[2] In the transmittal letter for the BCM results, the following statement appears: "The Joint Sponsors support the use of the BCM for the analysis of the targeting of explicit high cost support. They do not agree on its use for other purposes such as the setting of rates for telephone service." Letter from Glenn Brown, U S WEST, Inc., et al. to William F. Caton, FCC, filed Dec. 1, 1995, at n.*. And see Letter from Glenn Brown, U S WEST, Inc. to Richard Metzgar, FCC, dated April 16, 1996, at 2 (attached hereto as Appendix B).

[3] Telecommunications Act of 1996, Pub. L. No. 104-104, 110 Stat. 56 (1996) ("1996 Act" or "Act").

[4] Certainly, it would be inappropriate to argue that the Commission has only a "25% interest" in universal service matters, including high-cost funding.

[5] Below, the Commission will see references to the "BCM" without the reference to ARMIS. This will be within the context of discussing the original crafting and designing of the BCM and its basic model assumptions. See pp. 7-17, below.

[6] Comments on the BCM generally fall into two broad categories: those parties that support the use of proxy models for establishing a high-cost fund; and those parties that support the use of actual or book costs for establishing a high-cost fund. For example, BellSouth Corporation, et al. and Southwestern Bell Telephone Company fall into the latter category.

[7] See, e.g., NCTA Comments at iii (a proxy model should be used to determine USF support levels and the BCM is currently the best available model, although it might require certain corrections); 9-10 (BCM is superior to model proposed by Pacific Telesis);Teleport Communications Group, Inc. Comments at 7 (while the BCM is not without its shortcomings, those shortcomings are generally identifiable and potentially correctable); NASUCA Comments at 19 (the Commission should adopt a proxy cost model that calculates forward-looking verifiable cost estimates); 20-21 (BCM is superior to model proposed by Pacific Telesis); and see also AT&T Corp. ("AT&T") Comments at App. A.

[8] Of course, incumbent LECs are entitled to full recovery of their reasonable actual costs, through a combination of universal service funding and rates to customers.

[9] Some new entrants, such as NCTA, argue for a variation on the basic theme. NCTA supports the use of the BCM to determine CBG specific costs, but proposes that high-cost support be provided only at the wire center level. NCTA Comments at 10.

[10] In our opening Comments, U S WEST presented credible, persuasive evidence on the issue of why "wire center targeting" was not appropriate for the targeting of high-cost support. U S WEST Comments at 12 n.25 and App. A. Wire center targeting would allow a cable company, for example, to serve a "wire center" by actually providing service only to customers near the in-town central office. The cable company would receive "high-cost assistance" to serve customers that were not high cost to serve (perhaps even making a profit on the difference between its actual costs and the money it received for serving the "high-cost area"). And, it would not serve customers who in fact were high cost to serve, i.e., those at the outer edges of the wire center. Those customers would be left to the incumbent service provider -- who would also need high-cost funding to serve those customers. This is not a good economic or policy result. Wire center "targeting" is simply not the best targeting. Given the fact that nothing absolutely drives one to that level of targeting, it should be rejected as contrary to sound economics and prudent universal service public policy.

It is encouraging that some parties who do not even support the concept of "proxy models" see the benefits of targeting high-cost fund support at smaller geographic areas than wire centers. GVNW's Comments (at 13) would, for example, find the BCM an acceptable tool to disaggregate actual (as opposed to proxy) costs below the wire center level.

¹¹ This "give a little, get a lot" approach to universal service funding issues is, obviously, not one that strikes the most balanced economic or public policy resolution of the complex issues involved.

[12] MCI, through its work with Hatfield Associates ("Hatfield"), began this process somewhat early on. Until recently, however, MCI was cautious in promoting the more extensive revisions that Hatfield argued needed to be made to the BCM. For example, in the State of Washington, the Hatfield Study was introduced through the advocacy of AT&T, rather than that of MCI (although MCI "supported" the AT&T advocacy). Recently, however, in a proceeding in Pennsylvania and in ex parte filings, MCI itself has made clear its departure from the ranks of those who support the original BCM. In the Pennsylvania proceeding, AT&T and MCI sponsored the introduction of the Hatfield material. And see Letters from Leonard S. Sawicki, MCI to W. Caton, FCC, filed Mar. 28, 1996 and Apr. 9, 1996.

[13] In our opening comments, U S WEST indicated that a number of enhancements are currently planned to be made to the BCM/ARMIS. See U S WEST Comments at App. A. These changes will advance the BCM's ability to more accurately reflect distribution plant in urban areas.

[14] Telecommunications Act of 1996, 110 Stat. at 72, [[section]] 254(b)(5).

