In This Issue:
This special edition of the Communications Law Bulletin focuses on two major items, the Triennial Remand Order and the Special Access Notice, which the Federal Communications Commission ("FCC") released last week. Although these items are in the FCC’s Wireline docket, they are significant for wireless providers, Internet service providers, competitive local exchange carriers ("CLECs"), and all other users of services provided by incumbent LECs ("ILECs").
In the Triennial Remand Order, released on February 4, 2005, the FCC addressed for the fourth time its rules for the unbundling of ILEC facilities, or network elements, this time in the wake of the D.C. Circuit’s USTA II decision of 2004, which vacated major portions of the FCC’s Triennial Review Order of 2003 ("2003 TRO"). The Triennial Remand Order cuts back on the availability of unbundled network elements ("UNEs"), with the policy goal of stimulating carriers’ investment in facilities.
In the Special Access Notice, released on January 31, 2005, the FCC begins examining its regulation of the special access services offered by the ILECs. The special access services at issue in this notice are very similar to the UNEs, or combinations of UNEs, that the Commission has addressed in its Triennial proceedings. However, while UNEs are priced at rates based on Total Element Long Run Incremental Cost ("TELRIC"), special access rates are usually tariffed at substantially higher rates.
As UNEs become less available under the terms of the Triennial Remand Order, CLECs and other users of ILEC networks will have to rely even more on special access services. A hotly contested issue between ILECs and users has been whether significant competition exists in the provision of special access, and how FCC regulation should reflect the presence or absence of such competition. How the Commission resolves the Special Access Notice therefore will have major consequences for special access customers, many of which are wireless providers, CLECs and Internet service providers that compete in some way with the ILECs.
The Triennial Remand Order is the FCC’s latest attempt to establish network unbundling requirements for ILECs pursuant to Section 251 of the Communications Act (the "Act"), and it does so by severely limiting the availability of UNEs to CLECs and others. When Congress added Section 251 to the Act in 1996, it established three means by which CLECs could compete with ILECs in local telecommunications markets. First, CLECs could build their own facilities and enter into agreements with ILECs to connect the carriers’ networks and exchange customer traffic between them. Second, CLECs could acquire unbundled ILEC network elements – UNEs – at TELRIC-based rates in order to provide competing services. The FCC’s usual test for determining whether an ILEC must make a UNE available is whether a competitor’s market entry would be "impaired" if it does not have access to the UNE. Third, CLECs could purchase ILECs’ retail services at wholesale rates and resell those services to the CLECs’ customers. The Triennial Remand Order restricts the second means of local competition established in the Act, which has been a foundation of the CLEC industry since 1996.
The Triennial Remand Order is the FCC’s response to two major developments in telecommunications law in 2004. The first was the D.C. Circuit’s USTA II decision, which interpreted the unbundling requirements of Section 251 very narrowly and overturned many of the FCC’s UNE requirements adopted in the massive 2003 TRO. Although that order had limited the availability of many types of UNEs, USTA II found that the FCC had applied the unbundling provisions of Section 251 too broadly. The second development was the Justice Department’s decision not to seek Supreme Court review of USTA II. This meant as a legal matter that USTA II became settled law. As a policy matter, it signaled support from the Administration for the FCC to restrict the availability of UNEs further, consistent with USTA II.
Because, as summarized below, the Triennial Remand Order narrows the ILECs’ obligations to provide UNEs, most notably by eliminating the UNE for mass market local circuit switching, it can and should be viewed as a regulatory victory for the ILECs. The FCC also prohibited the use of UNEs for use exclusively in the long distance and wireless markets, finding that those markets are so competitive that ILEC unbundling is unwarranted for them. If there is a silver lining for CLECs, it is that the FCC declined to perform even more radical surgery on its list of UNEs, as urged by the ILECs. Rather than eliminate even more UNEs, the FCC refined its fundamental "impairment standard" for when ILECs must make UNEs available as well as its application of that standard to the UNEs for dedicated interoffice transport and high capacity loops.
