Southern New England Telephone Co. v. MCI Worldcom Communications, Inc.

353 F. Supp. 2d 287, 2005 U.S. Dist. LEXIS 1198, 2005 WL 217022

• Court: District Court, D. Connecticut
• Decided: January 28, 2005
• Docket: CIV.A.3:02 CV 274(SRU)
• Status: Published

Opinion

DECISION AND ORDER

UNDERHILL, District Judge.

On November 21, 2001, the Connecticut Department of Public Utility Control (“the DPUC”) issued a decision setting out the terms it required the Southern New England Telephone Company (“SNET”) and MCI WorldCom Communications, Inc., MCI Metro Access Transmission Services, Inc., and Brooks Fiber Communications of Connecticut (collectively “MCI”) to include in their proposed telecommunications interconnection agreement. SNET and MCI both challenge that decision, each arguing that several of the required terms are inconsistent with the Telecommunications Act of 1996 (“the 1996 Act” or “the Act”) and each seeking, by way of summary judgment, a ruling vacating the terms it challenges and upholding the terms its opponent challenges. The DPUC argues that all the challenged terms are permissible and must be left untouched. I conclude that all of the challenged terms are either inconsistent with federal law or arbitrary and capricious, and I remand the case to the DPUC for further proceedings.

BACKGROUND

I. The Telecommunications Act of 1996

Discarding the notion that telecommunications services are most efficiently provided by a state-regulated monopoly, the 1996 Act creates a federal regime designed to loosen the grip of incumbent local exchange carriers (“ILECs”) on the telecommunications market and to allow the entry into that market of competitive local exchange carriers (“CLECs”). Towards that end, the Act imposes a number of obligations on local exchange carriers in general and incumbent local exchange carriers in particular; three are relevant to this case.

Reciprocal Compensation. All local exchange carriers are required to “establish reciprocal compensation arrangements for the transport and termination of telecommunications.” 47 U.S.C. § 251(b)(5). In general, this requirement means that, for calls originating on one carrier’s network and terminating on another’s, the two carriers involved must arrange for the originating carrier, which typically bills the customer directly, to compensate the terminating carrier, which typically cannot bill the customer, for the use of the terminating carrier’s network.

Dialing Parity. All local exchange carriers must allow their competitors “to have nondiscriminatory access to telephone numbers, operator services, directory assistance, and directory listing, with no unreasonable dialing delays.” 47 U.S.C. § 251(b)(3).

Unbundled Access. An incumbent local exchange carrier must provide requesting competitive local exchange carriers with unbundled access to the elements of the incumbent’s telecommunications network. Unbundled access means “nondiscriminatory access to network elements on an unbundled basis at any technically feasible point on rates, terms, and conditions that are just, reasonable, and nondiseriminato-ry.” 47 U.S.C. § 251(c)(3).

ILECs and CLECs arrange to satisfy their federal obligations by entering into “interconnection agreements,” which govern all aspects of their relationship, including the issues of reciprocal compensation, dialing parity, and unbundled access. In the first instance, carriers attempt to voluntarily negotiate their interconnection agreement, but, if that fails, either party may seek compulsory arbitration by the relevant state public utility commission. 47 U.S.C. § 252(b). Once an agreement is reached, either voluntarily or through arbi tration, it is submitted to the state public utility commission, where it is reviewed for, among other things, compliance with federal law. 47 U.S.C. § 252(e). Any party aggrieved by the state commission’s decision may bring an action in federal district court. 47 U.S.C. § 252(e)(6).

II. SNET and MCI’s History

SNET is Connecticut’s ILEC. MCI is a CLEC. In January 2000, SNET and MCI began negotiating an interconnection agreement but failed to reach a consensus: MCI petitioned the DPUC for arbitration. The DPUC assigned the arbitration to one of its commissioners who issued a decision on May 15, 2001. Over SNET’s objection, the DPUC adopted the arbitrator’s decision as the commission’s final decision (“the Final Decision”).

On February 14, 2002, SNET brought this suit, in which it challenges four of the determinations made by the DPUC in the Final Decision. MCI defends the DPUC’s actions on those four issues, but claims that two other determinations in the Final Decision are incorrect. SNET denies that the two issues raised by MCI pose any problem. The DPUC maintains that its decision was correct in all material respects.

In August 2002, the DPUC voluntarily reopened proceedings in this case to consider whether the Final Decision required modification in light of changes in federal law. On November 13, 2002, the DPUC issued a ruling concluding that the Final Decision needed no change.

SNET and MCI have both moved for summary judgment on their respective claims and against the other’s claims. The DPUC opposes both motions.

STANDARD OF REVIEW

The 1996 Act does not explain how district courts are to review the determinations of state public utility commissions. Those courts that have faced the issue have uniformly concluded that a state commission’s interpretations of federal law are reviewed de novo, but its interpretations of state law and its findings of fact are reviewed under an “arbitrary and capricious” standard. See SNET v. DPUC, 285 F.Supp.2d 252, 258 (D.Conn.2003) (collecting cases); Global NAPS, Inc. v. Verizon New England Inc., 327 F.Supp.2d 290, 296 (D.Vt.2004) (collecting cases).

In reviewing an agency’s decision under the arbitrary and capricious standard, a court will uphold the agency’s decision if it can discern a rational connection between the facts found and the choice made. SNET v. DPUC, 285 F.Supp.2d at 258. The court may not, however, supply a reason for the agency’s decision that the agency itself has not given. Id.

When reviewing an agency’s interpretation of federal law, the court applies the law in effect at the time it conducts its review, even if that was not the law in effect at the time the agency made its decision. Pacific Bell v. Pac-West Telecomm, Inc., 325 F.3d 1114, 1130 n. 14 (9th Cir.2003) (citing U.S. West v. Jennings, 304 F.3d 950, 956 (9th Cir.2002)).

Though they filed motions for summary judgment, SNET and MCI effectively seek an administrative appeal of the DPUC’s determination under the standards just given. Accordingly, it is those standards, rather than thé familiar summary judgment standards, that govern this decision.

DISCUSSION

The six issues in dispute in this case relate to three of the obligations imposed by the 1996 Act: (1) reciprocal compensation, (2) dialing parity, and (3) unbundled access.

I. Reciprocal Compensation

Two of the DPUC determinations at issue in this case involve reciprocal compensation issues. SNET takes issue with the Final Decision’s determination that traffic bound for Internet Service Providers (“ISPs”) is subject to reciprocal compensation (“the ISP Issue”). MCI takes issue with the Final Decision’s determination that “foreign exchange” or “FX” traffic is not subject to reciprocal compensation (“the FX Issue”).

