WPC\ 2BJZ|{ X4#X\  P6G;ɒP#X01Í ÍX01Í Í#Xj\  P6G;ynXP# X'  ({ S'#&a\  P6G;u&P# Federal Communications Commission`g(#bFCC 99120 ă   yxdddy( I. A. 1. a.(1)(a) i) a) I. 1. 1. a.(1)(a) i) a) vf#X\  P6G;ɒP#Before the FEDERAL COMMUNICATIONS COMMISSION  yO'a7Washington, D.C. 20554#Xj\  P6G;ynXP#у  Xb4 In the Matter of hhCq) ` `  hhCq)  X44FederalState Joint Board onhhCq)ppCC Docket No. 9645  X4Universal Service hhCq) ` `  hhCq)  X4ForwardLooking Mechanism hhCq)ppCC Docket No. 97160  X4for High Cost Support forhhCq)  X 4NonRural LECs hhCq)  X ' FURTHER NOTICE OF PROPOSED RULEMAKING  Xe 4Adopted: May 27, 1999 `(# Released: May 28, 1999 ă By the Commission: Commissioner FurchtgottRoth dissenting and issuing a sta &F tement at a later dat &F e &F .  X4 Comment Date: July 2, 1999  X4 Reply Date: July 16, 1999  X4- TABLE OF CONTENTS ă  Xh4!para. (#(#\   X\I. INTRODUCTION p>"(# 1 X\II. PROCEDURAL HISTORY p>"(# 4  X4XX` ` A.` ` Universal Service Order ` p>"(# 4  X4XX` ` B.` ` Further Notice and the Input Value Development Process ` p>"(# 6  X4XX` ` C. ` ` Platform Order and Second Recommended Decision ` p>"(# 9 X\III. ESTIMATING FORWARDLOOKING ECONOMIC COST p!(# 11  X 4XX` ` A.` ` Designing a ForwardLooking Wireline Local Telephone Network ` p!(# 11  Xq!4XX` ` B. ` ` Synthesis Model ` p"(# 16  XZ"4XX` ` C.` ` Selecting ForwardLooking Input Values ` p"(# 20 X\IV. DETERMINING CUSTOMER LOCATIONS p"(# 23  X%4XX` ` A. ` ` Background ` p"(# 23"%0*&&``e#"Ԍ X4XX` ` B. ` ` Issues for Comment ` p"(# 25  X4XX` ` ` ` 1.  Geocode Data ` p"(# 25  X4XX` ` ` ` 2.  Road Surrogate Customer Locations ` p"(# 29  X4XX` ` X ` ` 3.  Methodology for Estimating the Number of Customer Locations p"(# 35 X\V. OUTSIDE PLANT INPUT VALUES p"(# 48  Xv4XX` ` A.` ` Background ` p"(# 48  X_4XX` ` B.` ` Copper and Fiber Cable ` p"(# 51  XH4XX` ` X ` ` 1. Background p"(# 51  X14XX` ` X ` ` 2. Issues for Comment p"(# 54  X 4XX` ` C.` ` Structure Costs ` p!(# 103  X 4XX` ` X ` ` 1. Background p!(# 103  X 4XX` ` X ` ` 2. Issues for Comment p!(# 106  X 4XX` ` D.` ` Structure Sharing ` p!(# 126  X 4XX` ` X ` ` 1. Background p!(# 126  X 4XX` ` X ` ` 2. Issues for Comment p!(# 129  X4XX` ` E.` ` Serving Area Interfaces ` p!(# 133  Xy4XX` ` X ` ` 1. Background p!(# 133  Xb4XX` ` X ` ` 2. Issues for Comment p!(# 136  XK4XX` ` F.` ` Digital Loop Carriers ` p!(# 142  X44XX` ` X ` ` 1. Background p!(# 142  X4XX` ` X ` ` 2. Issues for Comment p!(# 143 X\VI. SWITCHING AND INTEROFFICE FACILITIES p!(# 147  X4XX` ` A. ` ` Background ` p!(# 147  X4XX` ` B. ` ` Issues for Comment ` p!(# 151  X4XX` ` X ` ` 1.  Switch Costs p!(# 151  X4XX` ` X ` ` 2.  Use of the Local Exchange Routing Guide (LERG) p!(# 174  X|4XX` ` X ` ` 3.  Other Switching and Interoffice Transport Inputs p!(# 178 X\VII. EXPENSES p!(# 192  X74XX` ` A.` ` Background ` p!(# 193  X 4XX` ` B.` ` Issues for Comment ` p!(# 196  X 4XX` ` X ` ` 1. Plant Specific Operations Expenses p!(# 197  X4XX` ` X ` ` 2. Common Support Service Expenses p!(# 212 X\VIII. CAPITAL COSTS p!(# 228  X!4XX` ` A.` ` Depreciation ` p!(# 229  X"4XX` ` X ` ` 1.  Background p!(# 229  X#4XX` ` ` ` 2.  Issues for Comment ` p!(# 231  Xh$4XX` ` B.` ` Cost of Capital ` p!(# 237  XQ%4XX` ` C.` ` Annual Charge Factors ` p!(# 240"Q%0*&&``e#"ԌX\IX. OTHER ISSUES RELATED TO THE HIGH COST MECHANISM p!(# 243  X4XX` ` A. ` ` Alternatives to the ForwardLooking Cost Model ` p!(# 243  X4XX` ` B. ` ` Proposed Modification to Procedures for Distinguishing Rural and NonRural Companies ` p!(# 244  X4XX` ` X ` ` 1.  Background p!(# 244  X4XX` ` X ` ` 2.  Issues for Comment p!(# 246 X\X. PROCEDURAL MATTERS AND ORDERING CLAUSE p!(#256  XH4XX` ` A. ` ` Ex Parte Presentations ` p!(#256  X34XX` ` B. ` ` Initial Regulatory Flexibility Act ` p!(#257  X 4XX` ` C. ` ` Initial Paperwork Reduction Act Analysis ` p!(#272  X 4XX` ` D. ` ` Deadlines and Instructions for Filing Comments ` p!(#273  X 4XX` ` E. ` ` Ordering Clauses ` p!(#276 X\Appendix A (Proposed Input Values) p!(#A1 X\Appendix B (Description of PNR's Methodology to Estimate the Number of Customer Locations) p!(#B1 X\Appendix C (Outside Plant Structure and Cable Costs Survey) p!(#C1 X\Appendix D (Description of Proposed Methodology for Estimating Outside Plant Costs) p!(#D1 X\Appendix E (Description of Switching Costs) p"(#E1 X\Appendix F (Description of Proposed Methodology for Estimating Expenses) p"(#F1  X':M I. INTRODUCTION א\  X41. ` ` In the Telecommunications Act of 1996 (1996 Act),<ZD6 yO%'ԍ Pub. L. No. 104104, 110 Stat. 56. The 1996 Act amended the Communications Act of 1934, 47 U.S.C.  {O' 151 et. seq. (Act). Hereinafter, all citations to the Act will be to the relevant section of the United States Code unless otherwise noted.< Congress directed this Commission and the states to take the steps necessary to establish support mechanisms to ensure the delivery of affordable telecommunications service to all Americans. In response to this directive, the Commission has taken action to put in place a universal service support  XP4system that will be sustainable in an increasingly competitive marketplace. In the Universal  X;4Service Order , the Commission adopted a plan for universal service support for rural, insular, and high cost areas to replace longstanding federal subsidies to incumbent local telephone  X4companies with explicit, competitively neutral federal universal service support mechanisms.$D6 {O#'ԍ Federal-State Joint Board on Universal Service, Report and Order, CC Docket No. 96-45, 12 FCC Rcd  {Ot$'8776 (1997) (Universal Service Order), as corrected by Federal-State Joint Board on Universal Service, Errata, CC Docket No. 96-45, FCC 97-157 (rel. June 4, 1997), appeal pending in Texas Office of Public Utility Counsel v. FCC and USA, No. 97-60421 (5th Cir. 1997).  "0*&&``"  X4The Commission adopted the recommendation of the FederalState Joint Board on Universal Service (Joint Board) that an eligible carrier's level of universal service support should be based upon the forwardlooking economic cost of constructing and operating the network facilities and functions used to provide the services supported by the federal universal service  X4support mechanisms.D6 {O'ԍ Universal Service Order, 12 FCC Rcd at 8888, para. 199. The Commission also determined that high cost support for rural carriers should continue essentially unchanged and should not be based on forwardlooking costs  {O'until 2001, at the earliest. Universal Service Order, 12 FCC Rcd at 8889, para. 203. The Commission adopted the Joint Board's recommendation to define "rural carriers" as those carriers that meet the statutory definition of  {OA 'a "rural telephone company." Universal Service Order, 12 FCC Rcd at 8943, para. 310 (citing 47 U.S.C.  153(37)).  Xv42. ` ` Our plan to adopt a mechanism to estimate forwardlooking cost has proceeded  X_4in two stages. On October 28, 1998, with the release of the Platform Order, the Commission completed the first stage of this proceeding: the selection of the model platform. The platform encompasses the aspects of the model that are essentially fixed, primarily the  X 4assumptions about the design of the network and network engineering. FD6 {O'ԍ FederalState Joint Board on Universal Service, Fifth Report and Order, CC Docket Nos. 9645, 97160,  {O'13 FCC Rcd 21323 (1998) (Platform Order). In this Further Notice, we move toward completion of the second stage of this proceeding, by proposing input values for the model, such as the cost of cables, switches, and other network components, in addition to various capital cost parameters. For the most important inputs, we  X 4provide a description of the methodology we have used to arrive at the proposed values. D6 yO'ԍ Appendix A contains a complete list of the input values that we propose in this Further Notice. In addition, we seek to supplement the record regarding certain inputs to the model.  X{43. ` ` The forwardlooking cost of providing supported services estimated by the model will be used to determine high cost support for nonrural carriers beginning January 1,  XM42000.$M2 D6 {O0'ԍ FederalState Joint Board on Universal Service, Access Charge Reform, Seventh Report and Order and Thirteenth Order on Reconsideration in CC Docket No. 9645; Fourth Report and Order in CC Docket No. 96262; and Further Notice of Proposed Rulemaking, CC Docket Nos. 9645, 96262, FCC 99119 (rel. May 28,  {O'1999) (Companion Order). The Commission is adopting a companion Order and Further Notice that establishes the framework for determining federal high cost support levels and seeks comment on the  X4details of that mechanism.3D6 {O"'ԍ Id.3   X' II. PROCEDURAL HISTORY א\"0*&&``{"Ԍ  X4 A.Universal Service Order    X44. ` ` Prior to the 1996 Act, three explicit universal service programs were in place to provide assistance to small incumbent local exchange carriers (LECs) and LECs that served  X4rural and high cost areas: high cost loop support,[XD6 yO'ԍ Although the existing high cost loop fund has historically been known as the "Universal Service Fund," we will avoid this terminology because of the confusion it may create with the new universal service support mechanisms that the Commission has created pursuant to section 254 of the Communications Act.[ dial equipment minutes (DEM) weighting,  Xw4and the LongTerm Support (LTS) program.: wD6 {O 'ԍ The Commission's rules governing these programs are set forth at 47 C.F.R.  36.601 et. seq. (high cost  yO 'loop fund); 47 C.F.R.  36.125(b) (DEM weighting); and 47 C.F.R.  69.105, 69.502, 69.603(e), 69.612 (LTS).: Other mechanisms also have historically contributed to maintaining affordable rates in rural areas, including subsidies implicit in geographic toll rate averaging, intrastate rates, and interstate access charges. In the 1996 Act, Congress codified the Commission's longstanding commitment to ensuring universal service and directed that "[c]onsumers . . . in rural, insular, and high cost areas should have access to telecommunications and information services . . . that are reasonably comparable to those services provided in urban areas and that are available at rates that are reasonably comparable  X 4to [those] in urban areas."D BD6 yO'ԍ 47 U.S.C.  254(b)(3). D The 1996 Act also directed the Commission to reform universal service support mechanisms to ensure that they are compatible with the procompetitive goals of the 1996 Act. Section 254 required the Commission to institute a Joint Board on universal  X4service and implement the recommendations from the Joint Board by May 8, 1997.J D6 yO'Ѝ 47 U.S.C.  254(a).J After  Xz4receiving the recommendations of the Joint Board on November 7, 1996, zb D6 {O'ԍ FederalState Joint Board on Universal Service, Recommended Decision, CC Docket No. 9645, 12 FCC  {OW'Rcd 87 (1996) (First Recommended Decision). the Commission  Xc4adopted the Universal Service Order on May 7, 1997.  