U.S. Comments on MPT LRIC Study Group Report to the Telecommunications Council The United States appreciates the opportunity to express its views in response to the Telecommunications Council's request for comments on the introduction of Long Run Incremental Cost (LRIC)-based pricing for interconnection in Japan. Summary The introduction of LRIC-based interconnection rates within CY 2000 was one of the most important commitments made by the Government of Japan in the 1998 deregulation and competition policy discussions with the U.S. Government. Failure to implement this methodology in a thorough manner would call into serious question the Government of Japan's adherence to this pledge, and overall commitment to promoting an open and competitive telecommunications market. The United States believes that MPT's final LRIC model remains fundamentally flawed and does not meet this goal and that additional changes are needed to make it an accurate reflection of market-based interconnection costs, which is necessary to promote competition in the Japanese telecommunications market. Since MPT has presented two flawed options for introducing LRIC interconnection (A and B), it clearly needs an option C, reflecting true LRIC principles. Furthermore, MPT's suggestion that in implementing the model it should incorporate factors such as impact on NTT operations, end user rates, and universal service into interconnection pricing is unnecessary and inappropriate and runs directly counter to the basis for, and goals of, LRIC pricing. To the extent universal service funding is appropriate, these fees should be determined through a separate, transparent process. Inclusion of unrelated issues like user rates and NTT operations in this list suggest that MPT believes it is proper for NTT's competitors to bear the burden of NTT's inefficiencies on an ongoing basis - which, again is contrary to LRIC principles and to efficient functioning of the telecommunications sector. Concerns about LRIC-based interconnection rates leading to higher user rates or serious disruptions in NTT's operations do not appear credible and run contrary to experience in the United States. General Points As the Telecommunications Council deliberates on how to introduce LRIC-based interconnection rates in Japan, it should keep in mind the following general points. First, the introduction of LRIC interconnection rates within CY 2000 represents one of the most important commitments made by the Government of Japan in the 1998 deregulation and competition policy discussions with the U.S. Government. Failure to implement this methodology in a thorough and timely manner would call into serious question the Government of Japan's adherence to this pledge, and overall commitment to promoting an open and competitive telecommunications market. Second, elements of the model as currently configured do not reflect the internationally-accepted principles of a forward-looking LRIC model and are thus inconsistent with the Government of Japan's pledge. The Telecommunications Council's discussions on how to implement the model and translate costs into prices is an opportunity to address these deficiencies. Failure to do so will inevitably require this issue to be addressed at a senior government-to-government levels. Third, in presenting issues to be considered by the Telecommunications Council in the introduction of LRIC-based interconnection pricing, MPT has included issues that are not directly relevant in setting interconnection rates - e.g., the impact on NTT's operations, universal services, and user rates. The value of a LRIC model is to provide an objective and explicit basis for analyzing costs, and thus approximating market-based prices for interconnection. Combining consideration of the proposed non-relevant factors with the LRIC model will result in interconnection rates that do not accurately reflect the costs of interconnection and do not promote competition in the telecommunications market. MPT's emphasis on such factors indicates that promotion of competition is not a key objective in consideration of the model, and that MPT believes it is appropriate for NTT's competitors to bear the burden of NTT's inefficiencies on an ongoing basis - contrary to the efficient functioning of the telecommunications sector. While issues such as universal service may warrant consideration within Japan, this process should be clearly separated from interconnection pricing and conducted separately on an open, objective and transparent basis. In this light, it is extremely troubling that MPT would suggest that the Telecommunications Council examine NTT's "actual costs" as opposed to the "virtual costs" of a model. This reveals a fundamental misunderstanding on the part of MPT regarding the economics of regulated telecommunications markets and the need for forward-looking pricing: i.e. a failure to understand that a forward-looking model better replicates the actual economic costs faced by an incumbent than the artificially-manipulated accounting costs that NTT controls--costs that former Vice-Minister Igarashi himself termed "donburi kanjo." The introduction of ABC accounting has not significantly improved this situation. In addition to fairly compensating NTT, forward-looking LRIC-based