competition and must be provided to competing carriers under the Act. We further conclude that shared transport is essential to satisfy the requirements of Oregon law.

F. Elimination of Building Blocks.

USWC proposes eliminating certain building blocks in this proceeding. Joint Intervenors oppose USWC’s proposal. Staff opposes deleting some of the building blocks specified by USWC.

This docket was initiated to consider tariffs filed by USWC and GTE to implement the building blocks authorized in docket UM 351. The Commission did not notify interested persons that consideration would be given in this proceeding to eliminating building blocks altogether. Accordingly, it is inappropriate to consider USWC’s proposal at this time. If USWC wants to pursue this matter, it should submit a separate filing specifying the building blocks it believes should be eliminated from the current list.

G. USWC Cageless Collocation Proposal.

As described by Mr. Schmidt, cageless collocation allows carriers to place their equipment alongside USWC’s central office equipment, thereby eliminating the costs and delays associated with building collocation cages. Although many of the details associated with cageless collocation are still unclear, there are indications that the proposal may prove useful to requesting carriers. For example, Mr. Croan noted that TCG and USWC have agreed to implement a form of cageless collocation in Arizona.

In our view, the principal shortcoming of the cageless collocation proposal is USWC’s requirement that CLEC facilities must be routed through the SPOT frame. As we have emphasized, the SPOT frame is unnecessary and creates numerous problems that discriminate against new entrants. Given that cageless collocation allows CLECs to install and maintain equipment next to ILEC equipment, we do not understand why carriers should not also be permitted to connect their equipment directly to the MDF.

In addition to the SPOT frame issue, a number of other details must be resolved before we can evaluate the merits of cageless collocation. These include issues such as the amount and type of carrier equipment that can be placed in ILEC central offices, as well as methods and procedures governing carrier access, security, etc. We presume these matters will be addressed in future proceedings.

H. Integrated Digital Loop Carrier (IDLC).

There was substantial debate concerning how loops will be supplied to requesting carriers where customers are served by IDLC facilities. There are a number of technical problems associated with separating out loops from IDLC facilities so that requesting carriers may recombine them at a SPOT frame or collocation space. The problems stem from the fact that digital carrier bypasses the MDF and is connected directly to the ILEC switch.

USWC witness Mark Schmidt accurately described the dilemma created by decombining network facilities:

I think, to cut to the chase, the system was designed to work the way it is built today. To tear that apart and provide a CLEC with a NACC and a NAC in that scenario is very difficult, both for U S WEST and the CLEC. And there’s no guarantees, in my mind, that are tested and that we know about that once you do some things to tear that apart, depending on the way the CLEC would recombine it, who knows if it will work?

* * * * *

And I -- Mr. Wilson and I are in complete -- complete agreement that that’s not the way those systems were designed. They were not designed to be broken apart, they were designed to be kept together.

We agree with Mr. Schmidt and Mr. Wilson that disassembling the functions of an IDLC does not make sense from either a technological or an engineering standpoint. Nevertheless, the Eighth Circuit has held that the ILECs may separate existing building block combinations for recombination by the CLECs. As explained above, this process requires that the ILECs either locate spare copper facilities or install additional equipment to provision unbundled loops to requesting carriers.

Joint Intervenors maintain that the methods proposed by USWC and GTE to separate out IDLC-provisioned loops may impair the capabilities of the loops purchased by requesting carriers. USWC and GTE respond that the Commission should not mandate the specific technology (i.e., digital carrier or metallic facilities) used to provision the NAC building block. The ILECs affirm that all loops supplied to CLECs will meet the technical specifications established by the Commission.

We agree with the ILECs’ position with the following caveats: First, the NACs provided to requesting carriers must be capable of providing service equivalent to that received by ILEC customers in the same serving area. As Staff emphasizes, effective competition can only be realized if all carriers have access to facilities of the same service quality. Thus, if the ILECs choose to supply NACs using available copper facilities instead of digital carrier facilities, the CLEC customers served by those facilities must have access to the same type of telecommunications services available to ILEC customers in the same area.

Second, in instances where the ILECs cannot use copper facilities to provision NACs and must install the additional central office equipment described on pages 31-32 of this order, USWC and GTE must provide all of the NAC building blocks that can be provisioned through that equipment. Thus, if the equipment is capable of separating out DS1 NACs as well as DS0 NACs as Mr. Schmidt described, then both types of building blocks must be made available to requesting carriers.

A. Recurring vs. Nonrecurring Costs.

Recurring costs are the ongoing costs associated with providing a service or building block. These costs are generally investment-related and include both capital costs and operating expenses. Recurring costs are often presented as a cost per month or cost per unit of usage (e.g., minute of use) and are incurred throughout the time period the service or building block is provided to the customer.

Nonrecurring costs, on the other hand, are the one-time costs that are incurred at the time a customer establishes, disconnects, or changes service. These costs normally result from customer orders and are predominately labor-related. Nonrecurring costs include, for example, service order processing expenses associated with data entry and record keeping activities.

Both recurring and nonrecurring costs include labor expenses. Recurring costs include capitalized labor costs related to the placement of facilities, as well as labor costs associated with providing routine maintenance. An important issue relates to whether it is appropriate to capitalize a labor cost or whether the labor cost is required to process a service order or building block request. Generally, labor costs that are incurred to place any reusable plant should be capitalized, depreciated over a useful life and included in recurring costs, not levied as a one-time, up-front charge.

B. Nonrecurring Costs vs. Competition Onset Costs.

USWC, Staff, and the Joint Intervenors agree that nonrecurring costs must be differentiated from one-time start-up -- or competition onset -- costs. Whereas nonrecurring costs are typically incurred when a service is installed, modified, or disconnected, competition onset costs may be associated with an extraordinary event, such as the implementation of the Act. For example, competition onset costs may be incurred when operations support systems are modified to provide competing carriers with access to unbundled network elements.

Only GTE argues that nonrecurring costs should recover start-up or competition-onset costs. It contends that nonrecurring costs must include "expenditures to upgrade and improve operations support systems" and "to build and staff a wholesale order processing infrastructure." GTE witness Alfred Banzer further testified that some or all of the costs for "systems modifications, new systems, network rearrangements and augmentations and training" should be included in nonrecurring costs assessed to competing carriers. GTE claims that these costs are nonrecurring because they are "generally incurred on a one-time basis when service is ordered and provisioned" and are "properly recovered . . . through one-time charges to the CLEC when it orders network elements."

GTE’s claim that competition onset costs should be included in nonrecurring costs is incorrect. To begin with, competition onset costs, unlike nonrecurring costs, are not incurred each time a CLEC orders a building block. Rather, these costs are incurred before the CLEC places any orders with the ILEC. As AT&T/MCI witness Dr. Thomas Zepp explainss:

[C]ompetition onset costs are those costs that an ILEC must incur in order to comply with the most general pro-competitive provisions of the Act -- that is, the costs an efficient monopoly provider must incur to enable its network to be used efficiently by multiple carriers. These costs are not caused by competing carriers’ actual entry, but must be incurred to make entry possible. For example, U S WEST has claimed in this State and in other jurisdictions that the costs of designing electronic interfaces . . . fall into this category. The existence and magnitude of these costs . . . however, would not be attributable to a particular carrier’s request for services, unbundled network elements or facilities. These costs stem from the Act’s mandate that local exchange markets should be open to competition and that new entrants should have nondiscriminatory access to the incumbent’s network. (emphasis added)

We agree that competition onset costs -- including those associated with modifying operational support systems and infrastructure necessary to provide access to unbundled elements -- are not caused by new entrants, but rather result from the passage of the Act and prior decisions by this Commission. As Dr. Zepp points out, the lack of cost causality is evident by the fact that ILECs would still have to incur these costs in anticipation of competition even if no competitor entered an ILEC’s service territory. Consequently, we agree with Dr. Zepp that including competition onset costs in the up front, nonrecurring charges paid by new entrants for building blocks would violate the cost causation principle adopted in docket UM 351.

GTE’s position also conflicts with the pro-competitive goals of the Act. Dr. Zepp correctly notes that the social benefits of local exchange competition will inure to the public at large, not merely customers who obtain services from new entrants. It would undermine the objectives of the Act if prospective entrants are required to bear all of the ILECs competition onset costs because it would discourage the very entry that Congress intended.

The decision not to include competition onset costs in nonrecurring costs does not mean that such costs may not be addressed elsewhere. As staff points out, competition onset costs were addressed in docket UT 135 concerning USWC’s interconnection cost adjustment mechanism filing.

 C. TSLRIC Methodology.

All parties except GTE agree that nonrecurring costs should be based on the total service long-run incremental cost (TSLRIC) methodology adopted by the Commission in Phase I of docket UM 351. As noted above, the cost principles and methodology used to develop the recurring building block costs and prices adopted in Phase II of docket UM 351 are incorporated in the Telecommunications Cost Report, Vols. I and II.

GTE advocates that nonrecurring costs should recover the "actual costs" to provide wholesale services -- including building blocks -- to new entrants. "Actual costs" are defined as the actual labor and system costs that GTE can expect to incur to respond to CLEC requests for building blocks, as opposed to hypothetical costs or estimated costs of another firm. GTE argues that the Act, the Eighth Circuit, and Oregon case law all mandate ILEC recovery of actual costs.

Although GTE contends that its method of calculating nonrecurring costs corresponds with the incremental cost principles set forth in the Telecommunications Cost Report, the record indicates that it in fact seeks recovery of the embedded, or accounting, costs incurred to provide building blocks to CLECs. As discussed in greater detail below, the TSLRIC methodology is designed to compensate ILECs for efficiently-incurred, forward-looking costs. It is not designed to recoup 100 percent of an ILEC’s fully embedded cost of providing service. The costs produced by the TSLRIC methodology are not used to produce a specific ILEC revenue requirement or rate of return, as is the case in a general rate proceeding when the accounting or embedded costs of the utility are considered.

Contrary to GTE’s claim, the Act does not mandate recovery of "actual" or embedded costs. In fact, §252(d)(1)(A)(i) specifically provides that state commissions shall establish rates for interconnection and network elements based on cost determined without reference to a rate-of-return or other rate-based proceeding. In short, the Act prohibits the Commission from considering the ILEC’s revenue requirement when setting prices for unbundled elements and access to unbundled elements.

The decisions of the Eighth Circuit also cannot be read to require recovery of "actual" costs as that term is defined by GTE. To begin with, the Court did not find Section 252(d)(1)(A)(i) unlawful. Although it repeats the Act’s mandate that ILECs are entitled to recoup the costs of interconnection and unbundled elements, the Court did not specify a particular method of cost recovery or disagree with Congress’ decision that such costs must be calculated without reference to rate-of-return or rate-based proceedings. Also, while the Court rejected the FCC’s attempt to assert pricing jurisdiction under the Act, it did not conclude that the incremental cost methodology proposed by the FCC was unlawful. In fact, since the passage of the Act, numerous state regulatory commissions have held that unbundled element prices should be based upon incremental cost. In addition, more than one federal district court has sustained the use of incremental cost pricing.

GTE also relies on American Can Co. v. Lobdell to establish that it is entitled to recover its "actual" costs. American Can involved a revenue requirement proceeding to establish retail rates. A primary issue facing the Court was whether or not the total revenues produced by the retail rates of the utility would be sufficient to provide the utility with a reasonable opportunity to earn its revenue requirement.

American Can is not dispositive of the issues presented in this case. As we have explained, this is not a general rate case to establish retail rates or GTE’s overall revenue requirement, but rather a proceeding to consider issues relating to the ILEC’s wholesale building block tariffs. Although the American Can decision affirms the fundamental principle that a regulated utility must have the opportunity to earn its total revenue requirement, the principle does not apply to the revenue generated by every element, service, or facility provided by the utility. There is no obligation that each element, facility, or service recoup its fully embedded cost.

The arguments raised by GTE regarding "actual" cost recovery are similar to those raised by the company in docket UM 351 when recurring building block rates were established. In response to these concerns, the Commission requested that GTE file a rate rebalancing case so that we might review the company’s overall revenue requirement. Although we made that request more than two years ago, GTE not only failed to file a rebalancing case, it insisted that the Commission cannot require such a filing. As a consequence the Commission initiated an investigation at Staff’s request to review GTE’s earnings. On September 28, 1998, we issued Order No. 98-388, reducing GTE’s Oregon revenue requirement by $25 million annually, or approximately 11 percent. Under these circumstances, it is difficult to attach much weight to GTE’s present claim that its revenues are in jeopardy.

In summary, the Commission finds that that the nonrecurring costs adopted in this docket should be based on the same TSLRIC methodology used to develop recurring building block costs and prices. Nonrecurring costs should not be based on the ILECs’ historic, embedded costs.

 D. Cost Principle No. 2 -- Choice of Technology.

Although most parties agree that the nonrecurring costs for provisioning building blocks should incorporate the TSLRIC cost principles established by the Commission in docket UM 351, there is disagreement over how certain cost principles should be interpreted.

The most significant dispute involves the interpretation of Cost Principle #2 which addresses the choice of technology used in ILEC cost studies. It requires that the technology analyzed reflect the overall least cost technology that is currently available. The selection of the least cost technology is not based on the economics of adding to the current stock of telecommunications equipment providing service today, but rather assumes no equipment is currently in service and a completely new network is to be installed. The latter concept is consistent with Cost Principle #1--Definition of Long Run which defines the "long run" as a period long enough that all inputs are avoidable.

USWC states that nonrecurring cost studies should identify the forward-looking costs associated with nonrecurring activities necessary to provision building blocks. In determining these costs, it advocates that the Commission should use the least cost technology and methods of operation that are currently available today; that is, technology that is "commercially available to USWC and readily deployable on a mass scale." USWC witness Brigham emphasizes that forward-looking studies should not consider hypothetical future technologies, because it is impossible to estimate the cost or to determine how such technologies would be configured. Such an approach would also deny USWC the opportunity to recover its true forward-looking costs and prevent it from investing in necessary network infrastructure.

GTE affirms that NRCs should be based on forward-looking costs but, as noted above, seeks recovery of the "actual" costs it will incur to provision building blocks. According to witness Kevin Collins, the appropriate costs "must be those for the company’s actual activities and processes that will take place in responding to a wholesale service order [building block] request from a CLEC."

Staff and Joint Intervenors argue that USWC and GTE have redefined forward-looking least cost technology by utilizing cost studies that take into account the technology the ILECs actually use rather than the least cost technology currently available. These parties refute the claim that USWC and GTE currently utilize the most efficient technology available.

The Commission agrees with USWC that nonrecurring cost studies must rely on realistic assumptions about the time and costs associated with performing the nonrecurring activities necessary to provision building blocks. We also agree that nonrecurring costs and rates should not be based on hypothetical assumptions and future technologies. At the same time, the cost principles we have adopted require us to take into account currently available technology, not merely the technology USWC and GTE have deployed. As noted above, those principles require that cost estimates should not be driven by equipment selection choices that are influenced by the existing stock of equipment, but should assume investment in forward-looking technologies. Also, the equipment selected for inclusion in the cost studies should be based on a cost minimization approach with no constraints on the selection of current technology to serve customer demand for telecommunications services.

E. Nonrecurring Cost Studies.

Nonrecurring cost studies were prepared by USWC, GTE, Staff, and AT&T/MCI. The following paragraphs contain a brief description of the studies offered by the parties. A more detailed discussion of the technical assumptions included in the studies -- and the disputes regarding those assumptions -- is set forth in subsequent sections of this order.

1. USWC Studies.

USWC prepared eleven different studies to calculate the nonrecurring costs to provide building blocks. Each study incorporates the following six steps: (a) identifying the nonrecurring (one-time) activity; (b) estimating the work time necessary to complete the activity; (c) calculating the expenses for the activity by multiplying the work time by the appropriate labor rate; (d) aggregating the costs for each separate activity to obtain a subtotal; (e) loading the subtotal with service specific costs to produce a TSLRIC estimate; and (f) allocating shared and common costs to the TSLRIC estimate to obtain the nonrecurring cost of provisioning the building block.

