Evaluation Of The United States Department Of Justice

Table of Contents

Index of Full Citations

Introduction and Summary

1. Competitive Entry in Local Telecommunication Markets in Massachusetts
2. Verizon's Application Does Not Demonstrate That Verizon Provides Competitors Nondiscriminatory Access to DSL Loops
1. Verizon's Failure to Demonstrate Nondiscriminatory Access to DSL Loops
2. Verizon Has Not Established Reliable Performance Measures with Associated Benchmarks to Deter Backsliding When Providing Wholesale Services to DSL Competitors

3. Verizon's Wholesale Performance in Providing Competitors with the UNE-Platform
4. The Scalability of Verizon's OSS in Massachusetts Is Assured By Less Evidence Than in New York and By Less Effective Post-Entry Enforcement Mechanisms
5. Conclusions and Recommendations

Verizon's application to offer long distance services in Massachusetts follows its successful application for long distance entry in New York last year.⁽¹⁾ In Massachusetts, as in New York, progress in establishing the necessary preconditions for local competition has already produced important benefits to consumers, and still greater benefits can reasonably be expected as the market becomes more competitive. The Massachusetts Department of Telecommunications and Energy ("MA DTE") has conducted extensive proceedings to implement market-opening measures and has submitted a comprehensive evaluation of Verizon's application. Building on the successful efforts of Verizon and the New York Public Service Commission ("NY PSC"),⁽²⁾ the MA DTE has ensured that Massachusetts will benefit from competition, and Verizon is working with its wholesale customers and the MA DTE to continue to improve its systems and services.⁽³⁾

There is significant entry in Massachusetts by facilities-based carriers and resellers serving business customers, and the Department of Justice ("Department" or "DOJ") is particularly pleased to see a major commitment to facilities-based residential competition by AT&T Broadband and RCN. The Department evaluates applications under section 271 to determine whether a state's local markets are "fully and irreversibly open" to competition using three entry modes: full facilities, unbundled elements and resale.⁽⁴⁾ This application, in our judgment, shows that although Verizon has satisfied this standard in most respects, important issues remain inadequately addressed. The principal issue on which Verizon has failed to develop an adequate record is its provision of unbundled loops for digital subscriber line ("DSL") services.⁽⁵⁾ The Department has concluded that Verizon has not yet demonstrated (1) that it provides nondiscriminatory access to DSL loops, and (2) that suitable performance measures with unambiguous benchmarks are in place to deter backsliding. The Commission should not approve this application without such a demonstration.

1. Competitive Entry in Local Telecommunication Markets in Massachusetts

To determine whether Verizon has fully and irreversibly opened the local telecommunications market in Massachusetts to competition for both business and residential customers, the Department examines the three modes of entry contemplated by the Telecommunications Act of 1996⁽⁶⁾: facilities-based entry, the use of unbundled elements of the incumbent's network and resale of the incumbent's services.⁽⁷⁾ The Department first looks to actual competitive entry, because the experience of competitors seeking to enter a market can provide highly probative evidence about the presence or absence of artificial barriers to entry.⁽⁸⁾ The degree to which such existing competition is broad based determines the weight the Department places on it as evidence.⁽⁹⁾

For those entry modes where competitively significant entry is reasonably foreseeable but broad-based commercial entry is absent, the Department examines whether new technical and operational arrangements are available and working to support the entry mode, and whether performance benchmarks have been established to detect backsliding by the incumbent after long distance entry.⁽¹⁰⁾ Small market shares held by competitors or even the absence of entry, standing alone, are neither conclusive evidence that a market remains closed to competition nor a basis for denying an application under section 271.⁽¹¹⁾

According to Verizon, competitive local exchange carriers ("CLECs") serve 676,000 lines, or 11 percent of the total lines, in Massachusetts.⁽¹²⁾ This level of CLEC penetration is greater than the level in either New York or Texas at the time applications were filed in those states.⁽¹³⁾ The overwhelming majority of CLEC lines in Massachusetts are business lines.⁽¹⁴⁾ By contrast, slightly more than half of Verizon's lines serve residential customers.

The predominant mode of CLEC entry in Massachusetts is facilities based. Facilities-based CLEC lines serve approximately seven percent of the total lines in Massachusetts, or almost two-thirds of total CLEC lines.⁽¹⁵⁾ The substantial majority of these lines serve business customers.⁽¹⁶⁾ However, facilities-based CLECs have the potential to serve a significant number of residential customers. AT&T Broadband's cable facilities alone pass approximately 1.7 million of the 2.35 million homes in Massachusetts.⁽¹⁷⁾ AT&T Broadband is in the process of upgrading these facilities with digital telephony capabilities.⁽¹⁸⁾ Verizon estimates that AT&T Broadband currently provides local phone service to over 20,000 Massachusetts customers.⁽¹⁹⁾ RCN is building a fiber-optic network that will provide telephony service, high-speed Internet access, and broadband video distribution.⁽²⁰⁾ With this network, RCN plans to serve residential customers in the greater Boston metropolitan area.⁽²¹⁾ The development of these facilities-based alternatives for residential customers is encouraging, but it is unknown at this time how rapidly cable customers could choose to purchase telephony services from these providers once these services are available.