It is more than likely that the Triennial Remand Order will satisfy neither ILECs nor CLECs nor other interested parties. Indeed, the history of the FCC’s UNE rules has been one of almost continual litigation since 1996. But at some point the FCC will satisfy at least the appellate courts that its unbundling rules are consistent with the Act. The Triennial Remand Order is its current vehicle for doing so.
In general, the FCC upheld the overall unbundling framework that it adopted in the 2003 TRO, with one clarification of the impairment standard and three modifications to the unbundling framework.
The FCC also noted that alleged failings in other aspects of the FCC’s regulatory regime (such as excessive special access rates or scarce collocation space) should not result in compensatory unbundling; rather, these alleged problems should be addressed in the context of those rules.
Dedicated interoffice transmission facilities are dedicated to a particular competitive carrier’s use for transmission between or among ILEC central offices and tandem offices and to connect its local network to the ILEC’s network. The D.C. Circuit in USTA II remanded the transport analysis in the 2003 TRO because, due to the improper delegation to state commissions, the FCC’s findings of nationwide impairment for DS1, DS3 and dark fiber transport were inconsistent with the FCC’s acknowledgement that competitive alternatives are available in some locations. The court also faulted the FCC for not adequately considering where competitors could potentially deploy their own transport facilities.
CLECs use unbundled interoffice transport to aggregate end-user traffic. A large proportion of competitive transport facilities are located in dense business districts, and a large proportion of business lines are served by a relatively few wire centers. Because the revenues generated by transport facilities increase with the volume of traffic, and costs increase with distance, competitive carriers look first at the shortest routes with the greatest traffic volumes in considering where to deploy competitive transport, which are primarily in dense business districts.
Transport facilities are characterized by high fixed and sunk costs. Because these costs vary significantly from one route to another, no nationwide conclusions can be drawn, nor is the FCC able to construct a cost model to assess impairment. Thus, the FCC’s impairment analysis focuses on actual competitive transport deployment, which implicitly reflects competitors’ assessments that actual and potential revenues justify the deployment costs. Transport costs include the costs of collocation, fiber and electronic equipment and materials, as well as the costs of deployment, which include rights-of-way. CLECs, however, need not always install new conduit to deploy their own transport facilities.
In defining the relevant transport markets, the FCC measured impairment on a route-specific basis, taking into account inferences that can be drawn from actual competitive transport deployment on similarly situated routes. Based on those inferences, categories of routes can be established, defined by the economic characteristics of each end point, in order to identify routes with similar economic traits. The FCC rejected transport market definitions based on other characteristics, such as entire Metropolitan Statistical Areas ("MSAs") or an unbundling trigger based on a single transport route end point, because such definitions would encompass routes that are too varied in their economic characteristics. The FCC also defined the relevant markets on a capacity-specific basis, due to the significant differences in revenue potential from transport circuits of different capacities.
In order to identify similar routes, the FCC categorized similar end points and made impairment determinations for the various combinations of end point categories. Evidence of actual deployment was used to define the characteristics of wire centers where there is a lack of impairment, i.e., where reasonably efficient CLECs are capable of duplicating the ILEC’s transport facilities. This approach is based on an inference drawn from the economic conditions on both ends of a transport route, i.e., the ability to deploy facilities at each of the two end points of a route signals the ability to connect, directly or indirectly, the two end points via a transport facility.
The FCC found that the best and most readily administered indicators of the potential for competitive deployment are: (1) the presence of fiber-based collocations in a wire center; and (2) business line density in a wire center. Either one of these measures is a reasonable proxy for where sufficient revenue opportunities exist to justify the high fixed and sunk costs of transport deployment.
The FCC divided the ILEC wire centers into three tiers, based on actual competitive transport deployment or indicia of the potential revenues that can be derived from such deployment and the suitability of such wire centers for deployment:
The Triennial Remand Order applied these criteria differently for different types and capacities of transport facilities:
Entrance facilities (the transmission facilities that connect CLEC networks with ILEC networks) have unique operational and economic characteristics that justify separate impairment treatment from other transport facilities. CLECs are not impaired without access to ILEC entrance facilities because they are less costly, are more widely available from alternative providers and, because they aggregate all traffic between the CLEC and ILEC networks, have greater revenue potential than transport between ILEC central offices. Thus, CLEC self-deployment of entrance facilities is more likely to be economically justified. CLECs also have a unique degree of control over entrance facility costs because they can choose the locations of their own switches. These differences make application of the transport impairment tests to entrance facilities inappropriate.