A. Laiv of Reciprocal Compensation

The law governing the reciprocal compensation obligations of local exchange carriers is very much a product of history and is best understood by reviewing that history.

1. First Report and Order

The 1996 Act imposes on all local exchange carriers the “duty to establish reciprocal compensation arrangements for the transport and termination of telecommunications.” 47 U.S.C. § 251(b)(5).

In its first order implementing the Act, the FCC explained more exactly what the requirement entailed. “We conclude that section 251(b)(5) reciprocal compensation obligations should apply only to traffic that originates and terminates within a local area.” First Report and Order, 1 11 F.C.C.R. 15499, 16013 ¶ 1034 (1996). According to the FCC, reciprocal compensation arrangements were only needed in the situation where two competing local carriers work together to complete a local call. Id. When two carriers complete a local call, the originating carrier will recover its costs directly from the calling customer through payment of the customer’s local subscription. The terminating carrier, however, will not be able to recover its costs because, in general, customers are not billed for receiving calls. Of course, if the competing carriers exchange approximately the same volume of traffic, then originating and terminating costs will balance. When one carrier terminates many more calls than another, however, unless reciprocal compensation applies, the terminating carrier would be subsidizing its competitor by terminating the competitor’s calls for free.

By contrast, non-local calls, either interstate or intrastate, do not need reciprocal compensation because they already have a cost-recovery mechanism in place, namely, access charges. Completion of non-local calls typically involves three carriers — the originating LEC, the inter-exchange carrier (“the IXC”), and the terminating LEC. Id. Customers are ordinarily billed a per-call distance-based rate by the IXC, which in turn compensates both the originating LEC and terminating LEC by paying an access charge for the use of each LEC’s facilities, either to originate or terminate the call. Id. Because access charges provide appropriate compensation for non-local calls, the FCC concluded that they are not governed by section 251(b)(5)’s reciprocal compensation requirement, but by the appropriate rules concerning access charges — sections 201 and 202 of the Act for interstate calls and state law for intrastate non-local calls.

Having determined that only local calls were subject to section 251(b)(5)’s reciprocal compensation regime, the only other issue the FCC had to decide was what constituted a local call, or, more specifically, when was a call considered to originate and terminate in the same “local area.” The FCC resolved that issue by concluding “state commissions have the authority to determine what geographic areas should be considered ‘local areas’ for the purpose of applying reciprocal compensation obligations under section 251(b)(5).” Id. ¶ 1035.

These two holdings — that (1) reciprocal compensation was only due for calls that originated and terminated in the same “local area” and (2) states would determine what constituted a “local area” — were codified at 47 C.F.R. § 51.701.

2. The ISP Situation

Initially ILECs were the primary beneficiaries of reciprocal compensation arrangements. They had more customers than CLECs and, as a consequence, terminated more calls, entitling them to more reciprocal compensation payments. That situation changed as customers began to use telephone calls to connect to the internet via ISPs, because the calling patterns of ISP customers are not typical of-ordinary phone subscribers. ISPs almost'never originate calls, and the calls placed to them are usually longer than a typical phone call. CLECs began to target ISPs as customers and so began terminating a great many calls and, consequently, receiving a great deal of- reciprocal compensation.

In 1999 the FCC issued a declaratory ruling addressing the treatment of ISP calls. Declaratory Ruling, 2 14 F.C.C.R. 3689 (1999). It concluded that ISP calls were interstate in nature and therefore under FCC jurisdiction, not the jurisdiction of the states. Accordingly, despite the ruling of the First Report and Order, states were not authorized to determine whether ISP-bound calls were local. The FCC then went on to conclude that, because ISP calls were jurisdictionally interstate calls, they were also not “local” calls and therefore not subject to section 251(b)(5) reciprocal compensation. The decision was appealed to the D.C. Circuit where it was vacated. Bell Atlantic Telephone Co. v. FCC, 206 F.3d 1 (D.C.Cir.2000). The D.C. Circuit did not dispute that the FCC had jurisdiction to determine whether ISP calls were subject to reciprocal compensation, but it held that the FCC had not explained why the fact that ISP calls were jurisdictionally interstate was relevant to the question whether they were “local” for reciprocal compensation purposes. Id. at 6.

3. The ISP Remand Order

After the D.C. Circuit’s decision, the FCC issued another ruling on the treatment of ISP calls, the so-called “ISP Remand Order.” ISP Remand Order, 3 16 F.C.C.R. 9151 (April 27, 2001). In that order the FCC did not simply -revisit its analysis of ISP calls but instead took the opportunity to reconsider entirely the First Report and Order’s analysis of reciprocal compensation.

The FCC explained that its prior decision to classify certain traffic as “local” was inappropriate “because the term ‘local,’ not being a statutorily defined category, is particularly susceptible to varying meanings and, significantly, is not a term used in section 251(b)(5).” ISP Remand Order, 16 F.C.C.R. at 9167 ¶ 34. Accordingly, rather than reading a “local” restric tion into section 251(b)(5), the FCC concluded that, absent some other statutory limitation, section 251(b)(5) requires “reciprocal compensation for transport and termination of all telecommunications traffic, — i.e., whenever a local exchange carrier exchanges telecommunications traffic with another carrier.” Id. at 9166 ¶ 32 (emphasis supplied). Nevertheless, said the FCC, section 251(b)(5) does not apply to all such traffic because section 251(g) of the Act creates an explicit statutory “carve-out” from reciprocal compensation. Id.

The FCC went on to interpret section 251(g) as excluding from reciprocal compensation “exchange access, information access, and exchange services for such access.” In other words, said the FCC, rather than starting with the baseline assumption that all traffic was not subject to section 251(b)(5) and then attempting to define the limits of that section (as it had done in the First Report and Order), it would now start with the assumption that all traffic was subject to reciprocal compensation unless it was part of section 251(g)’s carve-out. Id. at 9167 ¶ 34.

Applying this rationale, the FCC concluded that ISP-bound traffic was an “information service” within the meaning of section 251(g) and therefore not subject to section 251(b)(5) reciprocal compensation. The FCC also noted that, under its new analysis, interstate calls — i.e., “long-distance” calls — were also not subject to section 251(b)(5) because they qualified as “exchange access” as that term is defined in 47 U.S.C. § 153(16). Finally, the FCC reasoned that, even though section 251(g) did not explicitly cover intra state exchange access calls, it was reasonable to conclude that they too were exempt from section 251(b)(5) because “ ‘it would be incongruous to conclude that Congress was concerned about the effects of potential disruption to the interstate access charge system, but had no such concerns about the effects on analogous intrastate mechanisms.’ ” Id. at 9168 n. 66 (quoting First Report and Order, 11 F.C.C.R. at 15869).