2. GTE Study.

GTE developed its own nonrecurring costs and prices for service ordering and provisioning functions. Nonrecurring costs and prices for other activities (e.g., special circuit design and special construction activities) mirror those in USWC’s tariff.

GTE’s nonrecurring cost study was developed by a team of company experts to estimate the costs of processing orders through a new national ordering center -- the National Open Market Center (NOMC) -- created for the purpose of providing unbundled services, resold services, and other services to CLECs as required by the Act and various state regulatory requirements. GTE’s nonrecurring costs are based on the service ordering costs of the NOMC plus the network provisioning costs of its Oregon operations.

3. Staff Study.

Staff analyzed the nonrecurring costs and prices developed by USWC and GTE and recommends several adjustments. The differences between the ILEC studies and Staff’s proposed nonrecurring costs and prices stem primarily from Staff adjustments which: (a) recognize additional efficiencies when certain building blocks are provisioned simultaneously; (b) reduce customer service representative times to account for electronic transmission of orders; (c) remove costs to engineer, furnish, and install facilities; and (d) make other adjustments, including increasing labor rates to 1999 levels, adjusting activity times, and eliminating double counting of costs between recurring, nonrecurring, and overhead costs.

4. AT&T/MCI Study.

AT&T and MCI recommend adoption of the Nonrecurring Cost Model (NRCM) to estimate the nonrecurring costs incurred to provision unbundled elements. The NRCM models individual tasks and activities for each type of building block request and calculates costs associated with preordering, ordering, and provisioning processes. Because permutations are probable for each service type, the NRCM develops scenarios that incorporate varying activities and costs. Basically, the model develops costs for three functions -- migration, installation, and disconnection.

Because the nonrecurring costs associated with the unbundled network elements produced by the NRCM do not correspond precisely with the nonrecurring costs for Oregon building blocks produced by the USWC cost model, AT&T and MCI have also remodeled USWC costs by incorporating assumptions included in the NRCM. Both the nonrecurring costs produced by the NRCM and the costs produced by the remodelling effort are substantially less than the nonrecurring costs proposed by USWC and GTE in this docket.

VII. NONRECURRING COST STUDIES -- DISPUTED ISSUES

In the following sections, we analyze several disputed issues pertaining to the nonrecurring cost studies prepared by the parties. The decisions on these issues shall be incorporated into the revised nonrecurring costs and charges filed in accordance with this order.

A. Service Order Processing Costs.

This issue involves the amount of human intervention necessary to process a CLEC order for unbundled elements. The USWC and GTE studies assume that all building block orders submitted by CLECs must be reviewed by ILEC representatives before the orders can be routed for further downstream processing. Staff and Joint Intervenors, on the other hand, assume that ILEC representatives will have to review CLEC orders a smaller percentage of the time. A greater amount of manual intervention increases the nonrecurring costs paid by competing carriers.

1. Flow through and Fallout.

Operations support systems (OSS) are electronic, software-driven computer systems and data bases used by telephone companies to manage preordering, ordering, provisioning, maintenance, repair, and billing functions for their retail and wholesale operations. By linking different functions together, OSS provide more efficient and effective control of ongoing network operations. OSS have been classified as a UNE by the FCC, and as a building block by this Commission.

At one time, the functions necessary to process a service order were extremely labor intensive, requiring constant human intervention to complete each order and update inventories. In recent years, OSS have been reengineered to reduce the need for direct human intervention by automating the processing of service order transactions, including service connections, disconnections, moves, and service changes. As these automated systems developed, the goal has been to achieve "flow through"-- meaning that a service request can be processed through several computer systems without the need for human intervention.

The opposite of "flow through" is "fallout." If an error occurs as data flows through computer systems, it may cause service orders to fall out of the system and manual activity may be required to correct or complete the order. Fallout is significant because it is a major cost driver underlying nonrecurring costs.

2. AT&T/MCI

The AT&T/MCI NRCM models the cost of processing a service order separately. It is designed to model the processes and procedures currently used by large ILECs and assumes fully automated OSS that allow maximum electronic flow through of CLEC local service order requests (LSRs) for building blocks. The NRCM assumes that CLEC LSRs will be processed with a fallout rate of two percent. In other words, it assumes that 98 percent of total CLEC orders will flow through the ordering process with no manual intervention by ILEC personnel.

AT&T/MCI witness Petti states that well managed and maintained OSS will have very little fallout. This is especially true in a competitive environment because fallout impacts delivery intervals, restoration/response times, and the overall cost of service. Companies thus have market incentives to continually improve customer service by minimizing fallout. Ms. Petti describes the service order process modeled by the NRCM as follows:

In the normal order scenario, the customer will directly contact the CLEC service representative. That same CLEC service representative will discuss the service desired by the customer and then determine what services are to be ordered. The CLEC service representative will obtain preordering information and access U S WEST’s OSS to obtain the customer service records, the availability of service, and other pertinent information. The CLEC representative will then input the request into the U S WEST Service Order Generator ("SOG") through the electronic gateway. There is no need for the ILEC service representative to be involved in any of these activities. If the order is rejected for some reason, then the order should return across the gateway to the CLEC for correction. The only instance in which the CLEC service representative would contact [the ILEC service representative] for assistance would be in the case of an order that requires investigation due to faulty records or other inconsistency. Such assistance would be minimal.

Ms. Petti notes that Southwestern Bell Telephone Company (SWBT) has implemented a system known as the Easy Access Sales Environment (EASE) system, that allows 99 percent flow through of CLEC orders for resold services purchased from SWBT. In addition, SWBT indicated that it expected to develop an "EASE-like" system by mid-1998 that will be capable of processing orders for unbundled elements. According to SWBT, the new system will have a flow through rate similar to that experienced with the EASE system.

USWC and GTE argue that the flow through/fallout assumptions embodied in the NRCM are unrealistic and assume a technology that is not currently available. GTE witness Francis Murphy argues that the NRCM does not properly model the processes and procedures used by GTE or reflect GTE’s experience with fallout. Mr. Murphy points out that there are two types of fallout. The first is due to errors on faxed or electronically submitted LSRs received from CLECs. These must either be returned to the CLEC for correction or the ILEC must intervene to correct or explain the error. In either case, manual intervention is required. Mr. Murphy asserts that GTE currently experiences a 50-80 percent error rate on orders submitted by new CLECs.

The second type of fallout occurs when LSRs are rejected by electronic systems. The rejected orders must either be corrected by the ILEC representative or returned to the CLEC for correction. Currently, GTE’s electronic systems reject about 37 percent of the LSRs submitted by CLECs.

Mr. Murphy asserts that the high flow through rate in the NRCM fails to recognize that ILECs must keep processes and procedures in place to serve new and/or smaller carriers that do not order building blocks electronically because of the capital expenditures required. These carriers place orders manually (e.g., via fax) and require substantial time commitments from the ILECs. Mr. Murphy states that the NRCM is flawed because it does not account for the fact that GTE and USWC must accommodate building block requests from all CLECs, not merely those using electronic interfaces.

GTE witness Kevin Collins emphasizes that it is important to distinguish between receiving orders electronically and processing orders electronically. He estimates that electronic OSS will not be fully implemented for another 2-3 years. At that time, Mr. Collins expects that GTE will electronically receive and process only 64 percent of the building block orders submitted by CLECs.

USWC witnesses Brigham and Buhler also maintain that the 98 percent flow through rate included in the NRCM cannot be achieved in the foreseeable future. To ensure that costs are forward-looking, nonrecurring cost studies should assume that some mechanized processes are used to perform certain nonrecurring functions. In most cases, however, mechanized processes can be used only for a certain percentage of orders, since no computer system can respond to all questions or be completely reliable. Nonrecurring cost studies must include realistic estimates of whether manual intervention will be required for each activity.

In addition to fallout resulting from incorrectly submitted orders, Mr. Buhler states that LSRs will fallout because they are designed to be completed manually. He notes that many types of CLEC orders cannot be processed using automated flow through because of their inherent complexity and constraints relating to the systems involved. For example, turning a USWC customer loop into an NAC that can be purchased by a CLEC requires a "disconnect" order to terminate the USWC retail service and a separate "connect" order to convert the retail customer loop into an NAC. In order to reuse these same facilities, the orders must be coordinated by a person. This is accomplished by inserting a "Map F" field identifier code. Mr. Buhler states that the NRCM understates the work ILECs must perform to provision building blocks by assuming these functions will be performed automatically.

Mr. Buhler also emphasizes that processes modeled by the NRCM do not correspond to the actual systems used by USWC to process service orders. As a result, the NRCM understates nonrecurring costs. For example, the NRCM assumes that systems used to process access service requests (ASRs) from IXCs will also be used for CLEC LSRs. Mr. Buhler asserts that LSRs for unbundled elements must be supported by different OSS because they involve more complex customer data and facilities. In their current form, the software systems used for ASRs do not have the necessary online interfaces to support the data acquisition and functionality requirements of preordering, ordering, and provisioning necessary for LSRs.

GTE and USWC contend that it is unreasonable to rely on the SWBT EASE system as evidence of the flow through rate that will be experienced for building block orders. They emphasize that the EASE system is only used for resale orders and that the new "EASE-like" system planned by SWBT is not ready for deployment today. Contrary to Ms. Petti’s claim, the ILECs also assert that the new system will have a much lower flow through rate for unbundled elements than experienced under the EASE system.

3. USWC and GTE.

The USWC and GTE nonrecurring cost studies assume that manual intervention will be necessary to process every CLEC building block order. In other words, the ILEC studies assume zero flow through of CLEC orders. According to Mr. Buhler, USWC representatives in the Interconnect Service Center must review all orders to ensure that the information on them is accurate and complete. The information must then be manually rekeyed into the USWC service order processor by the USWC service representative. As noted above, a CLEC order for unbundled loops will cause USWC to issue a disconnect order to unbundle the connection to USWC’s switch and the customer’s account, and a separate connect order to provide the loop to the CLEC’s collocated equipment. Both of these orders require manual intervention to coordinate the cut-over of circuits, even with the recent deployment of the Electronic Data Interchange (EDI) interface. According to Mr. Buhler:

The issue of EDI and the issue of flow through are independent of one another because EDI just allows you to communicate with an operational support system. It isn’t intended or built or designed in and of itself to do the conversion from the EDI messages into the proprietary -- the unique languages used by a company’s operational support system. So, in other words, EDI doesn’t in and of itself automatically convert the CLECs orders into USOCs [Universal Service Order Codes] and FIDs [Field Identifier Codes] and format the order for U S WEST’s downstream systems. We need another component with all of the business rules and all of the mappings that will accomplish that . . . . Until, then we still will have a manual step to input unbundled loop orders.

In its initial filing, USWC estimated that it would take ISC service representatives 45 minutes to process an order to connect an unbundled loop and 30 minutes to process an order to disconnect a loop. At the hearing, Mr. Brigham revised these estimates downward to 35 minutes and 15 minutes, respectively, to account for utilization of USWC’s IMA interface. These labor time savings lowered USWC’s proposed nonrecurring charge for the basic NAC from $117.49 to $97.55. The nonrecurring costs and prices of provisioning other building blocks were also reduced.

GTE’s labor time estimates for NOMC personnel are designated as confidential. We observe, however, that GTE’s cost study assumes that it will take NOMC representatives significantly more time to process a CLEC order to install an unbundled loop than estimated by USWC. On the other hand, GTE also allocates less time to disconnect activities than does USWC.

Mr. Buhler points out that USWC is in the process of modifying and updating its OSS to achieve some flow through of building block orders. The flow through rate that USWC will achieve is unknown, however, since the technology is still being developed. Moreover, predicting precise flow through levels is difficult because USWC and other ILECs have no previous experience processing building block orders.

Joint Intervenors challenge the assumptions underlying the USWC and GTE cost studies. In general, they maintain that the studies focus improperly on ILEC systems and facilities that have been deployed rather than more efficient technologies currently available. As a result, they claim that the USWC and GTE studies inflate CLEC service order costs because they assume that OSS functions require significantly more manual intervention than is necessary.

In particular, Joint Intervenors dispute USWC’s claim that all CLEC service orders must be individually reviewed for accuracy by an ILEC service representative and then manually rekeyed into the USWC service order processor. As noted above, Joint Intervenors maintain that there is no need for ILEC personnel to engage in these activities. They contend that a more efficient solution is to develop a software translation table that converts CLEC orders into the appropriate format for USWC’s downstream systems.

4. Staff.

Staff states that technology necessary to perform electronic processing of OSS functions is available now and will be fully implemented within the long run period contemplated by TSLRIC cost principles. Nonrecurring costs should therefore be based on the use of fully automated OSS.

Staff witness Jack Breen agrees that fully electronic interfaces make it possible to process CLEC service orders with minimal manual intervention. He points out, however, that not all carriers will order electronically and that some manual intervention will be required by ILEC personnel to process such orders. Mr. Breen recommends that service order costs be computed using a weighted average of electronic and manual processing costs. His analysis assumes that 64 percent of orders will be handled electronically, and 36 percent manually.

USWC and GTE maintain that the 64 percent electronic flow through assumption used by Staff does not accurately reflect the capabilities of currently available technology. They argue that Mr. Breen’s analysis incorporates flow through rates that GTE may achieve in the future only if it modifies its OSS to include additional electronic interfaces.

Joint Intervenors argue that Staff’s proposal will provide incorrect economic signals to CLECs. They maintain that a weighted average charge will force CLECs who have implemented electronic OSS to subsidize competitors who have not made such investments. In the alternative, Joint Intervenors recommend separate cost-based rates to encourage CLECs placing manual orders to deploy more efficient and economical OSS technology currently available.

5. Commission Decision -- Service Order Processing Costs.

Based on the evidence presented, the Commission is persuaded that the technological capability exists to process CLEC service orders for unbundled elements on an automated basis using electronic OSS. Although USWC and GTE have not fully implemented such systems to date, we agree with Joint Intervenors and Staff that the technology is available to produce computer programs that can process unbundled element orders with a minimum amount of human intervention.

In particular, we see no reason why it is necessary for ILEC service representatives to review all CLEC orders for accuracy and manually reenter the translation codes necessary to permit downstream processing of the order. As Ms. Petti explains in her testimony, properly designed software programs obviate the need for ILEC personnel to perform these tasks. In USWC’s case, for example, programs can be used which automatically convert the information entered by the CLEC into the FIDs and USOCs required by USWC’s OSS. This process will enable CLEC representatives to input service order requests into the ILEC’s OSS through the electronic gateway. If the order is rejected for some reason, the order should return across the gateway to the CLEC for correction. This process eliminates the "double ordering" procedure incorporated in the USWC and GTE cost studies. Although there will be instances where CLEC service representatives will need to contact ILEC representatives, we believe the frequency of manual intervention will be nowhere near the 100 percent level incorporated in the USWC and GTE nonrecurring cost studies.

Evidence concerning the OSS systems developed by Southwest Bell Telephone Company substantiates our finding that the technology is available to process unbundled element orders on a fully automated basis. Statements by SWBT personnel indicate that company will deploy electronic OSS interfaces this year which will enable SWBT to process unbundled element orders at flow through levels approaching 98 percent. SWBT’s plan to implement fully automated interfaces in the immediate future stands in stark contrast to the claims by USWC and GTE that all unbundled element orders must be reviewed manually. More importantly, it substantiates the position taken by Joint Intervenors and Staff that the technology to develop automated interfaces exists but simply has not been implemented by USWC and GTE.