For DS1 and DS3 facilities, CLECs have a 12-month period to transition to alternative options, including ILEC special access, and an 18-month transition period for dark fiber. These transitions apply only to the embedded customer base and do not permit CLECs to add new UNEs where the FCC has found no impairment. The FCC found that the 12-month transition is necessary to provide adequate time to make such decisions as where to deploy, purchase or lease facilities. Because ILECs generally do not offer dark fiber as a tariffed service and because it may take time for CLECs to negotiate Indefeasible Rights of Use or other arrangements, a longer transition is necessary for dark fiber. "Lit" DS3 or OCn services are sufficiently different from dark fiber not to qualify as a ready substitute. During these transition periods, any affected dedicated transport leased by a CLEC as of the effective date of the order will be available at the higher of 115 percent of the rate the requesting carrier paid on June 15, 2004, or 115 percent of the rate the state commission established or establishes between June 16, 2004, and the effective date of the order. These moderate price increases will mitigate the rate shock that could be suffered by CLECs if TELRIC pricing were eliminated immediately, while providing some protection for ILEC interests. This transition mechanism does not replace any commercial arrangements carriers have reached for the continued provision of transport facilities or services.
The FCC performed an impairment analysis for high-capacity "loops," the transmission facilities between a central office and a customer’s premises similar to that for dedicated transport, with some key differences. The USTA II decision never questioned the FCC’s conclusions that CLECs are impaired without access to the lowest capacity loops – voice grade DS0 copper loops – and are not impaired without access to the highest capacity loops. Thus, the analysis of the Triennial Remand Order is limited to DS1, DS3 and dark fiber loops. The FCC found that although the costs of deploying these high-capacity loops vary little among the different capacity levels, the revenue opportunities from a given customer location increase as the capacity level increases. CLECs deploying high-capacity loops face large fixed and sunk costs, which vary by the length of the loop, as well as substantial operational barriers in deploying fiber. The most significant costs in building a fiber loop result from deploying the physical fiber infrastructure into underground conduit. LECs do not typically construct fiber loop facilities at DS1 or DS3 capacity levels, but instead install higher-capacity fiber-optic cables and then use electronics to light the fiber at specific capacity levels.
The FCC found that operational barriers to the construction of competitive loop facilities include the six to nine months that are required to install local loops and additional delays caused by the need to secure public and private rights of way. Most of the costs of constructing loops are sunk costs because a loop can serve only a specific location and can generate revenue only by continuing to serve the customer. Thus, barriers to entry to competitive loop deployment are substantial.
According to the FCC, however, economies of scale can accrue when carriers construct loops to locations that are geographically close to the transport network, especially in urban areas with high concentrations of potential customer locations. CLECs try to build fiber rings close to as many large commercial buildings as possible, enabling them to use competitive DS3 loop facilities where buildings are either directly connected to a fiber ring or would require the construction of only a short "lateral" where buildings are either directly connected to the ring or are close to the ring. Thus, the highest concentrations of competitive loop deployment in central business districts are near where competitors have already deployed fiber rings. Competitive carriers are also able to economically serve lower capacity customers (e.g., customers at the DS1 level) in multi-tenant buildings. The incremental costs of providing channelized capacity over higher-capacity fiber loops are minimal when other customers in a building are already being served by competitive fiber or the likelihood of capturing customers at higher capacity justifies deployment of facilities that can be channelized to the DS1 level.
