5. What is the Appropriate Price for Use of the HFPL?

5.1. Parties' Positions

5.1.1 TURN's Position

TURN asserts that to determine the monthly rate, the Commission must first determine the cost of the loop (including the Commission's shared and common cost markup), then determine the amount of the loop cost that is reasonably subject to recovery from the service utilizing the high frequency portion of the loop. According to TURN, this provides the basis for a reasonable price. TURN proposes monthly recurring rates for the HFPL UNE of $2.0025 for Pacific and $2.3175 for Verizon.

According to TURN, the first issue to be addressed is what cost information the Commission should rely on in setting a reasonable monthly recurring charge for the HFPL UNE. Pacific's loop cost studies are based on 1994 data, and the Commission is currently reexamining those costs in its UNE Reexamination Proceeding.²⁴ Verizon presents a special problem because there are no approved cost studies for Verizon in California.

TURN's witness Dr. Roycroft used the FCC's Hybrid Cost Proxy Model (HCPM) that the FCC used to determine the cost of telephone loops for the purpose of calculating the amount of federal funding for universal telephone service provided to all ILECs, including Pacific and Verizon. The HCPM model yielded loop costs of $8.01 for Pacific and $9.27 for Verizon. According to TURN, the FCC's cost information provides a publicly available, reasonable, unbiased, and current basis for use in setting prices.

The second step, TURN states, is to determine a reasonable price for the HFPL UNE. Since the loop is a shared facility of DSL and other services--including local exchange service, vertical features, and toll service--the loop costs cannot be attributed to the production of any single service or product. TURN's witness Roycroft developed his recommended prices using an economic allocation tool known as the Shapley Value. The allocation that results is viewed by researchers to be fair and equitable. Also, the method ensures that the allocation components will always add up to the total cost associated with the shared facility. Also, the application of the Shapley Value is a very straightforward process that can easily be utilized in the future if prices need to be adjusted.

According to TURN, in recent years some have asserted that the term "cost allocation" is akin to an economic profanity. The reality is that the loop is a shared cost of DSL service and other services, and it is simply impossible to directly attribute the entire cost of the loop to any of these services. An allocation must take place. Some parties to this proceeding might argue that what TURN is proposing is to re-litigate every single case involving every service that uses the loop. The issue on the table is how to develop a price for the HFPL UNE. It is presumptively unreasonable to set a zero monthly recurring price for DSL line sharing's use of the loop. It is also unreasonable to require the HFPL UNE to bear the entire cost of the loop. So, some apportionment must be made. In developing his pricing recommendations, Roycroft took into account the fact that existing services or products also use the loop, and this fact was relied upon to determine what portion of shared loop costs should be borne by the new HFPL UNE product.

The Shapley Value method addresses the problem of recovering the cost of a shared facility by identifying possible groupings of service offerings that share facilities and assigning unbiased probabilities of each grouping utilizing the shared facility in all possible combinations with other services. According to TURN, the existence of these services was relied upon in TURN's effort to determine what portion of the shared loop costs should be borne by the new HFPL UNE product. TURN allocates the loop costs to four families of services: basic exchange service, toll/access, vertical services, and advanced services, and recommends setting rates at 25% of the total loop costs that result from application of the HCPM model, or $2.0025 for Pacific and $2.3175 for Verizon.

Pacific rebuts TURN's proposal saying that TURN's analysis is based on an improper starting point. TURN begins its analysis of loop costs by stating that first the cost of providing the shared input must be determined. TURN disregards the fact that the Commission set a loop price of $11.70 for Pacific in D.99-11-050, and instead develops its own loop cost using the FCC's HCPM model. The HCPM is based on the Hatfield model, which this Commission has rejected for identifying TELRICs of UNEs. Also, the FCC itself only used the HCPM to determine loop costs in high cost areas for determining universal service support and does not support that model for determining the TELRICs of UNEs.

