INTERPRETING AND APPLYING
THE RESULTS OF THE
BENCHMARK COST MODEL
7.1 The appropriate benchmark for determining USF eligibility and quantifying the level of support
For the cost results produced by the BCM to be put to actual use, a critical determination must be made as to the proper "benchmark" for determining USF requirements. An examination of diverse state universal service proceedings implicitly raises questions of (1) how to define the appropriate benchmark to determine eligibility for high cost support, and (2) how to determine the aggregate amount of high cost support to which an area is entitled. As discussed in the previous sections, the approach adopted by the BCM is to determine an area's eligibility for USF support by comparing the proxy cost of serving that area with a pre-established price. In the BCM, as presently structured, the entire difference between the average cost per access line for the area and the predetermined threshold price represents the universal service support requirement for that area.[1] Thus, the BCM determines eligibility for high-cost support and the amount of the entitlement essentially in a single step. The BCM also does not specify any jurisdictional responsibility for the allocation of these above-average costs, but rather is presumably intended to compute the entire USF support that is potentially required.
By contrast, the existing federal high cost program makes an initial determination as to eligibility for high cost support, before separately addressing the amount of support to be allowed. To determine eligibility for support, the existing federal high cost program compares the national average cost of the residential access line with an individual carrier's average costs for each of its study areas.[2] In order to determine the level of support that is to be provided to any given carrier, a sliding scale of assistance is determined based upon factors such as the number of lines in the study area and the degree to which the carrier's average line cost exceeds the national average by at least 15%.[3] Thus, the existing federal high cost fund does not make up the entire difference between the carrier's average cost and the national average cost, rather it makes up only a portion of that difference.
Identifying high cost areas is not the only way to get a perspective on universal service requirements, however. Rather than comparing cost data to some threshold level (either price as with the case of the BCM or average cost as is the case with the existing high cost fund), some states, such as Vermont and Wisconsin, have decided to make their universal service funding decisions dependent upon how the price of basic service compares to an affordability benchmark, without any direct[4] consideration of cost to serve. In a report recently submitted to the Vermont General Assembly, the Vermont Public Service Board included draft legislation that would define a high cost area as one in which "the price of basic service to customers is in excess of 150 percent of the state average."[5]
The Vermont Board is proposing a price-based approach for determining where high cost support should be targeted:
The price to the customer should be regarded as the fundamental variable of universal service. If the customer's price goes up, the customer will be less likely to keep his or her telephone service. The cost to the carrier is relevant only insofar as it affects the customer's price.[6]
In making this recommendation, the Board noted the link between existing prices and the incumbent LEC's embedded costs (particularly in the absence of competition), observing that:
In the absence of effective local competition, however, there is an important reason not to distribute high cost assistance based upon forward-looking costs. The local rate charged to a customer as the result of the embedded costs is far more significant in the customer's decision to keep or drop telephone service than is the theoretical cost of a new competitor.[7]
The following tables illustrate how these various benchmarking approaches compare. Table 7.1 compares how eligibility for high cost support is determined under each of the three approaches. Table 7.2 compares how (once the determination of eligibility is made) the amount of high cost support is determined. Finally, Table 7.3 shows how the existing federal high cost fund uses a sliding scale approach to target increasing levels of support to study areas with the highest costs. In addition, Appendix 7A displays the actual disbursements, by state, under the existing federal high cost fund.
Table 7.1 Benchmarks Vary for Determining Eligibility
for High Cost Support
Benchmark USF Plan
Does the LEC's average per-loop cost exceed 115% of Existing FCC
the national average per-loop cost? High Cost Fund
(in effect)
Does the average per-loop cost computed by the proxy Benchmark Cost
model exceed the price threshold? Model (proposed)
Does the actual price exceed 150% of the statewide Vermont
average price? (proposed)
Table 7.2 Methods for Computing Amount of High Cost Support Vary
Method USF Plan
Depends upon several factors, including number of lines in study Existing FCC
area and the amount by which the cost exceeds the national average* High Cost Fund
(in effect)
For all CBGs where the cost exceeds the price threshold, multiply Benchmark Cost
the number of households in the CBG times the difference between Model (proposed)
cost and price and add all of the individual CBG USF requirements
for a total USF requirement
Provide support for the difference between the actual price and Vermont
150% of the statewide average price (proposed)
*See Table 7.3 below.
