In the Matter of ) ) Federal-State Joint Board on ) CC Docket No. 96-45 Universal Service ) )
MCI REPLY
TABLE OF CONTENTS
EXECUTIVE SUMMARY i
I. MCI'S PROPOSAL 1
II. THE STUDIES OF THE BCM IN THE COMMENTS DO NOT IMPEACH ITS VALIDITY 4
III. THE JOINT BOARD AND COMMISSION SHOULD NOT ADOPT DIFFERENT MECHANISMS FOR
DETERMINING SUPPORT FOR DIFFERENT SIZES OF COMPANY 8
IV. DEPRECIATION HAS NOT BEEN UNDERSTATED 10
V. THE NEED FOR RATE REBALANCING SHOULD BE MINIMAL 12
VI. FUNDING SHOULD BE COLLECTED BASED ON ALL INTERSTATE REVENUES, NOT JUST END
USER REVENUES 14
VII. NECA SHOULD NOT BE THE ADMINISTRATOR OF THE UNIVERSAL SERVICE SUPPORT
SYSTEM ADOPTED IN THIS PROCEEDING 16
VIII. CONCLUSION 17
EXECUTIVE SUMMARY
The Benchmark Cost Model, with some refinements, is a useful tool for setting
the Universal Service support, and should be used for all carriers regardless
of size. In addition, LEC claims of the need for rate re-balancing and support
for past under-depreciation are overstated. Funding for universal service
support should be based on interstate revenues net of payments to others who
pay into the fund, not from only end user revenues, as some LECs advocate.
Because it is owned and controlled by the incumbent LECs and has not been an
effective administrator of its current responsibilities, NECA should not be the
administrator of the universal service support system the Joint Board and
Commission adopt in this proceeding. The Joint Board and Commission should
adopt MCI's proposals for handling the universal service support
Before the
FEDERAL COMMUNICATIONS COMMISSION
Washington, DC 20554
In the Matter of ) ) Federal-State Joint Board on ) CC Docket No. 96-45 Universal Service ) )
MCI REPLY
A properly designed universal service mechanism can contribute to the development of competition for exchange service, ensure that current high levels of subscribership are preserved and enhanced, and maintain high quality, affordable telephone service for all Americans. It is the competitive market which will be the best guarantor of affordable rates and high-quality service.
I. MCI'S PROPOSAL
In its comments, MCI stated that re-engineering a universal service support system for a competitive environment requires that the Joint Board and Commission "de-link" universal service from existing local exchange carrier (LEC) revenue requirements. The design of a new universal service mechanism must include the following steps:
o define universal service -- the provision of single-line residential access to the first point of switching in a local exchange network, unlimited usage within an exchange area, touch-tone service, white pages listings, and access to 911 and E911 service, operator services, directory assistance, and telecommunications relay service, at a rate no higher than the current nationwide average rate for basic telephony, about $20;[1]
o calculate the subsidy -- the difference between the total service long run incremental cost (TSLRIC) of basic universal service, determined separately for different geographic cost zones reflecting distinctive cost characteristics, and the revenues generated by rates set at the current nationwide average;[2]
o provide a "block grant" of the subsidy amount to the states and require the states to determine the distribution among eligible carriers; the Commission must then remove all subsidies that are built into interstate access rates;[3]
o mandate a "carrier of last resort" auction for any area that is or becomes unserved, allowing auction participants to determine at what subsidy level they would provide service.
Under MCI's proposal, the current system of support flows would be entirely replaced by explicit and targeted subsidy systems at the state and interstate level. Concurrent with identifying the universal service support necessary under this method, existing support flows would be removed from LEC rates. In the interstate jurisdiction, this would require reduction or elimination of the Carrier Common Line (CCL) charge, the current Universal Service Fund (High Cost Fund) charge, Long Term Support, triple-DEM weighting, the Subscriber Line Charge (SLC), and the Local Switching charge.