X745. ` ` In the Universal Service Order, the Commission adopted a forwardlooking economic cost methodology to calculate support for nonrural carriers. Under this methodology, a forwardlooking economic cost mechanism selected by the Commission, in consultation with the Joint Board, would be used to calculate nonrural carriers' forward X4looking economic cost of providing the supported services in high cost areas.{ $ D6 {OL#'ԍ #X\  P6G;ɒP#Universal Service Order, 12 FCC Rcd at 8890, para. 206. In the Universal Service Order, the Commission concluded that the federal universal service support mechanism would support 25 percent of the difference between the forwardlooking economic cost of providing the supported service and a nationwide  {O%'revenue benchmark. See Universal Service Order, 12 FCC Rcd at 8888, para. 201. In response to issues raised"% 0*&&%" by commenters and state Joint Board members, the Commission referred back to the Joint Board questions  {OX'related to how federal support should be determined. See FederalState Joint Board on Universal Service, Order  {O"'and Order on Reconsideration, CC Docket No. 9645, 13 FCC Rcd 13749 (1998) (Referral Order). See also  {O'FederalState Joint Board on Universal Service, Second Recommended Decision, CC Docket No. 9645, 13 FCC  {O'Rcd 24744 (1998) (Second Recommended Decision).{ " 0*&&``"Ԍ  X4 B.Further Notice and the Input Value Development Process    X46. ` ` In a July 18, 1997 Further Notice of Proposed Rulemaking, the Commission established a multiphase plan to develop a federal mechanism that would send the correct  X4signals for entry, investment, and innovation.]^D6 {O 'ԍ FederalState Joint Board on Universal Service, ForwardLooking Mechanism for High Cost Support for  {O 'NonRural LECs, CC Docket Nos. 9645, 97160, Further Notice of Proposed Rulemaking, 12 FCC Rcd 18514  {OU 'at 18519, para. 5 (1997) (1997 Further Notice).] The 1997 Further Notice divided questions  X{4related to the cost models into "platform design" issues and "input value" issues."{D6 yO'ԍ Generally, there is a platform component for each portion of the exchange network being modeled. Examples of platform design issues are the establishment of switch capacity limitations and the routing of feeder and distribution cables. Examples of input values are the price of various network components, their associated  {O*'installation and placement costs, and capital cost parameters such as debtequity ratios. See 1997 Further Notice, 12 FCC Rcd at 1851618, paras. 1718." The 1997  Xf4Further Notice subdivided each of the platform and input issues into four topic groups, and sought comment on each group separately in order to develop a focused dialogue among interested parties. The four groups were: (1) customer location; (2) outside plant design; (3)  X# 4switching and interoffice; and (4) general support facilities (GSF) and expense issues.R# X D6 {O,'ԍ See generally 1997 Further Notice.R  X 47. ` ` After reviewing the comments received in response to the 1997 Further Notice, the Common Carrier Bureau (Bureau) released two public notices to guide parties wishing to  X 4submit cost models for consideration as the federal mechanism. D6 {Od'ԍ Guidance to Proponents of Cost Models in Universal Service Proceeding: Switching, Interoffice Trunking,  {O.'Signaling, and Local Tandem Investment, Public Notice, CC Docket Nos. 9645, 97160, DA 971912 (rel. Sep.  {O'3, 1997) (Switching and Transport Public Notice); Guidance to Proponents of Cost Models in Universal Service  {O'Proceeding: Customer Location and Outside Plant, Public Notice, CC Docket Nos. 9645, 97160, DA 972372  {O '(rel. Nov. 13, 1997) (Customer Location & Outside Plant Public Notice).  X48. ` ` In addition to the 1997 Further Notice, the Bureau has solicited comment and allowed interested parties the opportunity to participate in the development of the input values  Xo4to be used in the forwardlooking mechanism. On May 4, 1998, the Bureau released a Public"o0*&&``3"  X4Notice to update the record on several inputrelated issues.N^D6 {Oy'ԍ Common Carrier Bureau Requests Further Comment On Selected Issues Regarding The ForwardLooking  {OC'Economic Cost Mechanism For Universal Service, Public Notice, CC Docket Nos. 9645, 97160, DA 98848  {O '(rel. May 4, 1998) (Inputs Public Notice).N The Bureau also issued data requests designed to acquire information that may be useful in determining the final input  X4values,D6 {Os'ԍ FederalState Joint Board on Universal Service, Order, CC Docket No. 9645, 12 FCC Rcd 9803 (1997)  {O='(1997 Data Request). and conducted a series of public workshops designed to elicit further comment from  X4interested parties in selecting final input values.\^JD6 {O 'ԍ Common Carrier To Hold Three Workshops On Input Values To Be Used To Estimate ForwardLooking  {O 'Economic Costs For Purposes Of Universal Service Support, Public Notice, CC Docket Nos. 9645, 97160, DA  {OL '982406 (rel. Nov. 25, 1998) (Workshop Public Notice).\ Finally, the Bureau conducted numerous ex  X4parte meetings with interested parties throughout this proceeding.@Zp D6 {O'ԍ See, e.g., Letter from W. Scott Randolph, GTE, to Magalie Roman Salas, FCC, dated March 2, 1999; Letter from Pete Sywenki, Sprint, to Magalie Roman Salas, FCC, dated, February 26, 1999; Letter from Chris Frentrup, MCI, to Magalie Roman Salas, FCC, dated February 9, 1999.@  X|4 C. Platform Order and Second Recommended Decision   XO49. ` ` In the Platform Order released on October 28, 1998, the Commission adopted the forwardlooking cost model to be used in determining federal universal service high cost  X# 4support for nonrural carriers.]# D6 {Of'ԍ Platform Order, 13 FCC Rcd at 21325, para. 4.] The model platform that the Commission adopted combined elements from each of the three models under consideration in this proceeding: (1) the  X 4BCPM, Version 3.0 (BCPM);X $D6 yO'ԍ Submission in CC Docket Nos. 9645 and 97160 by BellSouth Corporation, BellSouth Telecommunications, Inc., U S WEST, Inc., and Sprint Local Telephone Company (BCPM proponents), dated Dec. 11, 1997 (BCPM Dec. 11, 1997 submission). (2) the HAI Model, Version 5.0a (HAI);  DD6 yO'ԍ Letter from Richard N. Clarke, AT&T, to Magalie Roman Salas, FCC, dated Dec. 11, 1997 (HAI Dec. 11, 1997 submission). HAI was submitted by AT&T and MCI (HAI proponents). Versions of HAI filed before February 3, 1998, were known as the Hatfield Model. The proponents refer to the February 3, 1998 submission as HAI. We refer to this model as HAI throughout this Report and Order. and (3) the Hybrid  X 4Cost Proxy Model, Version 2.5 (HCPM). $ ,D6 yO"'Ѝ HCPM was developed by Commission staff members William Sharkey, Mark Kennet, C. Anthony Bush,  {O#'Jeff Prisbrey, and Commission contractor Vaikunth Gupta of Panum Communications. Common Carrier Bureau  {OM$'Announces Release of HCPM Version 2.0, Public Notice, DA 972712 (rel. Dec. 29, 1997). United States  yO%'Government Memo from W. Sharkey, FCC, to Magalie Roman Salas, FCC, dated Feb. 6, 1998.  In the Platform Order, the Commission also specified several issues that would be addressed in the inputs stage of this proceeding. These" 0*&&``4 "  X4issues include: (1) the geocode data source to determine customer locations;^D6 {Oy'ԍ Platform Order, 13 FCC Rcd at 21338, para. 34.^ (2) the road  X4surrogate method to determine the location of nongeocoded customer locations;^ZD6 {O'ԍ Platform Order, 13 FCC Rcd at 21341, para. 41.^ and (3) the use of the local exchange routing guide (LERG) to identify the existing hostremote switch  X4relationships.D6 {OX'ԍ Platform Order, 13 FCC Rcd at 21355, para. 76. The LERG is a database of switching information maintained by Bellcore that includes the existing hostremote relationships.  X4 10. ` ` On November 25, 1998, the Joint Board released the Second Recommended  Xx4Decision, in which it recommended that the Commission compute federal high cost support  Xc4for nonrural carriers through a twostep process.jcFD6 {OZ'ԍ Second Recommended Decision, 13 FCC Rcd at 24746, para. 5.j First, the Joint Board recommended that the Commission should estimate the total support amount necessary in those areas considered to have high costs relative to other areas. Second, the Commission should consider, in a consistent manner across all states, any particular state's ability to support high cost areas  X 4within the state.j D6 {O'ԍ Second Recommended Decision, 13 FCC Rcd at 24746, para. 5.j The Joint Board recommended that federal support should be provided to the extent that the state would be unable to support its high cost areas through its own  X 4reasonable efforts.m j D6 {O'ԍ Second Recommended Decision, 13 FCC Rcd at 2474647, para. 5.m In addition, the Joint Board recommended that the Commission continue to work with the Joint Board to select the input values to complete a forwardlooking cost  X 4model and to finalize the methodology for distributing federal high cost support.k D6 {OX'ԍ Second Recommended Decision, 13 FCC Rcd at 24757, para. 28.k  X}4 III. ESTIMATING FORWARDLOOKING ECONOMIC COST ׃  XO' A.Designing a ForwardLooking Wireline Local Telephone Network  X!4 11. ` ` To understand the assumptions made in the mechanism, it is necessary to  X 4understand the layout of the current wireline local telephone network.!"  