Massachusetts is also home to an active resale market.⁽²²⁾ A large number of the resold lines are used to serve business customers.⁽²³⁾ The extent of resale is likely due, in large part, to the relatively high discount rate (i.e., the relatively low wholesale prices) at which CLECs may purchase resale services in Massachusetts.⁽²⁴⁾

In contrast, CLECs have made little use of the UNE-platform in Massachusetts.⁽²⁵⁾ The limited use of UNE-platform contrasts sharply with the New York and Texas markets, where the use of UNE-platform accounts for rapid CLEC expansion into the residential market.⁽²⁶⁾

CLECs in Massachusetts are providing DSL service to a growing number of customers. After a late start, however, Verizon is the largest provider of DSL service in Massachusetts, adding four times as many DSL lines per month as all other CLECs combined.⁽²⁷⁾ A relatively large portion of DSL lines to both residential and business customers are provided by CLECs to Internet Service Providers ("ISPs"), which sell a bundled offering of Internet access through these DSL lines to end users. These ISPs typically deal with both Verizon and one or more of the data CLECs, shifting their business to the provider that offers the best combination of price and performance.

The Department presumes that opportunities to serve business customers by facilities-based carriers and resellers are fully available in Massachusetts, based on the substantial and successful entry efforts reflected in Verizon's application. There is significantly less competition at present to serve residential customers by facilities-based carriers and by resellers. With one possible exception, however, the record does not suggest that Verizon constrains these types of competition.⁽²⁸⁾

There is also significantly less competition by firms seeking to use unbundled network elements, including both DSL loops and the UNE-platform, combined with some indications that a failure by Verizon to satisfy all of its obligations may have constrained this type of competition. The Department's evaluation will focus on these issues.

2. Verizon's Application Does Not Demonstrate That Verizon Provides Competitors Nondiscriminatory Access to DSL Loops

Verizon's application reflects two serious deficiencies with respect to the provision of DSL loops.⁽²⁹⁾ First, Verizon has not yet satisfied its burden of demonstrating that it is providing nondiscriminatory access to DSL loops.⁽³⁰⁾ Second, to the degree that Verizon has failed to develop performance measures in conjunction with CLECs and the MA DTE that would provide a reliable indication of Verizon's performance, there is little assurance that future backsliding can be readily detected and addressed. The Commission has previously emphasized the competitive significance of DSL services⁽³¹⁾ and the importance of demonstrating nondiscriminatory performance through comprehensive and accurate reports of performance measures.⁽³²⁾

1. Verizon's Failure to Demonstrate Nondiscriminatory Access to DSL Loops

Verizon's performance reports indicate significant discrimination in the time Verizon takes to install DSL loops,⁽³³⁾ the quality of those loops at the time of installation,⁽³⁴⁾ and the manner in which Verizon repairs the loops.⁽³⁵⁾

Recognizing that these performance reports, on their face, indicate discriminatory performance, Verizon argues that the performance data it presents in Massachusetts do not provide an accurate indication of its performance. For example, Verizon contends that the installation interval measure is misleading because it is an aggregated measure of two kinds of orders -- those that need manual prequalification and those that have been prequalified.⁽³⁶⁾ Since the manual prequalification process requires extra time, and since Verizon does not manually prequalify its own orders, Verizon claims that the aggregated data do not provide a fair indication of its performance.⁽³⁷⁾ Verizon has submitted analyses that purport to show that when orders requiring manual prequalification are excluded, Verizon's performance for installing DSL loops for CLECs is nondiscriminatory.⁽³⁸⁾ However, it is difficult or impossible to verify Verizon's reformulated performance calculations and analysis because Verizon has not provided the data underlying its reformulated performance calculations and because Verizon has not given the CLECs their individual performance reports,⁽³⁹⁾ which would be necessary to permit CLECs to verify or refute Verizon's restated performance.

Verizon also contends that the measurement of the mean time to repair DSL loops provides a misleading indication of its performance. Verizon claims that it is much more likely to be unable to access CLEC customers' premises to repair DSL loops than to access the premises of its own retail customers,⁽⁴⁰⁾ and that the CLECs are less willing to schedule weekend appointments than are Verizon's retail customers.⁽⁴²⁾ Both of these factors, Verizon claims, lengthen the time needed to repair CLEC DSL loops.⁽⁴²⁾ CLECs deny that they avoid weekend repair appointments⁽⁴³⁾ and contend that they are unable to fully respond to Verizon's argument since Verizon has not provided CLEC-specific performance reports.⁽⁴⁴⁾

Verizon also claims that the measure of trouble reported within thirty days of installation and the measure reporting how long Verizon takes to repair DSL loops reflect certain inappropriate CLEC practices, rather than problems with Verizon's performance. First, Verizon claims that some of the CLECs accept loops that they know, or should know, are not working and then submit trouble tickets to have Verizon repair them.⁽⁴⁵⁾ In addition, Verizon claims that its missed repair appointment measure is overstated because the CLECs frequently submit trouble tickets without first properly isolating the problem to the Verizon network.⁽⁴⁶⁾ According to Verizon, this practice is indicated by the relatively high number of "no trouble found" reports Verizon receives from its technicians.⁽⁴⁷⁾ The unnecessary dispatches resulting from this practice allegedly tie up Verizon technicians and keep them from responding to real problems in the network.⁽⁴⁸⁾