In addition to concluding that the technology to produce fully automated interfaces is available, it is clear that such interfaces must be implemented if new entrants are to have a meaningful opportunity to compete in local exchange markets. The level of manual processing envisioned by the USWC and GTE studies not only increases the nonrecurring charges paid by competing carriers, it inhibits competitive entry by (a) increasing the time necessary to process CLEC service orders; (b) increasing the likelihood for error because information must be reentered by ILEC personnel; and (c) limiting the ability of CLEC’s to process a high volume of orders. To compete effectively in the local exchange market, new entrants must be able to perform services and interact with customers as quickly and efficiently as the incumbent LECs. Manual processing of unbundled element orders would significantly reduce the quality of service that new entrants can offer as well as the ability to market their services in a cost effective manner.

Because fully automated systems for processing unbundled element orders have not yet been deployed, we do not have information from other jurisdictions regarding actual flow though rates for such systems. We agree with Joint Intervenors, however, that efficient, well-managed, and fully automated operational support systems should result in very high flow through levels. As we have emphasized, such systems will enable CLEC representatives to query ILEC databases to obtain the information necessary to prepare properly documented service orders. Orders that are incorrectly submitted will be rejected by the system and returned to the CLEC for correction. ILEC representatives should become involved in the ordering process only infrequently: for example, where there is incorrect customer data or other system inconsistencies. In our estimation, it is reasonable to assume that a well-managed and maintained OSS will allow unbundled element orders to flow through at the 98 percent rate recommended by Ms. Petti. Accordingly, we find that the revised nonrecurring cost studies developed in accordance with this order should incorporate this level of flow through for all electronically submitted orders.

Notwithstanding our finding that the ILEC cost studies must incorporate flow through rates associated with fully automated OSS interfaces, we recognize that some CLECs may choose not to submit electronic orders for unbundled elements, at least initially. As we understand it, non-electronic orders are typically faxed or transmitted to ILEC service representatives over the telephone. Since these types of orders necessitate manual intervention by ILEC personnel, it is appropriate to develop a separate nonrecurring charge to insure that USWC and GTE are properly compensated for the costs they incur to process such orders.

We have elected not to adopt the weighted average flow through rate suggested by Staff witness Breen because we do not know what percentage of carriers will opt to process orders manually as opposed to electronically. Separate nonrecurring charges for manual orders will ensure that ILECs are not under or overcompensated for order processing costs. In addition, we agree with Dr. Zepp that separate nonrecurring charges for electronic and manual orders will send carriers the correct economic signals regarding the costs associated with the method of ordering they select. Carriers who place manual orders will be encouraged to deploy more efficient and economical OSS. This approach is consistent with the Cost Principle #3 which specifies that costs should be paid by those who are responsible for causing the cost to be incurred. Finally, separate charges will prevent CLECs that have deployed efficient OSS from subsidizing competitors who have not.

B. Provisioning Processes.

The parties dispute the manner in which CLEC orders for unbundled elements must be provisioned. The ILECs argue that building block orders must be treated as complex products requiring special handling and manual attention. Joint Intervenors, on the other hand, assert that building block orders may be provisioned in the same manner as residential or business basic exchange service -- as standardized products requiring minimal handling and lower nonrecurring costs.

1. POTS vs. Design Services.

The term "POTS" refers to "plain old telephone service." Generally, it relates to simpler products and services such as residential flat rate service or vertical features such as call waiting and speed dialing. According to USWC, POTS services are typically identified with a telephone number, are preengineered, and have a standard design and components. POTS service orders are supported by their own inventory OSS that configure the line based on the end-to-end service ordered by the customer.

"Design services" describe more complex, customized products and services. Examples include private lines, primary rate ISDN, interoffice trunks and feature group D service. These services generally require customized designs or individual configuration reviews, have quality and performance expectations, unique test characteristics, and may be associated with a telephone number or a circuit ID. Design circuits do not have to be associated with an end-user and, as a result, can be provisioned through the specification of meet points.

Design service orders can accommodate products that do not have a telephone number and require manual or coordinated handling. This type of order allows scheduling of a series of dates needed to process the order, such as record issue date, a design record layout, design verification assignment, and plant test date. Orders for design services are routed to OSS that contain inventory information concerning such services. Throughout the design service flow, orders are handled by employees specifically trained to provision and test design services.

2. USWC.

USWC witnesses Brigham and Buhler testified that unbundled elements, such as loops, must be provisioned using the more expensive design services flow. Unbundled loops, like private lines, are not switched or associated with a telephone number. Service orders for unbundled loops must be routed to OSS that contain inventory information about loops and to employees with the training and experience to provision unbundled elements. In addition, the design flow allows USWC to design specific services for CLECs. Unbundled loops can be configured in different ways depending on the method chosen by the CLEC to connect with USWC. Finally, only the design services flow accommodates coordinated cut-overs and the testing processes necessary to ensure connectivity between a CLEC’s collocated equipment and the point of interface to USWC’s loop.

Mr. Buhler emphasizes that manual intervention is required to accommodate CLEC building block requests, including processing disconnect and connect orders (described above) and coordinating circuit cut-overs. In addition, he observes that all software systems are subject to errors that require manual intervention on a periodic basis. The software necessary to improve reliability to the level now associated with modern switching equipment does not currently exist. Such a project would require total system replacements and drive up the cost of provisioning unbundled loops.

USWC is currently investigating whether the same loop facility can be used for both disconnect and new connect orders, thus eliminating some of the manual processing now required to provision unbundled loops. Improving the ability to associate related orders would ensure efficient loop engineering, but no implementation date for this process has been established. USWC is also working with Bellcore to determine whether the inventory for unbundled loops can be migrated to other OSS that support the POTS services flow. This would involve relocating loop inventory from TIRKS to SWITCH. However, coordinating the information that would remain in TIRKS with the unbundled loop information that would go to SWITCH is problematic and the advantages from a cost and technical standpoint are not apparent at this time. The ability to move inventory to SWITCH will not be available until the second half of 1998 at the earliest.

3. Joint Intervenors.

Joint Intervenors contend that USWC’s proposal to use the design services flow for unbundled element orders will disadvantage CLECs and their customers. According to Ms. Petti, CLEC customers who choose a CLEC as their local provider will experience longer waits for service and more technical problems than will ILEC retail customers. CLECs will also incur additional and excessive costs.

Whereas ILEC retail customer orders for simple business or residential service will flow through the POTS process, Ms. Petti emphasizes that all CLEC orders will require a separate set of OSS, special technicians, and procedures that are now required only for a minority of new telecommunications services provided by USWC. Instead of flowing through to the central office technicians and field installation crews normally used for basic services, CLEC orders will be sent to TIRKS. At that point, a special document known as a Work Order Record Document (WORD) will be created and sent to multiple locations for special handling, tracking, testing, and completion. These locations include the ISC (discussed above), the Circuit Provisioning Center (CPC), and the Special Service Test Center (SSC).

Ms. Petti emphasizes that it typically takes several days for an order to flow though the design services process. In contrast, new USWC retail customers can expect to receive service on the same day, or within a couple of days, from the time they place an order for service. As a result, CLEC customers will have to wait much longer than USWC customers to obtain the same service. Using the design services flow also means that maintenance and trouble reports will be handled by a smaller workforce than is assigned to USWC’s small business and residential users. This means that CLEC customers will have to wait longer for problems to be resolved. Finally, the CPC and SSC technicians responsible for design services flow are more highly paid than their counterparts in basic services because of their technical skills and knowledge. Because the higher labor rates for CPC and SSC technicians are factored into USWC’s nonrecurring cost studies, the cost of providing unbundled loops is overstated.

Ms. Petti also contends that there is no technical justification for provisioning unbundled loops through the design services flow. She disagrees with USWC’s claim that the POTS flow will not work because unbundled loops differ from ILEC loops. To the contrary, she maintains that a properly written and executed CLEC order should flow through without manual intervention as follows:

Using Uniform Service Order Codes (USOCs) and Feature Identification Codes (FIDs) unique to each CLEC to identify the unbundled loop, the Service Order Analysis and Control system (SOAC) could be triggered to query LFACS for the loop assignment as it does for services currently provided by USWC. By inventorying the CLEC tie cable appearance in the USWC SWITCH OSS as a pseudo OE/LEN/LU (originating equipment/line equipment number/line unit), the OSS would map the loop to the pseudo OE (EICT). SWITCH would then perform all of the inventory and assignments for the unbundled loop.

Ms. Petti states that pseudo identifiers have been used for years and are commonly used for switch cut-overs and line and station transfers. She maintains that USWC’s SWITCH OSS should be able to handle the inventory of pseudo OE. That process requires USWC’s data base manager or line assignment clerk to build pseudo tables for each switch. USWC witnesses have testified that Bellcore, the SWITCH vendor, has confirmed that the tie cable inventory for unbundled loop orders and any necessary programming may be available in the second half of 1998.

Ms. Petti also states that the design services flow envisioned by USWC will disadvantage CLECs because of the limitations of the TIRKS OSS. TIRKS is the primary OSS used by USWC to deliver high capacity business services. It is a costly, finite resource that may be compromised if large volumes of orders for unbundled loops must be processed. Ms. Petti contends that it is not feasible to use TIRKS to provide services to residential customers.

Ms. Petti agrees that CLECs will order a certain number of sophisticated and complex circuits. These types of circuits are usually ordered by large business customers and require the special handling associated with the design services process. Based on prior experience, however, Ms. Petti projects that the ratio of designed circuits to small business/residential orders will be only 10 percent of total orders.

4. Commission Decision -- Provisioning Processes.

The Commission agrees with Joint Intervenors that the nonrecurring charges paid by new entrants should not be based on the assumption that all orders for unbundled elements must be processed though the design services flow. The evidence demonstrates -- and USWC acknowledges -- that it takes longer and is more expensive to provision unbundled loops via the design services flow than it takes for USWC to provision basic retail services using the POTS services flow. This disparity means that CLECs will be unable to supply services in as efficient or as timely a manner as the services offered by the ILECs. As a result, new entrants are placed at a significant competitive disadvantage vis a vis the ILECs.

If competing carriers are to have a reasonable opportunity to compete with incumbent providers of local exchange service, they must be able to obtain unbundled elements in a manner that allows them to assemble services in approximately the same time it takes the ILEC to provision similar telecommunications services. Customers are likely to have little patience with competing carriers that take substantially more time than the incumbent to supply service, and only then at additional expense. USWC’s assumption that all CLEC building block orders must flow through the design service process produces a lack of parity from an ordering/provisioning standpoint and impedes the emergence of effective local exchange competition.

We are also unconvinced by USWC’s claim that it is necessary from a technical standpoint to provision all unbundled elements in the same manner as private lines and other complex telecommunications services. Ms. Petti testified persuasively that CLEC orders can flow through without manual intervention by using pseudo identifiers, an existing technology common to other telecommunications applications. USWC did not adequately refute this claim. In fact, Mr. Buhler acknowledged that USWC is currently working with Bellcore to determine if unbundled loop inventory can be migrated to OSS that support the POTS services flow. Although he indicated that problems could arise, he also confirmed that the ability to move loop inventory from TIRKS to SWITCH may be available in 1998. The record also discloses that USWC’s witnesses have testified elsewhere that the SWITCH OSS may be modified this year to accommodate the process contemplated by Ms. Petti.

We are also concerned with the ability of USWC’s TIRKS OSS to accommodate the demand for unbundled elements. As Ms. Petti explains, TIRKS was designed to handle orders for high capacity business services and may be compromised if large volumes of orders for unbundled loops must be processed through that system. USWC has only received a very limited number of orders for unbundled elements to date and has not demonstrated that TIRKS will be able to function adequately as the number of CLEC orders expands.

Although we reject the contention that all CLEC unbundled element orders must be provisioned as design services, we acknowledge that CLECs will order sophisticated and complex circuits that necessitate the specialized handling associated with the design services flow. According to Ms. Petti, design services currently provided to USWC customers include Digital Data Service, Foreign Exchange Service, Feature Group A and D, Voice Grade Analog Data, High Capacity Special Services and Interoffice Transport (Hi-Cap or DS1, DS3, STS-1, etc.), Frame Relay Service, and ISDN Primary Rate Interface Service. Basic residential and simple business services on the other hand, are not provisioned to USWC customers as design services. Based on her experience in a geographical area encompassing a very large urban center, and suburban and rural communities, Ms. Petti estimates that the ratio of design circuits to small business and residential orders will approximate 10 percent of total orders.

The Commission is persuaded by Ms. Petti’s testimony regarding this issue. Accordingly, we find that the nonrecurring costs of provisioning the Basic NAC, ISDN NAC, NACC Switched Lineside, NACC ISDN, Switching Features, Premium Listing, and Private Listing building blocks shall be based on the POTS services flow. While it is possible that these building blocks might be used occasionally for design services, they are used to provision POTS-type residential and business services the vast majority of the time. The nonrecurring costs associated with provisioning the remaining building blocks shall be based on the design services flow.

While it is impossible to accurately match all of the building blocks with the types of services that may be provisioned, we believe the approach we have taken to calculate building block provisioning costs is indicative of the provisioning processes that should exist in a forward-looking environment. More importantly, we conclude that USWC and GTE can and should develop provisioning processes which reasonably approximate those used by USWC and GTE to provision similar facilities and services for their own customers. As we have emphasized, USWC and GTE have a duty under §251(c)(3) of the Act to provide nondiscriminatory access to unbundled elements. The ILECs cannot meet this obligation -- and competition cannot succeed -- if we assume that building blocks can only be provisioned to CLECs through more expensive, time-consuming, and inefficient processes.

 C. Integrated Digital Loop Carrier (IDLC).

This issue concerns the percentage of IDLC systems, as opposed to the percentage of copper loops, assumed to be present in ILEC networks for purposes of calculating nonrecurring costs. In general, a greater percentage of IDLC results in a lower cost for provisioning loops and thus, lower nonrecurring costs.

1. Party Positions.

The AT&T/MCI NRCM assumes that all loops over nine kilofeet are provisioned by TR-303 IDLC systems. Ms. Petti claims that the TR-303 IDLC systems represent least cost forward-looking technology and "are made up of intelligent, processor-controlled network elements that can communicate over standard interfaces to the OSS systems in such a manner that little or no manual intervention is required for provisioning maintenance activities."

USWC and GTE oppose the IDLC assumptions incorporated in the NRCM and emphasize that IDLC technology is deployed in only very small parts of their current networks. It is unclear, however, what assumptions USWC has included in its cost studies regarding placement of IDLC systems. Mr. Schmidt testified that USWC "advocates a least-cost forward-looking network architecture that places TR-303 IDLC systems in the network beyond 12,000 feet in the highest density areas and TR-008 IDLC systems in the lower density areas." On the other hand, USWC’s reply brief indicates that USWC did not include any IDLC technology in its nonrecurring cost studies.

According to GTE witness Murphy, the IDLC assumptions in the NRCM are inappropriate for the provision of voice grade circuits. The model assumes that voice grade service will be handed off to the CLEC with multiple loops embedded in a DS1 handoff. Where loops are provided over fiber, however, it is necessary to install demultiplexing equipment and a plug in card, or channel unit, before voice grade loops can be unbundled. It is also necessary for a technician to make cross-connections at the MDF. Mr. Murphy claims that the NRCM fails to include the equipment and labor costs necessary to allow the handoff of voice grade loops.

2. Commission Decision -- IDLC.

As noted above, the cost studies approved in docket UM 773 were used to develop the recurring building block prices adopted in docket UM 844. The recurring cost studies incorporate specific assumptions regarding the percentage of digital carrier systems in USWC’s network using the overall least cost technology to calculate TSLRIC. The studies assume that 25 percent of the USWC network is supplied by IDLC systems and 75 percent is supplied by analog (copper or metallic) facilities.

For purposes of calculating nonrecurring charges, the Commission finds that the percentage of IDLC included in the nonrecurring cost studies should be consistent with the percentage of IDLC incorporated in the recurring studies. We realize that these percentages may change over time as cost studies are reevaluated and more efficient technologies become available. In the meantime, however, we believe it is important to maintain consistency between recurring and nonrecurring studies where the same issue is addressed.