Pacific states that TURN then determines that access to the HFPL should be priced at 25% of loops costs, based on an allocation of costs among four service families: basic exchange service, toll/access services, vertical features, and high-speed data services. The Commission has consistently rejected the position that costs of the voice grade loop are caused by toll and other services that use the loop. In OANAD, the Commission found,

[I]t would be inappropriate and contrary to the TSLRIC (TELRIC) principles adopted in D.95-12-016 to treat the loop as a shared cost (with usage services). (D.96-08-021, mimeo at 90-91.)

Pacific also cites similar language in Commission order D.94-09-065:

We concur with the general principle that NTS [non-traffic sensitive] costs (e.g., loop costs) should be assigned to subscribers' basic exchange services. (D.94-09-065 at 44.)

Pacific concludes that since allocation of loop costs among these four service families has been rejected by the Commission, Roycroft's determination that access to the HFPL should be priced at 25% of the unbundled loop cost should also be rejected.

Verizon points out that while TURN recommends rates based on the FCC's HCPM, neither the HCPM model nor TURN's witness Roycroft's work papers used to generate a price were introduced as exhibits into the record of this proceeding. According to Verizon, the Commission has established a separate phase of this proceeding to calculate costs and prices for line sharing. The current phase was limited to the narrow policy question of whether there should be a positive price for the HFPL, not what that price should be. Consequently, Verizon recommends that the Commission should establish the HFPL price in the cost and price phase of this docket, where the parties may file and fully analyze cost studies. Verizon proposes to present the evidentiary support for its HFPL-related costs in the cost study phase of this proceeding and has presented its cost methodology in this phase for illustrative purposes to demonstrate how direct costs associated with the HFPL may be calculated in the cost and price phase. Verizon recommends that the interim $3.00 monthly recurring rate for the HFPL remain in effect until a final rate is adopted.

Moreover, Verizon does not believe the HCPM is a valid tool to use to establish costs in a UNE docket. The HCPM was used by the FCC in a universal service cost docket. The FCC has warned against using this model in a UNE docket:

Our USF [Universal Service Fund] cost model provides a reasonable basis for comparing cost differences between states. We have previously noted that while the USF cost model should not be relied upon to set rates for UNEs, it accurately reflects the relative cost differences among states.

Verizon's witness Collins also expresses concerns with the manner in which Roycroft extracted loop cost estimates from the HCPM, which resulted in estimates that are biased downward. Roycroft indicated that in an attempt to focus on copper loops, he eliminated the Common-Language Location Identification (CLLI) codes that identified fiber feeder. As a result, the sample of CLLIs selected tends to be composed of very compact, densely-populated wire centers that have significantly lower loop costs than the statewide average. Roycroft's sample excluded the cost of the copper facilities in the core areas of the remaining lower density wire centers. While these wire centers serve some customers via fiber-fed DLC, customers located within a few miles of the wire center are served over 100% copper facilities. Ignoring these customers (served by copper) imparts a significant downward bias on the cost results.

As a comparison, Collins ran Verizon's Integrated Cost Model (ICM) to test for bias by placing the CLLIs with only copper into a single grouping of wire centers. Collins' results illustrate that the average loop cost for the CLLIs sampled by Roycroft was less than one half of the statewide average loop cost.

5.1.2 ORA's Position

ORA asserts that the price for the HFPL should be cost-based and set as an allocation of the unbundled loop charge. ORA recommends rates of $2.46 for Pacific and $3.00 for Verizon.

ORA makes its calculation by referencing Pacific's witness Scholl's Directory Assistance Decision example, and its allowable markup of 42%. ORA states, "Using this [the 42% markup], the joint and common costs assigned to the high frequency usage could justify a price of no more than $2.46, and therefore the price of the HFPL, could be no higher than that amount." (ORA Reply Brief at 8.)

For Verizon, ORA proposes retaining the $3.00 rate adopted in the Interim Line Sharing phase.

5.1.3 Pacific's Position

Pacific proposes that the Commission retain the $5.85 price for access to the HFPL that was adopted in the interim line sharing arbitration and take the opportunity to utilize these funds to help offset the shortfall in the cost of providing basic residential service.

Pacific states that in its Line Sharing Order, the FCC declared that one loop can actually comprise dedicated connections from a single customer to two different service providers-one providing the customer with voice service, and the other with data service. Either connection, on its own, requires the loop, and none of the loop costs on the shared line are attributable to only one of the two connections. Consequently, standard TELRIC methodology, which was designed for estimating direct costs, is not applicable to pricing access to the HFPL.

