Ex Parte Submission : Application By SBC Communications Inc., Southwestern Bell Telephone Company, And Southwestern Bell Communications Services, Inc. D/b/a Southwestern Bell Long Distance For Provision Of In-Region, InterLATA Services In Texas, CC Docket

	DEPARTMENT OF JUSTICE Antitrust Division
	City Center Building 1401 H Street, NW Washington, DC 20530 March 20, 2000

Magalie Roman Salas, Esq.
Secretary
Federal Communications Commission
445 12^th Street, S.W.
Washington, D.C. 20554

Dear Ms. Salas:

In its February 14, 2000 Evaluation, the Department of Justice recommended that the FCC deny this application. After reviewing SBC's Reply Brief, the third-party reply comments and ex parte submissions that post-date the Department's Evaluation, the Department continues to believe that this application should be denied for the reasons stated in our Evaluation. As we explain below, SBC's additional submissions have not provided persuasive evidence that it is providing non-discriminatory access to DSL-capable loops¹ or hot-cut analog voice-grade loops.²

I. DSL-Capable Loops

SBC has failed to demonstrate either that it is providing non-discriminatory access to DSL-capable loops, or that its proposed advanced services affiliate will adequately ensure the provision of nondiscriminatory access in the future.

1. SBC's Performance in Provisioning DSL-Capable Loops Is Discriminatory

As explained in detail in our Evaluation, SBC's initial submissions did not provide evidence that it is providing non-discriminatory access to unbundled loops for DSL services.³ SBC's Reply Brief faulted the Department's analysis for relying too heavily on "a single month's" performance results for the DSL and BRI metrics.⁴ But SBC's performance data for January 2000 (which became available after the Department's Evaluation was filed) confirm that the poor performance reflected in the December 1999 data was no aberration. Rather, SBC's performance in January for the DSL and BRI loops remained substandard and in some cases worsened.⁵

SBC has still failed to address two other deficiencies in its performance data: a large percentage of DSL orders are not being tracked in the average installation measure (Performance Measurement ("PM") 55.1), and data for DSL loops are not included in the measures for the return of firm order commitments, which makes it difficult or impossible to determine whether SBC is providing nondiscriminatory access to its OSS for order processing of DSL-capable loops.⁶

SBC attempts to discount the significance of the poor performance data on missed appointments by arguing that its "root cause analysis" showed that the high proportion of SBC-caused missed appointments was attributable to a "lack of facilities."⁷ SBC claims that loop installations which require installation of a new circuit (such as installing a DSL-capable loop for a CLEC) are more likely to experience missed due dates resulting from the absence of available facilities than are its ADSL installations where the DSL equipment is placed on an existing circuit. Thus, in its Reply Brief, SBC argues that the missed appointment comparison measured in PM 58 is not an appropriate one.⁸ SBC's Reply Brief instead suggests its own revision to PM 58 which purports to remove from the data instances in which the SBC-missed appointments were attributed to a "lack of facilities." SBC maintains that its analysis shows that the current disparity reflected in PM 58 should end when it begins to offer "line sharing" to the CLECs later this year.⁹

The Commission should reject this reasoning. Even after "line sharing" is available to the CLECs, they will still require the installation of a separate new circuit for a substantial portion of their services, including symmetric DSL services which are not able to share a loop with analog voice service.¹⁰ These symmetric DSL services are functionally similar to SBC's DS1 service, and thus a comparison between SBC's provision of DSL loops to CLECs and SBC's provision of retail DS1 service would be an appropriate comparison between competitive services.¹¹ We are not aware of any evidence suggesting that the installations needed in connection with SBC's retail DS1 service are less difficult than the installations needed for a new DSL circuit,¹² and it is this benchmark which was adopted for PM 58 as described in the SBC business rules.¹³ In January, SBC missed 15.5 percent of its due dates for DSL loops, while, in contrast, it missed only 5.1 percent of its retail customer due dates for DS1 service.¹⁴ This performance strongly suggests that SBC could achieve substantially better performance in providing DSL-capable loops, even in cases where there will be no line sharing on those loops.