Table 7.3 Existing FCC High Cost Fund* Study Area with Fewer than 200,000 Working Loops: Relationship of Loop Cost in Amount of Cost Covered by Total Interstate Loop Study Area to the National the HCF** Allocation Average Loop Cost below 115% 0% 25% between 115% and 150% 65% 90% above 150% 75% 100% Study Area with More than 200,000 Working Loops: Relationship of Loop Cost in Amount of Cost Covered by Total Interstate Loop Study Area to the National the HCF** Allocation Average Loop Cost below 115% 0% 25% between 115% and 160% 10% 35% between 160% and 200% 30% 65% between 200% and 250% 60% 85% above 250% 75% 100% *Under the existing system of separations, 25% of costs are allocated to the interstate jurisdiction. **The existing high cost fund covers the indicated percentages of the unseparated costs of loops. Source: CC Docket 80-286, Notice of Inquiry, 9 FCC Rcd 7404 (1994).
7.2 Recommendations regarding the establishment of a benchmark for determining eligibility and the method for computing USF support
We do not disagree, in principle, that the price charged for basic service is more relevant to a customer's decision as to whether to become or stay connected to the public switched network than is the theoretical cost of that connection. Nonetheless, there is a serious possibility that setting a price-based benchmark for the universal service support requirement will simply defer the cost issue to another day, i.e., when the regulator must review the reasonableness of the particular price being charged by a dominant local exchange carrier. Until such time as the provision of basic local exchange service in an area is sufficiently competitive, the price for that service presumably remains under scrutiny by the regulator and, absent cost data, the regulator has no way of evaluating the reasonableness of the price.[8] Therefore, while an examination of the existing prices throughout a state for telecommunications services and the corresponding subscribership levels is critical to establishing the level of "affordable" rates, the rates alone should not drive the determination of the amount of universal service support.
Regarding the method for computing assistance, it is important to recognize that a cost proxy model yields an objective measure of the high cost support that may be necessary in order to assure the provision of affordable service and to satisfy the Congressional goal that the price in rural areas be "reasonably comparable" to the price in urban areas. However, there are several offsetting sources of revenue in addition to the basic monthly access line rate itself that would cover the shortfall in most high cost areas. Once such revenue is properly recognized, it becomes apparent that either there is no need at all for explicit high cost support, or that the areas that qualify for high cost support are much more limited and less costly to fund than otherwise estimated, certainly far less than the total amount that has been, to date, computed by the BCM.
7.3 Federal and state policy makers should coordinate USF plans to ensure that in no event is more than 100% of "high cost" recovered by USF support
The effort to unravel existing sources of universal service support and then to size explicit universal service funds raises the issue of jurisdictional responsibility for high cost areas. As discussed above, the existing federal universal service fund covers costs over and above the "standard" 25% interstate allocation for above-average-cost regions of the country, with the eligibility for and amount of funding determined with reference to cost levels across entire study areas. When cost levels are examined at a more granular level, such as an individual wire center serving district or, as many incumbent LECs favor, at a CBG level, the sheer variability of costs would virtually guarantee that the number of "areas" that exceed any given high cost threshold will increase dramatically. All else being equal, greater granularity will in and of itself result in a larger funding requirement to be satisfied at both the federal and state levels.[9]
A key policy question yet to be addressed is which jurisdiction should bear the burden for supporting high cost areas, particularly where differing definitions of universal service and varying degrees of granularity are involved. At present, the federal high cost mechanism picks up the substantial majority of the unseparated line costs for study areas with costs exceeding the national average. With the comprehensive reexamination of universal service funding, the manner in which the federal and state jurisdictions split overall responsibility for universal service support is in flux. While the Telecommunications Act requires the FCC to establish a definition of universal service and a "national affordable" rate level, it is possible that a state PUC may have the discretion to adopt a different definition or affordability threshold.
