III. Interim Rates are Warranted

This decision affirms the Assigned Commissioner's and ALJ's September 28 joint ruling regarding the need for interim relief. We affirm the ruling's conclusion that the Commission has the authority to set interim rates and has done so on numerous occasions. Despite Pacific's argument to the contrary, interim rates need not be premised on an "emergency" alone, but can be adopted for other reasons, including procedural delays. The California Supreme Court addressed precisely this issue in TURN v. CPUC (44 Cal. 3d 870, 878 (1988)). In the underlying decision, the Commission granted an interim rate increase while expressly declining to make any finding that the "the interim rate increase was required by a financial emergency, or that the reasonableness of the pertinent costs was undisputed." (Id. at 875.) The Commission's decision was upheld by the Supreme Court, which found that the overriding circumstance was the prospect of many months and years of hearings and deliberations before a final rate could be determined. (Id. at 879.)¹⁰ The court affirmed that the Commission could set interim rates as long as the rate is subject to refund and sufficient justification for the interim relief has been presented. (Id. at 880.)¹¹

Our action today to set an interim rate for the loop and switching UNEs is not precluded by D.99-11-050. In that 1999 order, we stated that the rates adopted therein would remain in effect until changed by an order in the annual UNE reexamination. The 1999 order does not limit our ability to consider and establish interim UNE rates in this UNE Reexamination proceeding.

Interim rates are necessary due to delays in this proceeding caused by the need to examine competing cost models. The Assigned Commissioner and ALJ initially limited the scope of this proceeding to an examination of Pacific's updated cost studies. Despite repeated requests from Joint Applicants to allow them to submit their own cost studies, the Assigned Commissioner and ALJ rejected the notion of competing cost studies unless Pacific's filing failed to meet three criteria. Following an August 9 workshop to understand Pacific's "starting point" for its upcoming cost filings, Joint Applicants and other parties alleged that Pacific was not actually filing updated versions of the earlier cost models. These parties claimed that Pacific was merely filing adjustments to the outputs of the models used to set costs and prices in the prior OANAD proceeding. (See Joint Applicants Workshop Comments, 8/14/01; TURN Workshop Comments, 8/14/01.)

After review of these claims of insufficiencies in Pacific's filing, the ALJ and Assigned Commissioner determined that Pacific's filing did not meet the three criteria they had set forth. We agree that Pacific's "starting point" filed on July 26 and its updated cost studies filed on August 15 do not meet the criteria set forth in the July 11 ruling. Specifically, Pacific uses endpoints from OANAD and adjusts them rather than actually providing the former model with new inputs. Pacific's subsequent filing of a "linked version" does not correct this problem because it still does not provide the original model on which the calculations are based.

Pacific's filing fails the first and second criteria set forth in this proceeding because parties and staff cannot understand and replicate the calculations and the inputs of the prior OANAD models without the ability to run these models. Pacific itself is not replicating its prior OANAD models since it is not performing new runs of the SCIS model for switching investment, the Cost Proxy Model for loop investment, or other mainframe models used to calculate expenses and support investments. In other words, Pacific did not input changes to the prior OANAD model. Instead, as the Joint Applicants and other parties claimed, Pacific merely calculated the effect of estimated changes by adjusting the outputs of the prior OANAD model. While Pacific's "linked version" allows parties to trace through Pacific's calculations, it is not a model that constructs a forward-looking network. Finally, Pacific's filing fails the third criteria because parties cannot input their own numbers to Pacific's models and re-run them. Thus, it is impossible for parties to modify assumptions from the prior OANAD models.

Without the ability to modify assumptions and re-run the models, it is unclear how the evidence and assumptions that formed the basis for Joint Applicant's initial showing to open this proceeding can actually be tested, modified, and examined.

In their September 28 ruling, Commissioner Wood and ALJ Duda stated:

We are concerned that if we were to proceed only with the filing presented by Pacific, any resulting UNE prices might not be cost-based as required by Section 252 of the Telecommunications Act of 1996. We will have less confidence in the results of our efforts without the ability to run an actual model and test inputs and assumptions. It is not clear if Pacific can amend its filing to overcome the problems identified. Because Pacific's filing does not currently meet our criteria, we are faced with the option of allowing Joint Applicants and other parties to file competing cost models.

...

Because of the substantial delay in the case that would be caused by either allowing Pacific to amend its filing or by considering competing filings, we are persuaded to grant some form of interim relief. (9/28 Ruling, p. 5.)