The Triennial Remand Order found the area served by a wire center to be the appropriate geographic market for evaluating both actual and potential competition in the provision of high-capacity loops. The typical wire center is small, so its competitive characteristics will be fairly uniform throughout. As discussed in the previous section, competitive data concerning wire centers is readily available. A wire center-based approach also yields reasonably precise results linking impairment to the factor that most prominently determines whether construction of a competitive facility is economic – namely, the presence of extensive competitive fiber rings within an area. The FCC rejected suggestions of smaller geographic markets, such as individual buildings, and larger geographic markets, such as MSAs. The FCC also found that the feasibility of constructing DS1 and DS3 loops is too case-specific to allow nationwide impairment findings.
To identify which markets are likely suitable for competitive deployment of DS3 loops, the FCC derived proxies correlating to the evidence of actual DS3 deployment. These proxies indicate the circumstances in which competitive carriers can deploy short "laterals" to connect buildings to nearby splice points on competitive fiber rings, allowing an inference that DS3 or higher-capacity loops can be economically deployed. The presence of fiber-based collocators in a wire center service area is a good indicator of the potential for competitive deployment of fiber rings. The number of business lines served by a wire center is also indicative of its location in or near a large central business district with large commercial buildings, which is likely to house multiple competitive fiber rings with laterals to multiple buildings.
Unlike the test for dedicated transport, the FCC’s impairment test for high-capacity loops requires both a minimum number of business lines served by a wire center and the presence of a minimum number of fiber-based collocators to show that requesting carriers are not impaired. For loops, the FCC found, a test that would be satisfied by either criterion without the other (as is the case for dedicated transport) would deny unbundling where carriers are impaired, for two reasons. First, the presence of fiber rings without a high business line count might indicate a wire center service area that happens to fall along a ring that serves other high-revenue areas but that itself does not offer sufficient revenues to justify competitive deployment of high-capacity loops. Second, a high business line count in the absence of sufficient fiber-based collocators might indicate an area with high revenue opportunities but that is not close to existing fiber facilities or otherwise suitable for fiber ring deployment.
Because competitive carriers typically do not provision stand-alone DS1 loops (i.e., loops at the DS1 capacity provisioned either by the CLEC itself or a third-party provider), on account of their lower revenue potential, such facilities are usually offered competitively only where higher-capacity loops are already being offered to a building’s "anchor tenants." The impairment analysis for DS1 loops thus depends on whether it is likely that other competitive carriers have already deployed or will deploy DS3 facilities to buildings throughout the wire center service area, thus making DS1-level use of those facilities potentially viable. Accordingly, a higher business line count is necessary before determining that requesting carriers are not impaired without access to DS1 loops than is required for DS3 loops. The higher threshold accounts for the lower revenue opportunities afforded by DS1 loops and the importance of ensuring a greater likelihood of DS3 facilities available for channelization. The higher threshold also indicates higher revenue opportunities within the buildings in the service area, suggesting the likelihood of more extensive fiber rings.
Based on this economic analysis, the Triennial Remand Order adopted a proxy under which requesting carriers are not impaired without access to DS3 loops in any building served by a wire center with at least 38,000 business lines and four fiber-based collocators. These thresholds indicate fiber deployment and revenue opportunities sufficient to render competitive deployment of DS3 loops economic. In order to encourage competitive facilities-based deployment when it is economic to do so, required DS3 unbundling will be limited to one DS3 circuit in each building.
For DS1 loops, requesting carriers are not impaired in any building served by a wire center with at least 60,000 business lines and four fiber-based collocators. The line count test is higher because of the low revenue opportunities offered by DS1 loops and the corresponding need to ensure that DS3 or higher-capacity facilities are actually deployed and thus available for channelization. These thresholds will eliminate DS1 loop unbundling in wire centers accounting for about eight percent of all BOC business lines. Required DS1 unbundling will be limited to ten DS1 loops for any single CLEC in each building.
In addition, the Triennial Remand Order made a nationwide finding of non-impairment for dark fiber loops, explaining that any barriers to entry can be overcome through self-deployment of lit facilities at the OCn level. Making dark fiber available on an unbundled basis would undermine the incentives for CLECs to deploy facilities established by the "one DS3" limit. Moreover, because carriers will only construct fiber loops to the extent that building lit fiber loops would be economic, it can be inferred from the FCC’s prior determination of non-impairment as to lit OCn capacity loops that carriers are not impaired with regard to dark fiber, which is generally lit at the OCn capacity level.