2. The Proposed Advanced Services Affiliate Is An Insufficient Remedy For This Discrimination

The Commission has stated that a fully operational separate affiliate that provides all forms of advanced services would reduce the risk of discriminatory treatment of competitors. A requirement that the affiliate and CLECs use identical processes to order loops would address one possible form of discrimination.¹⁵ In addition, the use of identical processes under a fully implemented separate affiliate structure will permit the RBOC's performance for the separate affiliate to be used as a benchmark against which to measure its performance to DSL competitors, facilitating detection and deterrence of discrimination.¹⁶ These potential benefits of a separate affiliate structure, however, will be in jeopardy if the Commission accepts SBC's proposal for its separate affiliate arrangements. Those arrangements will not ensure nondiscriminatory behavior for three reasons.

First, SBC does not plan to offer all of the relevant services through its affiliate, Advanced Solutions, Inc. ("ASI"). SBC, not ASI, will provide retail ISDN and DS1 services. SBC's ISDN services are similar to the IDSL services of the CLECs, and SBC's DS1 service is functionally similar to the CLECs' symmetric DSL services. Thus, SBC's proposal does nothing to ensure that it will provide nondiscriminatory access to the loops used for such services.¹⁷

Second, the full benefits of a separate affiliate will not be available until ASI is "fully operational," which will not happen until long after the Commission must make its decision on SBC's Texas application. Until ASI is fully operational, the improved ability to detect and prevent discrimination will not be realized. With respect to potential discrimination by SBC against its DSL competitors, the critical dates are those relating to ASI's ADSL services.¹⁸ By SBC's own admission, its advanced services affiliate in Texas will not be fully operational with respect to ADSL services until October 2000.

According to SBC, ASI began providing advanced services in Texas on February 2, 2000, a month after SBC filed its 271 application.¹⁹ For ADSL services, ASI plans to continue to use a proprietary pre-order inquiry system -- a system not available for use by CLECs -- until April 5, 2000.²⁰ ASI does not plan to begin use of the EDI order submission system -- the system that CLECs may use -- until May 1, 2000, and it appears the transition to this ordering system will continue until June 15, 2000.²¹ We are not aware of SBC's reasons for delaying ASI's use of these systems until those times. If the systems that SBC currently uses (to which the CLECs do not have access) are superior systems, SBC would have strong incentives to delay the conversion process.²² If SBC's conversion schedule is dependent on the need for additional software development, we would be particularly concerned because of the risk that conversion may be delayed if SBC is unable to complete the software development on schedule.

In addition to its current use of pre-ordering and ordering systems that are not available to CLECs, ASI will continue to have access to SBC network planning and engineering resources that CLECs do not have until line sharing is provided to CLECs.²³ The date on which line sharing will become available to CLECs is not known, except that it will be sometime after June 2000.²⁴ ASI's ability to use these resources confers a significant competitive advantage on ASI, particularly in negotiating for collocation space, a scarce and valuable resource. Finally, SBC may receive advanced-services-related trouble reports on behalf of ASI, a service SBC does not provide to CLECs, until October 7, 2000.²⁵ Until that time, ASI may well be able to provide better trouble resolution service to its DSL customers than CLECs will be able to provide to their DSL customers. In sum, ASI will not be fully operational under the terms of the SBC/Ameritech merger conditions until sometime in the fourth quarter of 2000 at the earliest, and until ASI is fully operational the separate affiliate structure will provide little if any benefit in ensuring nondiscrimination.