D6 yOI"'ԍ We also note that technologies such as wireless services are likely to become more important over time in providing universal service. We will continue to review suggestions for incorporating such technologies into the  {O#'forwardlooking mechanism for future years. See, e.g., Letter from David L. Sieradzki, on behalf of Western Wireless, to Magalie Roman Salas, FCC, dated January 26, 1999 (submitting the "Wireless Cost Model"). In general, a telephone network must allow any customer to connect to any other customer. In order to"x!0*&&`` " accomplish this, a telephone network must connect customer premises to a switching facility, ensure that adequate capacity exists in that switching facility to process all customers' calls that are expected to be made at peak periods, and then interconnect that switching facility  X4with other switching facilities to route calls to their destinations. A wire center is the location of a switching facility. The wire center boundaries define the area in which all customers are  X4connected to a given wire center. The Universal Service Order required the models to use  Xz4existing incumbent LEC wire center locations in estimating forwardlooking cost."$zD6 {O'ԍ The Universal Service Order established ten criteria to ensure consistency in calculations of federal  {O'universal service support. Universal Service Order, 12 FCC Rcd at 8913, para. 250. Criterion 1 requires that a model must include incumbent LECs' wire centers as the center of the loop network and the outside plant should terminate at incumbent LECs' current wire centers.  XL4 12. ` ` Within the boundaries of each wire center, the wires and other equipment that  X54connect the central office to the customers' premises are known as outside plant. Outside plant can consist of either copper cable or a combination of optical fiber and copper cable, as well as associated electronic equipment. Copper cable generally carries an analog signal that is compatible with most customers' telephone equipment, but thicker, more expensive cables or loading coils must be used to carry signals over greater distances. Optical fiber cable carries a digital signal that is incompatible with most customers' telephone equipment, but the quality of a signal carried on optical fiber cable is superior at greater distances when compared to a signal carried on copper wire. Generally, when a neighborhood is located too far from the wire center to be served with copper cables alone, an optical fiber cable will be deployed to a point within the neighborhood, where a piece of equipment will be placed that converts the digital light signal carried on optical fiber cable to an analog, electrical signal that is compatible with customers' telephones. This equipment is known as a digital loop carrier remote terminal, or DLC. From the DLC, copper cables of varying gauge extend to all of the customer premises in the neighborhood. Where the neighborhood is close enough to  X4the wire center to serve entirely on copper cables, a copper trunk connects the wire center to a central point in the serving area, called the serving area interface (SAI), and copper cables will then connect the SAI to the customers in the serving area. The portion of the loop plant  X4that connects the central office with the SAI or DLC is known as the feeder plant, and the portion that runs from the DLC or SAI throughout the neighborhood is known as the  X4distribution plant.  XZ4 13. ` ` The model's estimate of the cost of serving the customers located within a given wire center's boundaries includes the calculation of switch size, the lengths, gauge, and number of copper and fiber cables, and the number of DLCs required. These factors depend, in turn, on how many customers the wire center serves, where the customers are located within the wire center boundaries, and how they are distributed within neighborhoods. Particularly in rural areas, some customers may not be located in neighborhoods at all but, instead, may be scattered throughout outlying areas. In general, the model divides the area" "0*&&``;"  X4served by the wire center into smaller areas known as serving areas. For serving areas sufficiently close to the wire center, copper feeder cable extends from the wire center to a SAI where it is crossconnected to copper distribution cables. If the feeder is fiber, it extends to a DLC terminal in the serving area, which converts optical digital signals to analog signals. Individual circuits from the DLC are crossconnected to copper distribution cables at the adjacent SAI.  Xa4 14. ` ` The model assumes that wire centers are interconnected with one another using  XJ4optical fiber networks known as Synchronous Optical Network (SONET) rings.`#XJD6 yO 'ԍ SONET is a set of standards for optical (fiber optic) transmission. It was developed to meet the need for transmission speeds above the T3 level (45 Mbps) and is generally considered the standard choice for transmission devices used with broadband networks. BCPM Dec. 11 submission, Model Methodology at 68. ` The  X34infrastructure to interconnect the wire centers is known as the interoffice network, and the  X 4carriage of traffic among wire centers is known as transport. In cases where a number of wire centers with relatively few people within their boundaries are located in close proximity to one another, it may be more economical to use the processor capacity of a single switch to supervise the calls of the customers in the boundaries of all the wire centers. In that case, a  X 4fullcapacity switch (known as a host) is placed in one of the wire centers and less expensive,  X 4more limitedcapacity switches (known as remotes) are placed in the other wire centers. The remotes are then connected to the host with interoffice facilities. Switches that are located in wire centers with enough customers within their boundaries to merit their own fullcapacity  Xl4switches and that do not serve as hosts to any other wire centers are called standalone switches.  X)415. ` ` There are also a number of expenses and general support facilities (GSF) costs  X4associated with the design of a forwardlooking wireline telephone network.u$D6 {O'ԍ See Platform Order, 13 FCC Rcd at 2135761, paras. 8191.u GSF costs include the investment related to vehicles, land, buildings, and general purpose computers. Expenses include: plant specific expenses, such as maintenance of facilities and equipment expenses; plant nonspecific expenses, such as engineering, network operations, and power expenses; customer service expenses, such as marketing, billing, and directory listing expenses; and corporate operations expenses, such as administration, human resources, legal,  X4and accounting expenses.a%zD6 {O 'ԍ Platform Order, 13 FCC Rcd at 2135758, para. 82.a  XZ' B. Synthesis Model   X,416. ` ` The "synthesis" model adopted in the Platform Order allows the user to estimate the cost of building a telephone network to serve subscribers in their actual" %0*&&``"  X4geographic locations, to the extent these locations are known.&D6 {Oy'ԍ Platform Order, 13 FCC Rcd at 21337, para. 33. See also discussion of customer location data, infra. To the extent that the actual geographic locations of customers are not available, the Commission determined that the  X4synthesis model should assume that customers are located near roads.'ZD6 {O'ԍ Platform Order, 13 FCC Rcd at 2134041, para. 40. See also discussion of road surrogating method,  {O'infra.  X417. ` ` Once the customer locations have been determined, the model employs a clustering algorithm to group customers into serving areas in an efficient manner that takes  Xv4into consideration relevant engineering guidelines.^(vD6 {O 'ԍ Platform Order, 13 FCC Rcd at 21342, para. 44.^ After identifying efficient serving areas,  X_4the model designs outside plant to the customer locations.^)_HD6 {OX'ԍ Platform Order, 13 FCC Rcd at 21346, para. 55.^ In doing so, the model employs a number of cost minimization principles designed to determine the most costeffective  X14technology to be used under a variety of circumstances, such as varying terrain and density.^*1D6 {O'ԍ Platform Order, 13 FCC Rcd at 21348, para. 61.^  X 418. ` ` The Commission concluded that the federal universal service mechanism should incorporate, with certain modifications, the HAI 5.0a switching and interoffice facilities  X 4module to estimate the cost of switching and interoffice transport.a+ l D6 {O'ԍ Platform Order, 13 FCC Rcd at 2135455, para. 75.a The Commission noted that it would consider adopting the LERG at the inputs stage of this proceeding to determine  X 4the deployment of host and remote switches.^, D6 {OV'ԍ Platform Order, 13 FCC Rcd at 21355, para. 76.^ In addition, the Commission adopted the HAI  X4platform module for calculating expenses and capital costs, such as depreciation.l- D6 {O'ԍ Platform Order , 13 FCC Rcd at 21357, para. 81. l  Xb419. ` ` The Commission noted that technical improvements to the cost model will continue, both before implementation of the model for nonrural carriers and on an ongoing  X44basis, as necessary.^.4"D6 {O"'ԍ Platform Order, 13 FCC Rcd at 21329, para. 13.^ The Commission therefore delegated to the Bureau the authority to make changes or direct that changes be made to the model platform as necessary and appropriate to ensure that the platform of the federal mechanism operates as described in the " .0*&&``]"  X4Platform Order.^/D6 {Oy'ԍ Platform Order, 13 FCC Rcd at 21329, para. 13.^  As contemplated in the Platform Order, Commission staff and interested parties have continued to review the model platform to ensure that it operates as intended. As  X4a result, some refinements have been made to the model platform adopted in the Platform  X4Order.0$ZD6 {O'ԍ Common Carrier Bureau To Post On The Internet Modifications To The ForwardLooking Economic Cost  {O'Model For Universal Service Support, Public Notice, CC Docket Nos. 9645, 97160, DA 982533 (rel. Dec. 15, 1998). All changes to the model platform have been posted on the Commission's Web site (http://www.fcc.gov/ccb/apd/hcpm).  