CLECs will have a 12-month transition period from the effective date of the order to switch to alternative DS1 and DS3 loop arrangements and an 18-month period to move to alternative arrangements for dark fiber loops, including the use of ILEC tariffed services. These transitions will apply only to the embedded customer base and do not permit CLECs to add new high-capacity loop UNEs where there is no impairment. These transition periods are necessary to perform the necessary tasks for an orderly transition, including decisions where to deploy, purchase or lease facilities. The longer transition for dark fiber loops is necessary for the same reasons as the longer transition for dark fiber transport. As in the case of the transport transition, high-capacity loops no longer required to be unbundled will be available at the higher of 115 percent of the rate the requesting carrier paid for the loop on June 15, 2004, or 115 percent of the rate the state commission established or establishes between June 16, 2004, and the effective date of the order.
The most prominent single policy change between the 2003 TRO and the Triennial Remand Order concerns mass market local circuit switching. Responding to the USTA IIdecision and the Justice Department’s decision not to seek Supreme Court review of USTA II, the Triennial Remand Order reexamined mass market local circuit switching and made a nationwide determination of no impairment; accordingly, subject to the transition period described below, this element will no longer need to be unbundled. As a result, unbundling of the "UNE-Platform" or "UNE-P" (which incorporates local circuit switching along with loops and transport) also will no longer be required (along with other elements contingent upon local circuit switching unbundling, such as shared transport, signaling, and call-related databases). By unbundling mass market circuit switching, the FCC also eliminated the need to resolve the issue of the number of lines that distinguish mass market customers from enterprise market customers (for which local circuit switching was unbundled in the 2003 TRO). In reaching this conclusion of no impairment, the FCC found that UNE-P has been a disincentive to CLEC investment and that many CLECs were making UNE-P a long-term business strategy rather than an interim step to a facilities-based strategy.
In changing its determination with respect to mass market local switching, the FCC noted that newer, more efficient switches (such as packet switches and soft switches) have been deployed throughout the country, which mitigates concerns expressed in the 2003 TRO. The FCC also declined to find impairment based upon other related barriers to entry, such as transport costs or lack of collocation space. The FCC noted that transport costs are based upon switch location and thus are largely within CLEC control. With respect to collocation, the FCC similarly found no impairment because CLECs no longer need to collate in every central office because the new switches can serve wider geographic areas. Any specific problems regarding collocation are to be addressed through the FCC’s collocation rules.
The FCC also found that the BOCs have made significant improvements in their batch hot cut processes, which also mitigates concerns that the FCC had expressed in the 2003 TRO. The FCC indicated, first, that the need for hot cuts is diminishing in light of VOIP and new commercial agreements, and, second, that any remaining inadequacy regarding hot cuts should be addressed through enforcement of interconnection agreements or (with respect to the BOCs) via complaints to enforce the Section 271 requirements upon which their long distance entry was conditioned. The FCC noted that any potential impairment that still might exist is outweighed by the investment disincentives resulting from a continued unbundling requirement.
For transitioning away from the previously available switching and UNE-P elements, CLECs must convert to alternative arrangements within twelve months of the effective date of the Triennial Remand Order – a longer period than initially proposed. Accordingly, interconnection agreements must be amended (and any corresponding change-of-law procedures must be completed) within twelve months of the effective date of the Triennial Remand Order. This transition applies only to embedded customers, so CLECs cannot add new customers using these elements. As proposed in the interim rules, during the transition period CLECs will continue to have access to UNE-P priced at the higher of (a) the rate paid by the requesting carrier on June 15, 2004, plus one dollar, or (b) the rate the state public utility commission establishes (if any) between June 16, 2004 and the effective date of the Triennial Remand Order, plus one dollar. The FCC found that these "moderate" price increases will assist in creating an orderly transition. The transition arrangements described here, however, are default mechanisms and do not supersede any commercial agreements reached between ILECs and CLECs.