Third, even after the separate affiliate is "fully operational" SBC will perform manual work in the processing, provisioning, and maintenance of DSL-capable loops. Each point of manual intervention creates opportunities for discriminatory treatment, and such discrimination will be difficult to detect and punish unless suitable performance measurement arrangements are in place.²⁶ Although requiring ASI to deliver DSL orders to SBC via the electronic interfaces available to CLECs could mitigate the potential for some discrimination in the ordering process, this requirement addresses only one potential point of discrimination. Other important manual processing steps include: (1) the DSL pre-order loop qualification procedure; (2) the order processing procedure in SBC's local ordering center,²⁷ (3) the provisioning of DSL-capable loops, an intensively manual process; and (4) the maintenance and repair of DSL-capable loops. In each of these areas performance measures designed to capture any preferential performance given by SBC to ASI over its DSL competitors are a necessary complement to the separate affiliate structure in order to ensure that SBC is providing DSL-capable loops to data CLECs in Texas in a nondiscriminatory manner.²⁸

For these reasons SBC's proposed separate affiliate arrangements provide no basis for finding that competitors will have nondiscriminatory access to DSL-capable loops.

II. Hot-Cut Voice Analog Loops

Since the filing of the Department's Evaluation, SBC has submitted additional and corrected data on its hot-cut provisioning performance.²⁹ Based on these data, SBC argues its performance on the three criteria discussed by the Commission in its New York order is "at least equivalent to Bell Atlantic's nondiscriminatory performance."³⁰ We disagree. SBC's additional submissions do not adequately address the shortcomings in its initial application which were discussed in the Department's Evaluation. SBC has not demonstrated that its hot-cut performance is as good as Bell Atlantic's hot-cut performance in New York, which the Commission found to be "minimally acceptable."

SBC's corrected data on the duration of its hot cuts, interim PM 114.1, indicate better performance than the data SBC originally presented, but the new data still appear flawed, thus non-responsive to the concerns expressed in DOJ's initial Evaluation.³¹ First, SBC's December-January data still describe the number of loops, rather than the number of orders, completed within the relevant time. This presentation based on loops likely overstates SBC's performance as compared to the Bell Atlantic performance analyzed in New York which was based on orders -- each order consisting of approximately four to five loops.³² Second, SBC still fails to measure a portion of the relevant CHC time period in its data yet provides no explanation or justification of its failure to do so. SBC's reported interval ends at the time the SBC technician notifies SBC's provisioning center that the cutover is complete, rather than the time that SBC notifies the CLEC.³³ Because of these flaws, SBC's data are not sufficient to show that SBC is completing its hot cuts with the same degree of timeliness as Bell Atlantic was in New York.

With regard to the extent of outages upon provisioning, SBC in its reply submitted evidence of ongoing CHC reconciliation work with AT&T;³⁴ however, AT&T, in its reply, stated that the November-December CHC data have not been reconciled.³⁵ Thus, the reliability of the new data is unclear.³⁶ If SBC's additional data are accurate, they indicate improved CHC performance for the month of November. But if all of the available order outage data are aggregated, SBC's order outage rate remains 6.5 percent, still higher than Bell Atlantic's 4.5 percent outage rate.³⁷ Thus, it appears that SBC's hot-cut performance in Texas is worse than Bell Atlantic's hot-cut performance in New York.³⁸

Conclusion

The Reply Comments and ex parte submissions that SBC has filed since the Department submitted its evaluation fail to show that SBC is providing adequate performance with respect to DSL-capable loops or with respect to hot cuts for voice-grade analog loops. For these reasons, SBC's application should be denied. The Department further requests that a copy of this correspondence be placed in the record of this proceeding.

FOOTNOTES

¹ The term "DSL-capable loops" includes both DSL and BRI loops.

² The Department's Evaluation also analyzed SBC's performance in providing interconnection trunking and Operational Support Systems ("OSS") issues relating to UNE-loop and UNE-platform. Evaluation of the United States Department of Justice, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4, at 36-53 (Feb. 14, 2000) ("DOJ Texas Evaluation"). As regards interconnection trunking, SBC has not persuaded us that it was providing adequate interconnection performance to Texas CLECs at the time SBC filed its application. The December and January data, if reliable, suggest that the Texas Public Utilities Commission's ("Texas PUC") close oversight has resulted in improved performance. As regards OSS issues, the Department expressed concern regarding the scalability of SBC's systems. A sustained demonstration of appropriate performance on the relevant loop order processing measures would be necessary in order to assuage this concern. With particular regard to the OSS UNE-platform issue, the Department recommended that the Commission reserve judgment until SBC had more commercial experience processing UNE-platform orders. This recommendation seems especially warranted in view of the present difficult experience with high-volume OSS processing of these orders in New York.