The recurring cost studies do not specify the type of IDLC systems which should be assumed for purposes of cost calculation. The evidence indicates that the TR-303 systems are currently available and represent more recent technology than TR-008 systems. Accordingly, TR-303 systems should be assumed for purposes of calculating IDLC costs.

We are not persuaded by the ILEC argument that IDLC should be excluded for purposes of calculating nonrecurring costs because digital carrier technology comprises only a small percentage of USWC’s and GTE’s existing network. As emphasized above, this position is inconsistent with the TSLRIC cost principles adopted in docket UM 351. See discussion of Cost Principles #1 and #2, supra.

D. Labor Times and Probabilities.

The nonrecurring cost studies identify the time required by ILEC personnel to complete each nonrecurring activity. The work time estimate is multiplied by the probability of occurrence and the appropriate labor rate to derive the cost for the activity. Labor times and rates are significant cost drivers for nonrecurring costs.

1. USWC.

USWC witness Brigham testified that the work time estimates and probabilities for each nonrecurring activity are based on special studies and analyses conducted by teams of service experts within USWC. The teams include product managers and experts from each functional area (e.g., Interconnect Service Center). The estimates and probabilities are usually updated each time a study is performed.

Joint Intervenors argue that USWC’s proposed labor times and probabilities are based upon outdated and unreliable personnel surveys -- known as "Task Oriented Cost" (TOC) studies -- conducted by USWC in the late 1980s and early 1990s. They further maintain that the TOC studies do not incorporate any of the efficiencies USWC has realized from reengineering its systems and from consolidating processes and functions. Joint Intervenors maintain that USWC has not supplied any documentation to show that a systematic review of the older TOC estimates has been performed. They also contend that USWC has not conducted any "time in motion" studies to substantiate the proposed labor times and probabilities.

USWC concedes that it relies on some TOC studies, but emphasizes that those studies are used merely as a starting point for developing time and probability estimates. USWC notes that the TOC studies are first reviewed by subject matter experts to determine if the studies reflect current practices. In some cases, the experts concluded that the TOC studies were outdated and could not be used. In other cases, the TOC studies were deemed current and continue to be used by USWC. In the NAC (loop) nonrecurring cost study, for example, USWC relied on the TOC studies only for work times relating to the loop and circuit provisioning centers. All other work times for the loop study are based on new analyses.

USWC contends that the activity times included in its nonrecurring cost studies represent the best available estimate of the work times that USWC is likely to experience on a forward-looking basis given currently available systems and methods of operation. USWC concedes that it has not conducted any time in motion studies to substantiate its work time and probability estimates, but emphasizes that AT&T/MCI NRCM is also deficient in that respect.

2. AT&T, MCI and GTE.

The AT&T/MCI NRCM and the GTE nonrecurring cost studies also include work time and probability estimates. Ms. Petti and Mr. Collins testified that those estimates are based on the opinions of subject matter experts. AT&T/MCI and GTE did not submit workpapers or documentation indicating the information that the experts relied upon to develop their work time and probability estimates.

3. Staff.

Staff witness Breen made a number of adjustments to the work time estimates included in the USWC and GTE cost studies. Staff’s adjustments are discussed below.

The Commission has a number of concerns relating to the work time and probability estimates included in the USWC nonrecurring cost studies. We acknowledge USWC’s efforts to update its old TOC studies with more recent estimates, but we do not believe that USWC has produced sufficient evidence to substantiate its proposed work times and probabilities. Our concerns include the following:

(a) For the most part, USWC did not identify the subject matter experts used to develop time and probability estimates. There are a number memos included with the TOC studies that we understand were prepared by experts, but we cannot tell how many other persons were involved in the review process for each work activity. Moreover, no information has been provided regarding the knowledge or experience possessed by the experts responsible for those decisions.

(b) It is not clear from the evidence when the TOC study results were modified by subject matter experts and when they were not. Although the attached memoranda indicate some modifications were made, we have no idea whether they represent all of the changes proposed. Nor can we determine with certainty whether the changes recommended were adopted and incorporated in USWC’s nonrecurring cost studies.

(c) In those cases where we have been able to discern that TOC studies were modified by a subject matter expert, we are often unable to determine why the changes were made. Even where the subject matter expert has made a better effort to document the reasons for proposed work times and probabilities, the underlying rationale is not always explained in sufficient detail to enable us to conclude that the decisions are reasonable.

(d) USWC does not specify the criteria used by the subject matter experts to evaluate work activities or estimate probabilities. Without such information, we cannot tell whether decisions to modify or verify the TOC studies are reasonable.

(e) The TOC studies list various work activities necessary to perform a given function. In some cases, these activities appear to be fairly well detailed. In other cases, however, the activities are not clearly described and there is insufficient information regarding how the associated work time estimates were developed. A properly constructed study should (1) list all of the steps a service representative or technician must take to complete a particular task; (2) include an adequate description of each step; and (3) include documented measurements of the time necessary to complete each step.

(f) Mr. Brigham testified the work time estimates included in the TOC studies are based on a sampling of results from throughout USWC’s service territory. The studies disclose that there are wide variations in the work time necessary to perform certain activities from state to state. USWC does not explain the reason for such variations. We are left to speculate whether the differences might be because of the type of central offices sampled, the type of equipment employed in the central office, the number of technicians used, etc.

(g) Mr. Brigham explained that the final work times produced by the TOC study are based on a weighted average of information from the states served by USWC. There is no information showing what factors were considered in computing the weighted average.

(h) For many work activities, the TOC studies include a line where the "minimum," "maximum" and "most likely times" are supposed to be entered. In the vast majority of cases, there is no entry for the minimum and maximum times. In these instances, it is impossible to tell how the "most likely time" was derived.

(i) Since the TOC studies were conducted several years ago, it is safe to assume that they do not incorporate any efficiencies derived from USWC’s recent reengineering efforts or new technological developments that USWC may have employed to reduce work times. Although USWC indicated that the subject matter experts took such issues into account during the course of their review, we are unable to discern from the record all of the instances where increased efficiencies or new technologies were considered. Nor are we able to tell how such efficiencies and technologies were factored into the work times and probability estimates included in USWC’s nonrecurring cost studies.

The work time estimates and probabilities included in the GTE cost study and the AT&T/MCI NRCM suffer from the same type of defects that afflict USWC’s cost studies. Both GTE witness Collins and AT&T/MCI witness Petti testified that the work times and probabilities are based on the judgment of subject matter experts. However, neither GTE nor AT&T/MCI supplied sufficient documentation to support the assumptions underlying the work time and probability estimates. Without a more detailed showing, the GTE and AT&T/MCI estimates are also unacceptable.

Because of the shortcomings in the proposals presented, it is extremely difficult to determine the appropriate work time and probability estimates for each nonrecurring activity. Although we could reasonably exclude such costs altogether because of a failure of proof, that approach would not acknowledge the fact that ILEC personnel must perform certain nonrecurring activities to provision building blocks for competing carriers.

In the following sections of this order, the Commission discusses certain nonrecurring activities and adopts interim work time and probability estimates for those activities. With respect to the remaining nonrecurring activities, we conclude that nonrecurring costs should be based on the minimum work time and probability estimates included in USWC’s TOC studies. These work times and probabilities shall remain in effect until such time as USWC and GTE file revised analyses that are approved by the Commission. Such analyses must be comprehensive and include full documentation. In order to avoid the problems associated with the studies presented in this case, we strongly encourage the ILECs to conduct time-in-motion studies. Such studies provide a more accurate indication of the work times and probabilities associated with the various nonrecurring functions necessary to provision building blocks.

E. Jumper Activity Times.

This issue involves the amount of work time necessary for an ILEC technician to install and connect or to disconnect and remove jumper wires at the ILEC’s main distribution frame. The parties discussed this issue at length.

USWC witness Schmidt testified that the physical unbundling of loops involves several manual, labor-intensive processes. Upon receipt of an order for an unbundled loop, USWC must dispatch a technician to the central office to disconnect a two-wire jumper (short jumper) from the loop at the MDF. The technician must then connect the loop to the CLEC using another short jumper that runs between the loop connection on the MDF and the tie cable that runs to the SPOT frame and then on to the CLEC’s collocated space. USWC estimates that 14 minutes are required to perform the physical jumper activity -- seven minutes at the MDF and another seven minutes at the distribution frame located near the carrier’s collocated space. The estimates are based on meetings held during 1997 by USWC personnel experienced in provisioning and maintenance activities. Mr. Schmidt indicated that USWC has investigated whether these manual processes can be mechanized, but there are no mechanized solutions available in the foreseeable future.

AT&T/MCI contends that USWC’s estimated jumper activity times are "very leisurely." According to Ms. Petti, low-profile COSMIC frames commonly have jumpers that are much shorter than those required for conventional MDFs. As a result, less time is required to run, install, or remove jumpers. The AT&T/MCI NRCM estimates that only two minutes are necessary to perform these activities.

The jumper time estimates included in the GTE cost study are less than half of those estimated by USWC and somewhat greater than those estimated by AT&T/MCI. The precise times are designated as confidential by GTE.

Staff argues that jumper activity times should not be included in the loop nonrecurring cost where switching is supplied by the CLEC. Mr. Breen emphasizes that the Commission did not distinguish between "short jumpers" (i.e., jumper wires connecting the loop and the switch) and "long jumpers" (i.e., tie cables between the MDF and a CLEC’s collocated space) when it developed the recurring cost for the "Jumper NAC" building block. He further maintains that the material and labor (or "EF&I") cost to place long jumpers was included in the recurring cost of the Jumper NAC building block approved in docket UM 351. As a result, Mr. Breen argues that it is inappropriate for USWC to charge the recurring Jumper NAC building block rate and also charge for the short jumper placement through a nonrecurring charge. He suggests that the distinction between long and short jumpers should be considered in future analyses of recurring and nonrecurring costs.

On the other hand, Staff recommends an increase in the NACC (switch port) building block. Mr. Breen observes that a short jumper must be provisioned as part of the NACC when switching is provided by the ILEC. Since the placement of short jumpers is typically expensed, it is appropriately treated as a nonrecurring cost. Mr. Breen recommends that the nonrecurring cost and price of the NACC should be adjusted accordingly. However, rather than use the seven-minute estimate suggested by USWC, Mr. Breen recommends that the Commission adopt the short jumper activity time included in USWC’s retail cost study. The jumper activities modeled in the retail study are the same as those in the nonrecurring cost study, but the work time estimates are significantly lower. Mr. Breen did not apply the lower retail jumper activity time to other USWC cost studies which include short jumper activities.

USWC does not object to increasing the NACC nonrecurring cost, but takes issue with Staff’s proposal to eliminate jumper activity costs from the nonrecurring cost of the loop. USWC claims that Mr. Breen does not properly distinguish between short and long jumpers. Mr. Brigham emphasizes that the short jumper must be installed each time a customer orders service and that short jumpers are neither capitalized nor included in recurring costs. By contrast, long jumpers (tie cables) are hardwired cables that run between the MDF and the CLEC’s collocated space. Long jumpers are permanent investments and are therefore capitalized and included in recurring costs. Mr. Brigham maintains that the cost data relied on by Mr. Breen to support the removal of jumper activity times from the nonrecurring cost of the loop pertain only to long jumper costs, not to the labor costs required to place the short jumper. He further argues that short jumper placement costs are not recovered in any other building block cost.

(a) How should short jumper-related costs be categorized?

Staff witness Breen correctly observes that the Commission did not distinguish between short jumpers and long jumpers when the Jumper NAC building block was established in docket UM 351. The recurring monthly price of the Jumper NAC building block is based on the investment and labor costs of installing tie cables within ILEC central offices.

Although short jumper-related costs were not considered in developing the recurring Jumper NAC rate, the evidence suggests that Jumper NAC rate would be more than sufficient to compensate the ILEC for the capital cost of the short jumper facility. Tie cables are permanently installed cables that include hundreds of wires. They are often more than several feet in length and may extend between different floors of an ILEC central office. Conversely, short jumpers are comprised of only one or two wire pairs and typically extend no more than a few feet between terminal connections on the MDF. Compared to tie cables, the capital cost of short jumpers is de minimis.

On the other hand, the record discloses that nearly all of the costs associated with short jumpers are labor-related expenses of connecting and disconnecting jumper wires. Moreover, short jumper wire is generally treated as a supply item. Since these costs are fundamentally expense items, it may be more appropriate to recover short jumper-related expenses through a nonrecurring charge, rather than create a separate short jumper building block. This approach allows the ILECs to assess a short jumper nonrecurring charge whenever a short jumper is required in conjunction with a building block. It also addresses our concern that a recurring monthly rate may not provide an adequate means of recovering the ILEC’s labor costs unless the CLEC retains the jumper connection for an extended period of time. For example, if the short jumper were priced at the current Jumper NAC building block rate of 12 cents per month, it would take many months for the ILEC to recoup its short jumper-related costs. In addition, we have noted that customer "churn" in a competitive local exchange market is likely to approximate the 40-50 percent annual rate now experienced in the long distance market. A high rate of customer churn will cause short jumpers to be connected and disconnected more frequently, lessening the chance that the ILEC will be able to recover its labor costs.

For these reasons, the Commission finds that short jumper-related costs incurred by USWC and GTE should be recovered as a nonrecurring charge. We are not persuaded that it is necessary to establish a short jumper building block or that short-jumper related costs can be reasonably recovered through recurring rates.

The Commission concludes in this order that the mandatory collocation proposals proposed by USWC and GTE contravene the nondiscriminatory access requirement in §251(c)(3) of the Act. We have also determined that, in light of the Iowa Utilities Bd. decision, nondiscriminatory access requires that requesting carriers must have direct access to the MDF and other ILEC facilities where building blocks may be combined. These decisions allow CLECs to self-provision short jumpers without any involvement by the ILEC. As a consequence, USWC and GTE may have limited occasion to connect and disconnect short jumpers and to assess nonrecurring charges for those activities.

Clearly, the nonrecurring charges assessed by USWC and GTE should not include short jumper-related costs except in circumstances where the ILEC performs short jumper activities and supplies the short jumper wire on behalf of the CLEC. Under the terms of this order, a CLEC may access the MDF to self-provision the short jumper between the loop termination location and the tie cable termination location on the MDF. This is true regardless of whether the CLEC purchases ILEC switching or supplies its own switching. Because the ILEC will not incur any labor costs to connect the short jumper in either instance, these costs may not be included in the nonrecurring charges paid by the CLEC.

Likewise, there will be instances where the ILEC will not incur any cost to disconnect short jumpers. For example, where a customer served by one CLEC seeks to transfer service to another CLEC, the ILEC will not incur any cost to disconnect the short jumper if the serving CLEC disconnects the short jumper itself at the MDF.

(c) How should short jumper labor costs be calculated?

The final issue relates to the work time estimate that should be used for purposes of calculating the nonrecurring charges in those cases where the ILEC performs short jumper activities. As emphasized above, the Commission encourages USWC and GTE to produce work time and probability studies that are more comprehensive and better documented than those presented in this case. In the interim, we agree with Staff that the short jumper activity time estimates incorporated in USWC’s retail cost studies should also be used for purposes of calculating nonrecurring costs where USWC technicians are required to connect or disconnect the short jumper. There is no significant difference between the short jumper work USWC performs for its retail customers and the work USWC may perform for the CLECs. USWC did not explain the substantial difference between the retail short jumper activity times and the seven-minute jumper activity time included in its nonrecurring cost studies.

Moreover, as in the case of the work time estimates discussed above, there is not enough information in the record to determine how USWC’s seven-minute jumper activity estimate was developed. For example, it is not clear whether the seven-minute estimate is based entirely upon expert opinion or whether measurements were taken of technicians performing jumper activities. Also, we do not know the time period over which the seven-minute estimate was developed, the location or size of the central offices under study, or the type of distribution frames used in the study. If we are to attach significant weight to such analyses, USWC must supply more complete and better organized data than that included in the nonrecurring cost studies presented in this case.