At a minimum, some kind of allocation of USF responsibility will be necessary between the federal and state jurisdictions. If one were starting with a blank slate, it is not readily apparent how the jurisdictional responsibility would be apportioned.[10] For example, suppose that the total per-line cost for a particular area is $35, and the FCC determines that the "supported" price for that area should be $30 but, for the same area, the state commission determines that the "supported" price should be $25. In this example, federal universal service support might be based upon a cost-price difference of $5 per line ($35 cost minus $30 price). If we assume an allocator of 25% to the federal side, the federal USF would therefore be responsible for supporting $1.25, which implies that a state USF would be responsible for supporting the remaining $3.75. In our example, however, the state also seeks to support another $5.00 per line per month, and therefore the total state USF requirements in this example would be $8.75 per line.[11] The Telecommunications Act considers this possibility and specifies, as a condition for allowing states to adopt definitions or regulations that lead to universal service support requirements that are broader than under the federal rules, that the state's regulations also "adopt additional specific, predictable, and sufficient mechanisms to support such definitions or standards that do not rely on or burden Federal universal service support mechanisms."[12] Thus, while the FCC is responsible for determining the federal USF support that is required, the federal guidelines clearly have direct implications for, and may work to constrain, state USF policy.
7.4 If a new universal service funding mechanism is established, other existing sources of support should be either eliminated or reduced
As we noted in Chapter 1, the Joint Sponsors of the BCM did not submit any information regarding how the results of the BCM would be used other than as a way to inform the high cost proceeding. The FCC, in implementing the recommendations of the Joint Board, is expected to adopt an explicit funding mechanism that meets the specific objectives set forth in the Telecommunications Act, and the BCM, ideally after the types of adjustments we recommend have been made, may form the core of this new universal service support mechanism. However, before any recovery of universal service costs begins to occur through a new universal service mechanism, the FCC must systematically identify all existing sources of universal service support (both explicit and implicit) that are redundant to this new fund, and must ensure that any duplication of support is eliminated from rates and/or funding mechanisms in the federal jurisdiction. A similar systematic approach will need to be undertaken with regard to any state sources of universal service support.
7.5 Revenues from other services linked to the provision of residential dial tone lines will offset much of the USF requirement
All of the foregoing discussion in the previous chapters of this report has focused on how to identify instances in which the costs to provide universal service to residential customers exceed a specified level of affordability; after all, a detailed analysis of costs and how they are modelled by the BCM is the core purpose of this report. However, it is important to recognize that costs are only one side of the equation. Before the BCM or any other costing study is used as a basis for dispensing support, it is critical that the revenue side of the equation be considered. There are a number of sources of revenue that LECs receive from their provision of primary residential dial tone lines. Some of these revenues are appropriately considered in determining the extent of any revenue shortfall that would require subsidization from the fund, while others are included in determining the relative contribution that each telecommunications provider will be required to make to the fund so as to make up the shortfall. What follows is a brief discussion showing how revenues can offset much of the universal service requirement.
It is widely acknowledged that, today, implicit subsidies within the LECs' rate structures and Yellow Pages revenues, an explicit subsidy, are sources of support for universal service. It is frequently asserted, however, that competition will erode the very revenues that incumbent LECs have relied on to support universal service at affordable rates. This argument ignores the fact that certain revenue sources available to incumbent LECs, such a Yellow Pages, are distinctly not affected by competition for local exchange service and that other revenues, from services linked to the provision of a dial tone line, remain tied to that service (and, thus, available to support universal service) whether the customer remains with the LEC or takes local service from a competing provider. It is quite possible that these revenues substantially, if not entirely, offset any universal service funding requirement derived on the basis of costs alone.
In most states, Yellow Pages directory revenues have long been used as a source of financial support for basic local exchange telephone service, principally (but not exclusively) the residential "dial tone" exchange access line. When the Court overseeing the divestiture of the Bell operating companies from AT&T in 1982 considered whether the Yellow Pages businesses should remain with the local companies or go to AT&T (along with various competitive businesses), the BOCs, many state public utilities commissions, and the National Association of Regulatory Utility Commissioners argued strongly for the retention of the Yellow Pages business by the BOCs expressly because of the enormous amount of revenue that was contributed by Yellow Pages to support basic exchange access services. Based on these arguments, the Court decided to retain the link that allowed Yellow Pages revenues and residential service, on the ground that Yellow Pages "provide a significant subsidy to local telephone rates [that] would most likely continue if the [BOCs] were permitted to continue to publish the Yellow Pages."[13] Nothing has occurred to change the public policy basis for this affirmative judicial finding that the profits from the publication of these directories should be used to defray a portion of the cost of providing basic local telephone service.