³ Id. at 10-27.

⁴ Reply Brief in Support of Application by Southwestern Bell, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4, at 22 (Feb. 22, 2000) ("SBC Reply Brief"). The Department's analysis emphasized the December data since the volume of orders was greater towards the end of the year.

⁵ The serious performance problems include a continuing lack of parity as measured by the DSL and BRI performance measures for missed due dates (PM 58), trouble reports within 30 days of installation (PM 59), and trouble report rate (PM 65), all Tier 2, competition-affecting measures as determined by the Texas PUC. These measures are not just slightly out of parity -- they disclose that CLECs may experience far more delays and service troubles than SBC's retail customers. SBC's performance has been substantially below the standard for all of these measures for two of the last three months for both DSL and BRI loops. SBC Communications Inc., Non-CLEC-Specific Performance Measure Tracking/Chart Results for February 1999 through January 2000 (aggregated by market area), Ex Parte Submission to the FCC, CC Docket No. 00-4, at 271-No.58b (BRI) to 58c (DSL), 271-No.59b (BRI) to 59c (DSL), and 271-No.65b (BRI) to 65-c (DSL) (Feb. 25, 2000) ("SBC January Performance Data"); Texas Public Utilities Commission, Performance Measures Business Rules (Version 1.6) at A-79, A-80, A-86 ("Business Rules 1.6"), Attach. A to Affidavit of William R. Dysart, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4, (Jan. 10, 2000).

⁶ See DOJ Texas Evaluation at 13-17.

⁷ Of course, SBC's "lack of facilities" argument does not address the substandard performance reports for the other DSL and BRI performance measures such as the troubles within 30 days of installation, or the trouble report rate.

⁸ In a recent ex parte, SBC explained that, although its business rule for PM 58 called for a comparison with its DS1 service, it thought this was a typographical error. Thus SBC used its retail ADSL service as a benchmark because it was a "logical comparison [that] accords with the comparison used in Performance Measurement 55.1". SBC Communications Inc., List of Prior Corrections to Information Filed with the Commission in this Proceeding, Ex Parte Submission to the FCC, CC Docket No. 00-4, at 3 (Mar. 8, 2000).

⁹ SBC Reply Brief at 24-26; Reply Affidavit of William R. Dysart on Behalf of SBC, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4 ¶ 31 ("SBC Dysart Reply Aff."). SBC has not adequately explained why it did not raise these issues either in proceedings before the Texas PUC or in its initial application, rather than postponing its explanation until it filed its Reply. The performance problem that SBC attempts to explain away was apparent starting with its September performance report, long before its application was filed. If SBC thought that the performance measure adopted by the Texas PUC was inappropriate, it could have suggested an alternative measure in its initial application, if not earlier. SBC's introduction of newly formulated performance data in its Reply Brief clearly limits the ability of other parties to evaluate and respond to SBC's claims. We have the same objection to SBC's Reply argument that its substandard performance in provisioning and maintaining BRI loops is the result of CLECs' use of nonstandard equipment -- an argument that the CLECs dispute but which would be difficult to resolve in the limited time following SBC's Reply. See SBC Reply Brief at 26; SBC Dysart Reply Aff. ¶ 43; Covad Communications Company, Points Regarding Covad's Initial Reaction to the Reply Comments of the Texas Commission in this Proceeding, Ex Parte Submission to the FCC, CC Docket No. 00-4, at 3-4 (Mar. 1, 2000) ("Covad Ex Parte").