According to Pacific, the FCC and this Commission have offered some guidance on the appropriate means of allocating loop costs on a shared line. One of the most fundamental principles of costing recognized by this Commission is the concept of "cost causation." As described in the Commission's Consensus Costing Principles, "Principle No. 2: Cost causation is a key concept in incremental costing...The basic principle of cost causation is that only those costs that are caused by a cost object in the long run should be directly attributable to that cost object." (D.95-12-016, Appendix C.) As described above, the single copper loop can provide both a dedicated voice connection and a simultaneous dedicated data connection. Either connection, on its own, requires the loop, and on a shared line, the two dedicated connections jointly cause the cost of the loop. Consequently, pursuant to this Commission's Consensus Costing Principles, allocation of costs to both the high-and low-frequency portions of the loop is appropriate.

According to Pacific, consumers have several options available if they wish to obtain high-speed access to the Internet. They may purchase DSL service, or they may choose broadband wireless, cable or satellite technologies. This Commission needs to bear in mind what impact an artificially low price for access to the HFPL would have on the broadband market in general.

Pacific states that the Commission should also consider the pro-competitive effect that a $5.85 price for access to the HFPL has had-and will continue to have-on the DSL market. If CLECs have to purchase an entire loop from Pacific, they would pay $11.70. Currently with line sharing, CLECs can purchase just the high frequency portion of that loop at an even more substantial discount-50 percent off the current loop price-down to $5.85. According to Pacific, this clearly provides a significant incentive for CLECs to enter the residential market and offer attractive prices.

Pacific's witness Dr. Fitzsimmons asserts that setting the price for access to the HFPL at 50% of the price of the unbundled loop will make a reasonable contribution to joint loop costs. The $5.85 price has been in effect for several months, and during that time CLECs have purchased increasing volumes of line-shared loops. Pacific sees that that price is spurring deployment of advanced services.

While TURN and ORA claim that an appropriate price for the HFPL is based on less than a 50% allocation of costs, Pacific asserts that TURN begins its analysis from the wrong starting point. TURN claims that the Commission-approved loop rate of $11.70 is too high, and recommends that the Commission set the price for the HFPL based on the loop rate TURN derived from the HCPM.

According to Pacific, TURN errs in applying the Shapley value methodology to reach a price for the HFPL that is 25% of the unbundled loop price because the Commission has consistently rejected the foundation upon which TURN bases this argument: that the loop should be treated as a shared cost with usage services.

Additionally, the HFPL is an appropriate source of contribution to the shortfall that currently exists in the provision of basic services. The Commission has in the past relied on Pacific's above-cost services to contribute to Pacific's losses incurred in the provision of basic service. The Commission's New Regulatory Framework (NRF) under which the Commission placed both Pacific and Verizon several years ago, does not guarantee Pacific price increases for basic service if it loses market share for those above-cost services. Instead, under NRF, Pacific's challenge is to increase its efficiency and introduce profitable new services.

Rhythms rebuts Pacific's and TURN's allegation that the HFPL is a joint product. DSL-based service is not available to a person who does not subscribe to basic exchange service. Because the stand-alone loop element is already available to serve customers not subscribing to basic exchange service from the ILEC, it is implausible to suggest that line sharing could be defined as anything but an enhancement to basic exchange service. According to Rhythms, the two arrangements could only be considered joint products if they were equally available on a stand-alone basis. That is not the case. Line sharing on a particular loop is available only to the specific customer whose analog voice service is provided over that loop.

Rhythms also rebuts Pacific's witness Fitzsimmons' assertion that the approach he advocates is an allocation of loop costs. As ORA witness Johnston has correctly observed, Pacific has not actually proposed to allocate loop costs among multiple uses of the loop. It has proposed additional revenues on top of those that it received prior to the requirement to provide line sharing. According to Rhythms, Pacific's allocation scheme is simply a mechanism to allow Pacific to recover more than the total cost of the loop from those customers who order both basic exchange and line shared DSL services over the same loop.