The September 28 ruling noted that Joint Applicants had provided an adequate initial showing in their initial April 20 filing in this case to support a reasonable presumption that costs for unbundled loops and unbundled switching had declined from previously adopted costs. Yet, Pacific's August 15 cost update filing does not allow the Commission staff, the Joint Applicants, or other parties to test this initial showing. For example, Joint Applicants provide ARMIS¹² data indicating that Pacific's switch investments have declined 40% on a per minute of use basis from 1994 to 1999 due to increases in minutes of use and insignificant increases in switching investments. (Pitts Declaration for Joint Applicants, 2/21/01, para. 12.) Further, they provide data indicating that the price for adding "growth lines" has declined significantly since 1996. (Id., para. 13.) Joint Applicants also indicate that based on service volume and cost data that Pacific reported to the FCC, Pacific's switching-related expenses and support investments have declined 23% (on an expense per line basis) and 32% (on an expense per minute basis) since 1994. (Murray Declaration for Joint Applicants, 2/20/01, para. 8-10.)

Regarding costs for unbundled loops, Joint Applicants assert lower capital costs due to Pacific's "Project Pronto," a large-scale upgrade of its fiber and digital loop carrier (DLC) facilities. (Murray/Donovan Declaration for Joint Applicants, 2/28/01, para. 24.) They also assert economies of density from a 48.5% increase in total access lines from 1994 to 1999. (Id., para. 18.) Joint Applicants' figures for access line growth are based on ARMIS data that Pacific reports to the FCC. In addition, Joint Applicants claim that certain DLC equipment costs have dropped to as low as 25% of the initial price. (Id., para. 30.) Pacific has not disputed a decline in DLC equipment costs,¹³ and it does not dispute the ARMIS data cited by Joint Applicants on volume and line growth. Furthermore, Pacific admits that it benefits from SBC-wide purchasing of switches and can obtain switches in California for a lower price than in Illinois.¹⁴ Joint Applicants point out that SBC has proposed UNE rates for switching in Illinois that are drastically lower than the current switching rates in California.

Considering that many of Joint Applicants' assertions begin with Pacific's publicly reported data, it is reasonably plausible that at least some of these factors will lead to decreases in UNE rates for loops and switching. Nevertheless, Pacific's filings have left the Commission and parties without the ability to test or examine the effect of these documented and undisputed changes involving line growth, corporate mergers, switching investments and DLC technology. Pacific generally states that many of the cost declines shown in the public data from 1994 to 1999 were actually considered and included in the forward-looking models that developed the costs the Commission adopted in 1998. Unfortunately, the Commission has no way to verify this claim without the ability to replicate the costs adopted in 1998 using a model provided by Pacific.

Essentially, Pacific has presented us with a "black box" that does not allow us to test the summary of evidence that initially persuaded the Commission to open the case. The Commission must either trust Pacific's "black box" without further scrutiny, or delay the case while the Commission investigates other models or a revised model from Pacific. Neither of these options is acceptable. This case was initiated based on a summary of evidence of cost declines. Delays in this case could lead to prolonging current rates at non-cost-based levels. Under Section 252 of the Telecommunications Act of 1996, this Commission is required to set UNE rates based on cost. (47 U.S.C. § 252 (d)(1).) We cannot in good conscience succumb to the delays caused by the inadequacies of Pacific's filing in the face of this preliminary evidence that costs have declined.

Our decision to set interim rates is in part supported by a recent order of the D.C. Circuit in Sprint Communications Company v. FCC.¹⁵ In the Sprint decision, the D.C. Circuit was asked to review the FCC's decision to grant in-region long distance authority to SBC for Kansas and Oklahoma. Appellants asked the D.C. Circuit to overturn the FCC's findings that UNE rates for Kansas were cost-based, claiming that the FCC could not properly find the rates in these states TELRIC compliant because "they are the product of a crude `settlement' method, trimmed by an arbitrary 25% `haircut.' " (Id., at *22.) In its decision on the appeal, the court declined to overturn the FCC's finding that Kansas UNE rates were cost-based and specifically noted that it could not find fault with the FCC "for approving the Kansas Commission's compromise resolution of an issue that the parties' behavior had left a muddle." (Id., at *25.) The court also discusses the difficulty in pinpointing TELRIC rates with exactitude and cites to a prior case where it stated:

This argument, however, assumes that ratemaking is an exact science and that there is only one level at which a wholesale rate can be said to be just and reasonable.... However, there is no single cost-recovery rate, but a [wide] zone of reasonableness.... (Id., at *12-13, citing Conway, 426 U.S. at 278.)