MCI proposed that the universal service subsidy be based on the difference between economic cost and the nation-wide average rate. A forward-looking model of the economic cost of the network such as the Benchmark Cost Model (BCM) could be used to determine the universal service support level. The BCM is an engineering cost model that computes the cost by Census Block Group (CBG) of serving every area in the country except Alaska, based on data on terrain and soil conditions and number of households.[4] Because the BCM is based on the cost of building the network today, it more closely reflects the economic cost of the network.
Some parties criticized the BCM in particular or the use of proxy cost models in general. Others argued that proxy cost models should not be used to set support levels for small telephone companies. Some LECs claimed that depreciation levels have been too low in the past, leaving them with under-depreciated plant that they must be allowed to recover. Others argued that rate re-balancing is necessary as competition develops. The LECs also advocated collecting universal service support only from services to end users. Finally, several parties nominated the National Exchange Carrier Association for administrator of the universal service support system. MCI responds to these issues below.
II. THE STUDIES OF THE BCM IN THE COMMENTS DO NOT IMPEACH ITS VALIDITY
Several commenters supported the use of a forward-looking proxy cost model such as the BCM.[5] MCI also supports the use of the BCM as a forward-looking proxy cost model to set the size of the universal service support. To ensure that the results of the BCM are based on forward-looking costs, the Commission must use the Hatfield/MCI factors to determine costs. These factors are based on the costs that a company will face to provide local exchange service, rather than on the relationship of past costs to investment.
Other parties opposed use of the BCM. For example, Southwestern Bell performed four analyses of the results of the BCM. In the first, Southwestern Bell compared the average loop investment per loop from the Universal Service Fund (USF) data filed by the National Exchange Carrier Association (NECA) to the BCM average loop investment per household by company.[6] This analysis found that investment from the BCM differed from the investment reported in the USF data, with some companies and states showing greater investment under the BCM and some showing less. The second analysis compared the USF average cost per loop to the BCM average cost per loop, and found the same result. The third analysis compared the Universal Service support payment under the BCM and the current USF plan. Finally, Southwestern Bell compared data by wire center for its operations in Texas with the BCM results, and found substantial differences.
It is not surprising that Southwestern Bell found differences in the loop investment per household. The BCM uses forward-looking technology and current prices of equipment, while the USF loop investments are based on companies' embedded costs. Since costs have been falling, it is not surprising when areas have greater investment under the embedded cost methodology. That some areas actually have less investment per loop under the USF would be surprising, except that Southwestern Bell appears to have used net plant rather than gross plant in its comparison.[7] The BCM results report gross plant; had Southwestern Bell compared the gross plant numbers from the USF data, it would have found few if any areas with greater investment under BCM. In addition, the states in which the BCM appears to overstate the USF investment is primarily the less densely-populated states. The Joint Sponsors are revising the BCM to more accurately reflect the fact that even in rural areas, households tend to cluster rather than being uniformly distributed. This should reduce the loop investment estimated by the BCM for those areas.
The results of Southwestern Bell's cost-per-loop analysis are basically the same as for its analysis of investment. This is to be expected, since the BCM derived costs by applying a factor to the estimated investment. The same explanation for the results of Southwestern Bell's analysis of investment therefore applies to its analysis of costs -- that Southwestern Bell has compared embedded cost to prospective cost, and that refinements to the BCM will more accurately reflect the necessary investment in rural areas.
It is also to be expected, as Southwestern Bell's third analysis shows, that the distribution among companies of support under BCM and under the current USF are different. The two methods represent different concepts of universal service support and are computed on completely different bases. The current USF assigns an additional amount of costs to the interstate jurisdiction based on the amount by which a LEC's reported embedded costs for its entire study area exceed the nation-wide average loop cost. The BCM reports the support as the difference between the LEC's forward-looking cost by CBG and a politically acceptable rate. It is not at all surprising that these two methods give different results, or that some large LECs get more support under the BCM methodology.
Southwestern Bell's final analysis imparts no information at all, because it is based on Southwestern Bell's own wire center data. The BCM investment will differ from booked cost simply because the BCM computes forward-looking investment. In addition, MCI understands that Southwestern Bell's wire center data is derived from Southwestern Bell's study area data through the use of factors rather than being based on the true booked wire center cost. When the embedded investment to which the BCM results are compared is itself of questionable validity, any comparison is meaningless. This analysis should be completely disregarded.