The Commission has similar problems with the two-minute short jumper activity estimate incorporated in the AT&T/MCI NRCM. Although, the NRCM purports to be based on subject matter expert opinion, AT&T/MCI did not provide sufficient documentation to substantiate the two-minute estimate.

GTE’s short jumper activity time estimates do not differ substantially from the USWC retail jumper activity times recommended by Staff, and are considerably less than the seven-minute estimate proposed by USWC. For those reasons, we are willing to use GTE’s short jumper activity time estimates to calculate GTE’s nonrecurring costs until the company develops better studies to document the labor time necessary to perform these activities.

F. COSMIC Frames.

This issue addresses the type of main distribution frame assumed to be used in ILEC central offices for cost study purposes. As discussed above, there are two types of main distribution frames generally in use: conventional MDFs and low-profile, or COSMIC, frames. The type of distribution frame is important for purposes of calculating nonrecurring costs because it takes less time to connect jumpers on COSMIC frames.

The AT&T/MCI NRCM assumes that COSMIC frames comprise 100 percent of the distribution frames installed in ILEC central offices. Ms. Petti states that COSMIC frames were included in the NRCM because they represent newer, forward-looking technology that is currently widely deployed in the industry. In Ms. Petti’s experience as a regional manager for Pacific Bell, all new central offices were equipped with COSMIC frames.

GTE objects to the COSMIC frame assumption included in the NRCM. It points out that neither Ms. Petti, nor the developers of the NRCM model know how many COSMIC frames are actually installed in Oregon.

2. Commission Decision -- COSMIC Frames.

We agree with Joint Intervenors that COSMIC frames represent forward-looking, least cost methodology consistent with TSLRIC principles. The 100 percent COSMIC frame assumption in the NRCM is therefore reasonable. The percentage of COSMIC frames actually installed by GTE or USWC is irrelevant from a TSLRIC standpoint.

Although the evidence indicates that COSMIC frames reduce the amount of time necessary for technicians to install and remove short jumper wires, there is not enough information in the record to determine the average number of minutes required to perform these functions on a COSMIC frame. The experts retained by AT&T/MCI have ascertained that two minutes is sufficient for these activities, but there is no additional support for this claim. For that reason, we find that the ILECs should perform studies to calculate the average work times to install and remove short jumpers on COSMIC frames. Once this information is produced, it should be incorporated into all applicable cost estimates. In the interim, all short jumper-related costs should be based on the short jumper work time estimates discussed above.

G. Disconnection Activities.

This issue deals with whether the nonrecurring charges paid by requesting carriers should also include costs associated with disconnecting service.

1. Party Positions.

GTE and USWC recommend that nonrecurring charges include both the cost of establishing service and the cost of disconnecting service in the future. In support of this position, USWC and GTE emphasize that it is often difficult to recover disconnection costs if charges are levied after the business relationship with a customer has ended. Mr. Brigham and Mr. Schmidt testified that customers whose accounts have been terminated are less likely to pay their telephone bills than are customers with active accounts. For that reason, it has become standard practice in the telecommunications industry and in other industries to recover disconnection costs as part of the up-front nonrecurring charge to establish service.

Staff agrees with USWC and GTE that any disconnection costs should be included in the initial nonrecurring charge paid by carriers purchasing building blocks.

Staff observes that it would be an administrative burden for the ILECs to have to track these costs through to the time of actual disconnection.

Joint Intervenors argue that disconnection charges should be separated out and paid by CLECs only when and if the disconnection charges are actually incurred by an ILEC. They maintain that including disconnection costs in up-front nonrecurring charge will: (a) make it significantly more expensive for CLECs to enter the market, creating a barrier to entry; (b) violate the principle of Cost Causation, by allowing ILECs to charge for activities before they are incurred; and (c) force CLECs to pay disconnection costs for activities that never occur, as in the case where loop facilities are transferred from one CLEC to another.

AT&T and MCI further argue that CLECs, unlike retail customers, do not pose a significant risk of defaulting on disconnection charges. They point out that the ILECs already require substantial deposits from carriers before supplying unbundled elements.

2. Commission Decision -- Disconnection Activities.

We agree with the ILECs and Staff that disconnection costs should be included in the upfront nonrecurring charge. Levying a separate disconnection charge would be difficult to collect and burdensome to administer. Although the ILECs are unlikely to experience problems recovering disconnection costs from large carriers such as AT&T or MCI, there will undoubtedly be many smaller, less-established carriers providing local exchange service. The ILEC would be exposed to a significant risk of nonpayment if these carriers purchase large quantities of building blocks and subsequently experience financial difficulties. Moreover, as Staff points out, it would be difficult to separately track, bill, and collect disconnection charges. The additional administrative costs associated with separate disconnection charges would increase the total cost of provisioning building blocks.

In view of the decisions the Commission has made regarding the calculation of nonrecurring costs, we do not believe that collecting disconnection costs upfront will discourage competitive entry as Joint Intervenors allege. Nor are we persuaded that our decision violates the cost causation principle adopted in docket UM 351. That principle addresses the assignment of costs to particular building blocks. It does not specify the precise mechanism by which those costs must be collected. Also, as USWC and GTE observe, it is standard practice in the utility industry to include disconnection costs in the nonrecurring charge of a given service.

We are not persuaded by Joint Intervenors’ claim that including disconnection costs in the initial nonrecurring charge will cause CLECs to pay for activities that may never occur. We recognize that ILECs may not have to perform any short jumper work or other physical activity to disconnect building blocks in certain circumstances. Also, because we have concluded that a high percentage of building blocks will be processed electronically, disconnection costs are likely to be minimal in many cases. Nevertheless, every building block transaction will require the ILEC to incur some cost at the time a building block is disconnected. At the very least, this will include the cost of processing the disconnection order to maintain a proper record of the transaction for billing and inventory purposes. Accordingly, we find that disconnection costs should be included in nonrecurring charges 100 percent of the time.

 H. Dedicated Inside Plant (DIP) and Dedicated Outside Plant (DOP).

DIP refers to the connection between the loop termination location on the MDF and the ILEC’s switch termination location on the MDF. By leaving this connection in place when a customer discontinues service, the ILEC is able to reduce expenses for technician time and copper wire, and connect the next customer more quickly and easily. DOP is the practice of allowing outside cross-connections to remain in place, leaving the customer’s loop in a "connect-through" state. Among other things, DOP reduces the need to dispatch outside field technicians for connect and disconnect orders.

When used together, DIP and DOP allow the ILEC to maintain a "warm dial tone" at premises that have been vacated by a customer. When a new customer moves in, that customer can simply pick up the telephone to order new service or to dial emergency numbers.

1. Party Positions.

Joint Intervenors emphasize that DIP and DOP significantly reduce the cost of provisioning, allow faster customer service, and eliminate errors that can arise from manual processes. According to Ms. Petti, these advantages have made DIP and DOP a common practice among ILECs. She notes that USWC’s retail cost studies assume that DIP and DOP will be used a substantial percentage of the time. The AT&T/MCI NRCM assumes that DIP and DOP will be available 100 percent of the time where customers migrate from ILEC service to CLEC service.

The ILECs argue that it is erroneous to assume DIP/DOP will be available 100 percent of the time. They point out that DIP/DOP cannot be used to activate a loop that is not connected to a switch. In other words, DIP/DOP assumes that the ILEC will combine the loop and switching building blocks for CLECs. USWC and GTE emphasize that this assumption is contrary to the Eighth Circuit’s holding that ILECs cannot be compelled to combine elements under the Act.

In addition, GTE and USWC observe that their networks use DIP/DOP significantly less than 100 percent of the time. GTE witness Murphy points out that the 100 percent assumption ignores the massive network growth now underway and understates the nonrecurring costs associated with outside plant activities by ILEC technicians.

2. Commission Decision -- DIP/DOP.

Under the Eighth Circuit’s decision, ILECs are not required to provide CLECs with building block combinations. Thus, an ILEC may disconnect the loop from the switch even where the CLEC buys both building blocks. Once the ILEC technician disconnects the jumper wire that runs between the loop and the ILEC switch, the DIP connection is broken. DIP is also broken in the instance where the ILEC technician disconnects the loop from the ILEC switch so that the loop can then be reconnected to a CLEC switch. For these reasons, the Commission agrees with USWC and GTE that it is unrealistic to assume DIP will be available when CLECs purchase unbundled elements.

On the other hand, we see no reason why an ILEC should disconnect existing outside cross-connections when an ILEC customer migrates to a CLEC or a CLEC customer migrates to another CLEC. Unlike the inside plant connections between the loop and switch, the outside plant connections do not involve a combination of separate building blocks and are not encompassed by the Eighth Circuit’s decision. In our opinion, it is reasonable to assume that DOP will remain in place the same percentage of the time that it is assumed to be available in USWC’s retail cost studies. This will take into account the customer migration scenario (where DOP connections should generally not have to be disturbed) as well as those cases where ILEC technicians still have to perform outside plant rearrangement activities (e.g., to connect new loops).

I. Number of Work Activities per Visit.

This issue involves the number of activities ILEC technicians perform during each visit to a nonstaffed central office or outside plant location. The more activities performed per visit, the lower the nonrecurring cost.

1. Party Positions.

AT&T/MCI witness Petti asserts that cost studies should assume efficient management practices. She observes that ILEC technicians do not perform work on a per order basis, but rather are sent to do several jobs at a time. These include such things as general maintenance, routines, and provisioning activities. Also, when technicians are dispatched, they are equipped with mechanized field access systems that allow them to complete orders, obtain new work assignments, close trouble tickets, update data bases, remotely assess test systems, and complete their work in a mechanized fashion. The NRCM assumes that an ILEC technician will complete an average of four work activities per trip to all work locations, including customer sites, outside plant locations, and unattended central offices.

USWC witness Schmidt argues that it is incorrect to assume that ILEC field technicians working on outside plant will perform multiple tasks at the same location. He emphasizes that the work required for outside plant technicians depends on the physical location of the customer requiring the work and the amount of time allotted to complete the work. USWC attempts to minimize travel time by allocating work within a geographic area, but technicians must often travel several miles between jobs.

GTE witness Murphy also maintains that the NRCM assumption is unreasonable. In order to provide service as quickly as possible, GTE technicians frequently complete only one work activity per trip. If technicians are required to wait until there were four activities per trip at a specific location, service could not be provided in a timely fashion. While it may be reasonable to assume multiple work activities at unattended central offices, Mr. Murphy contends that it is not reasonable to make the same assumptions for work activities at outside plant or customer locations.

2. Commission Decision -- Work Activities per Visit.

Although there is sufficient evidence in the record to allow the Commission to conclude that ILEC technicians often conduct more than one work activity per trip, none of the parties have submitted documentation that would enable us to calculate the average number of activities performed at the various work locations visited by ILEC technicians. We are inclined to agree with Messrs. Schmidt and Murphy that technicians visiting customer locations are likely to perform fewer activities because of the need to respond quickly to customer requests. On the other hand, visits to unattended offices, and perhaps also to SAIs, are likely to be somewhat less time-constrained and allow technicians to schedule more activities.

Without more specific information, we cannot find that the four activity per visit average in the NRCM is reasonable. The record does, however, support the finding that ILEC technicians frequently complete more than one activity per visit. For purposes of calculating nonrecurring costs, therefore, we will assume that ILEC technicians will complete two activities per trip on average to all work locations. As with the other cost study assumptions adopted in this order, the Commission will reevaluate this decision in subsequent proceedings once more complete information is produced.

J. Loop Unloading Activities.

USWC and GTE must place loading coils on copper loops over 18,000 feet long (18 kilofeet or 18kf) in order to supply adequate voice grade quality service. In such cases, resistance design standards require placement of three loading coils at distances of 3kf, 9kf, and 15kf from the central office.

Loop unloading is a type of circuit conditioning. If a customer with an 18kf loaded loop requests digital service, an ILEC technician must splice the line to remove loading coils at each of the three points on the loop where the coils are attached.

1. Party Positions.

Staff witness Breen testified that the costs associated with unloading loops are included in the maintenance factors used to develop recurring building block rates. To prevent double recovery of these costs, Staff argues that loop unloading costs should not also be included in the nonrecurring charges paid by CLECs. Mr. Breen contends that the costs associated with loop unloading and other outside plant activities are appropriately included in recurring costs and urges the Commission to refrain from establishing a policy of separately identifying and assessing nonrecurring charges for all of the outside plant rearrangement expenses that may be associated with the myriad of situations that arise. Mr. Breen suggests that the better approach is to continue the policy of recovering outside plant expenses and investments in recurring costs. The net effect of Staff’s recommendation is to reduce USWC’s and GTE’s proposed nonrecurring loop unloading charge from $597.61 to zero.

USWC concedes that the labor costs associated with unloading loops are currently included in the maintenance factor used to develop recurring costs. However, it claims that unloading costs are properly treated as nonrecurring costs because they are incurred each time a CLEC requests removal of coils and taps. USWC maintains that the appropriate solution is allow recovery of its proposed nonrecurring charges for loop unloading. It agrees to adjust its maintenance factor to avoid double recovery of these costs. USWC also notes that the Commission "ordered USWC to deload loops" by mandating loop conditioning charges in docket UM 351. GTE’s proposed nonrecurring charge for cable unloading mirrors USWC’s proposed charge.

Joint Intevenors agree with Staff that there should not be any nonrecurring charge for loop deloading. They claim that USWC’s proposal to adjust the recurring cost maintenance factor is improper because recurring costs are not at issue in this proceeding.

Joint Intervenors also argue that unloading costs should be deleted from both recurring and nonrecurring rates charged to CLECs because unloaded loops represent forward-looking technology. According to Ms. Petti, current outside plant design recognizes that customers, including those with loops greater than 18kf, require more sophisticated services. She states that the industry typically plans for unloaded cable in sufficient quantities to assure that the only effort required to serve a customer is to change to a spare loop that is not loaded. The method used most often is to place new cables as nonloaded and to change existing customers to the new cable. Another approach is to plan and incorporate the unloading of loops in routine cable maintenance. The unloaded loops would then be held as shelf stock, most efficiently in binder groups of twenty-five. Both of these methods contrast with USWC’s assumption that only one loop will be unloaded at a time.

USWC witness Schmidt contends that it is incorrect to assume that ILECs will maintain binder groups of "clean" or unloaded loops. He points out that telephone companies do not have the luxury of designing and engineering feeder plant to include binders of dedicated, unloaded loops. Moreover, he states that this type of excess capacity does not exist on USWC’s network.

2. Commission Decision -- Loop Unloading.

The Commission agrees that loop unloading and other similar outside plant activities should continue to be recovered through recurring charges for the reasons stated by Mr. Breen. Consequently, we decline to adopt nonrecurring charges proposed by USWC and GTE for loop unloading. USWC and GTE are not harmed by this approach because the costs of unloading will continue to be recovered through recurring rates.

USWC’s claim that the Commission "ordered USWC to provide loop deloading" in docket UM 351 argument is not on point. The conditioning building blocks established in Order No. 96-283 do not relate to loop deloading, but rather to other conditioning processes used to enhance the performance of voice grade and digital loops. Moreover, even if loop deloading was entailed in these processes, the costs associated with the conditioning building blocks are recovered through recurring charges, not through nonrecurring charges.

Joint Intervenors agree that the nonrecurring cost for unloading should be zero. We do not address their additional claim that unloading costs should be removed from the ILEC’s recurring cost because unloaded loops represent least cost technology. As Joint Intervenors emphasize with respect to USWC’s request to adjust the maintenance factor, recurring rates are not at issue in this case. If Joint Intevenors desire to pursue this issue, they must do so in another proceeding.

K. GTE Outside Facilities Connection Charge.

In Section 3.1.1(C) of its proposed tariff, GTE proposes an Outside Facility Connection charge that would apply each time a GTE service technician visits a customer’s premises. We decline to adopt this charge for the reasons discussed immediately above. That is, we are persuaded that outside plant activities should be recovered through recurring rather than nonrecurring charges.