¹⁰ Third Report and Order in CC Docket No. 98-147, Fourth Report and Order in CC Docket No. 96-98, In re: Deployment of Wireline Services Offering Advanced Telecommunications Capability and In re: Implementation of the Local Competition Provisions of the Telecommunications Act of 1996, CC Docket Nos. 98-147 and 96-98, FCC 99-355, at 16 n.51 (rel. Dec. 9, 1999).

¹¹ See generally Comments of Covad Communications Company, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4, at 23-25 (Jan. 31, 2000).

¹² Covad Ex Parte at 3.

¹³ Business Rules 1.6 at A-78 to A-79.

¹⁴ SBC January Performance Data at 271-No.58b (DS1) to 58c (DSL).

¹⁵ Memorandum Opinion and Order, In re: Application by Bell Atlantic New York for Authorization Under Section 271 of the Communications Act to Provide In-Region, InterLATA Service in the State of New York, CC Docket No. 99-295, FCC 99-404 ¶ 332 & n.1037 (rel. Dec. 22, 1999) ("FCC New York Order").

¹⁶ Memorandum Opinion and Order, In re: Applications of Ameritech Corp., Transferor, and SBC Communications Inc., Transferee, For Consent to Transfer Control of Corporations Holding 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 No. 98-141, FCC 99-279 ¶ 467 (rel. Oct. 8, 1999) ("FCC SBC-Ameritech Merger Order").

¹⁷ See FCC New York Order ¶ 332 n. 1037 ("We view it as critical that a BOC provide all forms of advanced services through a separate affiliate, and not just ADSL, so the affiliate would need to obtain stand-alone loops from the BOC in order to provide all varieties of advanced services.").

¹⁸ SBC defines non-ADSL services to include Frame Relay, Cell Relay, VPOP/DAS and Native LAN Plus. Reply Affidavit of Lincoln E. Brown on Behalf of SBC, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4 ¶ 5 (Feb. 22, 2000) ("SBC Brown Reply Aff.").

¹⁹ SBC Reply Brief at 18.

²⁰ Id. at 28.

²¹ SBC Brown Reply Aff. ¶¶ 4-6; see Reply Comments of Rhythms NetConnections Inc., In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4, at 42 (Feb. 22, 2000). For non-ADSL services, SBC states that ASI began to use the "same ordering and provisioning systems and procedures that CLECs use" on February 28, 2000. SBC Reply Brief at 4; SBC Brown Reply Aff. ¶ 5.

²² The Commission has tried to off-set this incentive by requiring SBC/Ameritech to offer CLECs a 25% discount on certain charges for DSL-capable loops until the SBC/Ameritech advanced services affiliate uses the EDI interface for pre-ordering and ordering of a "substantial majority of the facilities it uses to provide advanced services." FCC SBC-Ameritech Merger Order ¶ 372 & App. C. at 29 (Condition 18).

²³ SBC Communications Inc., Letter from Michael Kellogg to Carol Mattey, Ex Parte Submission to the FCC, CC Docket No. 00-4, at 2 & n.2, 3 (filed Feb. 16, 2000).

²⁴ SBC Reply Brief at 25 n.13 ("Under the Commission's Line Sharing Order, which became effective on February 9, 2000, SWBT has no obligation to unbundle the data channel of existing voice lines until June 2000.").

²⁵ Reply Affidavit of Sherry L. Ramsey on Behalf of SBC, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4 ¶ 4.i (Feb. 22, 2000) (assuming that "merger closing date" as used in this affidavit means Oct. 8, 1999, the date SBC and Ameritech filed a certificate of merger in Delaware).

²⁶ In this particular case, it is important to keep in mind that any possible discriminatory treatment is more than hypothetical. As discussed above and in our Evaluation, SBC's current performance reports reveal discrimination against CLECs in the provisioning of DSL-capable loops. DOJ Texas Evaluation at 17-23.

²⁷ While SBC may eventually provide for these order types to flow through without manual intervention, today all such orders are processed manually. Reply Affidavit of Carol Chapman on Behalf of SBC, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4 ¶ 49 (Feb. 22, 2000).