The National Cable Television Association (NCTA) filed a study of the BCM produced by Economics and Technology, Inc. (ETI study), as part of its comments.[8] While agreeing that the BCM can be " a valuable tool for achieving the universal service and local competition policy goals and mandates" of the 1996 Act, the ETI Study suggest several modifications that should be made before the BCM is used for this purpose.[9] The key variables that need adjusting according to the ETI Study are (1) the switch costs need to be updated and adjusted to reflect discounts that the LECs routinely receive; (2) support should be computed on a wire center rather than CBG basis to capture the available economies of scale and scope; (3) fill factors should be adjusted to reflect the fact that the subsidized service, single line residential local exchange service, does not require the excess capacity embedded in the LECs' networks; (4) the method for determining when to deploy fiber rather than copper is flawed, overstating costs; (5) the costs of digital subscriber loop do not reflect manufacturer discounts; and (6) the cost factor applied to the investment estimated by the BCM needs to reflect the forward-looking cost of providing residential local exchange service.
The Joint Sponsors of the BCM are currently working on modifications to the BCM based on comments previously received on the model. Several of the items the ETI Study raises are being modified in the revised BCM. For example, the Joint Sponsors are revising the switching module of the BCM to allow different switch architecture and sizes to be modeled, and the break points determining when copper and fiber plant will be deployed will be made a user input. Other items are available for modification as user inputs, such as if a user of the model has information on vendor discounts. Finally, the cost factor applied in the BCM which uses the Hatfield/MCI approach already reflects the forward-looking costs of providing residential local exchange service.[10]
The only modification that the ETI study advocates that the BCM will not accommodate is computing support at the wire center level. The BCM computes cost by CBG, partly because that is the geographic area for which data are available. The support needed for the wire center can be derived by summing up the data for the CBGs which make up that wire center.
III. THE JOINT BOARD AND COMMISSION SHOULD NOT ADOPT DIFFERENT MECHANISMS FOR DETERMINING SUPPORT FOR DIFFERENT SIZES OF COMPANY
Several of the smaller LECs ask that, even if the Commission and Joint Board adopt a proxy-cost methodology for setting the universal service support method, the support for small companies continue to be based on their reported costs.[11] They argue that a proxy-cost methodology will inevitably miss some special feature of the small LECs' territory and that this will result in the small LEC not receiving sufficient support. In addition, the small LECs have made substantial investment based on the promise of the current universal service support, and must be reimbursed for that investment.
The Commission and Joint Board should not adopt a bifurcated methodology for determining universal service support. First, there is no basis in the statute for different treatment of rural and urban areas. Indeed, since the statute specifically seeks to ensure availability of the same access to services in the rural areas as in urban areas, it would seem to support the use of one unified mechanism for determining universal service support.
Second, a properly-specified proxy model will capture the relevant aspects of even small LEC territories. The BCM currently includes data on terrain and population by CBG throughout the country. In addition modifications underway by the Joint Sponsors should further refine the model's treatment of small LEC and rural areas. Thus, the BCM will adequately capture the cost characteristics of small LEC service areas.
Any bifurcated treatment of rural and urban or large and small carriers will simply encourage the transfer of exchanges to the LEC that will receive the maximum support rather than to the LEC that can serve the area at lowest cost. This will result in a higher than necessary subsidy payment, and a higher than necessary overall cost of service.
The BCM's examination of cost characteristics at the CBG level of disaggregation will allow a reasonably accurate estimation of the cost of serving rural areas. If a company can show that the proxy cost model used to set universal service support misstates its cost, it can always seek a waiver. In addition, MCI's proposal that the state commissions be given a role in determining the distribution of support among the companies within their states would, if adopted, allow the small companies another mechanism for arguing that they should receive more support.