The Department has not been able to determine whether Verizon's objections to the performance measures are valid or whether Verizon is providing nondiscriminatory performance even under its suggested alternative methods of measuring performance.⁽⁴⁹⁾ We believe, however, that it is appropriate to insist that Verizon satisfy its burden of proof on these issues.⁽⁵⁰⁾ If the performance measures used in Massachusetts are inadequate or if they are being distorted by inappropriate CLEC practices, it is in the public interest for Verizon to address those problems through performance measure proceedings conducted by a state commission, rather than raising these issues for the first time in its section 271 application to the Commission.⁽⁵¹⁾ Similarly, if Verizon seeks to have its performance evaluated on the basis of measures that differ from the measures adopted by a state commission, Verizon at a minimum should be required to ensure that CLECs and other parties have adequate opportunity and sufficient data to assess and respond to Verizon's claims about the quality of its performance.

2. Verizon Has Not Established Reliable Performance Measures with Associated Benchmarks to Deter Backsliding When Providing Wholesale Services to DSL Competitors

To the extent that the Massachusetts performance measures do not accurately indicate whether Verizon is providing discriminatory or nondiscriminatory access to DSL loops, those deficiencies in the performance measures will substantially increase the difficulties of detecting and providing remedies for any discriminatory performance that may arise in the future. Both the Department and the Commission have previously emphasized the importance of reliable and objective performance measures that will minimize the difficulties of detecting and proving backsliding, and that by doing so will provide important incentives for incumbents to maintain nondiscriminatory performance after a 271 application has been approved.⁽⁵²⁾ This important objective will be jeopardized if the performance measures for which a BOC provides data for review in the 271 process are defined or implemented in ways that routinely provide either "false positives" or "false negatives" with respect to potential discrimination and thus keep the metrics from being meaningful.⁽⁵³⁾ Under those circumstances, detection, proof, and correction of discrimination is more likely to require costly and time consuming regulatory fact finding proceedings that may not effectively prevent serious competitive harm.

Verizon's claim that the Massachusetts performance measures produce "false positives" (i.e., that due to issues involving definition and implementation the reported performance appears to be discriminatory but in fact is nondiscriminatory) highlights only one of several possible weaknesses in the DSL performance measurement process. Another important weakness is the absence of independent checks on the accuracy of the performance that is reported. Although KPMG reviewed other Verizon performance metrics, it did not test the DSL metrics because they were implemented by Verizon after the initial testing period.⁽⁵⁴⁾ Moreover, Verizon has not provided individual CLECs reports that show its performance on their DSL orders.⁽⁵⁵⁾ We are not aware of any reason for this omission, and in fact Verizon provides such individual performance reports in New York.⁽⁵⁶⁾ Because Verizon has not provided such reports in Massachusetts, it is impossible for any CLEC to verify Verizon's performance reports or to adequately respond to Verizon's allegations that certain practices of the CLECs may be distorting the performance results.⁽⁵⁷⁾

Finally, Verizon does not provide performance reports on line sharing.⁽⁵⁸⁾ CLECs are currently placing line sharing orders in New York and Massachusetts, and Rhythms has complained that Verizon's failure to properly prepare its central offices to support the implementation of line sharing is impeding its ability to offer the service.⁽⁵⁹⁾ Because Verizon has not provided performance reports regarding the provisioning of line sharing orders, it is extremely difficult to resolve these disputes now, and it will be as difficult, if not more difficult, to effectively monitor Verizon's performance in the future.⁽⁶⁰⁾

In our view, the record at this time does not contain convincing evidence that (1) Verizon is providing nondiscriminatory performance to CLECs using DSL loops, and (2) suitable performance measures with associated unambiguous benchmarks are in place to promptly detect and remedy any future backsliding. The Commission should not approve this application without such evidence.

The Commission has emphasized its "strong preference for a record that contains data measuring a BOC's performance pursuant to state-adopted standards that were developed with input from the relevant carriers and that include clearly-defined guidelines and methodology."⁽⁶¹⁾ The usefulness of any attempt to measure performance will be enhanced if the reliability of the measures has been tested by informed public comment and if the measures used to assess section 271 compliance can be used for post-entry performance comparisons. In this proceeding, the Commission should reiterate its expectation that applicants will present such issues to state commissions in the first instance as the best means to ensure meansurements of performance that are useful for establishing a benchmark.

3. Verizon's Wholesale Performance in Providing Competitors with the UNE-Platform

As previously indicated, entry by competitors using the UNE-platform has been very limited in Massachusetts, especially in comparison with New York. This fact raises the question whether such entry has been impeded by Verizon's failure to meet its obligations under the 1996 Act. Although the Department has not reached any final conclusions on that question, there is substantial reason to believe that UNE-platform entry has been impeded by Verizon's failure, at least perhaps until quite recently, to make certain network elements available to competitors at cost-based prices.