The reasoning of the D.C. Circuit case supports our interim resolution of this proceeding which the deficient cost filing of Pacific has, in some ways, "left a muddle" for the Commission to unravel. Furthermore, given the acknowledgement by the D.C. Circuit's order that TELRIC ratemaking is not an exact science and involves a "zone of reasonableness," we find support in the order for this Commission's discretion to adopt interim UNE rates.

By setting interim UNE rates for unbundled loops and switching, the Commission is not violating Pacific's due process rights. Pacific was given ample opportunity to comment on the proposed interim rates through an additional round of comments that were solicited by the Assigned Commissioner and ALJ. The rates will be subject to adjustment once final rates are determined, either up for down. Thus, Pacific is not harmed by the interim rate levels.

In summary, we find that that interim relief is warranted based on the substantial delays looming in this case caused by the inadequacies of Pacific's cost filing. Interim relief is also warranted based on the summary of evidence initially provided by Joint Applicants indicating a reasonable presumption of cost declines for unbundled loops and unbundled switching.¹⁶

Before we turn to the substance of the relief requested, we must address Pacific's criticisms of the HAI model that underlies the interim relief request. Joint Applicants' proposed interim relief is primarily based on analysis and documentation involving the HAI model and the FCC's Synthesis model. We recognize that the FCC and other states have criticized aspects of prior versions of the HAI model, particularly HAI's assumption of uniform customer dispersion. (Tardiff Declaration for Pacific, 9/4/01, p. 3.) Pacific also criticizes the total investment and expense levels produced by HAI as too low when compared with actual figures. (Id., p. 2.) In addition, Pacific claims that HAI does not meet the three criteria for cost models and studies set forth in this proceeding.

Joint Applicants defend HAI, stating that the current version 5.2a is improved over all earlier ones. For example, Joint Applicants maintain that HAI's use of geocoded customer location data addresses the customer dispersion problem and is mirrored by other models currently in use, including the FCC's Synthesis Model. (Klick Testimony for Joint Applicants, 8/20/01, p. 21.) Joint Applicants also contend that they are not basing the requested interim relief on the absolute output of HAI, but on a trend analysis of its outputs from 1994 to 2000. Joint Applicants reason that any systematic bias in HAI's calculation of the absolute level of investments and expenses does not impact the discounts determined through the trend analysis. (Bryant Declaration for Joint Applicants, 9/7/01, p. 2.)

We agree with Joint Applicants that because HAI is used for a trend analysis in loop costs over the 1994 to 2000 time period, any criticism of its actual outputs are of lesser significance. While it is true that this Commission and the FCC have rejected prior versions of HAI, and the HAI model has its recognized shortcomings, it is the only actual "model" that has been filed thus far in the record of this proceeding to update UNE prices for loops. We disagree with Pacific's claims that HAI does not meet the criteria for cost models and studies. HAI meets two of our three criteria because we have been able to understand how HAI derived its results for unbundled loops and we have changed numerous model inputs and assumptions to produce our own results. While HAI does not exactly replicate the costs adopted in prior OANAD decisions, we find this is not necessary for purposes of a trend analysis because we are not using the absolute outputs of HAI to set rates. In addition, HAI does allow staff to replicate Joint Applicant's model runs.

Therefore, we will use the trend analysis based on the HAI model to set interim prices for loops, even if the model has elements that we disagree with, rather than relying on Pacific's cost filing, because we cannot adequately test and model all inputs with Pacific's filing. In other words, we will base the interim relief on the analysis presented using the HAI Model, but this does not prejudge the methodology or cost model we will use to set UNE rates in a later phase of this proceeding. We are not endorsing use of the HAI or the Synthesis models to set final updated UNE rates for unbundled loops or unbundled switching.

¹¹ The adoption of interim rates is not limited to energy matters. (See 80 CPUC 462, 465 (1976) D.86352, wherein the Commission approved "interim provisional rates" at the request of Pacific for its "Dimension PBX" service as a result of delays in the proceeding to establish permanent prices for the service.)

¹⁶ On October 9, Pacific filed an Appeal to the Full Commission of the September 28 ruling. We decline to entertain this interlocutory appeal and it is herein denied. On October 19, Pacific filed a motion to vacate the September 28 ruling on the grounds that a pending motion in R.93-04-003/I.93-04-002/R.95-04-043/I.95-04-044 ("Section 271 Proceeding") involving a proposed discount to unbundled switching prices moots the need for interim relief. The motion in the Section 271 proceeding proposes an approximately 40-44% discount to UNE switching rates, depending on usage assumptions, on the condition that the Commission approves Pacific's Section 271 application. We will deny Pacific's motion to vacate the September 28 ruling because we are not persuaded to amend the schedule of the UNE Reexamination based on a conditional proposal that is currently pending in another docket.