V. Comments on Draft Decision

The draft decision of the Administrative Law Judge (ALJ) in this matter was mailed to the parties in accordance with Section 311(g)(1) and Rule 77.7 of the Rules of Practice and Procedure. Comments were filed by AT&T/MCI and reply comments were filed by Verizon and XO California, Inc. (XO).

AT&T/MCI recommend the Commission deny Verizon's petition to modify interim rates for several reasons. First, they claim the draft decision is arbitrary and capricious because it only considers information regarding New Jersey rate changes, but ignores evidence of possible cost decreases, particularly to the shared and common cost markup. Specifically, AT&T/MCI note that the 22% shared and common cost markup included in the current interim prices is more than twice the 9.07% markup Verizon has proposed in the permanent phase of this proceeding. Furthermore, AT&T/MCI claim that 2004 Synthesis Model results indicate sharp cost declines in California that the Commission should consider when adjusting interim rates. Therefore, AT&T/MCI urge the Commission to consider this key evidence if it grants Verizon's petition.

Second, they claim the draft decision departs from Commission precedent regarding the cost of capital and the proper basis for interim prices without adequate explanation. For example, the Commission explicitly rejected the idea of a risk adder to the cost of capital in the SBC California UNE proceeding,⁴ but modifies Verizon's rates in the draft decision based on the newest Verizon New Jersey cost of capital which now includes a risk adder. In sum, AT&T/MCI request a denial of Verizon's petition to modify, or in the alternative, consideration of evidence of cost decreases if any adjustments are made to interim UNE rates.

XO agrees with AT&T/MCI that the draft decision legally errs in ignoring information of cost decreases, particularly Verizon's newest 9% markup proposal in the permanent phase of this proceeding.⁵

In response to AT&T/MCI's comments, Verizon maintains it is appropriate to adjust the interim UNE rates to reflect New Jersey rate changes, but inappropriate to consider other evidence of alleged cost declines, as AT&T/MCI suggest. Regarding the shared and common cost markup, Verizon claims it is improper to compare the 22% used in the interim rates with the 9% Verizon proposes in its latest study because these markup percentages were calculated with different methodologies. According to Verizon, shared and common costs that were once combined in the 22% markup are now broken out into different factors and loadings. Thus, Verizon does not concede that shared and common costs in its latest cost filings are limited to 9%. Verizon also maintains that AT&T/MCI's reference to 2004 Synthesis Model results is untimely, improper, and inappropriately diverges from the analysis originally used by the Commission in D.03-03-033. Furthermore, Verizon defends the draft decision as consistent with Commission precedent because D.03-03-033 adopted the New Jersey UNE rates as a starting point without independently reviewing the individual inputs, such as the cost of capital, that comprise the New Jersey rates.

The comments of AT&T/MCI are generally unconvincing, except with regard to the shared and common cost markup. On all other points, we agree with Verizon that rate changes to reflect adjustments to Verizon New Jersey rates are appropriate, and Commission precedent does not prevent this adjustment. It would be inappropriate to review individual inputs to the New Jersey rates, such as cost of capital, when we did not take this approach in D.03-03-033. We also agree it is untimely for AT&T/MCI to suggest that we consider 2004 Synthesis Model results, since the parties did not raise this issue in response to Verizon's petition to modify.

With regard to the shared and common cost markup, however, we acknowledge that the draft decision did mistakenly overlook AT&T/MCI's original comments to consider adjusting the 22% shared and common cost markup on the basis that Verizon itself is now proposing a 9% markup. When interim rates were first adopted in March 2003, AT&T/MCI pressed for a lower markup at that time. The Commission rejected the request, citing the 22% markup as California-specific, unlikely to be reduced to a 10% level, and similar to the 21% markup adopted for SBC California. (D.03-03-033, pp. 46-47.) Now, Verizon proposes a 9% common cost markup for California and the Ninth Circuit Court of Appeals has remanded the 21% SBC California markup back to the Commission.⁶ The Verizon New Jersey rates include an overhead markup of 10%, substantially below the 22% markup incorporated into Verizon's interim rates in California. In comments on the draft decision, AT&T/MCI claim it is legal error for the Commission to ignore their claims that much has changed since the Commission decided to add a 22% markup to interim rates.

Essentially, we now question the validity of using a 22% markup originally adopted in 1997, using pre-1997 data and analyses, when a more recent TELRIC study for Verizon operations in New Jersey indicates a 10% overhead rate is adequate. Given the admitted changes in methodologies in the last decade, it makes more sense to assume that New Jersey UNE rates incorporate a forward-looking common cost markup factor that can be relied on, and then adjusted based on Synthesis Model results as we did in D.03-03-033. Verizon contends that the 22% markup embedded in interim rates was calculated with an entirely different methodology than the 9% it now proposes, and it is improper to assume that shared and common costs are only 9% because Verizon now breaks out shared and common costs into new and different loading factors. If, as Verizon contends, current methodologies shift costs once considered shared and common into direct UNE charges, then our use of recently derived New Jersey rates and their embedded overhead markup as a starting point should adequately reflect all costs, both direct, shared and common. If Verizon's new methodologies assign more shared costs directly to UNEs, then our approach of adding a 22% shared and common cost markup based on an outdated methodology has the potential to double-count shared costs. In D.03-03-033, we did not consider it likely that a 22% markup in California could drop to as low as 10%. But given Verizon's newest proposal and its explanation that new methodologies assign more costs directly to UNEs, this rationale is no longer well supported. Therefore, we now reverse our earlier position and find that based on the overhead factors adopted in New Jersey and proposed by Verizon in California, it is more reasonable to use the New Jersey UNE rates, including their embedded markup factor, rather than combining 2004 New Jersey UNE costs with a 1997 California markup factor.

This approach is further supported given the recent Ninth Circuit remand of SBC California's 21% markup factor. Initially, we hesitated modifying the 22% markup factor, noting its similarity to the 21% adopted for SBC California. With the remand and the current uncertainty in the SBC California markup, we can no longer rely on this comparison.

In summary, we will modify the methodology we originally used in D.03-03-033 when calculating interim rates. Rather than removing the New Jersey 10% common cost markup and inserting our own 22% markup, we will simply use the New Jersey UNE rates without modifying their markup factor, and then apply our Synthesis Model adjustments as we did in D.03-03-033.