Of course, one factor that has limited the use of the UNE platform in Massachusetts has been AT&T's strategic decision to focus its entry efforts on the use of its cable facilities.⁽⁶²⁾ In New York, where AT&T did not own such facilities, AT&T's entry strategy relied to a significant degree on use of the UNE-platform.⁽⁶³⁾ It would not be appropriate to draw any inferences as to Verizon's market opening efforts merely because of the specific entry strategy chosen by AT&T in Massachusetts.

But AT&T's decision does not completely explain the difference between the two states. CLECs other than AT&T have made extensive use of the UNE-platform in New York and other states, but they have not yet done so in Massachusetts. For example, at the time of Verizon's New York application, even though AT&T had not yet entered that market using the UNE-platform, more than 152,000 lines were served through the UNE-platform, compared to approximately 12,000 lines served in Massachusetts at the time of this application.⁽⁶⁴⁾ Moreover, CLECs other than AT&T currently account for over 600,000 UNE-platform lines in New York.⁽⁶⁵⁾

The most plausible explanation for the limited use of the UNE-platform in Massachusetts appears to be the relatively high prices charged by Verizon for certain unbundled network elements,⁽⁶⁶⁾ and there are reasons to suspect that in some cases those prices have not been based on the relevant costs of the network elements.⁽⁶⁷⁾

In an effort to "eliminate pricing issues particularly regarding local switching in [Verizon's] Section 271 application now pending before the FCC," Verizon filed a short amended tariff proposal with the MA DTE⁽⁶⁸⁾ on October 13, 2000, twenty-one days after filing its 271 application.⁽⁶⁹⁾ The MA DTE approved Verizon's tariff proposal.⁽⁷⁰⁾ The tariff was not accompanied by any supporting documentation but, according to Verizon, the tariff lowers prices for certain UNEs to rates equivalent to those set by the NY PSC and approved by the Commission in its New York Order.⁽⁷¹⁾

The timing of Verizon's tariff filing is regrettable. If Verizon's previous UNE prices exceeded the relevant cost of those UNEs, the delay in correcting that defect likely slowed the development of competition in Massachusetts, particularly competition to serve residential customers. The timing of Verizon's tariff filing foreclosed opportunities of other parties to evaluate whether the new prices are appropriately cost based. Verizon submitted its proposal to the MA DTE one business day before initial third-party comments in this proceeding were due, effectively preventing interested parties from addressing the tariff proposal in their initial comments in this proceeding, and denying the Department the benefit of these parties' analyses before its own evaluation was due. Further, there is no underlying documentation to show that the listed rate reductions are, in fact, based on cost studies relied upon by the NY PSC or, more importantly, to show that the new rates are cost based in Massachusetts.

In these circumstances, the Commission could reasonably "restart the clock" in this proceeding, as of the date of Verizon's tariff filing, to ensure that there is fair and adequate consideration of this important issue. In any case, the Commission should carefully analyze Verizon's tariff filing to determine whether it in fact satisfies requirements for cost-based prices for unbundled network elements, at a minimum taking into consideration factual information or legal arguments presented by third parties in this proceeding that were not considered during the New York 271 proceeding.⁽⁷²⁾

4. The Scalability of Verizon's OSS in Massachusetts Is Assured By Less Evidence Than in New York and By Less Effective Post-Entry Enforcement Mechanisms

In order to verify the readiness and performance levels of Verizon's Operations Support Systems ("OSS"), the MA DTE, like the NY PSC, hired KPMG Consulting ("KPMG") as a third-party tester.⁽⁷³⁾ KPMG conducted extensive testing of Verizon's OSS, and that testing was extremely valuable in identifying and correcting OSS defects, and in providing significant evidence that Verizon's OSS can adequately handle CLEC transactions.

The KPMG test in Massachusetts, however, was less complete in several significant respects than the testing KPMG previously conducted in New York.⁽⁷⁴⁾ Moreover, as experience in New York has demonstrated,⁽⁷⁵⁾ some OSS deficiencies are recognized only under the stress of full-scale commercial entry, which has not yet occurred in Massachusetts.

Even without the post-entry order processing problems that occurred in New York, it might be troubling that Verizon offers less evidence of the scalability of its OSS than it provided in its New York application. Given that serious OSS problems arose as Verizon's OSS was scaled to meet increasing commercial use in New York, we lack a high degree of confidence in the predictive judgment that Verizon will not encounter future scalability problems in Massachusetts. Therefore we recommend that the Commission pay particular attention to the value of a strong, self-executing performance assurance plan, to ensure that any problems that might arise can be addressed quickly and effectively. In this regard, we are concerned that the effectiveness of the Massachusetts performance assurance plans may be compromised by the significant differences between these plans and the performance assurance plans that the Commission approved as part of Verizon's New York application. Most importantly, the Massachusetts performance assurance plans do not give the MA DTE explicit authority to reallocate the monthly distribution of bill credits among the provisions of the performance assurance plan or between that plan and the Change Control Assurance Plan.⁽⁷⁶⁾ It was precisely this mechanism that permitted the NY PSC to react quickly when it realized that Verizon's post-entry OSS problems were not captured by the performance measures contained in the performance assurance plan.⁽⁷⁷⁾ The Commission should ensure that effective remedies will be available to quickly resolve any post-entry performance problems in Massachusetts. These assurances are not present in the current record.

