*Pages 1--6 from Microsoft Word - 14662.doc* Federal Communications Commission DA 02- 286 FEDERAL COMMUNICATIONS COMMISSION Washington, D. C. 20554 February 6, 2002 Ms. Caryn D. Moir Vice President – Federal Regulatory SBC Telecommunications, Inc. 1401 I Street, NW, Suite 400 Washington, DC 20005 RE: SBC/ Ameritech Merger Order, CC Docket No. 98- 141, ASD File No. 99- 49 Dear Ms. Moir: This letter responds to SBC Communications, Inc. ’s (“ SBC”) January 4, 2001 letter regarding performance measurements payments under the SBC/ Ameritech Merger Order. 1 With this letter, I further explain the Common Carrier Bureau’s (“ Bureau”) views on the method for calculating payments under the Merger Order. In the SBC/ Ameritech Merger Order, the Commission adopted the Carrier- to- Carrier Performance Plan (“ Performance Plan”) that requires SBC to make payments to the United States Treasury should it fail to meet certain performance standards. 2 The Performance Plan prescribes the steps SBC must follow to calculate payments. 3 Before making its first payment, SBC orally asked the Bureau for direction on eleven payment issues arising from the Performance Plan. On December 11, 2000, the Bureau provided SBC a letter setting forth how the relevant payment provisions should be interpreted. 4 On January 4, 2001, SBC indicated it disagreed with the Bureau’s interpretation on four issues. 5 As explained below, based on my further review, I conclude that SBC’s position on three issues is reasonable and accordingly modify my prior guidance. On one issue, I decline to modify my prior interpretation, but note that the practical impact of the issue may be dwindling. 1 Letter from Sandra Wagner, Vice President, Federal Regulatory, SBC, to Carol E. Mattey, Deputy Bureau Chief, Common Carrier Bureau, FCC (Jan. 4, 2001) (“ SBC January 4th Letter”); Applications of Ameritech Corp., Transferor, and SBC Communications, Inc., Transferee, For Consent to Transfer Control of Corporations Holding Commission Licenses and Lines Pursuant to Sections 214 and 310( d) of the Communications Act and Parts 5, 22, 24, 25, 63, 90, 95, and 101 of the Commission’s Rules, CC Docket 98- 141, Memorandum Opinion and Order, 14 FCC Rcd 14712 (1999) (“ Merger Order”). 2 Merger Order at Appendix C, Attachment A. 3 See id. 4 See Letter from Carol Mattey, Deputy Chief, Common Carrier Bureau, to Sandra Wagner, Vice President, Federal Regulatory, SBC (Dec. 11, 2000) (“ Bureau Payment Calculation Letter”). 5 See SBC January 4th Letter. 1 Federal Communications Commission DA 02- 286 2 1. For measurements expressed as averages, should the “performance gap” – the extent to which SBC misses the performance standard – be capped at 100% if SBC misses the performance standard by a higher percent? 6 The Performance Plan does not, on its face, cap the difference between the level of service SBC provides to CLECs and the relevant performance standard (i. e., the “performance gap”). Accordingly, the Bureau instructed SBC to follow the formula spelled out in the Merger Order. 7 SBC disagrees with this guidance for two reasons. First, SBC contends that an uncapped performance gap will require the company to pay on more than the actual volume of activity. Second, SBC states that the Texas Commission subsequently imposed such a cap on the state’s performance plan, and therefore the Bureau should modify its interpretation of the relevant provisions in the merger conditions. 8 As explained below, I conclude there are public policy benefits in permitting SBC to make calculations under this provision in a uniform manner with the Texas plan, and therefore I authorize SBC to cap the performance gap at 100%. The Merger Order establishes a four- step method for calculating payments for this type of measurement (i. e., a measurement expressed as an average): 9 1) SBC calculates the “ideal value,” which is the minimum level of service SBC could provide CLECs without owing payments; 10 2) SBC calculates the percentage difference between the ideal value and the service it gave CLECs (i. e., the “performance gap”). 11 For example, if SBC provisions circuits to CLECs in nine days and calculates an ideal value of three days, the performance gap would be 200% (the difference of six days divided by the ideal value of three days); 12 6 See Issue Number 3 in the Bureau Payment Calculation Letter at 2- 3. 7 See Bureau Payment Calculation Letter at 2- 3. 8 SBC January 4 th Letter at 3. 9 In fact, there are three steps. Because the third step has two stages, I describe the process in four steps here. Stated simply, the formula is a function to the dollar value of the measurement multiplied by the number of data points multiplied by the average quality of SBC’s performance. 10 See Merger Order at Appendix C, Attachment A, A- 116, at “Step 1” for measurements expressed as averages or means. The “ideal value” is SBC’s term. SBC calculates the ideal value by translating the “critical- z” into the units being measured by the performance measurement (e. g., days, hours, and percentages). The ideal value is based partly on the service SBC provides its own retail customers (or a benchmark standard if SBC does not provide the service on a retail basis). See Merger Order at Appendix C, Attachment A, A- 116. For simplicity, I use a one- month example; in fact, SBC’s payments are based on chronic failures of either three consecutive months or six of twelve months in a calendar year. See Merger Order at Appendix C, Attachment A, ¶ 9. 11 See Merger Order at Appendix C, Attachment A, A- 116, “Step 2.” This step requires SBC to “[ c] alculate the percentage difference between the actual average and the [ideal value] . . .” 