The FCC also discussed conversions of tariffed services (such as special access) to UNEs, finding that an across-the-board bar on such conversions would be inappropriate. The FCC made this finding based upon several factors. First, the FCC’s other rulings have narrowed the purported problem with such conversions by prohibiting the conversion to UNEs where such UNEs would be used exclusively to provide long distance or mobile wireless service. Second, any such prohibition on conversions would be inconsistent with the FCC’s related finding that special access availability does not foreclose impairment and UNE access. Third, such a prohibition on conversions would require detailed line-drawing that would be administratively impracticable.
On a related note, the FCC declined to extend its Enhanced Extended Link (loops plus transport) eligibility criteria to stand-alone high-capacity loops as some of the BOCs had requested. The FCC found that high-capacity loops are not susceptible to misuse by long distance carriers seeking to avoid special access charges.
In addition, the FCC addressed the implementation of its new unbundling determinations by stating its expectation that both ILECs and CLECs will implement its new findings and amend their interconnection agreements in good faith without unreasonable delay. Because the new unbundling rules evaluate impairment based upon objective and fairly readily obtainable facts, the FCC requires the requesting carrier to undertake a reasonably diligent inquiry and self-certify that, to the best of its knowledge, its request is consistent with the new unbundling requirements. Once it has done so, the ILEC must immediately process the request. Any ILEC challenge to the request must be raised through interconnection agreement dispute resolution procedures after the request is processed.
Finally, in light of the need for certainty and prompt action, the Triennial Remand Order states that the new permanent unbundling rules shall take effect on March 11, 2005.
Released a few days before the Triennial Remand Order, the Special Access Notice is a comprehensive, searching examination of the FCC’s regulatory regime for the special access services provided by the ILECs. The issues raised in the Special Access Notice will face Chairman Michael Powell’s successor for resolution.
The Special Access Notice is extremely significant to special access users, which include wireless users, CLECs, and large businesses, because it signals the FCC’s concern that its assumptions and rules regarding special access may not be working as expected. The Special Access Notice notes that the FCC’s existing special access pricing rules (adopted in 2000 as one part of the so-called CALLS plan) will technically expire on June 30, 2005, although they will remain in place until the FCC adopts follow-up rules. The notice thus seeks comment on "what steps the Commission should take to ensure that rates for special access services remain just and reasonable after the expiration of the CALLS plan."
In starting this rulemaking, the Special Access Notice acknowledges that in its price cap regulations, the FCC typically does not examine a single "basket" or type of interstate service separately from other services, but that it is doing so here because of the importance of special access as a key input to many competitors’ service offerings. The FCC also points out that in 2003, interstate special access revenues constituted 45.4% of all BOC interstate operating revenues. The Special Access Notice seeks comment on the numerous price cap issues for special access that the CALLS plan covered, especially on how to establish a price cap regime for special access services that would be independent of that for other (switched) services.
The FCC also asks whether it should change its rules for granting special access pricing flexibility to ILECs subject to price caps. Those rules rely on regulatory "triggers" based on collocation by competitive carriers as predictive evidence of irreversible market entry. Some special access users have roundly criticized the pricing flexibility rules as being based on erroneous assumptions. The Special Access Notice revisits the economic assumptions about market definition and market power for special access that support the pricing flexibility rules. Acknowledging the similarities between special access service and UNEs, the Special Access Notice even asks about the relationship between the market power threshold used in the pricing flexibility rules and the impairment standards used in the Triennial Remand Order’s new rules governing UNEs.
The Special Access Notice denies requests by AT&T for interim relief from existing special access regulation. But the FCC states that it anticipates adopting an order prior to July 1, 2005, to establish an interim plan to ensure that special access rates remain just and reasonable while this rulemaking is pending. Any such interim order could be very important if resolution of the issues raised by the Special Access Notice is delayed.
Comments on the Special Access Notice will be due 60 days after its publication in the Federal Register, and reply comments will be due 90 days after Federal Register publication.