[15] See BellSouth Comments at Att. A (NERA Comments on Universal Service), p. 40.

[16] Id. at Att. A, p.38 ("the optimization process usually succeeds only at providing the lower bound on incremental costs").

[17] Id. at Att. A, p. 39 ("While such a model may well serve as a predictor of costs for a new network, it cannot possibly depict costs of an existing network with its inherent rigidities.") (emphasis in original).

[18] Even if the BCM/ARMIS were "corrected" with respect to certain fundamental Total Service Long-Run Incremental Cost ("TSLRIC") attributes (see note 37, infra), the BCM would not represent any individual company's TSLRIC for at least four reasons. First, the BCM uses national level cost data for the major network components, where individual companies' material prices are based on company-specific contracts. Second, the structure costs of the network are also based on national average contractor prices. Third, individual companies may use forward-looking technology or mix of technologies different than the BCM. Finally, the BCM utilizes a hypothetical network design and does not attempt to replicate any individual company's network arrangements.

[19] Comments of U S WEST at 12 n.25.

[20] Attachment to Letter from Glenn Brown, U S WEST, Inc., et al. to William F. Caton, FCC, filed Dec. 1, 1995, at I-2 (Executive Summary).

[21] See note 2, supra.

[22] See Ex Parte Letter from Leonard S. Sawicki, MCI, to W. Caton, FCC, filed Mar. 28, 1996 (The Cost of Basic Network Elements: Theory, Modeling and Policy Implications, prepared by Hatfield Associates for MCI, pp. 12, 16) and Comments of NCTA at Att. A (The Cost of Universal Service: A Critical Assessment of the Benchmark Cost Model, by Economics and Technology, Inc., p.180).

[23] See Surrebuttal of Peter B. Copeland, USWC, in Docket No. 95-2206-01, before the Utah Public Service Commission, May 1, 1996, at 6 (attached hereto as Appendix C) and Direct Testimony of Robert A. Mercer, AT&T, in Docket No. 95-2206-01, before the Utah Public Service Commission, Mar. 14, 1996, at 5 (attached hereto as Appendix D).

[24] Any model that utilizes the BCM or the loop portion of the BCM, as the Hatfield Model purports to do, also will produce results with a low calculation of urban distribution plant. For LRIC purposes, this underestimation would have to be corrected.

[25] Washington Utilities and Transportation Commission v. USWC, Docket No. UT-950200, Fifteenth Supplemental Order, at Part 5, II, A, 3.

[26] While Dr. Mercer has testified that the model incorporates three of the four BCM modules: 1) the BCM input data (which assigns CBGs to the closest central office, determines the CBG's spatial relationship to the CBG, and lists the USGS terrain data associated with the CBG); 2) a module which determines the feeder quadrant on which a CBG is served, the feeder plant distance, the distribution plant distance, and the terrain structure multipliers applicable to the CBG; and 3) a module which designs the feeder and distribution plant with the appropriate sharing of feeder plant, the associated structure and the total investments involved for the major cost drivers contained in the model (see Direct Testimony of Robert A. Mercer, supra note 23, at 6-7), the inability to independently verify any of this requires a leap of faith to assume that the Hatfield Study actually incorporates intact the original BCM modules.

[27] See Surrebuttal Testimony of Geraladine G. Santos-Rach, USWC, in Docket No. 95-2206-01, before the Utah Public Service Commission, May 1, 1996, at 6 (attached hereto as Appendix E). And see Rebuttal Testimony of Peter B. Copeland, USWC, in Docket No. 95-2206-01, before the Utah Public Service Commission, at 24-26 (attached hereto as Appendix F) for a discussion of what U S WEST has been advised was changed in the original BCM module and how those represented changes would affect the Study.

[28] See Surrebuttal of Peter B. Copeland, supra note 23, at 4.

[29] This subject matter expert is also a U S WEST witness in state rate, universal service, and interconnection cases.

[30] AT&T had a representative at the inspection site.

[31] In Docket Nos. 95-2206-01, 2202-01 and 94-999-01 (Utah), in response to a data request from U S WEST, Dr. Mercer, indicated that no documents describing the purpose and function of the Hatfield Model exist; no documents describing the methods and procedures used in the Model exist; and that Dr. Mercer's testimony was the sole source of the Model's assumptions. With respect to certain critical input data, Dr. Mercer indicated that the basis of certain Model assumptions were "conversations we have had over the years with LEC staff involved in [Digital Loop Carrier] procurement...." Direct Testimony of Robert A. Mercer, supra note 23, at 7. Those conversations were not, according to Dr. Mercer, recorded. Essentially, Hatfield appears to have made input changes based on educated guesses, rather than on first-hand knowledge or on sources that could be documented or validated.