L. Testing Activities.

This issue concerns whether nonrecurring charges should include costs associated with testing unbundled elements provided to requesting carriers.

1. USWC and GTE Proposed Tariffs.

USWC’s proposed tariff states that CLEC "will have responsibility for testing the equipment, network facilities and the Unbundled Loop facility. If USWC tests the unbundled loop at the carrier’s request and the fault is in USWC facilities, a charge shall apply."

GTE Advice No. 589 stated that "GTE will perform routine testing at the time of installation and will be responsible for its own facilities." This language was subsequently modified in Advice No. 611 as follows:

GTE will provision the NAC (loop) with the functionality specified by the [telecommunications carrier] and as described in the Tariff, for the particular building block ordered or combination of building blocks when channel performance building blocks are also ordered.

GTE will provision the NACC (port) testing for dial tone and complete a test call to verify the assigned telephone number. Switch features ordered by the [telecommunications carrier] will also be tested to ensure proper operation.

Currently automated loop testing of unbundled loops is not available. When such testing becomes available, it will be offered.

2. Staff.

Staff recommended that USWC revise its tariff to mirror the testing provisions in GTE Advice No. 589. Staff did not address the tariff modifications proposed subsequently by GTE in Advice No. 611.

Staff witness Breen testified that testing occurs during provisioning and on an ongoing basis. In both cases, the ILEC and CLEC should be responsible for testing, maintaining, and repairing their own facilities. Mr. Breen contends that the entity providing the switching should conduct the periodic testing required. He further observes that recurring charges already include costs associated with ongoing testing activities. These include, for example, situations where the ILEC tests a facility in response to a repair call and reports to the customer that the trouble is not associated with the ILEC facility.

Mr. Breen states that testing costs should not be included in nonrecurring charges where the ILEC is not required to visit a customer’s premises because those costs are already included in the recurring charge. On the other hand, a CLEC should be required to pay a trouble isolation charge where an ILEC technician travels to the customer’s premises and the trouble is not in the ILEC’s facilities. Mr. Breen proposes that USWC and GTE revise their building block tariffs so that CLECs pay trouble isolation charges in the same manner as USWC and GTE retail customers.

3. USWC.

USWC disagrees with Staff’s position regarding testing. Mr. Brigham states that the testing expenses included in USWC’s recurring costs studies are part of ongoing repair costs included in the maintenance factor. However, the maintenance factor is based on historical costs and does not consider the incremental expenses associated with testing loops. Mr. Brigham emphasizes that every time a CLEC requests USWC to test an unbundled loop, USWC must dispatch a technician to the frame to test the cable pair connected to the loop. Frequently, technicians must be dispatched to nonstaffed central offices to perform the test. This manual process is not normally required in the retail environment because USWC can perform automated testing through its switch.

USWC’s proposed tariff applies time and material charges when testing is requested by a CLEC. Mr. Brigham recommends that the "Testing Access" building block approved by the Commission in UM 351 should apply when the ILEC is required to test unbundled loops in the central office. He agrees with Mr. Breen that trouble isolation charges are appropriate where the CLEC requests USWC to visit an end user’s premises to test for trouble, and the trouble is not caused by USWC’s facilities.

4. GTE.

GTE witness Murphy states that the costs of testing associated with the initial establishment of all telephone services are properly categorized as nonrecurring costs. He emphasizes that the NRCM assumes minimal testing functions and an unreasonably low level of testing for unbundled elements. Mr. Murphy contends that new and smaller CLECs will not purchase testing equipment immediately, but will rely on GTE to do their testing for them. Larger CLECs may do some testing, but GTE will retain responsibility for network testing. Mr. Murphy claims that the NRCM does not account for these costs.

GTE is willing to provide CLECs with the same facility testing that GTE provides to itself and its retail customers. GTE witness McLeod indicated that GTE can provide automated testing where it also provides switching to the CLEC. GTE cannot provide automated testing of unbundled loops, however.

5. Joint Intervenors.

Joint Intervenors agree with Staff that testing should be performed by the party providing the switching. According to Ms. Petti, AT&T and MCI will use their own MLT OSS, no-test trunks and digital switches to test loops for provisioning, service activation, and maintenance. Consequently, the nonrecurring charges for the basic NAC should not include any testing charges.

Ms. Petti states that it is unnecessary for the ILEC to test the basic NAC prior to migrating an existing customer to a new entrant. The NRCM assumes that USWC’s local digital switches are equipped with Predictor/Automatic Line Insulation Test (ALIT) which conducts "automatic/non-intrusive proactive performance monitoring of the NAC to detect potential problems before the customer migration." Thus, it is assumed that the basic NAC will meet performance objectives prior to migration. After the NAC has migrated to the CLEC, it will perform ALIT-type testing. The only time reactive testing occurs is if the customer reports trouble.

Ms. Petti points out that each provisioning scenario included in the NRCM assumes some level of testing. The model contemplates that the ILECs will proactively maintain their networks to ensure proper operation and reliable customer service. It also assumes that the cost of basic maintenance is a recurring cost. Because ILECs must be prepared to respond to customer service inquiries, some level of testing will continue to be performed.

We agree with Staff that each ILEC and CLEC should bear the cost and responsibility for testing, maintaining, and repairing their own facilities. We also agree that the entity providing the switching facilities should bear the cost and responsibility for routine testing.

When building blocks are provided to CLECs, the ILEC must ensure that the building blocks meet the requisite technical parameters. This presumes that building blocks will be tested beforehand by the ILECs. The record indicates that it is unnecessary for USWC or GTE to manually test every loop before it is delivered to a CLEC. As Ms. Petti observes, ILEC local digital switches can perform ALIT-type testing to automatically monitor NAC performance and detect potential problems before the NAC is delivered to the CLEC.

Once building blocks are provided to the CLEC and combined to form a service, they must be tested to ensure that the service works as planned. Ongoing testing must also be provided to ensure that service quality is maintained. The record indicates that these testing functions can be performed on an automated basis by the party providing the switching. Thus, where the ILEC provides the switching function, it can test the loop on an automated basis. The CLEC can do likewise where supplies its own switching.

As Mr. Breen points out, ongoing testing expenses are already included in the maintenance factor of recurring building block rates. Mr. Brigham argues that this allowance is insufficient because of incremental costs associated with testing unbundled loops. We find this argument unpersuasive for the reasons discussed above. Moreover, even if Mr. Brigham were correct, we are persuaded that testing expenses should continue to be recovered through recurring costs because of the ongoing nature of testing activities.

An additional testing charge should only be required where a CLEC requests that an ILEC technician visit an end user’s premises to test for trouble, and trouble is not found in the ILEC facilities. We agree with Mr. Breen and Mr. Brigham that a trouble isolation charge should be assessed in that instance. We adopt Staff’s proposal to impose the same type of trouble isolation charge paid by USWC and GTE retail customers in this situation.

USWC’s proposal to impose the testing access building block charge when unbundled loops are tested within the central office is unnecessary for the reasons stated above. Furthermore, USWC’s proposed treatment of the testing access is inconsistent with the definition of that building block.

The nonrecurring charges recommended by Staff include several revisions to the USWC and GTE cost studies. Mr. Breen initially proposed excluding several circuit provisioning functions (e.g., ODAP/HCPC and IFCPC and CPC) and message trunk administration (e.g.,CAC) activities from the nonrecurring building block studies on the assumption that those costs were included in recurring EF&I costs. Based on further examination, Mr. Breen revised his nonrecurring cost estimates to add back certain costs originally thought to be included in the recurring building block costs. Mr. Breen’s revised estimates continue to exclude ODAP/HCPC and IFCPC costs associated with provisioning functions.

USWC witness Brigham asserts that ODAP/HCPC and IFCPC circuit provisioning work is required whenever a CLEC orders an NAC, NACC, multiplexing, or transport. He asserts that the costs associated with these network activities are not included in the recurring EF&I cost for these building blocks, because the activities do not occur when a building block (e.g., loop) is installed, but rather when the building block is ordered by a CLEC and an existing circuit must be provisioned. Mr. Brigham claims that Staff correctly included the costs in the basic NAC studies but incorrectly excluded those costs from the DS1 NAC, DS3 NAC, multiplexing, and transport studies.

Mr. Brigham also contends that Mr. Breen incorrectly excluded the CAC function from nonrecurring cost studies. CAC costs relate to message trunk administration (e.g., the assignment of trunk groups) and are not included in EF&I recurring costs. He asserts that transport building block nonrecurring costs must be adjusted upward to include this activity.

2. Commission Decision -- Circuit Provisioning Functions.

We adopt the adjustments proposed by Staff. To begin with, the record shows the CAC costs, which USWC claims were incorrectly excluded, were actually added back to the nonrecurring cost estimates by Mr. Breen. Second, the record shows that Mr. Breen included expenses associated with order handling, order screening, and order logging activities in the nonrecurring cost estimates for DS1 and DS3 NACs. Staff properly excluded ODAP/HCPC and IFCPC provisioning costs from the nonrecurring cost estimates because these types of expenses are already included in the building block recurring costs. As discussed previously, this type of outside plant expense should continue to be included in recurring costs.

N. Nonrecurring Cost Markup.

This issue deals with whether nonrecurring charges should be marked up over TSLRIC.

1. Party Positions.

The nonrecurring charges proposed by Staff incorporate the markup over TSLRIC approved for recurring building block rates in docket UM 844. USWC proposes a markup greater than that used in UM 844, but is willing to accept the markup recommended by Staff.

Joint Intervenors claim that the markup proposed by USWC is excessive and inconsistent with markups USWC has proposed for nonrecurring charges in the past. According to Dr. Zepp, the nonrecurring costs proposed in USWC's s retail cost studies are less than or equal to direct nonrecurring costs or a combination of direct and indirect nonrecurring costs. He alleges that the nonrecurring charge proposed by USWC for residential service in its recent rate case is less than the company’s estimate of TSLRIC. Similarly; USWC’s proposed nonrecurring charges for business service approximate TSLRIC plus shared costs. In addition, Dr. Zepp maintains that USWC’s current nonrecurring charges for both services are below USWC’s nonrecurring cost estimates. He asserts that it is improper for USWC to markup nonrecurring costs for competitors when it does not propose similar markups for USWC retail customers.

Joint Intervenors also oppose the markup authorized in UM 844 for recurring building block costs. They state that the decision to exclude nonrecurring costs from the markup authorized in that proceeding acknowledges USWC’s past practice of assessing nonrecurring charges which do not exceed the sum of direct and indirect nonrecurring costs.

Joint Intervenors assert that the markup over nonrecurring cost should be no greater than the 10.4 percent markup included in the AT&T/MCI NRCM. They maintain that such a markup will allow the ILECs a reasonable opportunity to recover their costs without denying the CLECs the opportunity to compete for local exchange customers.

2. Commission Decision -- Nonrecurring Cost Markup.

The Commission adopts Staff’s recommendation to use the same markup for nonrecurring charges that was authorized for building block recurring charges in docket UM 844. In prior decisions, we have held that building block rates should include a mark up over TSLRIC. Logically, there is no reason why nonrecurring charges should not also include a markup; provided, of course, that the resulting charges are not so high they discourage competitive entry. The Commission will monitor the progress of competition and may revisit this issue if it appears that the amount of markup included in the nonrecurring charges for building blocks creates a barrier to entry.

AT&T and MCI maintain that competition will suffer if CLECs must pay nonrecurring charges that include a significant markup over TSLRIC while the nonrecurring charges paid by ILEC retail customers include little or no markup over cost. Under such circumstances, CLECs will either have to absorb the difference or try to pass along the nonrecurring charges to their customers in the form of higher retail prices. Since AT&Tand MCI did not disclose how TSLRIC was calculated for the bundled retail services used in its comparison, we cannot confirm that the nonrecurring charges currently paid by USWC and GTE residential and business retail customers approximate TSLRIC.

Furthermore, the record in this case does not contain an extensive discussion of the relationship between building block and retail service nonrecurring charges. As a result, we do not have a sufficient basis to conclude that the markups included in building block nonrecurring charges should equal the markups in retail nonrecurring charges, or that a separate imputation test for nonrecurring charges should be established.

 VIII. TARIFF TERMS AND CONDITONS

A CLEC must obtain customer authorization before it may place building block orders on behalf of that customer. Staff witness Tom Harris recommends that CLECs obtain either a written letter of authorization (LOA) or electronic authorization from the customer.

The tariffs filed by USWC and GTE are consistent with Staff’s recommendation. Section 2.3 of GTE’s tariff provides that acceptable "confirmation documentation may be obtained through a written LOA from the end-user or an end-user’s electronic authorization by use of an 800/888 number." USWC revised §6.2.A.6.a of its tariff to state that an "LOA may be provided to USWC in written form or via electronic authorization by use of an 800 number."

Inexplicably, GTE’s post-hearing brief states that the company does not permit electronic letters of authorization. It further maintains that such authorization violates §222 of the Act, which permits disclosure of customer proprietary network information (CPNI) only after affirmative written request. GTE claims that a CLEC must obtain access to CPNI in order to ascertain the services a customer receives from the ILEC, and therefore which building blocks to order. GTE further maintains that electronic authorization is not a "written request," and, in fact, has not been adequately defined. In order to avoid "slamming" problems, GTE maintains that customers should be afforded a higher level of protection than that afforded by the "PIC change" process now used to authorize service transfers between interexchange carriers.

Staff responds that GTE’s concerns regarding slamming will be addressed in the Commission’s pending rulemaking proceeding in docket AR 324. It points out that those rules will apply to the tariffs authorized in this proceeding.

2. Commission Decision -- Letters of Authorization.

We agree with the authorization procedures recommended by Mr. Harris and set forth in the USWC and GTE tariffs. We are not persuaded by GTE’s claim that electronic authorization violates the Act or is inadequately defined. To begin with, we do not think that CLECs will require access to CPNI to order the building blocks necessary to serve a customer. When a customer agrees to take service from a CLEC, the CLEC knows what service the customer wants and the building blocks necessary to supply that service. Contrary to GTE’s claim, the CLEC does not have to ask the ILEC what services the customer is currently receiving to place a building block order. Since access to CPNI is unnecessary, the written authorization requirement in §222 does not apply.

We also disagree with GTE’s claim that electronic authorization has not been defined. Both the USWC and GTE tariffs provide that electronic authorization may be made by the customer over the telephone via an 800 or 888 number. We do not understand what additional explanation is required. Also, it is common practice for businesses to request permission to tape record electronic authorizations so that transactions can be verified. We believe that this additional precaution is reasonable and should be included in the tariffs of both ILECs.

Staff is correct that slamming issues will be examined in docket AR 324. Any rules adopted in that docket will apply to purchases under the building block tariffs.

Section 2.8 of USWC’s tariff lists a number of conditions where "Special Construction Charges" apply. Similar provisions are included in Section 4.6 of GTE’s tariff, which specifies "Time and Material Charges." Staff and Joint Intervenors oppose various elements of these tariff proposals.

Joint Intervenors allege that the tariff language proposed by USWC will allow the company to arbitrarily refuse to add facilities to provide building blocks for requesting carriers. They observe that the tariff states USWC will provide access to unbundled elements "to the extent existing facilities are available," that special construction charges will apply when building blocks "are not available," and that "all necessary construction will be undertaken at the discretion of USWC." Joint Intervenors claim the proposed language allows USWC to refuse to add facilities to provide building blocks to requesting carriers whenever it is competitively advantageous for USWC to so. Moreover, they allege that the tariff will allow USWC to assess special construction charges even where the costs of construction are already included in recurring costs, thereby resulting in double recovery.

USWC witness Mason testified that special construction charges should apply only "where USWC incurs additional costs to build or modify facilities for the benefit of a carrier and those costs are not included in existing rates." Mr. Mason also agreed that USWC will notify a requesting carrier if USWC declines to undertake construction on the part of the carrier. Joint Intervenors urge that these provisions be included in USWC’s tariff. They further recommend that USWC provide written notice of any refusal to construct facilities for a requesting carrier.