In light of its new analysis, the FCC amended the relevant regulations to (a) replace all references to “local telecommunications traffic”, with “telecommunications traffic” and (b) clarify that section 251(b)(5)’s reference to telecommunications traffic means: “Telecommunications traffic exchanged between a LEC and a telecommunications carrier ..., except for telecommunications traffic that is interstate or intrastate exchange access, information access, or exchange services for such access.” See 66 Fed.Reg. 26800, 26806; 47 C.F.R. § 51.701 (2001).

The ISP Remand Order was appealed to the D.C. Circuit. That court again rejected the FCC’s reasoning, this time concluding that section 251(g) was merely a transitional provision and could not justify the FCC’s decision to adopt entirely new regulations concerning ISP traffic. WorldCom Inc. v. FCC, 288 F.3d 429, 433 (D.C.Cir.2002). Nevertheless, the .D.C. Circuit did not vacate the FCC’s order (or its amendments to the regulations) but merely remanded the case for further consideration. Id. at 434. There has been no further action by the FCC on this issue.

4. The Transitional Regime of the ISP Remand Order

Because the D.C. Circuit did not vacate the FCC’s ISP Remand Order, that order’s. regulation of ISP traffic remains binding.

In the ISP Remand Order, the FCC expressed its opinion that ISP traffic should, ultimately, be subject to a “bill and keep” compensation .regime, that is, a regime where carriers do not compensate one another for terminating traffic but in stead collect their costs directly from their own customers. ISP Remand Order, 16 F.C.C.R. at 9155 ¶ 6. Nevertheless, the FCC did not impose “bill and keep” as the billing structure for ISP traffic but, instead, adopted a transitional regime designed to gradually decrease reliance on carrier-to-carrier payments for recovery of costs, while moving towards a system of recovering costs through direct billing. Id. at 9156 ¶ 7.

The principal feature of the transitional regime is the setting of caps on the amount of intercarrier compensation that can be required for ISP traffic. These caps only apply if an ILEC offers to exchange section 251(b)(5) traffic as well as ISP traffic at the same rate—known as “mirroring.” Id. at 9157- ¶ 8, 9193 ¶ 89. If an ILEC refuses to mirror, then the order requires it to exchange ISP traffic at whatever rate is already set for section 251(b)(5) traffic. Id. ■ The purpose of this provision is to “ensure[] that incumbent LECs will pay the same rates for ISP-bound traffic that they receive for section 251(b)(5) traffic.” Id. at 9194 ¶ 89. Nevertheless, even though the rates paid for ISP compensation are the same as the rates paid for reciprocal compensation, the ILEC is not paying “reciprocal compensation” for ISP traffic, i.e., section 251(b)(5) compensation, but is instead paying “ISP compensation,” a category that is sui gen-eris and, according to the FCC, is governed by section 201 of the Act. Id. at 9153 ¶1.

The ISP Remand Order’s, transitional regime only applies from its effective date, which is June 14, 2001. Id. at 9189 ¶ 82. The order does not override the reciprocal compensation terms contained in interconnection agreement made prior to the effective date. Id. It also does not preempt any existing state regulations to the extent they require compensation for ISP traffic exchanged prior to the effective date. Id. It simply prohibits states from addressing the issue of how to categorize ISP traffic when arbitrating or approving interconnection agreements or issuing regulations at any time after June 14, 2001. Id.

B. The ISP Issue

SNET takes issue with the following determination in the Final Decision:

Internet traffic is subject to local reciprocal compensation under the terms of this Agreement. The Parties agree that Internet Calls are not subject to reciprocal compensation under this Appendix nor under the Act. The Parties agree to make reasonable efforts to identify telephone numbers assigned to Internet Service Provers (ISPs) to each other for purposes of excluding the minutes associated with ISP traffic from compensation. The Parties agree this traffic will be meet point billed if and when the current FCC exemption is removed or modified to allow such billing.

Final Decision at 28. I conclude that this determination is both (a) arbitrary and capricious and (b) inconsistent with federal law.

1. Internal Inconsistency

An agency’s decision is arbitrary and capricious when the court cannot discern a’ rational connection between the facts found and the choice made. SNET v. DPUC, 285 F.Supp.2d at 258. Here, I cannot even discern the choice made, let alone a rational connection between that choice and the facts found. The Final Decision starts by saying “Internet traffic is subject to local reciprocal compensation,” but two sentences later declares that the parties shall “make reasonable efforts to identify telephone numbers assigned to Internet Services Providers (ISPs) ... for purposes of excluding the minutes associ ated with ISP traffic from compensation.” 4 In other words, the decision appears to say that ISP traffic both is and is not subject to reciprocal compensation. This irreconcilable language must be clarified.

2. Conflict with the ISP Remand Order

In spite of the contradictory language of the Final Decision, all parties agree that the DPUC intended to subject ISP traffic to reciprocal compensation. That is flatly prohibited by the ISP Remand Order, which explicitly held that ISP traffic is not subject to reciprocal compensation under section 251(b)(5), but is instead subject to its own form of compensation under section 201. ISP Remand Order, 16 F.C.C.R. at 9189 ¶ 82. The DPUC had no authority to issue a requirement to the contrary. “Because we now exercise our authority under Section 201 to determine the appropriate intercar-rier compensation for ISP-bound traffic, ... state commissions will no longer have authority to address this issue.” Id. The Final Decision was issued after the effective date of the ISP Remand Order and, therefore, the DPUC was not permitted to stray from the dictates of that order, which it did by declaring ISP traffic subject to reciprocal compensation.