X' C.Selecting ForwardLooking Input Values   Xe420. ` ` In the Universal Service Order, the Commission adopted ten criteria to be used in determining the forwardlooking economic cost of providing universal service in high cost  X94areas.j19FD6 {O0'ԍ Universal Service Order, 12 FCC Rcd at 891316, para. 250.j These criteria provide specific guidance for our selection of input values for use in the synthesis model. Rather than reflecting existing incumbent LEC facilities, the technology assumed in the model "must be the leastcost, mostefficient, and reasonable technology for  X 4providing the supported services that is currently being deployed."w2 D6 {O}'ԍ Universal Service Order, 12 FCC Rcd at 8913, para. 250 (criterion one).w As noted below, existing LEC plant does not necessarily, or even likely, reflect forwardlooking technology or design  X 4choices.F3 j D6 {O'ԍ See infra at para. 50.F Similarly, the input values we tentatively select in this Notice are not intended to replicate any particular company's embedded or book costs. Criterion three directs that "costs  X4must not be the embedded cost of the facilities, functions, or elements."y4 D6 {OE'ԍ Universal Service Order, 12 FCC Rcd at 8913, para. 250 (criterion three).y Rather, the model "must be based upon an examination of the current cost of purchasing facilities and  Xj4equipment."y5j D6 {O'ԍ Universal Service Order, 12 FCC Rcd at 8913, para. 250 (criterion three).y  X<421. ` ` As discussed in detail in sections VVIII below, we generally have proposed using nationwide, rather than companyspecific input values in the federal mechanism. In many cases, the only data for various inputs on the record in this proceeding are embedded cost, companyspecific data. We have used various techniques to convert these data to forwardlooking values. For example, we propose modifying the switching data to adjust for the effects of inflation and the cost changes unique to the purchase and installation of digital" 50*&&``"  X4switches.D6D6 {Oy'ԍ See infra para. 166.D We propose nationwide averages, rather than companyspecific values, to mitigate  X4the rewards to less efficient companies.Q7ZD6 {O'ԍ See, e.g., infra paras. 198, 214.Q  X422.` ` Although the BCPM sponsors have provided nationwide default values, they  X4and other LECs generally advocate companyspecific input values. For purposes of determining federal universal service support amounts, we believe that nationwide default values generally are more appropriate than companyspecific values. Under the new mechanism, support is based on the estimated costs that an efficient carrier would incur to provide the supported services, rather than on the specific carrier's book costs. There may be some categories of inputs, however, where companyspecific or state specific input values might be appropriate for use in the federal mechanism. We seek comment on specific  X 4alternatives to nationwide values for certain input values, as discussed below.N8 D6 {O'ԍ See infra paras. 122, 199200.N We make no finding with respect to whether nationwide values would be appropriate for purposes other  X 4than determining federal universal service support.$9X ~D6 yO'ԍ State commissions, for example, may find that it is not appropriate to use nationwide values in determining state universal service support or prices for unbundled network elements and may choose instead to use statewide or company specific values. $  X 4j  IV. DETERMINING CUSTOMER LOCATIONS ׃  Xy' A. Background   XK423. ` ` The determination of customer locations relative to the wire center heavily influences a forwardlooking cost model's design of outside plant facilities. This is because assumptions about the locations of customers will determine the predicted loop length, which  X4in turn will have a large impact on the cost of service.g:D6 {OU'ԍ See 1997 Further Notice, 12 FCC Rcd at 18535, para. 44.g Each of the models under  X4consideration in the Platform Order provided a methodology for determining customer  X4locations.^;0 D6 {O!'ԍ Platform Order, 13 FCC Rcd at 21337, para. 31.^ The Bureau sought comment on these proposals and solicited alternative  X4proposals for locating customers from interested parties.{<^ D6 {O6$'ԍ See, e.g., 1997 Further Notice, 12 FCC Rcd at 18535, para. 44; Inputs Public Notice at 34; Common  {O%'Carrier Bureau Seeks Comment On Model Platform Development, Public Notice, CC Docket Nos. 9645, 97160,  {O%'DA 981587 (rel. Aug. 7, 1998) (Platform Public Notice) at 24.{" <0*&&``"Ԍ X4ԙ24. ` ` In the Platform Order, the Commission concluded that HAI's proposal to use actual geocode data, to the extent that they are available, and BCPM's proposal to use road network information to create "surrogate" customer locations where actual data are not  X4available, provided the most reasonable method for determining customer locations.="D6 {O6'ԍ Platform Order, 13 FCC Rcd at 21337, para. 31. Although surrogating methods, and even customer location data provided by the Census Bureau, constitute geocode data, for purposes of clarity, we will use the term "geocode" data to refer only to actual precise latitude and longitude data, unless we specifically refer to the data as "surrogate geocode" data. The Commission concluded that "the source or sources of geocode data to use in determining  X4customer location will be decided at the inputs phase of this proceeding."a>D6 {O 'ԍ Platform Order, 13 FCC Rcd at 2133738, para. 34.a The Commission also concluded that "the selection of a precise algorithm for placing road surrogates pursuant to these conclusions should be conducted in the inputs stage of this proceeding as part of the  XJ4process of selecting a geocode data set for the federal mechanism."a?JDD6 {O?'ԍ Platform Order, 13 FCC Rcd at 2134041, para. 40.a  X ' B. Issues for Comment   X ' 1. ` ` Geocode Data   X 425. ` ` While we affirm our conclusion in the Platform Order that geocode data should be used to locate customers in the federal mechanism, we tentatively conclude that at this time we cannot adopt any particular source of geocode data because interested parties have not had adequate access or time to review such data. We tentatively conclude below that a road surrogate algorithm will be used to locate customers in the federal mechanism until a source of geocode data is selected by the Commission. We reiterate our expectation, however, that we will identify and select a source of accurate and verifiable geocode data in the future for use in the federal mechanism.  X426. ` ` In the Platform Order, we concluded that a model is most likely to select the leastcost, mostefficient outside plant design if it uses the most accurate data for locating customers within wire centers, and that the most accurate data for locating customers within  X4wire centers are precise latitude and longitude coordinates for those customers' locations.^@D6 {O7!'ԍ Platform Order, 13 FCC Rcd at 21337, para. 33.^ We noted that commenters generally support the use of accurate geocode data in the federal  X4mechanism where available.aAh D6 {O$'ԍ Platform Order, 13 FCC Rcd at 2133738, para. 34.a We further noted that the only geocode data in the record were those prepared for HAI by PNR Associates (PNR), but that "our conclusion that the model"k A0*&&``" should use geocode data to the extent that they are available is not a determination of the  X4accuracy or reliability of any particular source of the data."^BD6 {Ob'ԍ Platform Order, 13 FCC Rcd at 21338, para. 34.^ Although commenters support the use of accurate geocode data, several commenters question whether the PNR geocode data  X4are adequately available for review by interested parties.^CZD6 {O'ԍ Platform Order, 13 FCC Rcd at 21338, para. 34.^  X427. ` ` In the Universal Service Order, the Commission required that the "model and all underlying data, formulae, computations, and software associated with the model must be  Xa4available to all interested parties for review and comment."yDaD6 {O 'ԍ Universal Service Order, 12 FCC Rcd at 8915, para. 250 (criterion eight).y In an effort to comply with this requirement, the Commission has made significant efforts to encourage parties to submit  X34geocode data on the record in this proceeding. E3~D6 {Ob'ԍ See FederalState Joint Board on Universal Service, Protective Order, CC Docket Nos. 9645, 97160, 13  {O,'FCC Rcd 13910 (1998) (Protective Order). See also Inputs Public Notice at 34.  PNR took initial steps to comply with this  X 4requirement in December 1998 by making available the "BIN" filesyF D6 yO'ԍ BIN files are the output of the clustering routine in the synthesis model platform derived from the actual geocode customer locations and, as such, do not reveal the actual geocoded customer locations. The BIN files allow users to run all aspects of the model except for the clustering. PNR has made the BIN files available to  {O'interested parties for a fee of $25.00, pursuant to the terms of the Protective Order. See Letter from William M.  {O'Newman, PNR, to Magalie Roman Salas, FCC, dated December 17, 1998 (PNR Dec. 17 ex parte).y derived from the  X 4geocoded points to interested parties pursuant to the Protective Order.OG D6 {OD'ԍ See PNR Dec. 17 ex parte.O In addition, PNR has continued to provide access to the underlying geocode data at its facility in Pennsylvania.  X 4Several commenters, in petitions for reconsideration of the Platform Order, have argued that the availability of the BIN data alone is not sufficient to comply with the requirements of criterion eight, particularly in light of the expense and conditions imposed by PNR in  X4obtaining access to the geocode point data.H D6 {Og'ԍ See, e.g., Bell Atlantic Petition for Reconsideration at 56; BellSouth Petition for Reconsideration at 34; GTE Petition for Reconsideration at 21.  Xh428. ` ` We tentatively conclude that interested parties have not had an adequate opportunity to review and comment on the accuracy of the PNR geocode data. We note that a nationwide customer location database will, by necessity, be voluminous, relying on a variety of underlying data sources. In order to comply with criterion eight, all underlying data must be reasonably available to interested parties for review. In light of the concerns expressed by several commenters relating to the conditions and expense in obtaining data"zH0*&&``" from PNR, we find that no source of geocode data has been made adequately available for review. We anticipate that a source of accurate and verifiable geocode data can be selected for use in the federal mechanism in the future and we encourage parties to make continued efforts to ensure that all underlying geocode data are available for review. For example, we note that PNR has contacted its data vendors for the purpose of making additional underlying  X4data more freely available to parties in this proceeding.JID6 {O'ԍ PNR Dec. 17 ex parte at 1.J As noted in the Platform Order, we recognize that more comprehensive geocode data are likely to be available in the future and encourage parties to continue development of a data source that complies with the criteria  XJ4outlined in the Universal Service Order for use in the federal mechanism.