5. Conclusions and Recommendations

While competition in the market for local telecommunications services in Massachusetts is active, there are still issues left to be resolved before the market can be considered open. Based on our review of the record at this time, Verizon has not yet satisfied its burden of proving that it provides nondiscriminatory access to DSL loops or that adequate performance mechanisms are in place to deter backsliding. The Commission should also carefully examine the prices charged by Verizon for unbundled network elements to ensure that these prices are cost based. Finally, the Commission should give particular attention to the value of a strong performance assurance plan in Massachusetts to ensure that any post-entry OSS failures triggered by increasing volumes can be addressed promptly and effectively.

FOOTNOTES

1. On June 30, 2000, Bell Atlantic Corporation completed its merger with GTE Corporation, creating Verizon Communications. The Department will refer to the company as Verizon whether we are talking about events that occurred before or after the merger was completed.

2. See, e.g., MA DTE Letter Order (Jan. 14, 2000) (adopting the New York Carrier-to-Carrier metrics), attached to Verizon App. B, Tab 282.

3. See Z-Tel Comments at 4 (noting Verizon's efforts to improve GUI stability and line loss report accuracy, and Verizon's commitment to implement cut-through functionality); RNK Comments at 3 (stating that, despite having "its fair share of problems with Verizon," RNK has seen "reasonable improvement over the past ten months in the areas of resale billing, blockage, OSS reliability, claims adjustments" and billing accuracy).

4. DOJ Oklahoma Evaluation at vi-vii, 36-51.

5. In addition, the Commission should determine the extent to which prices for unbundled elements in Massachusetts are appropriately cost based, and should consider the significant differences between the Massachusetts performance assurance plans and the New York performance assurance plans.

6. Pub. L. No. 104-104, 110 Stat. 56 (1996) (codified as amended in scattered sections of 47 U.S.C.).

7. See DOJ Schwartz Aff. ¶¶ 149-192; DOJ Schwartz Supplemental Aff. ¶¶ 26-60; DOJ Oklahoma Evaluation at vi-vii, 36-51.

8. See, e.g., DOJ Oklahoma Evaluation at vi-vii, 41-42.

9. See, e.g., DOJ Schwartz Aff. ¶ 176.

10. DOJ Oklahoma Evaluation at 48-51.

11. See, e.g., id. at 29-30; DOJ Louisiana II Evaluation at 26-27.

12. See Verizon Taylor Decl. ¶ 25. Verizon serves 5.4 million lines in Massachusetts. Verizon Brief at 62 n.59.

13. In New York, CLECs served approximately 8.9% of total access lines. DOJ New York Evaluation at 9. In Texas, SWBT estimated that CLECs served 12.8% of total access lines, but the Department concluded that CLECs actually served closer to 8.0% of the market. DOJ Texas I Evaluation at 8-9.

14. See Verizon Taylor Decl. Attach. A at 1 (CLEC lines are 82% business and 18% residential.).

15. See Verizon Taylor Decl. Attach. A at 2 tbl.1. The number of facilities-based lines includes lines served by stand-alone UNE-loops and thus is not "pure" facilities-based lines. See id.

16. See id. Attach. A Ex. 2 (Eighty percent of CLEC facilities-based lines in Massachusetts serve business customers.).

17. AT&T Comments at 10.

18. Id.; AT&T Kowolenko Decl. ¶ 5.

19. Verizon Taylor Decl. ¶ 22 (citation omitted).

20. RCN Comments at 4.

21. Id. at 4-5.

22. See Verizon Taylor Decl. Attach. A ¶ 15. 246,000 of 676,000 CLEC lines are resale lines. These lines constitute four percent of total lines in Massachusetts. Id. Attach. A at 2 tbl.1.

23. See id. Attach. A Ex. 2 (Eighty-seven percent of CLEC resale lines are business lines.).

24. See MA DTE Evaluation at 392-93 (discussing resale discount rate). The Department notes, however, that a recent Eighth Circuit opinion makes the long-term availability of these rates uncertain. Iowa Utils. Bd. v. FCC, 219 F.3d 744, 754-56 (8^th Cir. 2000), petitions for cert. filed (U.S. Oct. 4, 2000) (No. 00-511) and (U.S. Oct. 10, 2000) (No. 00-555).

25. Approximately 12,000 UNE-platform lines are in use in Massachusetts, split evenly between business and residential customers. Verizon Taylor Decl. Attach. A Ex. 2. These lines constitute fewer than two-tenths of one percent of total lines in Massachusetts. Id. Attach. A at 2 tbl.1.

26. At the time of the New York application, approximately 152,000 lines were served through the UNE-platform. FCC New York Order ¶ 14. By July 2000, CLECs served one million additional customers over the UNE-platform, approximately 95% of them residential. See Verizon Taylor Decl. ¶ 21. At the time of the second Texas application, approximately 244,000 lines were served through the UNE-platform. FCC Texas Order ¶ 5. By September 2000, CLECs in Texas served 569,000 customers over the UNE-platform. SBC Kansas/Oklahoma Smith/Johnson Aff. ¶ 41.