This ruling does not mean that on remand SNET will be excused from paying any compensation for ISP traffic or that the DPUC will have no say in setting the rate for such compensation. On the contrary, under the transitional regime established by the FCC, though the DPUC may not declare ISP traffic subject to “reciprocal compensation” under section 251(b)(5), it may still set rates for ISP compensation so long as those rates do not exceed the caps imposed by the ISP Remand Order. 5 Those capped rates will only apply to SNET if it also agrees to exchange all section 251(b)(5) traffic at that rate, i.e., if it mirrors. If SNET refuses to mirror, then it must pay the same rate for ISP traffic as it pays for section 251(b)(5) traffic. 6

C. The FX Issue

Foreign Exchange—or FX—-service allows a customer to receive a ten-digit phone number with the first six digits— called the NPA/NXX number—different from those that would ordinarily be assigned to someone in that customer’s geographic area. For example, a customer in Hartford might request a phone number that would typically be assigned to a customer residing in New Haven. The result is that FX customers can receive calls from geographically remote customers without those remote customers incurring a distance-based toll. In the example just given, the Hartford customer could receive calls from New Haven customers without those customers incurring a distance-based toll. 7

The DPUC did not directly address the issue of compensation for FX traffic in the Final Decision, because the parties agreed that they would incorporate into their agreement whatever decision the DPUC reached in a separate docket — Docket No. 01-01-29, DPUC Investigation of the Payment of Mutual Compensation for Local Calls Carried, Over Foreign Exchange Service Facilities, (Jan. 30, 2002) (“the FX Decision”). Nevertheless, in late 2002, when the DPUC agreed to reopen the Final Decision, it heard argument on, among other things, the issue of FX traffic. Relying on its holding in the FX Decision, the DPUC ruled' as follows:'

Regarding FX service, the Department continues to believe that it is an interex-change service which is provided from an exchange that is different from the one that normally serves the area in which the subscriber is located .... While telephone calls made to FX service subscribers allow their customers to call without incurring toll charges, for purposes of payment of mutual compen.sation, such calls are non-local in nature; and therefore, they are not subject to mutual compensation.

Nov. 13, 2002 Decision at 3. This holding is contrary to federal law.

1. The FX Decision

The DPUC’s FX Decision served as the basis for its ruling after the reopening of the Final Decision in this case. The FX Decision was issued in response to a request by SNET that, among other things, the DPUC order CLECs to stop billing SNET for reciprocal compensation for FX calls and, instead, pay SNET originating access for such calls, just- as they would do for other intrastate toll calls.

Completely ignoring the change in the FCC’s interpretation of section 251(b)(5) set out in the ISP Remand Order and 47 C.F.R. § 51.701, the DPUC concluded that the question before it was “what geographic areas should be considered ‘local areas’ for reciprocal compensation obligations under 47 U.S.C. § 251(b)(5).” FX Decision, 2002 PUC LEXIS 23, Section IV.B. Undertaking that analysis, the DPUC concluded that only calls that originated and terminated within the same geographic local calling area- — -i.e., the geographic area in which ordinary, non-FX customers could call one another without incurring a distance-based charge — were “local” and therefore subject to section 251(b)(5) reciprocal compensation. Consequently, concluded the DPUC, FX calls were not subject to reciprocal compensation, and the CLECs must refund the compensation payments they had received for such calls — a process described as a “true-up.”

Nevertheless, the DPUC went on to hold that, although FX calls were not local calls, they were also not toll calls, and therefore it was inappropriate for CLECs to be required to pay access charges on those calls, because “[i]n the opinion of the Department, imposition of access charges on these calls would clearly not be in the public interest due to the level of customer confusion that would most likely be generated as well as the costs incurred by the CLECs in resolving those complaints.” Id., Section IV.C.

2. Waiver

As a preliminary matter, SNET- argues that MCI has waived its right to raise the issue of FX traffic in this proceeding. According to SNET, because the parties agreed to incorporate the holding of the FX Decision into their interconnection agreement, any argument MCI wished to make had to be made in the prior administrative proceeding. That contention is wrong for two reasons.

First, MCI was not “aggrieved” with respect to its agreement with SNET at the time the FX Decision was issued. . The FX Decision had no effect on the interconnection agreement at issue in this case until after the DPUC approved the its incorporation into that agreement. Accordingly, MCI was not permitted to dispute the FX issue until now. 8

. Second, regardless of whether the FX Issue was addressed in the DPUC’s Final Decision, it was explicitly raised, argued, and addressed in the DPUC’s decision on reopening. There is no reason why MCI cannot seek review of that decision in this proceeding.

3. Conflict with Federal Law

The DPUC misapplied federal law in the FX Decision. In that decision, the DPUC relied on the interpretation of section 251(b)(5) contained in the First Report and Order and not the more recent, binding interpretation contained in the ISP Remand Order and its implementing regulations. In the FX Decision, the DPUC set out its definition of a “local calling area” and concluded that because FX traffic did not fall within that definition it was not subject to reciprocal compensation. What the DPUC should have done was determine whether FX traffic is “interstate or intrastate exchange access, information access,, or exchange services for such access.” 47 C.F.R. § 51.701. If it is, then it is not subject to reciprocal compensation; if it is not, then it is subject to reciprocal compensation.

In many situations, adherence to the FCC’s old interpretation rather than the current interpretation would be harmless. Under both FCC interpretations the states are given considerable discretion to determine whether intrastate telecommunica tions traffic is covered by section 251(b)(5)’s reciprocal compensation requirement. The difference is that under the old interpretation such traffic was not subject to reciprocal compensation unless a state defined it as local, whereas under the current interpretation such traffic is subject to reciprocal compensation unless a state declares it to be intrastate exchange access. This distinction may often have little practical effect because, typically, telecommunications traffic that under the old regime was “local,” under the new regime is not exchange access, and, conversely, telecommunications traffic that was not “local” is either interstate or intrastate exchange access. The problem with the DPUC’s FX Decision is that it creates a classification of telecommunications traffic that, according to the DPUC, is not exchange access, but is also not subject to section 251(b)(5) reciprocal compensation. Federal law does not permit such a result.

“Exchange access” means “the offering of access to telephone exchange services or facilities for the purpose of origination or termination of telephone toll services.” 47 U.S.C. § 153(16) (emphasis supplied). “Telephone toll service” means “telephone service between stations in different exchange areas for which there is made a separate charge not included in contracts with subscribers for exchange service.” 47 U.S.C. § 153(48) (emphasis supplied). In short, as a matter of federal law, only telecommunications traffic that is either information access or toll service between exchanges (or exchange services for such access 9 ) is exempt from the reciprocal compensation of section 251(b)(5). 10

No one argues that FX traffic is an information access service. It is also difficult to see how FX traffic could be exchange access. FX service is not a toll service; 11 FX calls are billed exactly the same as local calls. Moreover, the DPUC has specifically declined to subject FX calls to access charges. Consequently, because it appears that FX traffic is not information access or exchange access (or exchange services for such access), it is not exempt from the reciprocal compensation requirement of section 251(b)(5), and the DPUC’s determination to the contrary is in conflict with federal law. 12

Finally, it is worth emphasizing that, unless and until the FCC changes its rules, ISP-bound traffic is never subject to section 251(b)(5) reciprocal compensation, regardless of whether or not it involves the use of FX service. The ISP Remand Order covers all ISP-bound traffic, without exception. See Global NAPS, Inc., 327 F.Supp.2d at 300 (“The FCC did not distinguish traffic between an ISP and its customer in different local calling areas from traffic between an ISP and its customer in the same local calling area.”). Consequently, because the effective date of the ISP Remand Order predates the Final Decision, any decision the DPUC makes on remand regarding FX traffic will have no direct effect on ISP-bound traffic.