^JJZD6 {OU 'ԍ Platform Order, 13 FCC Rcd at 21338, para. 34.^ We therefore seek further comment on a source of geocode customer locations that will comply with the Commission's criteria for use in the federal mechanism. In addition, we seek comment on the availability for review of the PNR geocode data, including any further measures necessary to  X 4ensure that the PNR geocode data are sufficiently available for review by the public.  X 4 2. ` ` Road Surrogate Customer Locations   X429. ` ` We tentatively conclude that the road surrogating algorithm proposed by PNR should be used to develop road surrogate customer locations for the federal universal service  Xf4mechanism. In the Platform Order, we concluded that, in the absence of actual geocode customer location data, BCPM's rationale of associating road networks and customer locations  X:4provides the most reasonable approach for determining customer locations.aK:D6 {O'ԍ Platform Order, 13 FCC Rcd at 2134041, para. 40.a As anticipated in  X#4the Platform Order, once a source of geocode data has been selected, the road surrogate  X4customer locations will be used only in the absence of geocode customer location data.aL~D6 {O='ԍ Platform Order, 13 FCC Rcd at 2134041, para. 40.a  X430. ` ` As noted in the Platform Order, "associating customers with the distribution of roads is more likely to correlate to actual customer locations than uniformly distributing customers throughout the Census Block, as HCPM proposes, or uniformly distributing  X4customers along the Census Block boundary, as HAI proposes."aMD6 {O^!'ԍ Platform Order, 13 FCC Rcd at 2134041, para. 40.a We therefore concluded in  X4the Platform Order that the selection of a precise algorithm for placing road surrogates should  Xq4be conducted in the inputs stage of this proceeding.^NqD6 {O$'ԍ Platform Order, 13 FCC Rcd at 21341, para. 41.^ "Z4 N0*&&``"Ԍ X431. ` ` Currently, there are two road surrogating algorithms on the record in this proceeding those proposed by PNR and Stopwatch Maps. On March 2, 1998, the HAI proponents provided a description of the road surrogate methodology developed by PNR for  X4locating customers.OD6 yO4'ԍ Letter from Michael Lieberman, AT&T, to Magalie Roman Salas, FCC, dated March 2, 1998 (AT&T  {O'March 2 ex parte). On January 27, 1999, PNR made available for review by the  X4Commission and interested parties, pursuant to the terms of the Protective Order, the road surrogate point data for all states except Alaska, Iowa, Virginia, Puerto Rico and eightyfour  Xx4wire centers in various other states.eP\x"D6 yOK 'ԍ Letter from William M. Newman, PNR, to Magalie Roman Salas, FCC, dated January 27, 1999 (PNR  {O 'Jan. 27 ex parte). PNR has made available by mail to interested parties the road surrogate point data for a fee of  {O '$25.00, pursuant to the terms of the Protective Order.e On February 22, 1999, PNR filed a more detailed  Xa4description of its road surrogate algorithm.QaFD6 yOX'ԍ Letter from Charles A. White, PNR, to Magalie Roman Salas, FCC, dated February 22, 1999 (PNR Feb.  {O '22 ex parte).  X3432. ` ` In general, the PNR road surrogate algorithm utilizes the Census Bureau's Topologically Integrated Geographic Encoding and Referencing (TIGER) files, which contain  X 4all the road segments in the United States.ER\ D6 {OV'ԍ PNR Feb. 22 ex parte at 1. A road segment is a length of road between two intersections. The Census Bureau classifies and numbers each of these road segments. PNR uses a slightly modified version of the Census  {O'Bureau road classifications. Id. at 2E For each Census Block, PNR determines how  X 4many customers and which roads are located within the Census Block.S D6 yOc'ԍ The PNR National Access Line Model is used to determine the number of residential and business  {O+'customer locations in a given wire center. See PNR Feb. 22 ex parte at 1. For each Census Block, PNR also develops a list of road segments. The total distance of the road segments within the Census Block is then computed. Roads that are located entirely within the interior of the Census Block are given twice the weight as roads on the boundary. This is because customers are assumed to live on both sides of a road within the interior of the Census Block. In addition, the PNR algorithm excludes certain road segments along which customers are not  Xd4likely to reside.LTdD6 {O3 'ԍ PNR Feb. 22 ex parte at 2. L For example, PNR excludes highway access ramps, alleys, and ferry  XM4crossings.JUMD6 {O"'ԍ PNR Feb. 22 ex parte at 2.J The total number of surrogate points is then divided by the computed road distance to determine the spacing between surrogate points. Based on that distance, the"6BU0*&&``U"  X4surrogate customer locations are uniformly distributed along the road segments.JVD6 {Oy'ԍ PNR Feb. 22 ex parte at 2.J  X4 33. ` ` Stopwatch Maps has compiled road surrogate customer location files for six  X4states suitable for use in the federal mechanism.WZD6 {O'ԍ See Letter from Pete Sywenki, Sprint, to Magalie Roman Salas, FCC, dated December 11, 1998 (Sprint  {O'Dec. 11 ex parte). We tentatively conclude, however, that until a more comprehensive data set is made available, the Stopwatch data set will not comply  X4with the Universal Service Order's criterion that the underlying data are available for review by the public. In addition, we note that the availability of only six states is of limited utility in a nationwide model.  X34!34. ` ` We tentatively conclude that the PNR road surrogate algorithm is a reasonable method for locating customers in the absence of actual geocode data. We note that PNR's methodology of excluding certain road segments is consistent with the Commission's  X 4conclusion in the Platform Order that certain types of roads and road segments should be  X 4excluded because they are unlikely to be associated with customer locations.^X D6 {O@'ԍ Platform Order, 13 FCC Rcd at 21341, para. 41.^ In addition, we note that PNR's reliance on the Census Bureau's TIGER files ensures a degree of reliability and availability for review of much of the data underlying PNR's road surrogate algorithm, in  X4compliance with criterion eight of the Universal Service Order.ZY\HD6 yO'ԍ We also note that PNR has made the road surrogate data points available to interested parties pursuant to  {OU'the provisions of the Protective Order in this proceeding. See PNR Jan. 27 ex parte; PNR Feb. 9 ex parte; PNR  {O'Feb. 22 ex parte.Z We note that the HAI proponents contend that use of a surrogate algorithm may overstate the amount of plant  Xh4necessary to provide supported services.Zhl D6 {O'ԍ See Letter from Chris Frentrup, MCI, to Magalie Roman Salas, FCC, dated February 19, 1999. We seek comment on the validity of this  XQ4contention. We also note that PNR has indicated that it intends to finalize a number of  X:4improvements to the road surrogate algorithm and data.[: D6 yO'ԍ Letter from William M. Newman, PNR, to Magalie Roman Salas, FCC, dated February 9, 1999 (PNR  {O'Feb. 9 ex parte). For example, PNR states that the new release will incorporate any new input requirements relating to an authoritative wire center list, housing units versus households, and treatment of phone penetration rates. In addition, the new release will include data for all fifty states, Washington, D.C., and Puerto  X4Rico.S\X D6 {O$'ԍ See PNR Feb. 9 ex parte at 1.S We seek comment on our tentative conclusion to adopt the PNR road surrogate algorithm to determine customer locations, and to adopt the PNR road surrogate data set for"\0*&&``" use in the model beginning on January 1, 2000. We also seek comment on any changes that should be made to the PNR methodology to improve the accuracy of the customer locations it generates.  X' 3. ` ` Methodology for Estimating the Number of Customer Locations   Xv4"35. ` ` In addition to selecting a source of customer data, we also must select a methodology for estimating the number of customer locations within the geographic region that will be used in developing the customer location data. We also must determine how demand for service at each location should be estimated and how locations should be allocated to each wire center.  X 4#36. ` ` In the Universal Service Order, the Commission concluded that a "model must estimate the cost of providing service for all businesses and households within a geographic  X 4region."u] D6 {O9'ԍ Universal Service Order, 12 FCC Rcd at 8915, para. 250 (criterion 6).u In the Inputs Public Notice, the Bureau sought comment on the appropriate method for defining "households," or residential locations, for the purpose of calculating the forward X4looking cost of providing supported services.L^ZD6 {O'ԍ Inputs Public Notice at 46.L Model proponents and interested parties have  X}4proposed alternative methods to comply with this requirement._}D6 yO'ԍ We note that the question of which residential and business locations should be included for purposes of estimating the forwardlooking cost of providing the supported services is distinct from the question of which  {O'lines should be supported. See Universal Service Order, 12 FCC Rcd at 8829, paras. 9596 (declining to adopt the Joint Board's recommendation to restrict universal service high cost support to primary residential and singleline businesses).  XO4$37. ` ` The HAI sponsors propose that we use the methodology devised by PNR, which is based upon the number of households in each Census Block, while the BCPM sponsors propose that we use a methodology based upon the number of housing units in each Census Block. A household is an occupied residence, while housing units include all  X4residences, whether occupied or not.`D6 {OB'ԍ These definitions reflect the Census Bureau's methodology for housing unit and household estimates.   {O 'See http://www.census.gov/population/methods/sthhmet.txt.  X4%38. ` ` Specifically, the HAI sponsors advocate the use of the PNR National Access Line Model to estimate the number of customer locations within Census Blocks and wire" `0*&&``"  X4centers.a"D6 yOy'ԍ HAI Dec. 11, 1997 submission, Model Description at 21. PNR has agreed to review and explain the process used in developing the National Access Line Model with interested parties, pursuant to the terms of the  {O 'Protective Order, at its location in Pennsylvania. See Letter from Charles A. White, PNR, to Thomas Mitchell, Collier, Shannon, Rill & Scott, dated April 29, 1999. The PNR National Access Line Model uses a variety of information sources, including: survey information, the LERG, Business Location Research (BLR) wire center boundaries, Dun & Bradstreet's business database, Metromail's residential database, Claritas' demographic database, and U.S. Census estimates. PNR's model uses these sources to estimate the number of residential and business locations, and the number of access lines demanded at each location. The model makes these estimations for each Census Block, and  Xv4for each wire center in the United States.`bvD6 yO 'ԍ HAI Dec. 11, 1997 submission, Model Description at 21.`  XH4&39. ` ` At the conclusion of PNR's process for estimating the number of customer locations: (1) PNR's estimate of residential locations is greater than or equal to the Census Bureau's estimate of households, by Census Block Group, and its estimate is disaggregated to the Census Block level, (2) PNR's estimate of demand for both residential and business lines in each study area is greater than or equal to the number of access lines in the Automated Reporting and Management Information System (ARMIS) for that study area, and the estimates are available by location at the Block level, and (3) each customer location is  X 4associated with a particular wire center.!cZ BD6 yO'ԍ Customer locations in unserved areas, as reflected by BLR wire center boundaries, are not associated with  {Ob'particular wire centers. See Letter from Charles A. White, PNR, to Magalie Roman Salas, FCC, dated April 12, 1999.!  Xy4'40. ` ` The BCPM sponsors rely on many of the same data sources as those used in PNR's National Access Line Model. For example, BCPM 3.1 uses wire center data obtained  XK4from BLR and business line data obtained from PNR.hdKd D6 yO`'ԍ BCPM April 30, 1998 documentation, Model Methodology at 2627.h In estimating the number of residential locations, however, the BCPM sponsors use Census data that include household and housing unit counts from the 1990 Census, updated based upon 1995 Census statistics regarding household growth by county. In addition, rather than attempting to estimate demand by location at the Block level, the BCPM model builds two lines to every residential location and at least six lines to every business.  X4(41. ` ` The synthesis model currently calculates the average cost per line by dividing the total cost of serving customer locations by the current number of lines. Because the current number of lines is used in this average cost calculation, the HAI sponsors argue that the total cost should be determined by using the current number of customer locations. The"e d0*&&``" HAI sponsors contend that "the key issue is the consistency of the numerator and denominator" in the average cost calculation. The HAI sponsors argue that other approaches are inconsistent because they select the highest possible cost numerator and divide by the lowest possible line denominator, and therefore result in larger than necessary support  X4levels.UeD6 {O'ԍ AT&T and MCI ex parte, Dec. 23, 1997.U The HAI sponsors argue that, in order to be consistent, housing units must be used  X4in the determination of total lines if they are used in the determination of total costs.fZD6 {O'ԍ Letter from Chris Frentrup, MCI, to Magalie Roman Salas, FCC, dated March 5, 1999 (MCI March 5 ex  {Ob 'parte). The HAI sponsors contend that "[i]f used consistently in this manner, building to housing units as  X_4GTE proposes is unlikely to make any difference in cost per line."Vg_D6 {O 'ԍ MCI March 5 ex parte (Issues 1 and 2).V  X14)42. ` ` In contrast, the BCPM sponsors and other commenters contend that the total cost should include the cost of providing service to all possible customer locations, even if  X 4some locations currently do not receive service.h HD6 {O'ԍ See, e.g., BCPM Joint Sponsors Inputs Public Notice comments at 7; GTE Inputs Public Notice  {O'comments at 9; RUS Inputs Public Notice comments at 2. Furthermore, the BCPM sponsors contend that if total cost is based on a smaller number of locations, support will not be sufficient to enable carriers to meet their carrieroflastresort obligations. The BCPM sponsors also argue that basing the estimate of residential locations on households instead of housing units will  X 4underestimate the cost of building a network that can provide universal service.ii D6 {O'ԍ BCPM Joint Sponsors Inputs Public Notice comments at 67.i The BCPM sponsors, as well as some other commenters, contend that residential locations should be  Xy4based on the number of housing units whether occupied or unoccupied.jy6 D6 {O`'ԍ See, e.g., BCPM Joint Sponsors Inputs Public Notice comments at 7; GTE Inputs Public Notice  {O*'comments at 9; RUS Inputs Public Notice comments at 2. These commenters contend that only this approach reflects the obligation to provide service to any  XK4residence that may request it in the future.kK D6 {O'ԍ See, e.g., BCPM Joint Sponsors Inputs Public Notice comments at 7; GTE Inputs Public Notice  {OX 'comments at 9; RUS Inputs Public Notice comments at 2.  X44   X4*43. ` ` We tentatively conclude that PNR's process for estimating the number of customer locations should be used for developing the customer location data. We also tentatively conclude that we should use PNR's methodology for estimating the demand for"k0*&&``"  X4service at each location, and for allocating customer locations to wire centers.lD6 {Oy'ԍ See Appendix B for a complete description of the PNR methodology for estimating the number of customer locations. We believe that the PNR methodology is a reasonable method for determining the number of customer locations to be served in calculating the cost of providing supported services. To the extent that the PNR methodology includes the cost of providing service to all currently served households, we tentatively conclude that this is consistent with a forwardlooking cost model, which is designed to estimate the cost of serving current demand. As noted by the HAI sponsors, adopting housing units as the standard would inflate the cost per line by using the highest possible numerator (all occupied and unoccupied housing units) and dividing by the  XH4lowest possible denominator (the number of customers with telephones).UmH"D6 {O 'ԍ AT&T and MCI ex parte, Dec. 23, 1997.U  X 4+44. ` ` In addition, we do not believe that including the cost of providing service to all housing units will promote universal service to unserved customers or areas. We note that there is no guarantee that carriers would use any support derived from the cost of serving all housing units to provide service to these customers. Many states permit carriers to charge substantial line extension or construction fees for connecting customers in remote areas to their network. If that fee is unaffordable to a particular customer, raising the carrier's support level by including the costs of serving that customer in the model's calculations would have no effect on whether the customer actually receives service. In fact, as long as the customer remains unserved, the carriers would receive a windfall. We recognize that serving unserved customers in such circumstances is an important universal service goal. As discussed in the companion Order and Further Notice adopted today, we will initiate a separate proceeding in  X4July 1999 to investigate the issue of unserved areas. nD6 yO'ԍ After developing more fully the record on this issue, we will be better able to determine whether such  {OJ'unserved areas should receive federal universal service support. Companion Order at para. 92.   X4,45. ` ` If we were to calculate the costs of a network that would serve all potential customers, it would not be consistent to calculate the cost per line by using current demand. In other words, it would not be consistent to estimate the cost per line by dividing the total cost of serving all potential customers by the number of lines currently served. We note, however, that the level and source of future demand is uncertain. Future demand might include not only demand from currently unoccupied housing units, but also demand from new housing units, or potential increases in demand from currently subscribing households. We also recognize that population or demographic changes may cause future demand levels in some areas to decline. Given the uncertainty of future demand, we are concerned that including such costs may not reflect forwardlooking costs and may perpetuate the system of implicit support. " n0*&&``"Ԍ X4ԙ-46. ` ` We recognize, however, that additional comment would be helpful with regard to certain issues. For example, if a currently vacant unit will again receive service in the near future, one might argue that it should be included in the calculation of total cost. It is also possible that housing stock is subject to a type of churn that could inflate the number of households used in determining total cost without affecting the total number of lines. That is, a certain percentage of housing units may be repeatedly vacated and then reoccupied, with the specific households involved constantly changing. At any given time, a certain number of housing units might be unoccupied as a result. Under the Census definition, such units are not considered households and therefore may not be included in the number of residential  X14locations estimated by PNR.'o1D6 yO 'ԍ As explained in Appendix B, PNR uses two databases, Metromail and Claritas, to estimate the number of residential locations and uses whichever number is greater. Claritas uses updated Census estimates of the number of households, so in cases where the Claritas number is larger, PNR's estimate would not include unoccupied housing units. In cases where the Metromail number is larger, PNR's estimate could include unoccupied housing units, but these housing units would have an associated telephone number.' We seek comment on whether the costs associated with providing service to these housing units should be included in the total cost by identifying an additional number of unoccupied units. The PNR methodology may provide an estimate of the number of residential locations that is greater than the number that currently receive  X 4telephone service, however.p xD6 yO'ԍ As explained in Appendix B, the Metromail counts used by PNR have an associated telephone number. The Claritas household counts, on the other hand, are not restricted to households with telephones. Therefore PNR's methodology may already account for at least some portion of housing units subject to this type of churn. We seek comment on this issue.  