27. See Verizon Guerard/Canny Decl. Attach. E at 38 (2-wire x-DSL services: PR 2-01 & PR 2-02).

28. RCN's allegation that Verizon is not providing nondiscriminatory access to its poles, RCN Comments at 28-35, deserves careful attention because the alleged failure, if true, could substantially delay the emergence of an additional facilities-based provider. The Department has not been able to fully evaluate these allegations because Verizon did not address them fully in its application. See Verizon Brief at 34-35.

29. The New York Order stated that the Commission would "examine carefully" the state-adopted performance standards measuring the average provisioning interval, the number of missed installation appointments, and the applicant's maintenance and repair functions in future applications. FCC New York Order ¶¶ 316, 333, 335; see also FCC Texas Order ¶ 282. The D.C. Circuit Court affirmed the Commission's Order, but stated, "We . . . expect, as did the FCC, that as DSL-capable loops become a larger proportion of unbundled loops, and as performance standards are developed, checklist compliance will require a separate and comprehensive evidentiary showing with respect to the provision of DSL-capable loops." AT&T Corp. v. FCC, 220 F.3d 607, 624 (D.C. Cir. 2000) (quoting FCC New York Order ¶ 330).

30. The MA DTE submitted a detailed evaluation of Verizon's DSL performance concluding that Verizon provides nondiscriminatory access. On several issues, however, the basis for that conclusion is not clear from the MA DTE's submission. For example, is it unclear to what extent the MA DTE based its conclusions about Verizon's DSL installation timeliness and maintenance and repair performance on Verizon's studies of POTS lines. See MA DTE Evaluation at 298-99, 320. It is also unclear whether the MA DTE's conclusion about the adequacy of Verizon's missed installation appointments and maintenance and repair performance relied on newly implemented, but as yet unproven, process improvements including the enhanced capability of Verizon's mechanized database, new cooperative testing procedures, and recently ordered (but not yet tariffed) substitutes for copper facilities. See MA DTE Evaluation at 309-10, 315. In addition, the Department does not know whether the MA DTE's conclusions on Verizon's missed installation appointments performance were based, in part, on the misconception that Verizon retail does not provide the largest share of DSL loops in Massachusetts. See MA DTE Evaluation at 307 n.965. Further, the Department is uncertain how much weight the MA DTE gave to its finding that CLECs did not respond to Verizon's August 2000 assertions that CLECs were accepting non-working loops when it appears that the remaining opportunity for comment may have been limited to oral argument and that CLECs have disputed Verizon's assertion in their initial comments to this Commission. See MA DTE Evaluation at 312; Rhythms Comments at 32-33; Covad Comments at 51-52. The Department is also uncertain whether the MA DTE concluded that CLEC practices had distorted Verizon's current performance data (for loop installations and maintenance and repair) solely on the basis of CLEC statements in December 1999 (before the DSL joint testing procedures were fully implemented), or whether there is more recent evidence of those CLEC practices. See MA DTE Evaluation at 313-14, 320. Finally, it is unclear how the MA DTE will be able to effectively monitor Verizon's future performance on missed installation appointments without having an established measurement method in place. See MA DTE Evaluation at 307-08. It is our hope that the MA DTE can provide further clarification on these issues.

31. FCC New York Order ¶ 330 (noting "the critical importance of the provisioning of DSL loops to the development of the advanced service marketplace"); FCC Texas Order ¶ 282.

32. FCC Texas Order ¶ 282; FCC Louisiana II Order ¶ 198; FCC Michigan Order ¶¶ 141-142.

33. For July, 2000 the reported difference in installation intervals offered (PR 1-02) is about one day. The difference in average completion intervals (PR 2-02) is about 1.2 days. The specified interval for provisioning DSL loops is six days, and the metric measuring compliance with this standard (PR 3-10) shows that Verizon completed 83.12 percent of its own orders on time but only 51.45 percent of its competitors' orders on time. Verizon Guerard/Canny Decl. Attach. E at 38.

34. The performance measure that is designed to monitor installation quality is the percent of troubles reported within 30 days of installation (PR 6-01). For July, this measure shows that Verizon's retail rate was only 2.97 percent, while the corresponding rate for CLECs was 8.46 percent. Verizon Guerard/Canny Decl. Attach. E at 38.

35. Verizon's performance in providing repair services to its DSL competitors' loops is measured by the metric groups titled "Missed Repair Appointments" and "Trouble Duration Intervals." Missed Repair Appointments in July (MR 3-01) for the CLECs was 19.19 percent, compared to 16.62 percent for retail. Verizon Guerard/Canny Decl. Attach. E at 39. Mean Time to Repair - Loop Trouble (MR 4-02) was 26.58 hours, compared to 49.78 hours for the CLECs. Id. Mean Time to Repair - Total (MR 4-01) was 45.37 hours for the CLECs and 24.93 hours for retail, suggesting almost a day's difference in repairing DSL loops. Id. The percent of CLEC customers out of service for more than twenty four hours (MR 4-08) was 51.05 percent for the CLECs and 37.38 percent for Verizon customers. Id.