Staff witness Breen recommends the Commission reject tariff language requiring special construction charges when "the facilities to provide network elements are not available and, at the request of the Carrier, and agreement by USWC, USWC constructs facilities to provide the network elements for the Carrier and there is no other requirement for the facilities so constructed." Mr. Breen asserts that special construction charges should not apply when facilities are constructed to provide defined building blocks. This is because recurring building block rates already reflect the cost of placing that investment. Furthermore, the language proposed by USWC and GTE permits the ILECs to refuse to construct facilities to provide authorized building blocks.

At the same time, Staff states that it is reasonable for ILECs to restrict the purchase of building blocks to the same terms and conditions that restrict purchases under the ILEC retail tariffs. For example, if a specific building block requires fiber optic outside plant facilities and those facilities are not installed in the serving central office, the ILEC should not have to install fiber facilities to provision that building block. On the other hand, if the ILEC has installed facilities in the serving central office, it should be required to extend, expand, or reinforce those facilities to provision the building block with a priority equal to that for its retail operations. Staff recommends that such a provision be included in the USWC and GTE building block tariffs.

Staff also recommends that §2.8.C of USWC’s tariff be deleted. That section states that "all necessary construction will be undertaken at the discretion of USWC consistent with budgetary responsibilities and consideration for the general body of users."

Relying on Iowa Utilities Bd., USWC states that it is not required under the Act to construct or modify facilities on behalf of CLECs. It agrees, however, to construct facilities to the limited extent necessary to accommodate interconnection and access to unbundled network elements. USWC asserts that its special construction charges are designed to "prevent CLECs from conscripting USWC to build their facilities." It maintains that requesting carriers should bear the risk of facilities that are constructed or modified on their behalf by paying the upfront costs associated with those activities.

USWC further maintains that its tariff provisions are designed to ensure that CLECs do not overwhelm USWC with construction demands that prevent the company from meeting its obligations to other customers. USWC points out that its tariff mirrors the provisions applicable to private line customers, whose service more closely resembles the building blocks requested by CLECs than does the service supplied to POTS customers.

In response to concerns regarding double recovery of costs, USWC maintains that constructing facilities for competitors without advance payment essentially amounts to bankrolling its competitors’ operations. USWC states that it will work with Staff to establish a policy to ensure that no double recovery occurs once an appropriate construction policy is adopted.

GTE also challenges Staff’s recommendations. Like USWC, it observes that ILECs are not required under the Act to construct new facilities for CLECs. Rather, it maintains that §251(c)(3) of the Act "requires unbundled access only to an ILECs existing network -- not to a superior unbuilt one." GTE notes that the Commission has also interpreted Oregon law to hold that ILECs "have no obligation to build additional facilities unless there are adequate financial assurances that their costs can be recovered." GTE emphasizes that the facilities at issue here will be ordered out of a tariff and could be abandoned at whim, providing the ILEC with no assurance that one-time construction costs will be recovered by monthly recurring rates. It urges that any new construction costs be recovered through nonrecurring charges.

GTE also maintains that Staff’s position will create the wrong economic incentives. If CLECs can compel ILECs to construct facilities to serve CLEC customers, CLECs will be able to seek business without risking their own capital resources. GTE asserts that such a result does not occur in competitive markets. Economically correct decision making requires each party to evaluate the risk associated with providing service to a particular customer. GTE emphasizes that decision making will be inappropriately skewed if one market participant is able to shift a substantial portion of its capital risk to its competitor.

2. Commission Decision -- Special Construction Charges.

The Commission adopts Staff’s recommendations. Contrary to the claims of USWC and GTE, we do not find that the Staff proposals violate federal law or impose unreasonable risks upon the ILECs. Staff does not recommend that the ILECs construct facilities to provide requesting carriers with superior service quality. Rather, it proposes that USWC and GTE be required to extend, modify, or construct facilities where necessary to provide requesting carriers with access to authorized building blocks. This is essentially the position articulated by USWC, which agrees to "construct facilities to the limited extent necessary to accommodate interconnection or access to unbundled elements." It is also consistent with the Eighth Circuit’s interpretation of §251(c)(3) of the Act.

In addition, Staff’s recommendation only requires the ILECs to provide such building block facilities where the ILEC’s retail customers also have access to those facilities. Thus, in supplying building blocks to competing carriers, USWC and GTE must only assign the same construction priority that they assign to providing those same facilities to their retail customers. This policy will enable CLECs to offer their customers services equivalent to those that the ILEC supplies to its customers. Parity in the provision of building block facilities is essential to meet the nondiscriminatory access requirements of §251(c)(3).

USWC and GTE maintain that the Staff proposal subjects them to the risk that they will not recoup their investment costs under recurring rates, because CLECs may prematurely abandon service. We do not believe the risk that the ILECs face is any greater than that which they now face in providing service to their retail customers, particularly large business customers. Moreover, special construction charges continue to apply in a number of cases where USWC and GTE are requested to provision facilities in a manner that might entail additional risk. Joint Intervenors acknowledge that special construction charges are appropriate where a carrier requests special or unique arrangements.

USWC and GTE argue that special construction charges are necessary because construction costs may not be recovered through the recurring charges assessed on CLECs for building blocks. Conversely, Joint Intervenors contend that special construction charges may lead to double recovery because the expenses are already collected in the recurring building block rates. The Commission addressed this issue previously in Order No. 97-003, an arbitration proceeding involving USWC, AT&T, and MCI. We see no reason why the same policy should not apply in this case:

As a general matter, the costs of providing a network element or service are included in the TELRIC-based price of that element or service. However, where an ILEC incurs additional costs to build or modify facilities for the benefit of a requesting carrier, and those costs are not included in existing rates, the ILEC is entitled to recover such additional costs. The ILEC has the burden of showing that any claimed additional costs are not already recovered through its existing rates.

If an ILEC demonstrates that it is entitled to recover additional costs to provide facilities on behalf of a requesting carrier, it may propose to recover those costs through nonrecurring charges. However, because large up-front charges tend to discourage competition, the Commission will attempt to spread cost recovery over a reasonable period of time and allocate such costs among all requesting carriers. This approach is consistent with that approved by the FCC in 47 C.F.R. §51.507(e). See also, Order No. 96-283 at 13-14, Order No. 96-325, Appendix A at 11; First Report and Order at ¶682, 743-752.

As a final matter, we agree with USWC that notice should be given in those circumstances where a carrier has requested an ILEC to provide building block facilities and the ILEC refuses to grant the request. We further agree with Joint Intervenors that the reasons for rejecting a carrier’s building block request should be put in writing by the ILEC. If the carriers are unable to resolve the dispute, they may request the Commission to resolve the matter.

C. Tariff Terminology. (Issue VIII.A)

USWC’s tariff uses the terms "NAC," "NACC," and "building block," as well as the terms "loop," "port," and "network element." Staff witness Williams recommends that USWC cross reference these definitions so that any differences in terminology may be identified. He also recommends that the definitions correspond with those used in docket UM 351. USWC agreed to revise Section 2 of its tariff to incorporate Staff’s recommendations.

GTE’s proposed tariff does not include the terms "loop" or "network element," but does define the term "port." Mr. Williams recommends cross referencing the terms "port" and "NACC" for the reasons noted above. GTE did not indicate whether it opposes Staff’s recommendation.

In its opening brief, Staff also recommends several additional definitions for inclusion in the USWC and GTE tariffs. USWC and GTE did not address Staff’s recommendations in their reply briefs.

The Commission agrees with Staff that the Oregon building block terms -- NAC, NACC, and building block -- should be defined and cross referenced with the unbundling terminology used by the FCC -- loop, port, and network element -- so that any differences may be identified. We also agree that the terminology used in the USWC and GTE compliance tariffs should be consistent with the definitions used in docket UM 351. However, because USWC, GTE, and other parties have not had an adequate opportunity to respond to definitions included in Staff’s post-hearing brief, we do not adopt those at this time. USWC and GTE should review Staff’s recommended definitions to determine if they should be incorporated in the revised tariffs filed in accordance with this order. If Staff or other parties object to the definitions included in the revised tariffs, they may raise those objections during the tariff review process.

D. NAC Capabilities and Provisioning. (Issue VIII.B)

The basic two-wire NAC may be provisioned as either an analog or digital connection. GTE witness Banzer testified that a CLEC must specify the appropriate channel performance building block on the LSR if the carrier wishes to use the basic NAC to support digital service.

Staff witness Jack Williams asserts that the basic NAC must be capable of providing satisfactory analog service. It may also be capable of providing satisfactory digital performance and other services without modification. The CLECs should be able to test the NAC and determine if performance improvements are necessary. If the CLEC does not specify or pay for any performance above the analog requirements, the ILEC should not be required to provide increased performance.

Section 6.1.B of USWC’s proposed tariff states that "USWC will provide the Carrier with all of the functionalities of a particular element, so that the Carrier can provide any telecommunications service that can be offered by means of the element." Mr. Williams recommends that GTE include a similar provision in its tariff. He also recommends that the ILEC tariffs specify the procedures and charges necessary to enhance the performance of the NAC beyond its guaranteed analog capabilities. Finally, Mr. Williams proposes that CLECs be allowed to obtain the basic NAC as a metallic (copper) facility if the ILEC has those facilities available. USWC revised §6.2.5.F of its tariff to correspond with this recommendation.

GTE contends that the Staff recommendations are vague and may conflict with services GTE can legitimately offer. GTE is also concerned that other customers may incur service problems unless CLECs are required to notify the ILEC when service-enhancing copper cable technologies are employed. In addition, CLECs should be required to certify that any such technology will not interfere with the ILEC’s existing or future technology.

We agree with Mr. Williams that the ILEC tariffs should specify circumstances which require enhancements to the NAC, the procedures CLECs must follow to obtain such enhancements, and any charges for such enhancements. If a loop is capable of digital performance or other capabilities without performance enhancements, that information should be available to the CLEC. Without such information, CLECs will be unable to submit accurate and complete LSRs and may order unnecessary services. Section 6.1.B of USWC’s tariff contemplates that it will provide CLECs with the information required to select the proper building blocks. GTE’s tariff should be revised to include similar language.

We also find that CLECs should be able to purchase metallic facilities if such facilities are available from the ILECs and can be provided without significant problems. On the other hand, if copper facilities are not available, the ILECs should not be required to install them. For example, GTE witness Banzer noted that GTE received approval to remove certain metallic facilities several years ago. We agree that it does not make sense to require the company to reintroduce those facilities.

Finally, we concur that CLECs who purchase building blocks to supply enhanced copper cable technologies should provide advance notice to the ILECs. This procedure was been adopted in the GTE/MCI interconnection contract. It permits the ILEC to determine if the technology proposed by the CLEC will adversely impact ILEC facilities or services provided to other customers. USWC and GTE should propose a list of such services for inclusion in their tariffs.

After the CLEC notifies the ILEC that it intends to use a particular service-enhancing technology, the ILEC should evaluate the proposed service and provide the CLEC with a timely response. If the ILEC concludes that the proposed service will have an adverse impact upon other facilities or customers, the ILEC and CLEC should explore methods for resolving the problem. The carriers may also seek assistance from the Commission.

E. NAC Conditioning. (ISSUE VIII.D)

The issue of whether cable unloading costs should be included in nonrecurring charges is discussed above. This issue deals with whether additional costs for conditioning the basic NAC should be included in nonrecurring charges.

Staff states that the basic NAC should be capable of analog service at 300 to 3000 hertz without needing any conditioning. It also maintains that an ILEC should not assess a conditioning charge unless a CLEC requests conditioning.

In response to Staff’s recommendations, USWC revised §6.2.A.2 of its tariff to provide that "no conditioning charge is required for standard analog voice grade service over a basic NAC." That section also provides that basic unbundled loops are available within the analog voice frequency range of 300-3000 Hz. In addition, USWC revised §6.2.A.2.a of its tariff "to provide unbundled loops to carriers with the same level of conditioning as exists within the USWC network." USWC witness Schmidt further stated that USWC will assess conditioning charges only when conditioning is requested by a CLEC.

GTE indicated that this issue is applicable only to USWC. Staff disagrees and recommends that GTE’s tariff be revised to reflect the changes Staff has proposed.

The Commission agrees with Staff that conditioning charges should only apply when the CLEC requests performance capabilities in excess of the minimum 300-3000hz analog capability. USWC's tariff revisions to Section 6.2.A.2 are consistent with this approach We also agree with USWC that CLEC loops should not have any lesser level of conditioning than those generally available on the ILEC’s network. GTE should include similar provisions in its tariff.

F. Network Interface Device (NID). (Issue VIII.E)

At the initial stage of this proceeding, the parties agreed not to address this issue because the costs and rates for the NID building block had not yet been developed. GTE agreed to remove the NID building block from its proposed tariff. During the hearings, however, an access issue arose regarding the need for a stand-alone NID in circumstances where a CLEC provides an NAC to a customer’s premises and seeks to use the ILEC’s NID. Joint Intervenors allege that USWC, unlike GTE, will not provide access to a stand-alone NID.

In a situation where a CLEC provides its own loops to a customer, USWC has stated that it will allow the CLEC to place its own NID at a customer’s premises. The CLEC may connect its NID to the USWC NID to gain access to the customer’s inside wiring. USWC will also allow a CLEC access to terminate its loops within the USWC NID in this situation, provided there is spare capacity inside the NID housing. The CLEC will be responsible for the drop installation and for adhering to the installation recommendations and requirements prescribed by the NID supplier. In addition, the CLEC must meet all requirements of the FCC and National Electrical Code.

GTE agrees to provide similar access to the NID. It is willing to allow the CLEC to terminate its loops at the GTE NID where spare capacity exists within the NID. CLECs may also gain access to end-user inside wiring via a NID to NID connection. In addition, a CLEC supplying its own NAC may move all inside wiring terminating on a GTE NID to a CLEC NID. In that case, an NID to NID connection is not required.

We find that USWC and GTE have developed reasonable terms for providing access to ILEC NIDs where the CLEC provisions its own loops. While the provisions relating to NID access may have to be refined as competition proceeds, we are satisfied that the arrangements described above will provide CLECs with adequate access to USWC and GTE NIDs for the present.

G. NACC Definition. (Issue VIII.H)

Two issues were raised regarding the NACC definition.

(a) Staff recommends that the §6.2.2.D of the USWC tariff be revised to eliminate any reference to minutes of use charges in the definition of the NACC ISDN building block. USWC witness Mason responds that the NACC definition includes billing for minutes of use because a portion of the recurring costs for the NACC ISDN includes the cost of switching over and above the average usage for a flat-rated business line. Mr. Mason observes that ISDN processors are more expensive than analog ones. ISDN lines also utilize more of the switch because they establish and control more channels than an analog line. Thus, the additional minutes of use costs were included in the cost study.

Although Mr. Mason agrees that the minutes of use reference should be eliminated from the NACC ISDN definition, he suggests that an additional building block should be introduced to cover the incremental costs incurred by USWC to provide the NACC building block. He points out that the NACC ISDN rate adopted by the Commission in Order No. 97-239 does not include usage and billing costs.

The Commission finds that the definition of the NACC ISDN in USWC’s tariff should be revised to delete the language which indicates that minutes of use will be billed to the requesting carrier. In Order No. 97-239 at 8, we held that the recurring NACC ISDN rate should not include usage and billing costs because those costs are already included in the recurring building blocks rates for switching and billing. If USWC believes that the Commission should revisit this issue and revise the NACC ISDN building block price to take into account additional costs, it should file a new recurring cost study. Recurring building block rates are not at issue in this proceeding.

(b) MCI witness Hydock recommends that §4.5 of GTE’s tariff be revised to eliminate the requirement that carriers purchasing unbundled switching also purchase an unbundled NACC. Mr. Hydock states that this requirement is contrary to the Commission’s decision to unbundle the NACC and switching building blocks.