prices send the correct signals and provide the proper incentives for competitors to invest in the telecommunications in the most efficient manner. Specific Points Below is additional elaboration on these points. Deficiencies of the existing model: In its report, the Working Group describes two scenarios based on different assumptions, ("A" and "B") that result in radically different results. The report states that the significantly higher costs produced by scenario A are the result of including non-traffic sensitive costs in the cost of switching. There is absolutely no justification for such an approach; indeed, it would betray one of the essential foundations of a forward-looking model, which is to accurately model costs based on the manner in which they are incurred. Accepting a model which provides a basis for recovering non-traffic sensitive costs on a usage-sensitive basis would perpetuate inefficient and anti-competitive structures the model is intended to eliminate. Passed on to consumers, these usage-sensitive prices result in per-minute prices that limit usage of the network for voice, and increasingly Internet. Thus, as a starting point, scenario A is completely unacceptable. Furthermore, it appears that even scenario B seriously overstates incremental costs of interconnection. This is evident from at least two perspectives. First, contrary to numerous recommendations by commenting parties, there is no evidence that the revised model fully eliminated non-traffic sensitive costs from the switching module (e.g. vertical switching features such as touch-tone dialing and other ancillary services that end-users choose to buy are incorrectly attributed as a cost to be borne by competing interconnecting carriers.) Experts estimate that such costs can account for up to 20 percent of the fixed cost of a switch; since these costs are directly recovered from end-users, they have no business being included in interconnection costs or charges. The Telecommunications Council should instruct the working group to revise the model to make such adjustments eliminating ALL non-traffic sensitive costs from the switching module. Second, the report includes an entirely unsatisfactory description of attempts to address depreciation issues. While the Report notes the wide variations between NTT, working group, BT and U.S. depreciation rates, it concludes by stating that using foreign data as a reference would "harm the credibility of the model... because lives of equipment vary among countries." While there may be legitimate variations among depreciation rates in different countries, the fact that the depreciation rates proposed by the MPT model are significantly higher than those for the U.S. and BT should immediately signal a problem in the MPT model. If MPT is unwilling to accept foreign data it must credibly demonstrate that its domestic data is superior based on an objective study and subject to comment and rigorous audits. Such a study must go beyond NTT and must go beyond the NCCs in Japan if they have not been in business long enough to provide reliable long-term data. Absent such a wide-ranging objective study, MPT should adjust its depreciation periods to reflect data in foreign countries. At a minimum, the MPT could use foreign data as an interim reference, while introducing a more systematic methodology for determining accurate depreciation rates. It is commonly understood that regulations that guarantee cost recovery based on the rate-of-return methodology to which NTT has been subject skew depreciation decisions by regulated entities. Therefore, usage of depreciation data from countries using non cost-based regulations (i.e. price caps) or competition more accurately reflects forward-looking LRIC principles. One category alone--the service life of fiber-optic cable--should be enough to indicate how flawed MPT's current methodology is. There is no manufacturer the United States is aware of--in the United States, Europe, or Japan--which expects fiber optic cables to only last 11.2 years-- the MPT rate chosen. Experts typically expect fiber to often have a service life of over twenty years. The advent of dense wave-division multiplexing may further extend the economic life of fiber, by permitting increases in capacity to be achieved by use of electronics, rather than additional fiber. In this regard, the Telecommunications Council should be extremely skeptical of suggestions that adoption of Scenario B would require that end-user rates be raised. This scenario currently does not take into account the additional reductions in costs through incorporating the above measures in the model. Instead, it appears that this is an attempt to simply recuperate the opportunity cost incurred by competition. In addition, NTT already charges its customers extremely high local service fees, recovering its costs through a variety of means--an extraordinarily high up front line connection charge; monthly basic fees, and high per-minute end user charges. It is likely that consideration of NTT's full cost-recovery mechanisms--combined with incorporation of more accurate depreciation rates and correct attribution of user-incurred charges into the LRIC model would eliminate any NTT justification for raising user rates - and that the enhanced