36. Verizon Brief at 24; Verizon Lacouture/Ruesterholz Decl. ¶ 100; Verizon Guerard/Canny Decl. ¶ 78.

37. Verizon Brief at 24; Verizon Lacouture/Ruesterholz Decl. ¶ 100; Verizon Guerard/Canny Decl. ¶ 78.

38. Verizon Guerard/Canny Declaration ¶¶ 79-80 & Attach. K. The MA DTE finds Verizon's analysis persuasive but states that since the study was not presented in the 271 Docket, it would not comment on its substance. MA DTE Evaluation at 300 n.947, 308.

39. See, e.g., Covad Comments at 13; Rhythms Comments at 33.

40. Verizon Brief at 25; Verizon Lacouture/Ruesterholz Decl. ¶ 106.

41. Verizon Lacouture/Ruesterholz Decl. ¶¶ 73-74 & Attach. G (discussing the effect of not accepting weekend repair appointments on the UNE POTS repair metrics).

42. Verizon Lacouture/Ruesterholz Decl. ¶ 106.

43. Rhythms Comments at 31-32 (noting that Rhythms does not decline weekend repair appointments); see also Covad Comments at 20-22 (stating that Verizon adds to the "no access" problem by assigning "all day" appointment windows); Network Access Solutions Comments at 3-4 (same).

44. Rhythms Comments at 32.

45. Verizon Brief at 25-26; Verizon Lacouture/Ruesterholz Decl. ¶¶ 103-105. The CLECs deny that they are improperly accepting loops and say that it would be foolish for them not to have the loops installed properly. Rhythms Comments at 32-33; Covad Comments at 18. Several CLECs argue that the fault lies with Verizon. Digital Broadband alleges that some Verizon DSL loops pass initial testing, but not subsequent testing. ALTS Comments at 40. The change in result, Digital Broadband claims, is due to Verizon's post-installation alteration of loops. ALTS McMillan Decl. ¶ 10. Covad argues that if Verizon does not test the loop at the NID, a non-working loop could pass initial acceptance testing. Covad Comments at 19. Network Access Solutions claims that Verizon has declined to engage in cooperative loop testing. Network Access Solutions Comments at 6.

46. Verizon Lacouture/Ruesterholz Decl. ¶ 78 & Attach. I (discussing the effect of failure to isolate troubles on UNE POTS repair metrics).

47. Verizon Lacouture/Ruesterholz Decl. ¶ 78.

48. Id.

49. Although it is not directly relevant to our analysis here, we note that in some contexts, a standard of parity or nondiscrimination would not adequately protect competition. When a CLEC offers a service that competes against a different (but substitutable) incumbent local exchange carrier ("ILEC") service, the ILEC may be able to constrain such competition by providing inferior access to inputs needed by the CLEC, even if it provides such inputs in a "nondiscriminatory" manner.

50. See FCC Michigan Order ¶ 44 ("[T]he BOC applicant retains at all times the ultimate burden of proof that its application satisfies section 271.").

51. Rhythms states that several of the performance measure issues that Verizon raises in this proceeding have not been previously brought to the state regulators responsible for the performance reporting measures. Rhythms Comments at 28; see also FCC New York Order ¶ 326 ("The absence of a New York performance benchmark or Commission reconciliation of conflicting data claims makes it difficult for this Commission to decide between the competing statistics.").

52. See, e.g., DOJ Louisiana II Evaluation at 38; FCC New York Order ¶¶ 11-12.

53. DOJ Texas I Evaluation at 5 ("Meaningful metrics require clear definitions that will allow measurement of activities or processes in a way that has real-world, practical significance.").

54. Rhythms Comments at 29-30 (quoting KPMG Technical Session Tr. 5185-89).

55. Id. at 28; Covad Comments at 13.

56. Rhythms Comments at 28.

57. Id. at 33 (stating that the lack of CLEC-specific data prevents CLECs from investigating Verizon's allegations of bad CLEC behavior).

58. See Verizon Guerard/Canny Decl. Attach. A at 7; id. Attach. E. Performance data covering 23 CLEC line sharing orders provisioned by Verizon in New York during July 2000 were submitted with this application. Id. Attach. N. However, significant numbers of line sharing orders have since been placed and Verizon does not state when regular performance reports will be available for New York and Massachusetts.

59. Rhythms Comments at 36; see also Rhythms Williams Decl. ¶¶ 36-39 (discussing shortcomings of Verizon's pre-wiring of central offices and updating of cable and pair inventory system).

60. KPMG did not test Verizon's provision of line shared DSL loops. ALTS Comments at 41.

61. FCC New York Order ¶ 334.

62. AT&T's "preferred strategy for entering local markets is through the use of its own facilities," and "[b]ecause AT&T owns a significant amount of cable plant in Massachusetts, it is thus focusing its efforts to provide local telecommunications service to residential customers in Massachusetts on the cable facilities it owns in that State." AT&T Comments at 9.

63. See AT&T Taps Bell Atlantic for Local Residential Service in N.Y., Wash. Telecom Newswire, Apr. 21, 1999, available in 1999 WL 7295977.