12 Stated differently, in this example it took SBC three times longer to provision CLEC circuits than its retail circuits (nine days versus three). 2 Federal Communications Commission DA 02- 286 3 3) SBC multiplies the performance gap by the number of data points. 13 Continuing with the example, SBC would multiply 200% by the number of times it provisioned circuits to CLECs, e. g., 150 provisioned circuits to yield 300; 14 then 4) SBC multiplies the product of Step 3 by a fixed- dollar amount based upon the measurement’s designation in the Performance Plan as “High,” “Medium,” or “Low.” 15 In the example, SBC would multiply 300 by the pre- set dollar amount, e. g., $900 for a “Medium” measurement. SBC’s final payment amount for this measurement would thus be $270,000. SBC first argues that the performance gap calculated in the second step should be limited to 100%. To do otherwise, SBC claims, would require the company to pay on more than the actual number of data points, i. e., applying a 200% performance gap to 150 data points would cause the company to pay on 300 data points. 16 Capping the performance gap at 100% would reduce the example payment to $135,000. 17 I find this argument unpersuasive. Failing the performance standard by a wide margin, which is often within SBC’s control, creates a large performance gap. A large performance gap does not mean SBC pays on more than the actual number of data points, as SBC argues. Rather, SBC would simply be paying for a larger disparity on the specified number of occurrences. 18 SBC also suggests that the Bureau should accept its position because the Texas Commission subsequently modified the Texas plan to cap the performance gap at 100%. 19 As SBC notes, part of the Performance Plan was modeled on the Texas plan. While the Commission was explicit that it was not bound to any future state change, 20 the fact that the Texas Commission chose to modify this aspect of the state performance plan warrants a consideration of whether there are public policy benefits in applying the calculation in the same fashion for the federal plan. The Commission is committed to the goal of working closely with the states in developing and applying national performance measurements as a general matter, and I believe that objective should guide our interpretation and application of the relevant merger conditions in this instance. I conclude that administrative efficiency would be served if SBC were permitted to apply 13 Although SBC’s January 4 th Letter uses the term “occurrences,” SBC stated orally to Bureau staff that it uses “occurrences” and “data points” synonymously. SBC and the Bureau thus agree that the “total number of data points” refers to the total volume of CLEC activity for the measurement, e. g., the number of circuits provisioned to CLECs. 14 In other words, 150 provisioned circuits times 200% to yield the number 300. 15 See Merger Order at Appendix C, Attachment A, A- 116 at “Step 3.” 16 SBC January 4 th Letter at 3. 17 In other words, 100% performance gap times 150 data points times the $900 pre- set dollar amount. 18 See Merger Order, 15 FCC Rcd at 14867, ¶ 377; see also id. at ¶ 378 (stating that SBC’s payments will vary according to the “level and significance of the discrimination detected”). 19 SBC January 4 th Letter at 3. 20 Merger Order at Appendix C, Attachment A, ¶ 4 (stating that the Bureau will decide if state changes should be made to the Performance Plan). 3 Federal Communications Commission DA 02- 286 4 this payment calculation in a fashion that mirrors the Texas performance plan. Accordingly, SBC may follow the 100% cap approach for measurements expressed as averages under the federal performance plan. 2. Should SBC report z- scores and calculate payments for performance measurements with 10 or fewer data points? 21 The Performance Plan does not, on its face, exclude any performance measurements from either reporting or payment based on volume. Accordingly, I stated in the Bureau Payment Calculation Letter that SBC should report and pay on measurements with 10 or fewer data points. 22 SBC disagrees. Specifically, SBC argues that the Performance Plan’s trebling of damages for volumes between 10 and 100 suggests by implication that volumes of 10 or fewer should be excluded. 23 SBC also has orally indicated its concern that it not be required to make payments for situations in which there are so few data points that a meaningful statistical conclusion cannot be made. Second, SBC suggests that the Texas Commission’s exclusion of such low volume measurements in its state plan should guide the Bureau’s decision. At the outset, I note that the business rules expressly describe in detail the types of data SBC should exclude. 24 Nowhere among these exclusions are low- volume measurements. The fact that the Performance Plan trebles damages for volumes between 10 and 100 does not mean that the Commission wished to exclude volumes of 9 or fewer, given that the Commission was clear to exclude other data. To do so would be inconsistent with the Performance Plan’s goal of completely capturing SBC’s performance (except for limited, explicitly stated circumstances) and, where necessary, establishing payment obligations. 25 Second, the Performance Plan already addresses low- volume situations. The Commission adopted a specific statistical test for use with measurements with “29 or fewer” observations. 