Rhythms states that because line-shared access to the loop creates no loop cost, Fitzsimmons focuses on the asset value of that access, and in so doing, proposes a charge to replace the profit that Pacific could have generated with that asset, were it not for the requirement to allow competitive access. However, this loss of profit occasioned by allowing competitive access is a private opportunity cost to a monopolist, not a cost to society as a whole. According to Rhythms, the FCC specifically rejected "opportunity cost" pricing for UNEs at paragraphs 708 and 709 of the Local Competition First Report and Order.

Rhythms disagrees with Pacific's conclusion that a positive HFPL rate is needed to subsidize basic service. In the Commission's Universal Service docket, the Commission has already created an explicit universal service funding mechanism that is designed to deliver all of the subsidy that the Commission deems necessary to support residential basic exchange service. In addition, the Commission also required Pacific to offset that additional revenue by reducing prices for other services.

Moreover, states Rhythms, Pacific has not presented any convincing evidence to establish the need for a subsidy of its retail local exchange prices. Pacific's witness Scholl relies on the premise that the combination of residential basic exchange prices plus California High Cost Fund - B (CHCF-B) funding should recover the entire cost of basic exchange service plus a 46% allocation of Pacific's retail shared and common costs. This allocation of shared and common costs far exceeds the amount the Commission found to be a reasonable allocation to basic exchange service in its Universal Service decision, D.96-10-066. In that decision, the Commission concluded that:

As TURN points out, Congress recognized that potential in the Telco Act, which contemplates that universal service should bear no more than a reasonable share of joint and common costs, and in the Conference Report, which suggests that the cost of universal service bear less than a reasonable share. Consistent with that direction, we have reduced common costs per line from $2.91 to $2.00 to safeguard against these possible competitive problems. We note that the revised common costs are a more reasonable allocation. The reduced amount represents approximately an 11% mark-up over direct and shared costs, which is commensurate with the overhead factors experienced in the local exchange industry.

Rhythms also asserts that Pacific's proposal that a charge equal to 50% of the UNE loop price, yields a rate of $5.85, is not entirely correct. The FCC requires deaveraged prices for the UNE loop, and Pacific has agreed to deaveraged loop prices on an interim basis in its ICA with MCImetro Access Transmission services, L.L.C. Hence, if the Commission adopted Pacific's proposed 50% of the loop price, the correct price for access to the HFPL would be 50% of the UNE loop price in each deaveraged zone.

ORA rebuts Pacific's argument that the proposed price of $5.85 will be pro-competitive because CLECs have purchased increasing volumes of line shared lines during the 13 months the interim price has been in effect. ORA points out that the "CLEC" that is purchasing the increased volumes is SBC's ASI, not an unaffiliated CLEC. According to ORA, ASI has purchased more than 95% of those line-shared loops.

5.1.4 Verizon's Position

Verizon asserts that there are direct costs associated with providing the HFPL UNE. Only by investigating this cost as compared to the cost of unbundled POTS (Plain Old Telephone Service) loops can the appropriate cost-based HFPL share be determined. Until this matter is resolved in the costing phase, Verizon recommends that the interim rate of $3.00 per month be continued.

Verizon states that in the interim phase, the company had not identified any incremental loop costs caused by providing the HFPL over home-run copper loops,²⁵ and consequently did not propose a positive price for the HFPL. Since that time, Verizon has identified "embedded constraint" incremental costs associated with providing the HFPL over copper loops. Providing the HFPL over home-run copper loops will take place on Verizon's existing network, which has many copper loops that are 12-16 kft in length. In a forward-looking environment, those same loops may well be converted to hybrid fiber/copper loops. If Verizon is providing the HFPL on the existing all-copper loop, it cannot efficiently introduce fiber into this loop or convert that customer to a hybrid fiber/copper loop.

ORA disagrees with Verizon's sudden conversion to finding a direct cost associated with providing the HFPL. No TELRIC rationale can be found to justify recovery of "embedded" costs. Further, Verizon's disinclination to migrate DSL customers to fiber is an artificial one, not a legal or technical requirement.