X4.47. ` ` We also note that locations outside of existing wire centers will not be included  Xy4under the PNR methodology.BqyD6 {O'ԍ See supra note 99.B Therefore the accuracy of the wire center boundaries is of importance in estimating the number of customer locations. PNR currently uses BLR wire  XK4center information to estimate wire center boundaries.brKb D6 yO^'ԍ HAI Dec. 11, 1997 submission, Model Description at 21. b As noted above, the BCPM model also uses BLR wire center boundaries, as does Stopwatch Maps in its road surrogate customer  X4location files.is D6 {O'ԍ See Sprint Dec. 11, 1998 ex parte, attachment at 1.i PNR has indicated its intent to evaluate alternative sources of wire center  X4boundaries to be used in the customer location data.t D6 yO;"'ԍ Letter from Charles A. White, PNR, to Magalie Roman Salas, FCC, dated April 12, 1999. We therefore seek comment on the accuracy of the BLR wire center boundaries and any possible alternatives to establish more accurate wire center boundaries.  X4  V. OUTSIDE PLANT INPUT VALUES ׃"t0*&&``\"Ԍ X'ԙ A.Background  X4  /48. ` ` As the Commission noted in the Platform Order, outside plant, or loop plant,  X4constitutes the largest portion of total network investment, particularly in rural areas.^uD6 {O6'ԍ Platform Order, 13 FCC Rcd at 21335, para. 27.^ Outside plant investment includes the copper cables in the distribution plant and the copper and optical fiber cables in the feeder plant that connect the customers' premises to the central office. Cable costs include the material costs of the cable, as well as the costs of installing  Xa4the cable.vaZD6 yOl 'ԍ As discussed below, cable installation costs for buried cable often are included with the structure costs.  X34049. ` ` Outside plant consists of a mix of aerial, underground, and buried cable.w3D6 yO 'ԍ The phrase "plant mix" refers to the ratio of outside plant that is aerial, underground, or buried in a network or particular area. Aerial cable is strung between poles above ground. Underground cable is placed underground within conduits for added support and protection. Buried cable is placed underground but without any conduit. A significant portion of outside plant investment consists of the poles, trenches, conduits, and other structure that support or house the copper and fiber cables. In some cases, electric utilities, cable companies, and other telecommunications providers share structure with the LEC and, therefore, only a portion of the costs associated with that structure are borne by the LEC. Outside plant investment also includes the cost of the SAIs and DLCs that connect the feeder and distribution plant.  XM4150. ` ` The Universal Service Order's first criterion specifies that "[t]he technology assumed in the cost study or model must be the leastcost, most efficient, and reasonable  X!4technology for providing the supported services that is currently being deployed."mx!BD6 {O'ԍ Universal Service Order, 12 FCC Rcd at 8913, para. 250.m Thus, while the synthesis model uses existing incumbent LEC wire center locations in designing  X4outside plant,yD6 {Ox'ԍ See supra para. 11; Universal Service Order, 12 FCC Rcd at 8913, para. 250. it does not necessarily reflect existing incumbent LEC loop plant. Indeed, as  X4the Commission stated in the Platform Order, "[e]xisting incumbent LEC plant is not likely to  X4reflect forwardlooking technology or design choices."Xz\f D6 {O!'ԍ Platform Order, 12 FCC Rcd at 21350, para. 66. "Instead, incumbent LECs' existing plant will tend to reflect choices made at a time when different technology options existed or when the relative cost of equipment  {Op#'to labor may have been different than it is today." Id.X If the prices of fiber cable and DLCs have decreased over time relative to the cost of copper cable, for example, the synthesis model would design outside plant with more fiber and DLCs and less copper cable than has" z0*&&``F"  X4been deployed historically in an incumbent LEC's network.{ D6 yOy'ԍ If we look at current deployment, an incumbent LEC may be deploying even more fiber and DLCs today than the optimizing routines in the synthesis model would predict. For example, a LEC building a network that is capable of delivering video and broadband services may deploy less copper than the synthesis model would estimate is the optimum amount needed to provide the services supported by the federal mechanism.  X' B.Copper and Fiber Cable   X' 1.` ` Background   Xv4251. ` ` In the 1997 Further Notice, the Commission sought comment on the input  Xa4values that the model should use for cable material and installation costs.Y|aD6 {O 'ԍ 1997 Further Notice, 12 FCC Rcd at 18544.Y The Commission specifically sought comment on the accuracy of the default values in the BCPM and HAI  X34models and encouraged companies to submit data to support their positions.}3BD6 {O&'ԍ 1997 Further Notice, 12 FCC Rcd at 18544. The BCPM and HAI default values are the default input values for the useradjustable input values in the BCPM and HAI models, respectively. Although we have chosen a model platform and therefore are no longer considering adoption of the BCPM and HAI models, we continue to consider the BCPM and HAI default input values in this phase of the proceeding. For some inputs, these are the only values on the record. Although the BCPM model includes nationwide default values, the BCPM sponsors generally advocate companyspecific values and in some cases have proposed such values. The Commission tentatively concluded that cable material and installation costs should be  X 4separately identified by both density zone and terrain type.Y~ D6 {Or'ԍ 1997 Further Notice, 12 FCC Rcd at 18544.Y Because the Commission had received no documentation confirming that feeder and distribution cable installation costs should differ, the Commission tentatively concluded that the federal mechanism should adopt  X 4HAI's assumption that such costs are identical.Y N D6 {O'ԍ 1997 Further Notice, 12 FCC Rcd at 18544.Y  X4352. ` ` In the Inputs Public Notice, the Bureau sought comment on the analysis of David Gabel and Scott Kennedy on data from the Rural Utilities Service (RUS) regarding the  Xf4cost of installing cables.y\fD6 {O 'ԍ Inputs Public Notice at 7. See David Gabel & Scott Kennedy, Estimating the Cost of Switching and  {O!'Cables Based on Publicly Available Data (National Regulatory Research Institute NRRI 9809, April 1998) (NRRI Study). Dr. Gabel and Mr. Kennedy are consultants for the Commission in this proceeding.y On December 11, 1998, the Bureau held a public workshop  XO4designed to elicit comment on the input values for materials costs.OD6 {O%'ԍ See Workshop Public Notice. The December 1, 1998 workshop addressed issues relating to switching and expenses. At the workshop, Dr."O^0*&&``Q" Gabel presented the methodology used by the Commission staff to derive preliminary values for cable costs based on his earlier analysis of the RUS data.  X4453. ` ` Commission staff sought to supplement the record with respect to outside plant structure and cable costs by requesting additional data from LECs, including competitive  X4LECs.D6 yO'ԍ After numerous discussions with industry during development of the survey, staff distributed a final version on December 14, 1998, and requested responses by January 14, 1999. A copy of the outside plant structure and cable cost survey is attached in Appendix C. Ten companies eventually responded to the voluntary survey, somewhat fewer than the  X_4number that had indicated they would be willing to provide data.d_ D6 yO0 'ԍ BellSouth, Ameritech, Pacific Bell, Nevada Bell, Southwestern Bell, Sprint, GTE, Aliant, SNET, and AT&T submitted data in response to the structure and cable cost survey. Several companies requested additional time to complete and submit their data. After receiving and reviewing the data, staff found that, despite detailed survey instructions, further discussions with a number of companies were required before staff could assemble the data for comparison and analysis. In a number of cases, respondents filed revised data or clarified the data they had submitted.d Because of the delay in receiving the data and the time necessary to review and revise the data, staff has not completed its analysis of the survey data.   X ' 2.` ` Issues for Comment   X 4554. ` ` We now examine the inputs needed to determine outside plant cable costs in the synthesis model. The synthesis model uses several tables to calculate cable costs, based  X 4on the cost per foot of cable, which may vary by cable size (i.e., gauge and pair size) and the  X4type of plant (i.e., underground, buried, or aerial). There are four separate tables for copper distribution and feeder cable of two different gauges, and one table for fiber cable. The engineering assumptions and optimizing routines in the model, in conjunction with the input values in the tables, determine which type of cable is used.  