II. Dialing Parity

SNET takes issue with the DPUC’s decision regarding access to SNET’s Directory Access Listing or “DAL,” a database containing customer names, addresses, and telephone numbers. The DPUC made the following ruling regarding DAL:

At no time will SNET offer the DAL to a third party at rates lower than those offered to [MCI] or under terms and conditions better than those offered to [MCI]. If SNET offers (through either tariff, contract, agreement or other arrangement) any of the DAL services available under this Agreement to third parties at more favorable rates, terms or conditions, SNET shall immediately offer such rates, terms or conditions to [MCI], effective as of the date SNET first made such more favorable rates, terms or conditions available to a third party.

Final Decision at 2-3. I conclude that this ruling is arbitrary and capricious

Section 252(i) of the 1996 Act imposes the mandatory requirement that:

A local exchange carrier shall make available any interconnection, service, or network element provided under an agreement approved under this section to which it is a party to any other requesting telecommunications carrier upon the same terms and conditions as those provided in the agreement.

That rule applies regardless of whether it is expressly incorporated into the parties’ interconnection agreement. If, as SNET suggests, the DPUC’s ruling is meant as an implementation of section 252(i), then it goes too far. Section 252(i) only applies to agreements “approved under this section,” namely, interconnection agreements between local exchange carriers. The DPUC’s ruling, by contrast, applies to SNET’s agreements with any third party. 13

The DPUC, however, maintains that its decision was not an implementation of section 252(i), but rather an attempt to enforce the “dialing parity” requirements of section 251(b)(3) of the Act. That section imposes on all local exchange carriers the “duty to permit all such providers to have nondiscriminatory access to ... directory listing, with no unreasonable dialing delays.”

Without a doubt, when issuing an arbitration ruling, the DPUC is entitled to impose appropriate conditions upon the parties in order to implement the requirements of, among other things, section 251(b)(3). See 47 U.S.C. § 252(b)(4)(C). Nevertheless, the “dialing parity” requirement of section 251(b)(3) applies only to “providers of telephone exchange service and telephone toll service,” a point the FCC made abundantly clear in its 2001 ruling on Directory Listings. Directory Listing Order, 14 16 F.C.C.R. 2736, 2758 ¶ 50 (2001) (“under section 251(b)(3) of the Act competing directory assistance providers that provide telephone exchange or telephone toll service are entitled to nondiscriminatory access to all directory assistance”) (emphasis supplied). In that ruling, the FCC clarified which competing DAL providers qualified as providers of telephone exchange service. Nothing in that ruling held that the nondiscrimination provisions of section 251(b)(3) applied to anyone other than providers of telephone exchange or toll service. Accordingly, because the DPUC’s regulation of SNET’s agreement with any third-party purchaser of DAL services does not appear rationally related to the goals of the Act 15 it is arbitrary and capricious.

III. Unbundled Elements

Three of the issues decided by the DPUC concern SNET’s obligation to provide MCI with access to unbundled network elements (“UNEs”) at cost-based prices. First, SNET takes issue with the DPUC’s general requirement that SNET, at MCI’s request, combine UNEs not ordinarily combined in SNET’s own network (“the Combination Issue”). Second, both SNET and MCI take issue with the DPUC’s decision regarding specific UNEs. SNET argues that the Final Decision improperly requires it to provide the Operator Services and Directory Access (“OS/ DA”) UNEs using MCI’s preferred signaling protocol (“the OS/DA Issue”). MCI argues that the Pinal Decision improperly restricts the ways it can use SNET’s Calling Name (“CNAM”) database and Line Information Database (“LIDB”) UNEs (“the CNAM/LIDB Issue”).

A. FCC Unbundling Regulations

Like the rules governing ISP-bound traffic, the rules governing the unbundling obligations of local exchange carriers are very much creatures of history. In the eight years since the enactment of the 1996 Act, the FCC’s attempts to implement the unbundling requirements of section 251(c)(3) of the Act have been partially rejected three times, twice by the D.C. Circuit and once by the Supreme Court. In 1996, the FCC issued the First Report and Order, only to have that order rejected in part by the Supreme Court in 1999. See AT & T Corp. v. Iowa Utilities Board, 525 U.S. 366, 389-90, 119 S.Ct. 721, 142 L.Ed.2d 835 (1999). The FCC tried again in 1999, issuing the so-called “UNE Remand Order,” 16 only to have a large part of that order rejected by the D.C. Circuit in 2002. See United States Telecom Association v. FCC, 290 F.3d 415 (D.C.Cir.2002). In 2003, the FCC issued its “Triennial Review Order,” 17 which set out ILECs’ unbundling obligations. Again the D.C. Circuit rejected a substantial part of that order. United States Telecom Association v. FCC, 359 F.3d 554 (D.C.Cir.2004). Most recently, on December 15, 2004, the FCC announced that it had adopted yet another set of rules to replace those rejected by the D.C. Circuit. See “FCC Adopts New Rules For Network Unbundling Obligations of Incumbent Local Phone Carriers,” available at http://hraunfoss.fcc.gov/edocs_public/atta-chmatch/DOC-255344Al.pdf (Dec. . 15, 2004). The current rules have not yet been published, though the FCC’s announcement gives a general outline of their content.

The end result of this process is that the current state of the rules governing un-bundling is unclear. Some issues are governed by surviving parts of the FCC’s prior orders and regulations, while some issues remain unresolved, though they will apparently be addressed by the forthcoming FCC rules.

B. The Combination Issue

SNET believes that the Final Decision’s requirement that SNET combine UNEs at MCI’s request is impermissibly broad. The text in dispute is as follows:

Consistent with the terms and conditions in this Appendix, the Act and regulations thereunder, SNET shall offer each Unbundled Network Element individually and in ordinary combination with any other Unbundled Network Element or Unbundled Network Elements in order to permit SNET to combine such Unbundled Network Element or Unbundled Network Elements with another Unbundled Network Element or other Unbundled Network Elements obtained from SNET or with network components provided by itself or by third parties to provide Telecommunications Services to its customers. SNET will combine Unbundled Network Elements made available by SNET with other contiguous SNET Unbundled Network Elements.

******

At [MCI’s] request, SNET shall cross connect or combine Unbundled Network Elements either manually or where technically feasible or available electronically.