Rhythms asserts that Verizon's "embedded constraint" theory violates the FCC's rules for pricing UNEs. First, Verizon's witness Collins' premise is false because line sharing arrangements need not be limited to all-copper loops. The very schedule for this docket demonstrates that the Commission intends to develop prices for an arrangement to provide access to the HFPL over a forward-looking, fiber-fed network architecture. According to Rhythms, Verizon is actively engaged in upgrading the DLC equipment in its local exchange affiliates' networks to facilitate the provisioning of DSL-based services over fiber-fed loops. Also, Verizon has publicly announced an agreement with Alcatel to purchase an estimated $800 million of ADSL electronics. Collins fails to suggest what constraint requires Verizon to continue to provide access to the HFPL over all-copper loops.

According to Rhythms, Collins' proposal violates the requirements of the FCC and this Commission that UNE prices be based on forward-looking economic cost. Collins incorrectly suggests that the price for access to the HFPL UNE should be based on the difference between Verizon's estimate of the forward-looking cost of the loop and an entirely different cost standard. In other words, the cost assigned to access to the HFPL would be, by definition, an amount above and beyond the forward-looking cost of the loop or any measure of forward-looking economic cost. This is a clear violation of the current pricing standard for UNEs.

TURN points out that the Washington Utilities and Transportation Commission (WUTC) order²⁶ Verizon cites in support of adopting a monthly recurring charge for the HFPL includes sections that Verizon did not quote. Specifically, the WUTC also includes its rationale for supporting its determination that the loop is a shared cost of both voice exchanges services and DSL service in a line-sharing environment. According to TURN, the WUTC decision affirms the obvious: in a DSL context, the loop is a shared input to both voice and data services, and it is a shared cost of both services.

5.1.5 Discussion

We begin our discussion of the proper price for the HFPL by stating that we are making a policy determination, not analyzing TELRIC cost studies or other scientific data to determine the HFPL price. We point to the assigned ALJ's Ruling of July 19, 2001, which denied Rhythms' motion to strike testimony filed by other parties. Rhythms asserted that TURN's witness Roycroft's testimony exceeded the defined scope of the HFPL pricing phase. According to Rhythms, the discussion at the May 2, 2001 PHC made it clear that such testimony was not to include cost studies reexamining the underlying loop rates, but was to be limited only to a discussion of whether any portion of the already existing UNE loop rate should be allocated to CLECs' use of the HFPL for DSL service.

In ruling on Rhythms' motion to strike, the assigned ALJ concluded as follows:

Rhythms has taken much too narrow a view of the scope of the HFPL proceeding. While other parties agree that this is a policy issue, they are not precluded from submitting cost data which serves as the basis for their policy positions.

We reiterate that we are making a policy decision on the proper charge for the HFPL, based on the record evidence presented in this proceeding. We do not purport to base our adopted rate on a detailed cost study for the HFPL.

TURN uses the FCC's HCPM model to develop loop rates for both Pacific and Verizon. As Pacific points out, the Commission has already adopted an $11.70 loop rate for Pacific.²⁷ Also, Verizon claims that the way TURN extracted loop cost estimates from the HCPM was biased. Verizon then recalculates those loop costs, using its own ICM model. As Verizon states, neither the HCPM or TURN's workpapers are included in the record of this proceeding. We note that the same holds true for Verizon's ICM model and workpapers. We are left with two models before us, with different conclusions, and no way for us to validate either model. Therefore, we will not rely on either the HCPM or Verizon's ICM in making our determination of the proper price for the HFPL. In addition, TURN acknowledges that the HCPM was developed by the FCC to develop costs for determining universal service support, not for developing costs for UNEs. Pacific points out that the FCC stated in the Kansas/Oklahoma 271 Order that the HCPM model "should not be relied upon to set rates for UNEs."²⁸ As Pacific states, contrary to the FCC's explicit direction, TURN now asks the Commission to do just that. We have adopted an interim loop rate of $9.93 for Pacific, and the Commission-adopted rate will form the basis for our determination of the proper price for the HFPL.