26 One- time low- volume situations will not, of themselves, lead to payment; instead, SBC would only make payments in low- volume situations when it misses the established standard three months in a row (or six months in a year). This aspect of the Performance Plan should protect SBC from having to make payments for random events. SBC states that the Texas Commission clarified the Texas plan to exclude low- volume measurements for payments to the state. 27 As noted above, however, changes at the state level are not automatically made to the federal Performance Plan. Moreover, as SBC concedes, the Texas Commission excluded low- volume measurements only for payments directly to the state, not for payments to CLECs. 28 21 See Issue Number 9, Bureau Payment Calculation Letter at 4- 5. 22 See Bureau Payment Calculation Letter at 4- 5. 23 See SBC January 4 th Letter at 5- 6. 24 See Merger Order at Appendix C, Attachment A, A- 12 – A- 111. 25 See Merger Order, 15 FCC Rcd at 14867, ¶ 377 (stating that the goal of the Performance Plan is to ensure that quality of service to CLECs will not deteriorate as a result of the SBC/ Ameritech merger). 26 See Merger Order at Appendix C, Attachment A, A- 112 – A- 114. 27 SBC January 4 th Letter at 6. 28 Id. Under the Texas plan, payments fall into two tiers, i. e., payments to the Texas Commission and payments to CLECs. 4 Federal Communications Commission DA 02- 286 5 This is an important distinction between the two plans. The Texas Commission is still assured that SBC has an incentive to improve performance even in low- volume situations under its plan. If we were to accept SBC’s proposal to exclude low volume measurements here, there would be no such assurance under the federal plan. I therefore believe this is an instance where the benefits of applying divergent federal and state approaches could outweigh the potential administrative costs. For these reasons, I decline to modify my prior interpretation. I note, however, that the effect of this issue may, as a practical matter, be less significant in light of the increase in CLEC activity over the last year. As CLEC business increases, measurement volumes increase, and SBC should encounter fewer low- volume situations. 3. If SBC is required to make a payment for failure to meet a standard (i. e., failing to meet the monthly ideal value for that standard for three consecutive months), how should the second component of the payment calculation – the extent to which the performance standard was missed – for parity measurements expressed as averages or means be calculated for purposes of determining SBC’s payment obligation? 29 The Performance Plan states that SBC should compare the level of service SBC provides CLECs (the “actual CLEC service”) for each of the three months analyzed to the ideal value for the most recent month (i. e., the third month). 30 The Bureau instructed SBC to use this methodology in the Bureau Payment Calculation Letter. 31 SBC nevertheless observes that the Performance Plan’s approach could result in a negative performance gap and payment amount because the months are not comparable. 32 SBC suggests that a more appropriate approach would be to compare each month’s actual CLEC service to the ideal value for the same month. 33 Upon further review, I agree that SBC’s suggested approach will avoid the unintended results SBC describes and is consistent with the Performance Plan’s methodology for other types of measurements. 34 SBC therefore may use its proposed approach on this issue. 4. In conducting parity tests, should SBC use the variance computed for ILEC- to- CLEC data in months when there are no SBC retail data? 35 (Continued from previous page) 29 See Issue Number 4, Bureau Payment Calculation Letter at 3. 30 Merger Order at Appendix C, Attachment A, A- 116 – A- 117. As described above, the ideal value is the minimum service SBC could give CLECs without being liable for payments. In addition, the relevant period of analysis could be six of twelve months. 31 See Bureau Payment Calculation Letter at 3. 32 SBC January 4 th Letter at 4. For example, assume the ideal value for September is three days. Assume further an actual CLEC service level of two days and one day for July and August, respectively. The performance gaps for July and August would be negative (two days minus three days and one day minus three days). 33 SBC January 4 th Letter at 3- 4. 34 See Merger Order at Appendix C, Attachment A, A- 116 – A- 117 (prescribing the month- to- month comparison for measurements expressed as percentages, ratios, and proportions). 35 See Issue Number 7, Bureau Payment Calculation Letter at 4. 5 Federal Communications Commission DA 02- 286 6 The Performance Plan is silent on this issue. In the Bureau Payment Calculation Letter, I stated that SBC should use the SBC retail variance of an adjacent month, preferably the next most recent month. 36 SBC disagrees, stating instead that it should instead calculate a pooled variance estimate using the SBC and CLEC results for the current month. 37 Because the Performance Plan does not directly address this issue and SBC’s proposal is reasonable, I conclude SBC may use its proposed approach on this issue. I appreciate the opportunity to work through these issues with you and your staff. If SBC disagrees with our interpretation of the Merger Conditions, it should file an application for review with the Commission pursuant to section 1.115 of the Commission’s rules. 38 Please do not hesitate to contact me if I can be of further assistance. You may also contact Mark Stone in the Common Carrier Bureau directly at (202) 418- 0816 for further information on this matter. Sincerely, Carol E. Mattey Deputy Chief, Common Carrier Bureau CC: Al Syeles, SBC 36 See Bureau Payment Calculation Letter at 4. 37 SBC January 4 th Letter at 5. 38 47 C. F. R. § 1.115. 6