IV. DEPRECIATION HAS NOT BEEN UNDERSTATED
Several parties claim that the investments which the LECs made in past years to meet their universal service obligations are legitimate costs which they are entitled to recover, and that regulators have not allowed the LECs to depreciate their investment at rates anywhere near the pace at which technology has changed.[12] Together these facts imply, these parties argue, that an appropriate combination of rate rebalancing and explicit support mechanisms must be part of the appropriate solution to the preservation of universal service as the market conditions in the local exchange market are radically transformed by the 1996 Act. MCI does not agree that there is a substantial under-depreciation problem.[13] Regulators have been setting depreciation rates based on the plant retirements actually adopted by the LECs for some years now. In fact, most of the LECs have only recently completed an amortization of depreciation reserve deficiencies that corrected the very problem about which the parties complain. At the same time it adopted these amortizations, the Commission adopted the remaining life method of setting depreciation rates, which prevents large depreciation reserve deficiencies from re-occurring.
Even if there were an under-depreciation problem, this would not by itself justify an explicit recovery mechanism for that under-depreciation. The LECs have not been constrained to earn a risk-free rate of return. The interstate rate of return is set at 11.25 percent, nearly 5 percentage points above the (risk-free) ten-year Treasury security interest rate recently cited by the Commission.[14] In addition, the price cap LECs earned nearly 14.1 percent on their interstate services for 1995. Thus the price cap LECs have earned almost 8 percentage points above the risk-free rate of return.
A risk-premium is justified only if the LECs do indeed face some risk of not recovering their full investment. The LECs are not entitled to a guarantee of completely recovering their past investments, any more than any other competitive firm is. To insure the LECs against all risk of what turned out to be poor investment decisions would be to weaken the very discipline that competition is supposed to provide on prices.
The LECs themselves, in a recent proceeding before the Commission, have stated that their current cost of capital, even after taking account of the additional uncertainty of the 1996 Act, remains around 11.25 percent.[15] This estimate was based on a comparison of the costs of capital of firms that face competitive risk that the LECs argued was similar to that faced by them under the 1996 Act. Thus, by their own reckoning, the LECs' earnings are sufficient to reimburse them for the risk they face.[16] No additional funding is needed due to any alleged under-depreciation.
V. THE NEED FOR RATE REBALANCING SHOULD BE MINIMAL
In addition to their purported need for reimbursement for their alleged under-depreciation, some LECs claim that significant rate rebalancing will be necessary as competition develops.[17] This will be necessary, they claim, because they have been constrained to charge high rates to one group of customers or to one part of the area they serve to ensure low rates for another. Thus, for example, high rates for local service for businesses allows low rates for local service for residential customers, or high rates for cities allow lower rates for suburban and rural areas.
MCI does not believe that the need for rate rebalancing is as pervasive as these LECs paint it. Most LEC rates are currently well in excess of LEC costs; for example, only in the very sparsely-populated rural areas do the LECs' costs of providing local service exceed their rates.[18] It will serve no legitimate purpose to allow the LECs to raise rates in some areas when those rates already exceed their true costs. Competition should drive those costs out of the LECs' prices; allowing the LECs to raise rates will merely insure the LECs against the forces of competition, to the detriment of ratepayers. Any LEC claims of the need for rate rebalancing should be allowed only if the LEC makes a showing that the economic cost of the service is not being covered by the current rates.
This is especially true in the case of SLCs. Several parties in this proceeding advocate an increase in the SLC, to more closely align the recovery of costs with the cost-causer.[19] While MCI agrees in principle with the notion that the cost should be borne by the cost-causer, MCI does not agree that an increase in the SLC is justified. As the industry moves into a competitive market, the regulators should not be looking for ways to insulate the LECs' embedded costs from competition. This will result in both higher rates to end users and harm to competition, since new entrants will have to pay these higher SLCs when they resell local service. Rather than shifting cost recovery between different customers of the LECs, the Commission should concentrate on getting all rates reduced to economic cost, either by setting the rules so that competition can drive the rates down, or by regulation where competition will not be effective in driving rates down.[20]
VI. FUNDING SHOULD BE COLLECTED BASED ON ALL INTERSTATE REVENUES, NOT JUST END USER REVENUES
Several of the LECs propose that funding should be based on interstate revenues from only end users.[21] This would prevent double counting, they say, and would avoid giving an incentive to bypass the service used as an input to the service offered to the end user.