[32] See In the Matter of Implementation of the Local Competition Provisions in the Telecommunications Act of 1996, CC Docket No. 96-98, Notice of Proposed Rulemaking, FCC 96-182, rel. Apr. 19, 1996, at [[paragraph]] 134 ("Interconnection NPRM").

[33] Later, on November 1, 1995, the Joint Sponsors filed data for 17 additional states; and, on December 1, 1995, data for all remaining states, except Alaska. The December 1st filing included a written summary of data for 49 states, plus Washington, D.C., as well as CD ROMs that included all the detailed computer runs for each state.

[34] "The BCM represents what may be the most comprehensive attempt to date to develop an objective national model of the cost of providing basic residential local exchange service....The BCM satisfies many of the essential attributes of a useful cost proxy model, and can be a valuable tool for achieving the universal service and local competition policy goals and mandates of the Telecommunications Act of 1996." ETI, The Cost of Universal Service: A Critical Assessment of the Benchmark Cost Model, supra note 22, at iii-iv.

[35] See id. at 112 n.166. See also Letter from Attorney Judson D. Cary, U S WEST, Inc. to L. Selwyn, Economics and Technology, Inc., dated Apr. 26, 1996 (attached hereto as Appendix G).

[36] ETI, The Cost of Universal Service: A Critical Assessment of the Benchmark Cost Model, supra note 22, at 15.

[37] The BCM is not a TSLRIC study of basic local service, although it uses forward-looking technology for its investment base. There are a number of areas where the BCM methodology departs from general TSLRIC principles. The two most important areas of difference are: 1) The BCM does not include long-run demand for local service that matches a long-run planning horizon generally defined by TSLRIC studies as a period of time long enough so that cost estimates are based on the assumption that all inputs are variable (see ETI, The Cost of Universal Service: A Critical Assessment of the Benchmark Cost Model, supra note 22, at 15; AT&T Comments at App. A.), and 2) The BCM's annual expense factors are based on historical relationships, not on forward-looking studies for the provision of basic service (see Rebuttal Testimony of Peter B. Copeland, supra note 27 at 13). And see note 18, supra.

[38] Additionally, those who attack the BCM from a LRIC perspective do not themselves proffer what U S WEST would characterize as sound LRIC arguments or analyses, in any event. See discussion immediately below.

[39] In the Executive Summary of the BCM results transmission it was stated that the purpose of the study was "to identify areas where [the] cost of service can reasonably be expected to be so high as to require explicit high cost support for the preservation of universal service." Note 20, supra.

[40] See, e.g., MCI Comments at 4-5; BellSouth Comments at 9.

[41] U S WEST Comments at ii, 1-2.

[42] Telecommunications Act of 1996, 110 Stat. at 86, [[section]] 104, amending 47 USC [[section]] 151. See also id., 110 Stat. at 72, [[section]] 254(b)(3).

[43] See id., 110 Stat. at 71-72, [[section]] 254(b).

[44] Telecommunications services, by their nature, exist without regard to state boundaries. Indeed, many of these services are provided via airspace above and across multiple state boundaries. While the retail delivery of telecommunications services might well exist in a locality or a state, the business of telecommunications, the planning for telecommunications services rarely -- if ever -- is state-specific. Compare the Commission's recently issued Interconnection NPRM wherein the Commission acknowledges (albeit in a somewhat different context) that "[i]t would make little sense, in terms of economics, technology, or jurisdiction, to distinguish between interstate and intrastate components." Interconnection NPRM [[paragraph]] 37.

[45] See, e.g., AT&T Comments at 8 ("A surcharge on all retail telecommunications services, both interstate and intrastate, creates a fair, simple and efficient recovery mechanism."); GTE Comments at 16 (proposing that funding should be obtained on the basis of a single, uniform surcharge applied to all end-user transactions).

[46] See AT&T Comments at 8 n.10.

[47] It appears evident from AT&T's Comments that it believes it will have to pay a facilities-based carrier a surcharge on services AT&T buys in order for it to sell those services to an ultimate end user. AT&T Comments at 8 n.10 ("to avoid any double count, resellers would certify the portion of the telecommunications services that they purchased which are used for resale and apply to the NUSF administrator for a surcharge credit for those exempted purchases"); 22 ("surcharge credits to resellers for the surcharge they pay to their facilities-based carriers" would have to be part of the USF funding process).

[48] See MCI Comments at 16; Sprint Comments at 17; see, e.g., MFS Comments at 16.

[49] U S WEST Comments at 18-20 and App. B.