²⁸ Indeed, as the Commission has stated, SBC's "provision of inputs to its advanced services affiliate will serve as an important benchmark against which to measure its performance to unaffiliated carriers." FCC SBC-Ameritech Merger Order ¶ 467.

²⁹ See SBC Reply Brief at 38-42; SBC January Performance Data at 271-No. 114a to 115d; SBC Communications Inc., Charts Providing Corrected Hot Cut Data for PMs 114, 114.1, and 115 for the Months of December 1999 and January 2000, Ex Parte Submission to the FCC, CC Docket No. 00-4 (Mar. 2, 2000) ("SBC Hot Cut Ex Parte").

³⁰ SBC Hot Cut Ex Parte at 1.

³¹ DOJ Texas Evaluation at 31-32, nn.84-85.

³² Thus, the benchmark of 90 percent timeliness which the FCC deemed minimally acceptable for the completion of hot cut orders may not be the appropriate benchmark to establish comparable performance if timeliness is measured as a percentage of completion of hot cut loops.

³³ DOJ Texas Evaluation at 31 n.84; see also Affidavit of Candy R. Conway on Behalf of SBC, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4 ¶¶ 84 ("Once the CHC is completed, the central office technician then calls the LOC technician. The time of this call is logged by the LOC NCT as the completion time of the CHC. The LOC NCT then notifies the CLEC that the CHC has been completed."), 87 (Jan. 10, 2000).

³⁴ See Reply Affidavit of Candy R. Conway on Behalf of SBC, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, InterLATA Services in Texas, CC Docket No. 00-4 ¶¶ 12-13 (Feb. 22, 2000) ("SBC Conway Reply Aff.").

³⁵ AT&T states that it and SBC have been unable to agree on the base numbers of loops and orders for November and December, and thus are unable to calculate the outage rates. Reply Declaration of Sarah DeYoung on Behalf of AT&T, In re: Application by SBC Communications Inc. et al. for Provision of In-Region, Inter LATA Services in Texas, CC Docket No. 00-4 ¶¶ 25, 71 (Feb. 22, 2000); see also AT&T Corp., Response to SBC's Reply Comments and Ex Partes Filed in this Proceeding, Ex Parte Submission to the FCC, CC Docket No. 00-4, at 5-6 (Mar. 6, 2000) ("AT&T Hot Cut ExParte I"); AT&T Corp., Response to SBC's Hot Cut Ex Parte Filing on March 2, 2000, Ex Parte Submission to the FCC, CC Docket No. 00-4, at 2 (Mar. 13, 2000) ("AT&T Hot Cut Ex Parte II").

³⁶ After filing its Reply Brief, SBC acknowledged that it had previously incorrectly reported the total numbers of loops for all CLECs for December and January, stating that it had undercounted loops for those months. SBC Hot Cut Ex Parte at 2. SBC has also made conflicting representations regarding AT&T's December loop totals. AT&T Hot Cut Ex Parte I at 6 (comparing AT&T loop numbers presented in SBC Conway Reply Aff. ¶ 12 with numbers listed in SBC's AT&T-specific performance reports).

³⁷ The FCC in its New York Order aggregated all of the available reconciled data in calculating the 4.5 percent order outage rate. FCC New York Order ¶ 302 & n.961 (aggregating outage data for June 21, 1999 through August 31, 1999). If the analysis is restricted to the three most recent months for which order outage data are available, September through November, the order outage rate is 7.01 percent. See SBC Conway Reply Aff., Conf. Attach. A at A-1. Although SBC submitted additional data on December CHC outages upon provisioning, these data are presented on a loop, rather than order, basis. SBC Conway Reply Aff. ¶ 12.

³⁸ As regards the equally-important FDT [Frame Due Time] hot cut process, SBC's timeliness data are again presented in terms of loops not orders, possibly overstating the timeliness of its performance, and SBC has presented no additional data on the overall order outage rate. See SBC Hot Cut Ex Parte.