GTE witness Banzer responds that GTE’s tariff merely reflects the fact that a NACC standing alone provides no functionality; the functionality is provided by the switching elements.

The Commission concurs with GTE. Although we have unbundled switching from the port, GTE correctly observes that purchasing an ILEC port provides no functionality unless the carrier also purchases ILEC switching elements. As we have noted elsewhere, building blocks do not have to function on a stand alone basis. GTE’s tariff reflects that fact.

H. LSR Limitations.

Section 3.1.1.A of GTE’s proposed tariff provides that "only one service type (e.g., NAC or NACC) can be ordered per LSR for each end-user customer." Staff and Joint Intervenors oppose this restriction. Mr. Breen observes that the restriction may impose added costs on CLECs to process separate orders for individual building blocks. Mr. Hydock claims there is no apparent reason for the restriction.

GTE witness Banzer explains that the proposed restriction does not limit the number of identical services that can be listed on the LSR. In other words, a carrier can "order any amount of the same service between the same locations for installation on the same date on a single LSR." According to Mr. Banzer, this process is designed to correspond with the manner in which GTE’s ordering and provisioning processes are set up.

As we understand GTE’s proposal, an LSR submitted by a CLEC must be limited to a single type of building block. Thus, a CLEC could include on the same LSR an order for 10 NACs to be installed at the same location on the same date. However, if the CLEC also wanted to purchase 10 NACCs to be installed at the same time and place, a separate LSR would be required. Likewise, separate LSRs would also be necessary for any additional building blocks the CLEC wanted to install together with the NACs and NACCs.

We agree with Joint Intervenors and Staff that the proposed restriction should be deleted from GTE’s tariff. GTE has not provided an adequate explanation why it is necessary to restrict the type of building blocks that can be ordered on an LSR. We do not understand why GTE’s ordering and provisioning processes cannot be adjusted to accommodate a more acceptable method of ordering building blocks. Furthermore, the proposed restriction would increase the CLEC’s cost of providing service as well as the possibility that errors will be made in processing CLEC service orders. Both of these results would limit the ability of new carriers to compete effectively in local exchange markets.

I. Bona Fide Request Process.

Section 2.7 of USWC’s tariff and Section 2.12 of GTE’s tariff contain substantially similar provisions relating to the bona fide request (BFR) process. The BFR process allows competing carriers to "request access to unbundled elements not already available" in the ILEC tariffs. MCI and Staff raise a number of issues regarding the BFR procedures proposed by USWC and GTE.

(a) MCI witness Hydock argues that the BFR tariff language is vague and overly broad. He argues that the BFR process should be limited to cases where a building block is not offered by the ILEC, rather than situations where the building block is "not already available" as described in the USWC and GTE tariffs. USWC witness Don Mason responds that the BFR process does not apply to all unbundled element requests; only those unbundled elements that are not already available in USWC’s tariff or through an interconnection agreement.

The Commission concurs with MCI that the language in the USWC and GTE tariffs is subject to misinterpretation. Although USWC intends that the BFR process should apply only to unbundled elements that are not authorized, the language quoted above could be construed as applying to unbundled elements that are authorized but "not available" from the ILEC for some reason. Accordingly, we find that paragraphs A. and B. of §2.7 of USWC’s tariff and the first paragraph of §2.12 of GTE’s tariff should be revised to indicate that the BFR process applies only to building blocks not already authorized by the Commission.

(b) Paragraph 6 of the GTE’s proposed BFR process states that disputes relating to any bona fide request should be handled through mediation or arbitration under §252 of the Telecommunications Act of 1996. Mr. Hydock argues that such a process could last well over a year, thereby allowing GTE to tie up competitors in protracted proceedings over a difference of opinion regarding the mechanics and process of the BFR. While both ILECs and CLECs may wish to maintain legal and regulatory relief mechanisms, Mr. Hydock asserts that some sort of accelerated dispute resolution should be allowed.

We agree that referring BFR disputes to the mediation and arbitration procedures under the Act would unnecessarily prolong the resolution of these disputes. USWC’s tariff provides that such disputes may be arbitrated under the dispute resolution provisions of its tariff. GTE should delete paragraph 6 of §2.12 of its proposed tariff and insert language that permits GTE and CLECs to resolve BFR disputes through dispute resolution procedures similar to those included in USWC’s tariff.

(c) Mr. Hydock alleges that the timetables included the USWC BFR process may preclude competitors from making timely commitments to their customers if USWC decides certain building blocks are not available. USWC responded to these concerns by revising its tariff to shorten the time for responding to BFRs. The revised version provides that USWC will:

(1) acknowledge a BFR within 14 days after receiving the request from a carrier;

(2) prepare a preliminary analysis within 30 days after receiving the request, indicating whether the requested interconnection or building block access meets applicable unbundling requirements; and

(3) within 90 days after notifying the CLEC that the request meets unbundling requirements, provide a quote specifying the terms and conditions under which USWC will supply the requested building blocks, including quantity, cost, and any minimum volume and term commitments.

Under USWC’s proposal, a CLEC will know within 30 days whether a request for a new building block meets the unbundling requirements specified in the tariff. If USWC objects to the request for some reason, the CLEC may pursue dispute resolution procedures immediately. If USWC does not object, then the total time to process the request will not exceed 120 days.

Although it may be necessary to revisit this issue at a later date, 120 days does not seem an excessive amount of time to evaluate requests for building blocks that have not been previously authorized. It is also consistent with the time period set forth in OAR 860-035-0070 of our Open Network Architecture rules. We therefore adopt USWC’s revised tariff.

GTE’s tariff sets forth initial timeframes for acknowledging receipt of a BFR and for preparing a preliminary analysis. It does not specify any deadline for the GTE to submit a final quote to the CLEC, however. The Commission finds that GTE should revise its tariff to correspond with the 120 day period discussed above. This will ensure that CLEC building block requests are processed within a time certain and eliminate the potential for prolonged delays in responding to such requests.

(d) In addition to the concerns raised by the parties, the Commission has two additional concerns with the proposed BFR tariff language. First, the USWC tariff states at various places that the company will evaluate BFR requests to determine if they comply with "the Act." This language should be changed to specify compliance with the unbundling requirements established by the Commission rather than the requirements of the federal Act. Second, both the USWC and GTE tariffs make reference to "unbundled network elements." While unbundled network elements are roughly synonymous with building blocks, they are not identical in all respects. Therefore the tariff language should be changed to substitute "building block" for "unbundled network element."

J. Recovery of Nonrecurring Costs.

Joint Intervenors argue that up-front nonrecurring charges create barriers to entry and stifle the growth of competition. For that reason, they maintain the ILECs should have the burden of showing: (a) why nonrecurring costs should not be capitalized into recurring charges; (b) that the nonrecurring cost is not already included in recurring costs; and (c) that no other future customer could benefit from the nonrecurring activity. Dr. Zepp emphasizes that there is a potential for double recovery of costs when the ILEC assesses recurring charges for investments which include capitalized costs for engineering and installing facilities and nonrecurring charges that recover similar costs.

The ILECs oppose recovering nonrecurring costs through recurring charges. USWC witness Brigham emphasizes that nonrecurring costs such as the labor costs necessary to establish service for a customer are appropriately treated as operating expenses and should be recovered through nonrecurring charges. He further maintains that, although USWC has identified a minimal level of double counting in its maintenance cost factors, there is no basis for Dr. Zepp’s claim that nonrecurring costs should be recovered through recurring charges.

In Order No. 96-283, the Commission agreed with the approach taken by the FCC in 47 C.F.R. §51.507(e). That rule stated, in part, that "State commissions may, where reasonable, require incumbent LECs to recover nonrecurring costs through nonrecurring charges over a reasonable period of time." In making that determination, we had in mind substantial upfront nonrecurring charges that would discourage competitive entrants from purchasing building blocks on an unbundled basis. Our intention was that the impact of such charges could be mitigated by allowing competing carriers to pay nonrecurring charges in equal installments over a period of several months, in a manner similar to the way recurring charges are assessed. We did not, however, intend that properly calculated nonrecurring costs should be capitalized and rolled into recurring charges. Our reading of the First Report and Order does not disclose that FCC intended such a result either.

Accordingly, we find that nonrecurring costs should be recovered through nonrecurring charges. If we conclude that any given nonrecurring charge may discourage competitive entry, we may authorize recovery of that nonrecurring charge over a period of several months to mitigate the financial barrier to entry. Since this order does not adopt any specific nonrecurring charges and we do not know the precise level of nonrecurring costs and charges that will result from the determinations made herein, it is unnecessary to comment further on this issue at this time.

MCI objects to §2.4 of GTE’s tariff which provides that "all telecommunications carrier services shown in this tariff are available in GTE serving areas except where prohibited by technological and economical limitations and where facilities and conditions permit." Mr. Hydock argues that, because building block rates are cost based and include a reasonable markup, the reference to "economical limitations" is unnecessary. MCI further asserts that the Commission has already determined that the building blocks authorized in UM 351 are technically feasible. Consequently, GTE should not be allowed to refuse to provide a building block unless it obtains a specific waiver from the Commission.

According to GTE witness Banzer, building blocks will generally be available on a statewide basis. The tariff language in Section 2.4 merely provides carriers with notice that there may be instances where a given building block cannot be provided. For example, the GTE equipment in use at a particular location may be incapable of providing the requested building block for technical reasons.

We agree with MCI that the reference to "economical limitations" is vague and should be deleted from the GTE tariff. We interpret the remaining language to have the limited scope described by Mr. Banzer. That is, it addresses those situations where GTE cannot supply a building block because of technical limitations in the facilities serving the area. The ILEC duty to supply building blocks to requesting carriers is addressed in our discussion of special construction charges.

As we have also ordered, USWC and GTE must provide written notice where they cannot provide building blocks requested by a CLEC. We presume that such a notice will precipitate discussions regarding other possible methods of accommodating the CLEC’s requirements. If the parties disagree over the provision of a building block, the dispute resolution processes set forth in the tariff would be triggered. If dispute resolution is unsuccessful, the parties may seek resolution from the Commission.

2. Section 2.4 Paragraph 2.

This paragraph requires that "any services to which a [carrier] has subscribed under this tariff that are not being used by the carrier to provide service to its customer (end-user) must be discontinued upon request by [GTE] where the related facilities or equipment are needed by [GTE]." MCI claims this provision is anticompetitive because it is prudent business practice for CLECs to maintain an inventory of facilities. It argues that GTE should not be able to require competitors to turn over inventory they hold and are paying for.

On first impression, it seems reasonable that carriers should be obligated to return building block facilities that are unused and needed by the ILEC. Such a situation might occur, for example, where a CLEC is paying for and maintaining a large inventory of unused loops that could otherwise be sold to other CLECs or used by GTE to serve its customers. Requiring the carrier to return the unused loops in that instance would prevent GTE from having to install additional loop facilities where unused capacity already exists.

However, it would be extremely difficult to ensure that the proposed tariff language is implemented in a fair and reasonable manner. In the first place, we do not know how GTE would be able to ascertain that building block facilities purchased by CLECs are not being used. This is particularly true in the case where a CLEC buys loop facilities and supplies its own switching. Secondly, we agree with MCI that it is reasonable for CLECs to be able to maintain an inventory of loops or other building blocks in order to meet anticipated demand for service. Requiring CLECs to relinquish all unused facilities eliminates this ability and places competitors at a disadvantage because GTE would have first right to those facilities.

We are persuaded the proposed tariff language will not produce a balanced solution to questions involving the allocation of unused building blocks and, therefore, should not be adopted. We acknowledge that these issues are complex and will require additional study as local competition progresses. In the interim, we expect that CLECs and ILECs will negotiate solutions to matters involving the allocation of unused capacity. Carriers have successfully negotiated issues relating to interoffice transport capacity, and we have no reason to believe that similar results can not be achieved in this context. In the event disputes cannot be resolved informally, the Commission will decide the matter.

3. Section 2.9.

This section of GTE’s tariff requires a carrier to maintain all facilities under the guidelines of GTE’s Technical Interface Reference Manual. MCI does not object to this provision, but states that GTE must make the Manual available to CLECs and provide a process for resolving disputes over the interpretations of the Manual.

GTE and USWC should take steps to ensure that all carriers requesting building blocks also have access to necessary technical manuals and publications. Disputes regarding interpretations of the manuals should be handled under the dispute resolution procedures in the ILEC tariffs.

4. Section 2.10.

This section of GTE’s tariff states that it is not responsible for changes in protection criteria or facilities that have the effect of rendering a purchasing carrier’s equipment obsolete. MCI requests that GTE provide a notification period and develop a process for resolving the impact of the changes.

Whenever possible, GTE should provide notice of changes that may have an adverse impact on the equipment used by other carriers. We expect that carriers will work together to discuss methods for minimizing the impact of such changes.

This section makes reference to deposit requirements. MCI argues that USWC should waive deposit requirements for CLECs with sound credit histories.

There is insufficient information in the record to allow us to adopt MCI’s proposal. Since MCI did not propose any tariff language for consideration, we are left to speculate what issues must be addressed to formulate reasonable deposit requirements for requesting carriers. Our experience demonstrates that deposit issues involve a more comprehensive analysis than MCI has presented.

2. Section 2.7.C

MCI objects to this tariff provision because it requires CLECs to specify that building blocks will be used to provide "telecommunications services." MCI claims that the Commission has never limited the use of building blocks to telecommunications services.

Our decisions in docket UM 351 unbundled ILEC telecommunications services into building block components so that requesting carriers could purchase those components to provide competing telecommunications services. We did not entertain the idea that CLECs might desire to use building blocks to provide some other type of non-telecommunications service, and MCI has not specified what types of non-telecommunications services might be provisioned. That being the case, we decline to modify this section of USWC’s tariff.

3. Section 4.4

Among other things, this section provides that USWC will provide interconnection to carriers that is equal to that of USWC. MCI states that the tariff should state that the USWC standards will conform to the Commission’s decisions in its pending quality of service rulemaking in AR 324.

As noted above, the quality of service rules the Commission adopts in AR 324 will apply to all telecommunications tariffs. It is unnecessary to modify the proposed tariff to include that fact.

4. Section 6.1.B

This section states that USWC will not restrict a CLEC from combining USWC building blocks with any technically compatible equipment the CLEC owns. MCI recommends that the section be revised to include technically compatible equipment that is leased by the CLEC from a third party. We agree.

5. Section 6.2.A and 6.2.C

These sections make reference to technical reference publications. As noted above, we agree with MCI that all technical publications and manuals referred to in the tariff must be identified and made available to requesting carriers.

6. Section 6.2.A.3

This section provides that CLECs are responsible for providing facilities that are compatible with USWC’s unbundled loops. MCI complains that CLECs are not given any guidance regarding how this determination should be made. It states that the tariff should be revised to require USWC to provide adequate documentation and cooperate with CLECs regarding compatibility issues.

MCI’s proposal may have merit, but there is not enough information in the record for the Commission to specify the type of documentation that must be provided or the level of assistance that is appropriate under these circumstances. As we have emphasized, USWC must provide CLECs with available technical publications to assist them to interconnect with USWC loops and other facilities.

7. Miscellaneous USWC Tariff Sections.

MCI also disputes USWC tariff language regarding liability, indemnification, and audit processes (§§2.12-2.14 and §2.25); points of interface (§4.5); and service interruptions (§4.7). MCI did not place sufficient evidence in the record for the Commission to make a determination regarding these issues.

 ORDER

IT IS ORDERED that:

U S WEST Communications, Inc., and GTE Northwest Incorporated shall submit revised tariffs specifying building block access arrangements, nonrecurring charges, and tariff terms and conditions consistent with the findings set forth herein. The revised tariffs shall be filed within 60 days of the date of this order.

 Made, entered, and effective ________________________.