Pacific proposes a rate of $5.85 for the HFPL, and suggests that that amount will assist Pacific in making up some of the shortfall associated with providing residential basic exchange service.²⁹ In this case we are pricing the HFPL as a UNE and must follow the FCC's rules for pricing UNEs. CFR Rule 51.505(d) lists the factors that may not be considered in calculation of the forward-looking economic cost of a network element. Subsection (4) reads as follows:

Revenues to subsidize other services. Revenues to subsidize other services include revenues associated with elements or telecommunications service offerings other than the element for which a rate is being established.

We find that Pacific's proposal to collect $5.85 or 50% of our adopted loop rate violates Rule 51.505(d)(4), which bars states from setting UNE rates which include revenues to subsidize other services, which is precisely what Pacific is proposing. While the Commission in the past has relied on above-cost services to subsidize below-cost services, that sort of cross-subsidization is not appropriate in the pricing of UNEs.

Nor do we agree with Pacific's argument that the CLECs are lucky to pay only $5.85, because if they have to purchase the entire loop, they would pay $11.70. This argument is spurious because CLECs who utilize the HFPL are competing with Pacific's separate affiliate ASI, which supplies service over line-shared loops. It is not economically feasible for a competitor to pay $11.70, and then attempt to compete against ASI with its lower loop cost.

Next we examine Verizon's proposal that the current $3.00 HFPL rate which was adopted in the Interim phase be continued until final pricing. Verizon has the mistaken impression that this phase of the PLS proceeding is scheduled to address only the policy issue of whether there should be a positive price for the HFPL, not what that price should be. Verizon is mistaken. This phase of the PLS proceeding is scheduled to set a permanent price for the HFPL, to replace the interim rates adopted in the Interim Line Sharing phase in D.00-09-074. At the PHC on May 2, 2001 in the PLS proceeding, Rhythms counsel indicated that this phase is to determine "on a permanent basis what the monthly loop recurring price should be, if any, for the HFPL."³⁰ The assigned ALJ cited this section from the transcript in her July 19, 2001 Ruling denying Rhythms' motion to strike certain testimony filed by other parties. This phase of the PLS proceeding will set the permanent HFPL rate; that issue is not scheduled to be addressed further in the costing phase of this proceeding.

Verizon indicates that in the costing phase, it intends to propose a rate of $7.32 based on what it terms its "embedded constraint" theory. Since this represents Verizon's proposal for a permanent HFPL rate, we will examine Verizon's proposal here. As Rhythms and ORA point out, Verizon's proposal to recover the costs of retaining its home-run copper network to provide HFPL to customers since it cannot efficiently introduce fiber into a loop or convert that customer to a hybrid fiber/copper loop, violates the FCC's rules on factors that may not be considered in pricing unbundled network elements. CFR Rule 51.505(d) (1) reads as follows:

Embedded costs. Embedded costs are the costs that the incumbent LEC incurred in the past and that are recovered in the incumbent LEC's books of accounts.

Verizon's investment in its copper network clearly fits in this category of embedded costs that cannot be considered in setting a price for UNEs. Verizon's embedded constraint theory violates CFR 51.505(d)(1). We reject Verizon's proposal for setting a permanent HFPL rate of $7.32.

In the 1999 Pricing phase of our Open Access and Network Architecture Development (OANAD) proceeding, we adopted a loop rate of $11.70 for Pacific. However, shortly after the release of the DD in this proceeding, the Commission set new interim UNE loop and switching rates for Pacific in the UNE Reexamination proceeding. In Decision 02-05-042, we approved an interim rate of $9.93, which we will use as the basis for determining the permanent price for the HFPL for Pacific. However, we will set a procedure in place so that if Pacific's loop rate changes as a result of setting permanent UNE rates in the UNE Reexamination proceeding, or any other proceeding, that new loop price will then be used to readjust the HFPL rate as well, without further proceedings before this Commission.