Yellow Pages was not a competitive business at the time of divestiture and, fourteen years later, it still remains a monopoly business of the incumbent LECs. Despite the absence of legal barriers to competition, Yellow Pages is the type of business activity that typically has only a single provider, due to the presence of formidable, perhaps even insurmountable, economic entry barriers. Shortly after the divestiture, there was a flurry of competitive activity in the Yellow Pages area.[14] Significantly, and notwithstanding the fact that these ventures were initiated by well-financed firms with considerable experience in and knowledge of the directory publishing business, none of them has succeeded in making any consequential inroads into this market.
LECs obviously resist the imputation of Yellow Pages revenues to support basic regulated telephone services, as it is an extremely profitable business. Recently, some LECs have argued that this practice raises competitive issues, since no other competitors have a similar requirement placed on them although these other companies also have may unregulated affiliates that profit from their corporate relationship. However unlike potential publishing competitors or local exchange service competitors with affiliated businesses, the directory publishing businesses of incumbent LECs enjoy a unique and direct benefit by being associated with the LECs' regulated monopoly telecommunications services.
In addition to the historic basis for using Yellow Pages revenues as a source of universal service support, there are compelling economic justifications for this policy as well. Yellow Pages directories serve primarily local markets. That is, the Yellow Pages directory for a given city, which contains listings of businesses and professionals offering products and services in that community, are utilized almost exclusively by local telephone subscribers in that immediate area. The local Yellow Pages directory is distributed without charge to all households and businesses in the coverage area. Like other advertising media, the willingness of advertisers to pay for their listings and ads is directly related to the circulation of the book; put another way, the value of the Yellow Pages directory to the incumbent LEC is directly related to the total number of telephone subscribers in the coverage area.[15] Moreover, that value is not diminished merely because some individual subscribers may elect to take their dial tone service from a competing local carrier. As Yellow Pages telephone directory derives its value, both to advertisers and to the LEC that publishes it, directly from the existence of near universal connectivity to the local network, and it is both appropriate (from a policy standpoint) and economically reasonable for the substantial profits generated by the incumbent LEC from Yellow Pages advertising to support the overall universal service goal. For all these reasons, ratepayers should continue to receive the benefit of Yellow Pages revenues, and such benefit should be used in particular as an explicit offset against any universal service requirement for basic telephone service.
In order to maintain the Yellow Pages contribution as support for affordable residential service generally and to prevent it from simply becoming a means of enhancing the incumbent LEC's competitive position in the residential dial tone market, the contribution (expressed on a per-primary-access-line basis) should be spread across all primary residential access lines, irrespective of which certificated local carrier provides the service. Non-dominant CLECs (who do not publish their own directories and whose customers are included in the incumbent's White and Yellow Pages directories) should be credited with an amount that reflects the Yellow Pages contribution, expressed on a per-line basis, multiplied by the number of primary residential access lines that they provide.[16] CLECs should be allowed to use this per-line "credit" to offset its universal service fund payment obligations or, alternatively, to pay any interconnection charges or other payments to the incumbent LEC that it is required to make.
In addition, costs associated with the provision of universal service should also be adjusted to reflect revenues produced by services and rate elements that are an integral part of the basic residential service offering, but that are often separately priced (although in no sense "optional" or "discretionary" with respect to the central connectivity purpose of a primary residential access line). Examples of rate elements that meet this definition include the basic monthly residential primary line exchange service rate, flat or measured local usage and Extended Area Service (EAS) charges, and any intrastate end user common line (EUCL) charge, as well as the interstate Subscriber Line Charge (SLC) and Carrier Common Line Charge (CCLC), so long as they continue to exist.
It is appropriate to recognize and offset these revenues from the costs of providing universal service, because consumers pay for telephone service pursuant to a complex rate structure, in which individual elements bear widely varying relationships with cost. The "price" that a consumer pays for his telephone service is not the amount of any one, individual rate element (like the basic monthly dial tone line rate), but is instead the total charge for the package of basic residential service components, no matter how these may be structured. That "price" consists of many elements, including the monthly residential (1FR or 1MR) access line rate, the interstate SLC, local calling usage charges, directory assistance charges, and the interstate SLC and CCLC switched access rate elements. The relationship between the rate applicable for each of these rate elements and the cost of each such element (to the extent that such a cost can even be separately identified) is highly variable, reflecting a mixture of public policy and marketing considerations. It would be fundamentally incorrect to compare the costs of providing primary residential access line service with the revenues associated with only selected rate elements of a multi-element "price." Instead, the comparison must be made holistically to the aggregation of all components of the "service" that is furnished to the residential subscriber irrespective of how that service happens to be priced. All that is relevant is the totality of the rate and service elements that comprise the service that is actually furnished to and purchased by the residential subscriber; extraction of individual components in isolation can provide a seriously distorted -- and quite irrelevant -- picture of the cost/revenue relationship that is being examined.