Final Decision at 30-31. I conclude that this language is broader than federal law permits.

1. Rules on Combinations

The FCC’s pronouncements on ILECs’ obligations to combine network elements have survived the appellate process relatively unscathed. The 1996 Act states that: “An incumbent local exchange carrier shall provide such unbundled network elements in a manner that allows requesting carriers to combine such elements in order to provide such telecommunications service.” 47 U.S.C. § 251(c)(3). The FCC’s implementing regulations further specify that:

Upon request, an incumbent LEC shall perform the functions necessary to combine unbundled network elements in any manner, even if those elements are not ordinarily combined in the incumbent LEC’s network, proved that such combination: (1) is technically feasible; and (2) would not undermine the ability of other carriers to obtain access to unbundled network elements or to interconnect with the incumbent LEC’s network.

47 C.F.R. § 51.315(c).

In Verizon Communications, Inc. v. FCC, 535 U.S. 467, 122 S.Ct. 1646, 152 L.Ed.2d 701 (2002), the Supreme Court faced the question whether the FCC regulations just quoted were a permissible exercise of the Commission’s authority; in other words, whether the regulations passed the test of Chevron U.S.A. Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). Under the first step of Chevron’s two part test, the Court turned to the language of the statute and concluded that section 251(c)(3) left open whether the ILEC or CLEC was required to do the work of combining. Verizon, 535 U.S. at 534, 122 S.Ct. 1646. Accordingly, the Court held that the regulations did not conflict with the plain language of the statute and went on to the second Chevron step — determining whether the regulations were reasonable.

In deciding whether the regulations were reasonable, the Supreme Court made a number of preliminary observations. First, the Court noted that the FCC had clarified the regulations in the First Report and Order by stating, “[i]f the carrier is unable to combine the elements, the incumbent must do so.” Id. at 535 (citing First Report and Order ¶ 294). Second, the Court interpreted the regulations to require only that the incumbent “perform the functions necessary to combine, not necessarily to complete the actual combination.” Id. Third, the Court pointed out that, as with all unbundling, the entrant must pay a reasonable cost-based fee for the incumbent’s services. Id. Fourth, the Court noted that the regulations themselves only required the combinations to be done when technically feasible and nondiscriminatory. Id. at 535-36.

Given these limitations on the incumbent’s obligation, the Court had little trouble concluding that the regulations were reasonable under Chevron.

In sum, what we have are rules that say an incumbent shall, for payment, perform the functions necessary to combine network elements to put a competing carrier on an equal footing with the incumbent when the requesting carrier is unable to combine, when it would not place the incumbent at a disadvantage in operating its own network, and when it would not place other competing carriers at a competitive disadvantage. This duty is consistent with the Act’s goals of competition and nondiscrimination, and imposing it is a sensible way to reach the result the statute requires.

Id. at 538. In the Triennial Review Order, the FCC briefly addressed the issue of UNE combinations, but did not dispute the Supreme Courts interpretation of the regulations. (“We affirm our existing rules regarding UNE combinations.”). Triennial Review Order, 18 F.C.C.R. at 17339 ¶ 573.

Accordingly, I conclude that existing FCC regulations require ILECs, upon the request of a CLEC, to perform the functions necessary to combine UNEs, when the requesting carrier cannot perform those functions, provided that such a combination is technically feasible and will not undermine the ability of other CLECs to access UNEs or achieve interconnection. As with all UNEs, the ILECs are permit ted to charge a cost-based fee for the combination.

2. Conflict with Federal Law

SNET argues that the DPUC’s decision is inconsistent with FCC regulations, as interpreted by the Supreme Court in Verizon. I agree.

First, the DPUC’s decision states, “at [MCI’s] request, SNET shall cross connect or combine Unbundled Network Elements.” That is inconsistent with Verizon’s interpretation of FCC regulations, under which SNET need only perform the functions necessary to combine, as opposed to completing the actual combination. Moreover even this obligation only applies when MCI is unable to perform those functions on its own. 18

Second, the sentence, “At [MCI’s] request, SNET shall cross connect or combine Unbundled Network Elements either manually or where technically feasible or available electronically,” makes it unclear whether technical feasibility is an absolute requirement or merely one of several alternative requirements. The sentence would be clearer, and more obviously in line with federal law, if it were written with the following punctuation: “At [MCI’s] request, SNET shall cross connect or combine Unbundled Network Elements either manually or, where technically feasible or available, electronically.” Even so punctuated, it is still not clear what the DPUC means by “technically feasible or available.” The regulations do not make availability an alternative to technical feasibility.

Third, the DPUC decision does not include the necessary limitation that combination is only required when it “[w]ould not undermine the ability of other carriers to obtain access to unbundled network elements or to interconnect with the incumbent LEC’s network.” 47 C.F.R. § 51.315(c)(2).

Because the DPUC decision fails to clearly impose the limitations required by federal law, its ruling on the Combination Issue is in error. On remand, the DPUC should modify its decision to make clear that SNET has an obligation to perform the functions necessary to allow MCI to combine elements when MCI cannot perform those functions, but SNET is not required to make such a combination if SNET demonstrates 19 that the requested combination is either not technically feasible or would undermine the ability of other carriers to obtain access to unbundled network elements or to interconnect with SNET’s network.

C. Specific Elements

Both SNET and MCI raise issues with the DPUC’s rulings on unbundling of specific elements, namely, OS/DA services and CNAM7LIDB databases. SNET believes the Final Decision’s OS/DA unbundling requirement goes too far. MCI believes the CNAM/LIBD unbundling requirement does not go far enough.

1. Unbundling Rules for Telecommunications-Related Services

The Triennial Review Order significantly reduced the circumstances under which various telecommunieations-related services need to be offered as UNEs. In the Triennial Review Order, the FCC concluded that the only time an ILEC is ever required to unbundle most of its telecommunications-related services—including, relevant to this case, OS/DA and CNAM7 LIDB—is when it is also required to offer local switching as an unbundled element. Triennial Review Order, 18 F.C.C.R. at 17328, 17333 ¶¶551, 560. The circumstances under which local switching is to be unbundled, according to the Triennial Review Order, depend on determinations to be made by state public utility commissions. Id., at 17237 ¶ 419.

In light of the D.C. Circuit’s invalidation of part of the Triennial Review Order, the FCC has recently stated its intention to completely eliminate ILECs’ obligation to provide unbundled access to local switching. That rule will not take effect immediately, but will be preceded by a twelvemonth transitional period. As a result, it appears likely that, once the transitional period is over, there will be no circumstances under which ILECs are required to provide unbundled access to OS/DA or CNAM/LIDB.