We need to determine how to allocate costs between two UNEs that both utilize the loop-voice service which uses the low-frequency portion of the loop and the HFPL. Parties have made a number of comments about the allocation of costs between the high and low frequency portions of the loop. At the heart of this issue is the question of whether the loop is a shared cost. In the Commission decisions cited by Rhythms and Pacific,³¹ the Commission looked at this issue in a different context. In those proceedings, we were looking at the issue of whether basic exchange service or the loop is a shared cost, such that the costs of providing it should be recovered through the various services that use it. The issue centered on whether some of the costs of the loop should be recovered in the prices of services that use the loop. That issue is not on the table here - no price changes are being proposed for any existing retail service. Instead, the Commission is setting a price for a new UNE-the HFPL. In those earlier decisions we decided that it is not appropriate to treat the loop as a shared cost.

What we are dealing with here is clearly distinguishable from the issues in these earlier decisions. For one thing, we are not being asked to include toll or vertical services costs within the price of the loop, which in any event, would violate the FCC's rules for pricing UNEs. The Commission has already spoken on that issue, and we will not revisit the issue in this proceeding. However, this case is different because we are dealing with setting a rate for a new UNE-the HFPL-which was created from an existing UNE --the loop--which includes both high and low frequencies. Parties do not dispute that the loop is a shared physical resource. In other words, we have voice and data service that utilize different portions of the loop, and we need to allocate costs between them

The FCC recognized the need to make some sort of allocation in its Line Sharing Order:

We note that the TELRIC methodology that the Commission adopted in the Local Competition First Report and Order does not directly address this issue. More specifically, the Commission in that order noted that the TELRIC methodology was designed to price `discrete network elements or facilities,' rather than services. In the case of line sharing, however, the facility in question is, by definition, also used for two incumbent LEC services (local exchange service and interstate access service). We are thus presented with the question of how to establish the forward looking economic cost of unbundled bandwidth on a transmission facility when the full embedded cost of that facility is already being recovered through charges for jurisdictional services. Accordingly, we must extend the TELRIC methodology to this situation and adopt a reasonable method for dividing the shared loop costs.³²

The FCC clearly points to the need to come up with a reasonable method of allocating the shared loop costs. As we stated in our decision in the Interim Arbitration phase of this proceeding, the FCC acknowledges that the FCC-adopted TELRIC methodology does not directly address the issue of pricing a line-shared loop.³³ In other words, we need to allocate prices to voice service and the HFPL, using our interim adopted $9.93 loop price as the ceiling for the adopted rates, since that amount has been determined to recover all costs-including shared and common costs-associated with the loop. Pacific asks that we allocate 50% of the price of the loop to the HFPL, but we have already rejected Pacific's proposal since Pacific proposes to use that revenue to make up some of its alleged shortfall in revenues from residential basic exchange service, which violates the FCC's rules for the pricing of UNEs. Also, Pacific's proposal does not divide the shared loop costs between the high and low frequency portions of the loop, as the FCC directs in ¶ 138 above. Instead, it proposes to add a price for the HFPL, in addition to the price for the loop as a whole.

TURN proposes an allocation of 25% based on the Shapley Value, based on the theory that four major services utilize the loop: basic exchange service, toll/access, vertical services, and the HFPL. Rhythms criticizes TURN's allocation saying that there are many other services that utilize the loop, including 911, 800 and 976 services, directory assistance and operator services. While Rhythms is correct that other services utilize the loop, we find that TURN has identified the four major users of the loop, and indeed 800 service is a subset of toll service. Therefore, we will adopt TURN's proposal that 25% of total loop costs be allocated to the HFPL. This allocation yields a rate of $2.48 for Pacific.

In its Reply Comments, Pacific asserts that the DD correctly held that the loop is a shared facility, the cost of which should be divided among those services using the loop. However, Pacific disputes the DD's conclusion that there are four major services that utilize the loop, rather than two.³⁴ Pacific argues that it is not appropriate to apportion the shared cost of the loop across four main types of services when not all customers use all four types of service. Verizon asserts that the result of a Shapley allocation is an "arbitrary pro rata share" of costs (Verizon at 6). TURN responds to theses charges stating that it is indisputable that voice services use the loop and that the Shapley Value approach identifies groups of services that make market and regulatory sense. The result of applying the Shapley Value is to make the users of the shared facility better off than if they were not sharing. TURN describes the shared cost problem as "thorny" but reiterates that the Commission must set a price for the HFPL UNE. According to TURN, the DD adopts the Shapley Values economic tool to develop the monthly price because it provides the Commission with an unbiased, impartial, straightforward analytical means of addressing the problem of how to determine a just and reasonable monthly recurring price for the HFPL UNE.