Finally, in addition to the two classes of revenues discussed above, the LECs also receive revenues from optional and discretionary services and from services that can exist independently of the residential dial tone line, including all nonresidential services. Examples of revenues in this category include intraLATA and interLATA toll, custom calling and CLASS services, inside wire maintenance, interstate and intrastate switched access charges (other than the CCLC), and voice mail services. While such revenue sources are not so tied to the provision of basic residential service as to require that they be used directly to offset universal service costs, these revenues should be included in the valuation base used to determine the LEC's share of the aggregate universal service fund contribution.
[2]A study area is the area within a single state jurisdiction where a LEC provides local telephone service. No changes have been permitted in study area boundaries since November 15, 1984, except with special permission of the Commission. See 47 C.F.R. Part 36, Appendix-Glossary (1993).
[3]See Table 7.3, below.
[4]As discussed later in this section, cost is considered indirectly when the rate is set or approved.
[5]Vermont PSB, Universal Service in a Competitive Era: A Report to the Vermont General Assembly by the Vermont Public Service Board, January, 1996, Attachment: Draft Legislation, at 3. The rate would include dial tone, the end user common line charge, and average local measured service charges. Id., at 64, note 141.
[6]Id., at 61.
[7]Id., at 60.
[8]If, instead, the dominant carrier could set price without any regard whatsoever to cost then there would be a clear incentive to raise prices above the statewide average.
[9]The rationale that incumbent LECs advance in support of increased granularity in reporting and applying universal service costs is that this is the level of deaveraging that is confronted by competitors when they consider entry into a particular geographic area. This "cream skimming" accusation has been made by incumbent telecommunications utilities for decades, but the fears have never been borne out in actual practice. The various entities that are now in the process of pursuing entry into the local exchange service market are large firms that are counting upon their ability to amass a large, geographically diverse market base, not isolated market niches that would only be profitable for small, specialized providers.
[10]The existing High Cost Fund provides some kind of precedent. See Table 7.3 above for details on the existing jurisdictional responsibility for high cost support.
[11]If the state establishes an excessively low supported price though, such a decision could thwart the competitive goals of the Telecommunications Act (by creating a relatively larger burden for new entrants). Note, too, that while the Telecommunications Act allows a state to "provide for additional definitions and standards to preserve and advance universal service within that State only to the extent that such regulations adopt specific, predictable, and sufficient mechanisms to support such definitions or standards that do not rely on or burden Federal universal support mechanisms." Telecommunications Act, Sec. 254(f). This means (as the example demonstrates) that if a state decides to support a lower affordability threshold or a broader set of services than those supported under the federal mechanism, the state cannot allocate a share of the additional funding burden to the federal jurisdiction.
[12]Id.
[13]United States. v. American Tel. & Tel. Co., 552 F. Supp. 131, 193-194 (D.D.C. 1982).
[14]For example, Southwestern Bell's directory publishing affiliate tried to compete with New York Telephone by offering its own Manhattan Yellow Pages. Donnelley Directory, a division of the R. H. Donnelley Company, a firm that had long been in the business of publishing and marketing Yellow Pages directories under contract with Bell and non-Bell local telephone companies, attempted to enter the Yellow Pages business in several markets.
[15]The use of advertising revenues to "pay" a portion of the costs of products and services used by consumers is extremely common: Newspapers, magazines, radio and television broadcasters, and other media use these revenues to subsidize the content of their publications; consumers pay for that content indirectly, through the prices of the various goods and services they purchase from the companies that advertise.
[16]Since a customer will be entitled to purchase one and only one primary residential access line per legal dwelling unit, if a customer takes service from two or more local service providers, the customer will be required to designate one carrier as the provider of the subsidized primary access line service.