Additionally, in its announcement, the FCC indicated that it has adopted a new rule that will, among other things, prohibit the use of UNEs for the provision of telecommunications services in the long-distance markets. Accordingly, even if OS/DA and CNAM/LIDB are required to be unbundled, it appears that their use will be limited.

2. The OS/DA Issue

Operator Services assist customers in billing and completing calls. Directory Assistance allows customers to retrieve other customers’ telephone numbers. The Final Decision contains the following provisions, regarding OS/DA:

SNET shall offer unbundled operator services where SNET does not provide [MCI] with customized routing or a comparable signaling protocol (e.g., Feature Group D).

* * Ht * * *

SNET shall offer unbundled DA Services where SNET does not provide [MCI] with customized routing or a comparable signaling protocol (e.g., Feature Group D).

* :|: * * :]: *

[MCI] designates Feature Group D as its compatible signaling protocol for Operator Services and Directory Assistance (OS/DA). Currently SNET does not provide Feature Group D signaling protocol. Until SNET offers Feature Group D signaling protocol, or [MCI] designates another signaling protocol which SNET provides, SNET will offer operator services and directory assistance as an unbundled network element in accordance with Section 6.0. At such time [as] SNET offers Feature Group D signaling protocol for customized routing, the parties will negotiate an amendment to this section based on [MCI]’s requirements.

Final Decision at 12-13. SNET contends that requiring it to provide OS/DA as a UNE until it provides a compatible signaling protocol is impermissibly restrictive.

It appears that the combined effect of the Triennial Review Order and the FCC’s soon-to-be-released new order will be to eliminate the need for SNET to provide OS/DA as a UNE. Accordingly, at the very least, it makes sense to vacate the DPUC’s decision to allow it to consider the effect that those orders — both of which post-date the Final Decision — have on the Final Decision. Nevertheless, because it is possible that at least during the transitional period SNET may still have to provide OS/DA, it is worth briefly addressing some problems with the DPUC’s current ruling.

The current regulations governing OS/DA read as follows:

An incumbent LEC shall provide a requesting telecommunications carrier nondiscriminatory access to operator services and directory assistance on an unbundled basis, in accordance with section 251(c)(3) of the Act and this part, to the extent that local circuit switching is required to be unbundled by a state commission, if the incumbent LEC does not provide that requesting telecommunications carrier with customized routing, or a compatible signaling protocol, necessary to use either a competing provider’s operator services and directory assistance platform or the requesting telecommunications carrier’s own platform.

47 C.F.R. § 51.319(d)(4)(ii). The quoted regulation is the result of the Triennial Review Order, where the FCC revisited the question of unbundling OS/DA. In that order the FCC explained that “we require incumbent LECs to provide unbundled access to competitive carriers purchasing the switching UNE, if the incumbent LEC does not provide customized routing necessary to use alternative providers.” Triennial Review Order, 18 F.C.C.R. at 17333 ¶ 560. (emphasis supplied). In a footnote responding to a petition for clarification by MCI WorldCom and AT & T, the FCC explained further that “such customized routing must be provided in a manner that allows competitive LECs to efficiently access either a third party’s OS/DA platform or their own OS/DA platform.” Id. n. 1735.

Accordingly, aside from whatever impact the FCC’s forthcoming order will have, there are at least two problems with the DPUC’s current ruling on the OS/DA Issue. First, it requires SNET to offer unbundled OS/DA whenever it does not offer customized routing or a compatible 20 signaling protocol, when in fact current regulations only require SNET to unbundle OS/DA when it does not offer customized routing or a compatible signaling protocol and it provides the unbundled local switching element. 21 Second, the Final Decision requires SNET to unbundle OS/DA unless it provides customized routing using MCI’s preferred signaling protocol, viz., Feature Group D. Apparently, the reason for this requirement is that MCI cannot use customized routing to access its own OS/DA service unless the calls are routed using Feature Group D. The DPUC misunderstands the appropriate standard. SNET is required to offer customized routing that allows MCI to efficiently access either MCI’s own OS/DA platform or a third party’s OS/DA. Consequently, unless the DPUC makes a finding that Feature Group D signaling is the only way customized routing can be used to access either MCI’s OS/DA or any third-party’s OS/DA, it cannot require SNET to implement the Feature Group D signaling protocol.

3. The CNAM/LIDB Issue

The CNAM Database, or Caller ID with Name Database, provides caller-ID services. LIDB refers to the Line Information Database, a database used to validate calls made using non-standard billing, such as calling-cards or collect calls. The DPUC made the following ruling regarding CNAM and LIDB:

SNET provides LIDB Service and/or CNAM Database as set forth in this Appendix only as such services are used for [MCI’s] activities on behalf of [MCI’s] local service customer where SNET is the incumbent local exchange carrier. [MCI] agrees that any use of LIDB, for the provision of LIDB Service Applications and/or CNAM Query by [MCI] outside the area where SNET is the incumbent local exchange carrier, will not be pursuant to the terms conditions, rates, and charges of this Appendix or Agreement.

Final Decision at 17. MCI argues that this ruling constitutes an impermissible restriction on its use of a UNE.

As with the QS/DA issue, it appears that the combination of the Triennial Review Order and the forthcoming FCC order will drastically limit the circumstances under which SNET will need to offer unbundled CNAM/LIDB services at all. Moreover, even in those circumstances where SNET must offer unbundled CNAM/LIDB services, it appears that the FCC intends to change the rules regarding permissible uses of UNEs. Accordingly, rather than try to guess at whether the DPUC’s ruling will survive the FCC’s upcoming order, the most efficient course of action is to remand this issue, along with the others, back to the DPUC for further consideration, presumably after the FCC issues its new rules.

CONCLUSION

SNET’s Motion for Summary Judgment (doc. # 43) is GRANTED. MCI’s Motion for Summary Judgment (doc. # 47) is GRANTED. The DPUC’s rulings on all disputed issues are vacated, and the case is remanded to the DPUC for further proceedings consistent with this decision.

It is so ordered.

1 . The full name of this decision is Implementation of the Local Competition Provisions in the Telecommunications Act of 1996, First Report and Order.

2 . The full name is Implementation of the Local Competition Provisions in the Telecommunications Act of 1996, Intercarrier Compensation for ISP-Bound Traffic, Declaratory Ruling in CC Docket No. 96-98 and Notice of Proposed Rulemaking in CC Docket No. 99-68.

3 . The full name is Implementation of the Local Competition Provisions in the Telecommunications Act of 1996, Intercanier Compensation for ISP-Bound Traffic, Order on Remand and Report and Order.