We concur with TURN's reasoning. The allocation using the Shapley Value is not "arbitrary" as Verizon suggests. It is based on analysis of the major uses of the loop, and therefore provides us with a way to allocate costs between the HFPL and other uses of the loop. Again we reiterate that this is a policy determination since, as the FCC acknowledges, we cannot use standard TELRIC methodology to price the HFPL.

Pacific and Verizon assert that the DD overturns D.96-08-021 without factual or evidentiary support. According to Pacific neither TURN nor any other party has presented any testimony in this proceeding supporting the position that the Commission erred in its prior determination that it is improper to allocate loop costs to "toll and other services that use the loop." TURN responds saying that the plain language of the Pub. Util. Code § 1708 gives the Commission ample authority to adopt the conclusions set forth in the DD. The Commission has the authority to alter a prior precedent if warranted by the record in the case. Indeed, the cases relied on by Pacific, Verizon and the Joint CLECs were themselves a change of prior precedent. TURN points out that prior to D.94-09-065 the Commission had determined that loop costs were shared costs, caused by all of the services that use the loop, and rejected the argument that loop costs are a direct cost of local service as "nonsensical." (D.84-06-113 at 455). In D.94-09-065, the Commission changed its mind.

TURN and ORA conclude that they do not believe that the Commission would be altering precedent if it adopts the DD, but they point out that the Commission has the authority to do so, based on consideration of the facts before it. TURN and ORA assert that this is particularly important given the provision of advanced services. Indeed, it is speculated that the HFPL will carry the vast majority of voice and data service in the not too distant future. The Commission needs, and has, the authority to reach decisions that reflect changed circumstances. We concur with TURN/ORA's argument. We have the authority under Pub. Util. Code § 1708 to change a prior Commission order, once parties are given notice and an opportunity to be heard. The telecommunications sector has changed since the passage of the Telecommunications Act in 1996, and we must be able to set policies that reflect those changes.

We need to set up a process to address future changes in circumstances so that minor changes do not have to be brought back to the Commission for review. At any time that the Commission's adopted loop rate for an ILEC changes, the rate for the HFPL will also recalculated, based on 25% of that adopted loop price. Telecommunications technology is changing at a rapid pace, and we want to take into account the fact that another major revenue stream could be developed that uses the loop. This new future service could cause us to change our policy of allocating 25% of the loop price to the HFPL. Any party to this proceeding may file a motion in this docket to open the proceeding to re-examine the allocation issue as a result of changes in technology. Such motion should include specific information on new and significant uses of the local loop that warrant changing the allocation factor.

Verizon presents a special challenge, since the Commission has not yet adopted a loop price for Verizon. Both ORA and Verizon propose continuation of the current $3.00 rate, although for different reasons. Verizon made that proposal as an interim rate, which would be adjusted in the final costing phase of this proceeding. However, as we stated above, this is the proceeding to set a final permanent rate for the HFPL for both Pacific and Verizon, and we do not intend to revisit this issue in the costing phase.

Since we currently have no adopted loop rate for Verizon, and we have rejected use of rates obtained from using the FCC's HCPM, we cannot at this time employ the Shapley Value to determine an appropriate HFPL rate for Verizon. However, once we adopt a UNE loop rate for Verizon, the rate for the HFPL portion will be set at 25% of that adopted loop rate. In the interim, we will continue the $3.00 HFPL rate adopted in the Interim Line Sharing phase of this proceeding.

Rhythms raises the issue of geographic deaveraging of loop prices. We recently adopted interim geographically deaveraged loop rates for Pacific in D.02-05-042. Since the HFPL rate we have adopted is set as a percentage of the adopted loop costs, that rate would vary by geographic zone. The HFPL rate shall be set at 25% of the adopted loop rate for each geographic zone. Once we adopt geographically deaveraged loop rates for Verizon, the HFPL prices in each zone will also be set using the 25% allocation we have adopted here.