MCI agrees that the Commission needs to avoid double-counting in the assessment of funding amounts. However, a more equitable method for avoiding double-counting would be to adopt MCI's proposed approach, which is to assess funding based on total revenues, net of payments to other telecommunications providers for input services. This method would collect money from all interstate providers, including those whose interstate service is access. The LECs' self-serving proposal would let them avoid any responsibility for funding universal service.
While any universal service support the LECs have to pay based on their access revenues may be passed on in higher access charges to their customers, this is not a guaranteed outcome. The amount of universal service payment that customers of any carrier will have to pay depends on the conditions in the marketplace. Depending on the relative elasticities of supply and demand, customers of any particular carrier might end up paying all, some or none of that carrier's universal service obligation. The LECs' shareholders should be placed at the same risk of having to fund the LECs' universal service obligation as other telecommunications carriers' shareholders are of their companies' obligations. If the Commission were to exempt input services from a funding obligation as suggested by the LECs, it would ensure that the retail companies' shareholders and customers would fund universal service, rather than all telecommunications companies' shareholders and customers as Congress intended.
In addition, commercial mobile radio service (CMRS) providers should contribute to the interstate universal service fund.[22] Because implementation of the universal service fund that will be adopted in this proceeding is made by the Joint Board and the Commission, contributions to the universal service are a federal decision. Therefore, the CMRS providers can and should be required to contribute to the federal universal service fund or funds, based on their revenues net of payments to other carriers for input services. Similarly, they should be eligible to receive support payments in areas where they are providing the services that receive universal service support.[23]
VII. NECA SHOULD NOT BE THE ADMINISTRATOR OF THE UNIVERSAL SERVICE SUPPORT SYSTEM ADOPTED IN THIS PROCEEDING
Several parties urge the Commission to appoint NECA as the administrator of any universal service fund set up by this proceeding.[24] MCI urges the Commission not to use NECA for this purpose. NECA is an entity set up by the Commission, but owned and controlled by the incumbent LECs. As such, it would not be an impartial administrator of any fund which must be paid to any LEC. Any outside party selected by the Commission or the states to administer the fund should be chosen only after bids have been submitted in response to requests for proposal, as suggested by various parties.[25] This method would ensure that the fund or funds would be administered at lowest cost.
The Commission has previously described how poorly NECA has performed its most fundamental responsibilities.[26] In the Show Cause Order, the Commission, among other things, found it necessary to scrutinize further NECA's core operations because it believed that it was obligated to "ensure that NECA is discharging its responsibilities under [Commission] rules."[27] NECA has not shown itself to be a trustworthy executor of its current responsibilities; it should not be entrusted with administration of the universal service fund or funds that are so vital to the development of competition.
VIII. CONCLUSION
As discussed infra, the Benchmark Cost Model, with some refinements, is a useful tool for setting the Universal Service support, and should be used for all carriers regardless of size. In addition, LEC claims of the need for rate re-balancing and support for past under-depreciation are overstated. Funding for universal service support should be based on interstate revenues net of payments to others who pay into the fund, not from only end user revenues, as some LECs advocate. Because it is owned and controlled by the incumbent LECs and has not been an effective administrator of its current responsibilities, NECA should not be the administrator of the universal service support system the Joint Board and Commission adopt in this proceeding. The Joint Board and Commission should adopt MCI's proposals for handling the universal service support.
Respectfully submitted,
MCI TELECOMMUNICATIONS CORPORATION
Chris Frentrup
Senior Regulatory Analyst
1801 Pennsylvania Avenue, NW
Washington, DC 20036
(202) 887-2731
May 7, 1996
STATEMENT OF VERIFICATION
I have read the foregoing and, to the best of my knowledge, information, and belief, there is good ground to support it, and it is not interposed for delay. I verify under penalty of perjury that the foregoing is true and correct. Executed on May 7, 1996.
Chris Frentrup
1801 Pennsylvania Avenue, NW
Washington, D.C. 20006
(202) 887 2731