64. FCC New York Order ¶ 14 (citation omitted); Verizon Taylor Decl. Attach. A at 1.

65. See Michael McDonald, Rivals Swipe Verizon's Residential Users, Crain's N.Y. Bus., Oct. 16, 2000, available in 2000 WL 9441651.

66. Ascent Comments at 2; WorldCom Comments at 31; Z-Tel Comments at 3; see also AT&T Comments at 3; MA AG Comments at 3.

67. First, there has been a disparity between the prices of UNEs in Massachusetts and the prices of those same elements in the other states. While the prices of UNEs may well differ somewhat from state to state, as each state commission sets UNE prices independently, there is no obvious reason for the magnitude of the disparity between Massachusetts and other states in the Verizon region. Second, CLECs have raised several facially reasonable arguments suggesting that the UNE rates were incorrectly calculated in the MA DTE's 1996 order, the most striking of which concerns the failure to incorporate the initial switch vendor discount. See, e.g., WorldCom Comments at 12-18. Third, in July 2000, Verizon reduced rates for UNE-platform used to serve residential lines. See Verizon Mudge Decl. Attach. A Ex. 1. We are not aware of evidence that the cost of these elements differs depending on whether the end user is a business customer or a residential customer, and the price differential could be interpreted as evidence that the UNE rates for business platform lines at that time were not cost based.

68. Letter from Robert Mudge, Verizon President-Massachusetts, to the MA DTE dated Oct. 13, 2000, Ex Parte Submission to the FCC, CC Docket No. 00-176 (Oct. 13, 2000); MA DTE Evaluation at 213.

69.

71. This filing does not appear to accord with the Commission's previous statement "that a section 271 application, as originally filed, will include all of the factual evidence on which the applicant would have the Commission rely in making its findings thereon." FCC Michigan Order ¶ 49; accord FCC New York Order ¶ 35.

70. MA DTE Evaluation at 213.

71. Letter from Robert Mudge, Verizon President-Massachusetts, to the MA DTE dated Oct. 13, 2000, Ex Parte Submission to the FCC, CC Docket No. 00-176 (Oct. 13, 2000).

72. In light of the limited information available to the Department at this time, as well as the Commission's greater experience with rate-making issues, the Department will not independently attempt to assess whether the prices in Verizon's recent tariff filing are appropriately cost based.

73. KPMG Final Report at 5. KPMG also tested Verizon's documentation, interfaces and processes for CLECs to access and use Verizon's systems. Id.

74. For example, KPMG was not asked to conduct a volume test that assumed full-scale commercial entry in Massachusetts, to conduct a volume test of LSOG 4, to strictly adhere to the military-style test philosophy, to fully retest fixes, to conduct root cause analysis of all identified problems, to volume test manual ordering processes, to fully examine the help desk, or to analyze the availability of Verizon's back-end systems when assessing the overall availability of the OSS interface.

75. Immediately after its December 1999 authorization to offer long-distance service in New York, Verizon's order processing software failed. Verizon resolved the problem after the NY PSC reallocated bill credits within the performance assurance plan and added a special provision to the plan supported by $24 million additional bill credits. Order Directing Improvements to Wholesale Service Performance, Nos. 00-C-0008 & 00-C-0009 (NY PSC Feb. 11, 2000); Order Directing Market Adjustments and Amending Performance Assurance Plan, Nos. 00-C-0008, 00-C-0009 & 99-C-0949 (NY PSC Mar. 23, 2000). At the same time, Verizon entered into a consent decree with the Commission, which contained a $3 million voluntary contribution to the United States Treasury. Order and Consent Decree, In re Bell Atlantic-New York Authorization Under Section 271 of the Communications Act to Provide In-Region, InterLATA Service In the State of New York, No. FCC 00-92 (Mar. 9, 2000).

76. Other unexplained differences that could reduce Verizon's potential liability for poor performance under the plan include: the elimination of scoring measurements with a sample size less than ten, which may decrease the likelihood that individual CLECs will be compensated for poor performance where Verizon's overall performance on that critical measure was satisfactory; the lack of a requirement for Verizon to issue refund checks for bill credits owed to CLECs that no longer do business in Massachusetts; and a change in the domain clustering rule, which appears to weaken the potential for additional penalties when poor performance is concentrated in the pre-order, order, provisioning or maintenance domains. See Verizon Guerard/Canny Decl. Attach. C Ex. 4 at 2; Attach. C at 17, 23-25. Additionally, Verizon did not file a Massachusetts-specific Change Control Assurance Plan. Instead, it filed a copy of its New York Change Control Assurance Plan, the enforceability of which in Massachusetts is unclear. See Verizon Guerard/Canny Decl. Attach. C Ex. 9.

77. Moreover, the Massachusetts performance assurance plan does not contain the special measures for electronic data interface (EDI) notifiers that were added to the New York performance assurance plan, which leaves Massachusetts unable to identify such problems quickly. The importance of these measures is highlighted by record evidence that in Massachusetts, Verizon had difficulties returning billing completion notices in a consistent and timely manner in June and July 2000. MA DTE Evaluation at 115. Verizon asserts that the error was resolved in August 2000. Id. at 115-16. This fix, however, has not been verified by either KPMG or the MA DTE.