4 . I assume, though it is far from clear, that the intervening sentence, which states "Internet Calls are not subject to reciprocal compensation under this Appendix nor under the Act,” is merely an acknowledgment of the FCC’s decision in the ISP Remand Order.

5 . It does not appear that MCI and SNET's relationship is governed by paragraph 81 of the ISP Remand Order, which requires "bill and keep” exchange for interconnection agreements between carriers who have never previously exchanged traffic. ISP Remand Order 16 F.C.C.R. at 9187 II 81.

6 . The fact that under the transitional regime the same rate applies to reciprocal compensation and to ISP compensation does not at all justify the DPUC’s declaration that "Internet traffic is subject to local reciprocal compensation.” Internet traffic is not subject to reciprocal compensation, it is—according to federal law—subject only to ISP compensation, although the ISP compensation rate is, under the transitional regime, tied to the reciprocal compensation rate. Even if the DPUC's statement were modified to say "internet traffic is subject to compensation at the same rate as traffic that is subject to reciprocal compensation” it would still be impermissible because that rule can only apply if SNET refuses to mirror. If SNET mirrors, the reverse would be true—traffic subject to reciprocal compensation would, by agreement, be subject to compensation at the same rate as internet traffic.

7 . There are actually two kinds of FX service: (1) Traditional FX, which is set up by running a private line from the customer’s premises to the geographically remote switch. (2) Virtual FX, in which, although a foreign NPA/NXX number is assigned to the customer and therefore no toll is charged for calls between that customer and customers on the foreign switch, the local switches are programmed to route such calls just as they would route non-local calls. In other words, certain calls appear to be (and are billed as) local calls, but are physically routed like non-local calls. ILECs typically offer traditional FX, and CLECs typically offer virtual FX.

8 . To the extent SNET is arguing that MCI agreed to be bound by whatever the DPUC would hold in the EX Decision, I am not persuaded. The DPUC had an obligation not to accept an interconnection agreement that was contrary to federal law, regardless of whether the parties agreed to it. Either party is free to challenge the DPUC’s failure to fulfill that obligation. Moreover, the record indicates only that MCI agreed to incorporate the FX Decision, not that it waived its right to challenge that decision as inconsistent with federal law.

9 . “Exchange services for such access” is not a defined term, but the FCC has indicated that it "is closely related to the provision of exchange access and information access.” ISP Remand Order, 16 F.C.C.R. at 9168 n. 65.

10 . It is probable that section 251(b)(5) does not apply to telecommunications traffic that states have traditionally subjected to intrastate access charges, even if such traffic is not "toll” traffic. See ISP Remand Order 16 F.C.C.R. at 9168 n. 66 ("it is reasonable to interpret section 251(b)(5) to exclude traffic subject to parallel intrastate access regulations”); Global NAPS, 327 F.Supp.2d at 297 ("reciprocal compensation requirements of the 1996 Act exclude traffic already subject to interstate and intrastate access regulations.”). That exemption has no relevance here because the DPUC has explicitly declared that FX traffic is not subject to access charges.

11 . See also SNET v. DPUC, 285 F.Supp.2d at 262 n. 4 ("Virtual FX calls ... do not appear to be 'interLATA toll' calls because no separate charge is made to SNET's subscribers for this traffic.”).

12 . MCI offers a different argument in support of this conclusion. It argues that the FCC issued a binding ruling that FX traffic is subject to reciprocal compensation when its Wireline Competition Bureau ("the WCB”) issued the so-called "Virginia Arbitration Order.” That argument has no merit. In the Virginia case, the WCB stood in the stead of the Virginia State Corporation Commission, which had refused to arbitrate an interconnection agreement. Virginia Arbitration Order, FCC Docket No. 00-218 177, F.C.C.R. 27039, 27042. The WCB did not base its decision regarding FX traffic on either a requirement of federal law or an interpretation of federal law. Id. at 27181; see also Applica tion by Verizon Maryland Inc., for Authorization To Provide In-Region InterLATA Services in Maryland, Washington, D.C., and West Virginia, Memorandum Opinion and Order, 18 F.C.C.R. 5212, 5300 ¶ 151 n. 601 (2003) ("The Bureau did not, however, address the legal question of whether incumbent LECs have an obligation under [the 1996 Act] to provide reciprocal compensation for virtual FX traffic.”).

13 . Additionally, since the issuance of the Final Decision, the FCC has altered its interpretation of section 252(i) to prohibit CLECs from opting into only certain parts of interconnection agreements. See Review of the Section 251 Unbundling Obligations of Incumbent Local Exchange Carriers, Second Report and Order, 19 F.C.C.R. 13494 (2004) (changing from the "pick-and-choose” rule to the "all-or-nothing” rule).

14 . The full name is Provision of Directory Listing Information under the Telecommunications Act of 1934, As Amended, First Report and Order.

15 . MCI argues that the DPUC’s regulation was an implementation of the DPUC's obligations under state law. The DPUC has not argued that position, but, if MCI’s reading is correct, the DPUC will no doubt malte that clear on remand.

16 .The full citation is Implementation of the Local Competition Provisions of the Telecommunications Act of 1996, Third Report and Order and Fourth Further Notice of Proposed Rulemaking, 15 F.C.C.R. 3696 (1999).

17 . The full citation is Review of the Section 251 Unbundling Obligations of Incumbent Local Exchange Carriers, Implementation of the Local Competition Provisions of the Telecommunications Act of 1996, Deployment of Wireline Services Offering Advanced Telecommunications Capability, Report and Order and Order on Remand and Further Notice of Proposed Rulemaking, 18 F.C.C.R. 16978 (2003).

18 . The Final Decision also contains the ambiguous statement that "SNET shall offer each Unbundled Network Element individually and in ordinary combination ... in order to permit SNET to combine such Unbundled Network Element ... with another Unbundled Network Element ... obtained from SNET or with network components provided by itself” (emphasis supplied). If this was typographical error and the second reference to SNET was intended to be to MCI, then that appears consistent with the requirement that SNET only perform the functions necessary to allow MCI to combine elements (although it would then be inconsistent with the Final Decision's later language requiring SNET to do the combining). If the reference to SNET was intentional, then the sentence is, aside from being ambiguous, inconsistent with federal law.

19 . The regulations impose on SNET the burden of demonstrating the existence of either of the limiting circumstances. 47 C.F.R. § 51.315(e)-(f).

20 . All parties agree that the Final Decision’s reference to a "comparable signaling protocol" was an error, and the DPUC meant to say "compatible signaling protocol.”

21 . Something it will apparently not have to do once the transitional regime comes to an end.