We now turn to adopting a markup for shared and common costs41 to apply to the nonrecurring costs we adopted for Verizon in D.98-12-079 and to apply to the interim UNE rates we adopt today.
Verizon states that the Commission adopted a 22% markup when it established Verizon's current UNE rates in D.97-01-022. (D.97-01-022, mimeo., p. 7.) Verizon proposes that to simplify this interim proceeding, the Commission should use this same 22% markup, both with regard to interim loop and switching rates and non-recurring costs.
In contrast, Joint Commenters contend that a forward-looking shared and common cost markup should be substantially lower than the 22% embedded in Verizon's rates, because the 22% markup emerged from GTEC's 1996 TSLRIC cost study. They contend that a 22% TSLRIC markup bears no relation to a TELRIC-based shared and common cost markup. Joint Commenters explain that when the Commission converted Pacific's TSLRIC studies into TELRIC results, estimates of Pacific's shared and common costs were cut approximately 50% because a TELRIC study involves far fewer "shared" costs and eliminates retail only costs. (Joint Commenters, 8/20/02, p. 14.) They further note that when GTEC merged with Bell Atlantic to create Verizon, four years after GTEC's 1996 cost studies were filed in California, GTEC itself represented that the merger would lead to more than $2 billion in expense savings, including significant overhead cost savings. (Murray Declaration, 7/30/02, para. 29.)
Therefore, instead of the 22% markup proposed by Verizon, Joint Commenters recommend using a markup of 8% based on their own analysis of average RBOC overhead costs. When the Commission sets Verizon's interim rates, Joint Commenters recommend the Commission adjust the shared and common cost markup embedded in Pacific's rates from 19% to 8%, or the markup embedded in the Verizon New Jersey rates from 10% to 8%. Likewise, they recommend adding 8% to Verizon's nonrecurring costs adopted in D.98-12-079 in order to set nonrecurring prices for Verizon.
Joint Commenters derive their 8% markup proposal by aggregating ARMIS data reported to the FCC related to expenses and revenues for all RBOCs nationwide. Using this data, they calculate the ratio of total regulated corporate operations expenses to total costs (other than overheads) for all RBOCs in 2001. (Murray Declaration, 7/30/02, paras. 33-36.) Joint Commenters contend that the resulting 8% can be used as a conservative estimate of Verizon's shared and common costs because Verizon, as the nation's single largest RBOC, is well-positioned to achieve a lower level of overhead costs than the average RBOC.
In response, Verizon opposes Joint Commenters' request to lower the markup to 8%. First, Verizon maintains the calculation is flawed because it fails to include all shared and common costs, such as network support expenses and engineering expenses. Second, Verizon states the Commission has already rejected a similar proposal to base the markup calculation for Pacific on ARMIS data. (D.99-11-050, mimeo., pps. 69-70.) Third, Verizon contends that when the Commission adopted GTEC's 22% markup, it made substantial adjustments to reduce GTEC's proposal from 46% to the 22% level. Finally, Verizon notes that the Commission recently rejected the identical suggestion by Joint Commenters to adjust Pacific's markup to 8% using ARMIS data and instead increased Pacific's markup to 21%. (D.02-09-049, mimeo., pps. 12 and 17.)
Z-Tel supports use of the 8% markup factor recommended by Joint Commenters.
We are not persuaded by Joint Commenters that we should abandon the 22% markup that we adopted for Verizon in D.97-01-022. We acknowledge that the 22% markup was set in the context of a TSLRIC proceeding and that once we conduct a TELRIC analysis for Verizon, the markup percentage may be reduced. However, we are not persuaded that the markup will be reduced by the large percentage suggested by Joint Commenters, particularly when Verizon's markup was already reduced from a proposed level of 46% to its current 22%. Moreover, we should not abandon the calculation of the markup that we made using Verizon specific information in D.97-01-022 for an overly simplistic methodology that is based on aggregate ARMIS data from all RBOCs. To arrive at their 8% markup, Joint Commenters' aggregate revenue and expense data for all RBOCs and fail to even consider the individual variations between RBOCs in overhead percentages. This is an overly simplistic methodology using nationwide, aggregated RBOC data rather than company-specific data, and for those reasons, we do not find it reasonable.
Although we deny Joint Commenters' 8% proposal, we could, of course, opt to use the 10% overhead rate embedded in the Verizon New Jersey rates. We decline to use this approach because we are not comfortable simply lifting an overhead rate from New Jersey and applying it to California. While we use New Jersey information to compare costs for loops and switching, the information is adjusted using the Synthesis Model for network differences between California and New Jersey. We do not think the same approach is valid for comparing overhead costs between differing states. To the best of our knowledge, the Synthesis Model was not intended for overhead comparisons. We have no basis to support applying the Verizon New Jersey markup of 10% to Verizon California's interim UNE rates and nonrecurring costs.
The only other approach offered was Verizon's proposal to retain the 22% markup adopted in D.97-01-022. We find it more reasonable to maintain the 22% markup in the interim rates we adopt today because it is based on a Verizon-specific analysis, even if it is somewhat dated. Pacific's current markup of 21%, as recently adopted in D.02-09-049, provides a useful guidepost. We cannot ignore the huge disparity between an 8% markup based on a methodology that averages ARMIS data for all RBOCs, and the 21% that is currently in place for Pacific based on a company-specific analysis. We acknowledge that Pacific's markup was calculated based on cost studies that were filed in the 1997 time frame, as was Verizon's. That is certainly a reason to review the markup for both companies in the context of UNE pricing proceedings, but it is not sufficient justification to abandon Verizon's company-specific markup for one that is based on nationwide averages.
In comments on the Draft Decision, Joint Commenters contend that the Commission inappropriately relies on Pacific's 21% markup because that number is outdated given recent mergers. Further, they argue that the draft order errs in not addressing the flaws in Verizon's 22% markup raised by Joint Commenters. They request that the Commission revise the order to adopt the 10% markup embedded in Verizon New Jersey UNE rates. Verizon responds that there is no evidence on the record to support adoption of the 10% New Jersey markup. Further, Verizon reiterates that the Commission already reduced Verizon's proposed 46% markup to the 22% level when the markup was adopted in D.97-01-022 and that no further adjustments are warranted.
There are no substantive changes to the Draft Decision based on these comments. The order does not rely on Pacific's 21% rate to set the markup for Verizon, but merely notes this is a useful benchmark for comparison. The order acknowledges that while the markup level is an area of UNE ratesetting that will be explored further for Verizon, the Commission should take a more conservative approach in maintaining a markup derived from a Verizon-specific analysis at this time.