competition caused by lowering interconnection rates would provide long-term pressure to lower NTT's local service rates. Need to Look at Actual Versus Virtual Costs The MPT has suggested that the Telecommunications Council should take into consideration the difference between the model and the actual operation of the network--to compare "actual" versus "virtual" costs. This point is based on a fundamental misperception concerning the purpose of a forward-looking model. A forward-looking model ignores embedded, historical costs because, from an economic point of view, these costs do not reflect the incumbent's real pricing constraints--the costs a competitive market would permit a company to recover. Embedded costs are sunk costs. While the incumbent may try to recover these costs, permitting it to recover all such costs by imposing them on competitors eliminates market-based incentives to operate efficiently, and imposes non-market-based burdens on the competitors. From an economic point of view, forward-looking costs are more accurate and more "real" than accounting costs. NTT's accounting costs include inflated personnel, equipment, and depreciation expenses. Where recurring costs (e.g. personnel) do not reflect market-based costs, it is NTT's responsibility to reduce them, not impose them on competitors; where they are sunk costs (e.g. equipment), they do not represent a cost NTT must recover to operate. NTT's recent introduction of discount plans demonstrates that NTT has broad freedom to price services irrespective of past investments. Failure to extend to competitors the much lower actual price floor established by forward-looking economic costs exposes competitors to widespread "price squeezes", which appears already to have begun. MPT's Request for Consideration of the Effects of LRIC-based Pricing on Universal Service, End-user Rates, and Business Operations of the Incumbent Local Exchange Carrier MPT has asked the Telecommunications Council to consider the impact introducing LRIC-based prices may have on disrupting universal service and the destructive effect such introduction could have on end-user rates and business operations of NTT East and West. The introduction of a an explicit universal service program in Japan may merit consideration, but this issue is not directly related to interconnection and funding for universal services should not be incorporated into interconnection rates. To determine whether competitors should have to contribute to the maintenance of universal services, MPT should first define universal service goals, fully analyze whether any shortfalls in meeting such goals exist, determine the costs of funding possible shortfalls, and consider a competitively-neutral system for recovering such costs. In addition, competitors should be able to benefit from such a system when they provide service to customers receiving universal service subsidies, providing incentives to extend competition to all customer segments and regions. Such a process should be conducted in a transparent manner. Given the absence of preparation in these areas, the idea of addressing this issue by simply manipulating interconnection rates is completely inappropriate and will lead to interconnection rates that undermine competition-- as well as universal service objectives. Regarding the potential of LRIC-based interconnection to be "destructive" of end-user rates, the Telecommunications Council should bear in mind that "destruction" of rates is not entirely a bad thing, and that, in general, "price destruction"--i.e. reduction in rates--is one of the key consumer benefits of deregulation and competition policy. It is curious that the MPT would oppose reducing end-user rates, or the expansion of competition that would have precisely this long-term effect. Regarding the potential of LRIC-based rates to be "destructive" of NTT's business operations, we disagree with the implicit assumption that promoting competition through lowering interconnection rates will negatively impact NTT's operations. In fact, the U.S. experience has been the contrary - the introduction of competition has actually strengthened incumbents. While an incumbent carrier's market share may fall, the expansion of market demand and greater efficiency caused by competition increase the incumbent carrier's total revenues. For example, while the percentage of long distance traffic carried by AT&T fell from 80% in 1986 to 52% in 1997, its actual traffic minutes increased significantly from 141 billion minutes to 250 billion minutes. In revenue terms, while AT&T's share of total long distance revenues declined from 65% in 1990 to 45% in 1997, its total revenues increased by over $5 billion dollars during this same period. Similarly, while the U.S. Regional Bell Operating Companies (RBOCs) have lost over a billion dollars of income annually because of reductions in interstate access charges that have recently declined at a rate of over 20% a year, they are still extremely healthy, and the increased volume in long-distance origination and termination has significantly cushioned the impact of these reductions. These companies have typically reported rates of return of over 10%, showing that they have prospered in an increasingly competitive environment 1 1