X!4655. ` ` After the synthesis model has grouped customer locations in clusters, it determines, based on cost minimization and engineering considerations, the appropriate technology type for the cluster and the correct size of cables in the distribution network. Every customer location is connected to the closest SAI by copper cable. The copper cable used in the local loop typically is either 24 or 26gauge copper. Twentyfour gauge copper is thicker and therefore is expected to be more expensive than 26gauge copper. Twentyfour gauge copper also can carry signals greater distances without degradation than 26gauge copper and, therefore, is used in longer loops. In the synthesis model, if the maximum distance from the customer to the SAI is less than or equal to the copper gauge crossover point, then 26gauge cable is used. Feeder cable is either copper or fiber. Fiber is used for loops that exceed 18,000 feet, the maximum copper loop length permitted in the model, as";0*&&``"  X4determined in the Platform Order.aD6 {Oy'ԍ Platform Order, 13 FCC Rcd at 2135253, para. 70.a When fiber is more cost effective, the model will use it to replace copper for loops that are shorter than 18,000 feet.  X'` ` a. Engineering Assumptions and Optimizing Routines  X4756. ` ` Before we consider our proposed input values for cable costs, we discuss certain input values related to the engineering assumptions and optimizing routines in the synthesis model that affect outside plant costs. Specifically, we must determine: (1) whether optimization in the synthesis model should be turned on or off; (2) whether the model should use T1 technology; and (3) whether the model should use rectilinear or airline distances and the value of the corresponding "road factor."  X '` `  i.Optimization  X 4857. ` ` In the synthesis model, the user has the option of optimizing distribution plant  X 4routing via a minimum cost spanning tree algorithm discussed in the model documentation. ZD6 {O'ԍ The synthesis model always optimizes feeder plant. See C.A. Bush, et al., The Hybrid Cost Proxy Model  yO~'Customer Location and Loop Design Modules, Dec. 15, 1998 (HCPM Dec. 15, 1998 documentation) at 13. The algorithm functions by first calculating distribution routing using an engineering "rule of thumb" and then comparing the cost with the spanning tree result, choosing the routing that  Xd4minimizes annualized cost.:dD6 {O'ԍ Id. at 11.: The user also has the option of not using the distribution optimization feature, thereby saving a significant amount of computation time, but reporting network costs that may be significantly higher than with the optimization. In addition, the user has the option of using the distribution optimization feature only in the lowest density zones.  X4958. ` ` We tentatively conclude that the synthesis model should be run with the optimization turned on when the model is used to calculate the forward looking cost of providing the services supported by the federal mechanism. We point out that the optimization approach represents what a network planning engineer would attempt to accomplish in developing a forward-looking network. This approach also complies with criterion one's requirement that the model must assume the leastcost, most efficient, and reasonable technology for providing the supported service that is currently being deployed. We note, however, that the optimization can substantially increase the model's run time. Preliminary staff analysis of comparison runs with full optimization versus runs with no optimization indicate that, for clusters with line density greater than 500, the rule of thumb" F0*&&``"  X4algorithm results in the same or lower cost for nearly all clusters.XD6 yOy'ԍ Since, under full optimization, the model chooses the least cost of the full optimization algorithm or the rule of thumb algorithm, a comparison run as described above can show how well the full optimization performs as a function of density. We seek comment on whether an acceptable compromise to full optimization would be to set the optimization factor  X4at "-p500," as described in the model documentation.D6 {Ok'ԍ See C.A. Bush, et al., The Hybrid Cost Proxy Model Customer Location and Loop Design Modules  {O5'(Dec. 15, 1998) at 3031; see also Design History of HCPM, April 6, 1999 at http://www.fcc.gov/ccb/apd/hcpm. With this setting the model will optimize distribution plant whenever the density of a cluster is less than or equal to 500 lines per square mile. For purposes of further analysis of the proposed input values, we also anticipate that parties may wish to run the model without optimization turned on to save computing time. After staff has completed its analysis of comparison runs, we intend to make available a spreadsheet showing the estimated percentage change, for each nonrural study area, between running the model with the distribution optimization disabled and running the model with the distribution optimization enabled.   X '` ` ii.T1 Technology  X 4:59. ` ` A user of the synthesis model also has the option of using T1 technology as an alternative to copper feeder or fiber feeder in certain circumstances. T1 is a technology that allows digital signals to be transmitted on two pairs of copper wires at 1.544 Megabits per second (Mbps). If the T1 option is enabled, the optimizing routines in the model will choose the least cost feeder technology among three options: analog copper, T1 on copper, and  Xb4fiber.]bDD6 yOW'ԍ HCPM Dec. 15, 1998 documentation at 10.] For serving clusters with loop distances below the maximum copper loop length, the model could choose among all three options; between 18,000 feet and the fiber crossover point, which earlier versions of HCPM set at 24,000 feet, the model could choose between fiber and T1; and above the fiber crossover point, the model would always use fiber. In the HAI model, T1 technology is used to serve very small outlier clusters in locations where the copper distribution cable would exceed 18,000 feet. The BCPM sponsors and other LECs contend that T1 is not a forward looking technology and, therefore should not be used in the synthesis model. The HAI sponsors contend that current advertisements show that T1 is  X4being used currently.D6 {O/!'ԍ See Letter from Richard N. Clarke, AT&T, to Magalie Roman Salas, dated March 17, 1999, at Attachment A.  X|4;60. ` ` As noted, a number of parties contend that the T1 on copper technology is not forward looking. Other sources indicate that advanced technologies, like HDSL, potentially"e. 0*&&``"  X4can be used on T1 technology to transmit information at T1 or higher rates.KZD6 yOy'ԍ HDSL (high data rate digital subscriber line) transmits 1.544 Mbps or 2.048 Mbps in bandwidths ranging from 80 kilohertz (kHz) to 240 kHz, rather than in a bandwidth of 1.5 megahertz (mHz) required for traditional  {O 'T1 services. See www.adsl.com/general_tutorial.K We seek comment on this issue. We also seek comment on the extent to which HDSL technology presently is being used on T1.  X4<61. ` ` The only input values for T1 costs on the record in this proceeding are the  X4HAI default values.D6 yO( 'ԍ The HAI sponsors provide default values for T1 technology including the cost of repeaters and remote  {O 'T1 terminals. See HAI Inputs Portfolio at 4548. Because the synthesis model and the HAI model use T1 differently, we tentatively find that the HAI default values would not be appropriate for use in the synthesis model. In light of the fact that T1 may not be a forward looking technology and  XH4the lack of appropriate input values, we tentatively conclude that we should not use the T1 option in the synthesis model. We seek comment on our tentative conclusion. We ask that parties who disagree with our tentative conclusion and recommend that the T1 function be used in the synthesis model propose input values that will accurately estimate the cost of this technology, including what values are needed for the costs of shielded copper, repeaters, and terminals.  X '` ` iii.Distance Calculations and Road Factor   Xy4=62. ` ` We tentatively conclude that the synthesis model should use rectilinear distance,  Xb4rather than airline distance, in calculating outside plant distances ,bDD6 yOW'ԍIn short, this means that telephone plant will be built on northsouth and eastwest routes, rather than "as the crow flies." because this more accurately reflects the routing of telephone plant along roads and other rights of way. In fact,  X44research suggests that, on average, rectilinear distance closely approximates road distances.4D6 {O'ԍSee Robert F. Love, James G. Morris, and George O. Wesolowsky, Facilities Location: Models and  {OK'Methods, Chapter 10 (Elsevier Science Publishing Co. 1988) (Facilities Location Models). As a result, we tentatively conclude that the road factor in the model, which reflects the ratio between route distance and road distance, should be set equal to 1. We seek comment on these tentative conclusions.  X4>63. ` ` We also note that airline distance could be used in the model, if we were to derive accurate road factors. We seek comment on this alternative. Specifically, we seek" 0*&&``"  X4comment on whether we should use airline miles with wire center specific road factors.mBD6 yOy'ԍ Such a road factor could be calculated as follows. First, using the wire center boundaries from the customer location data, we would determine which Census Blocks are contained within each wire center. Second, we would extract the TIGER files containing road coordinates and distances for each of these Census Blocks. Third, we would create a database matching a sample set of coordinates of road intersections within the wire center, the road distance to an adjoining intersection, and the coordinates of the adjoining intersection.  {Oa'Using a formula for approximating airline distance, see Facilities Locations Models at 270, we could create a column containing airline distances. Fourth, we would regress road distance on airline distance to obtain the appropriate road factor for the wire center.m Research has shown that the airline distance metric with an appropriate road factor is more  X4accurate than the rectilinear metric.D6 {OU 'ԍSee Facilities Location Models, Chapter 10. The authors find that the goodnessoffit of a model similar to airline distance is significantly better than one incorporating rectilinear distance. We seek comment on this alternative approach.  X'` ` b. Cost of Copper Cable  Xv'` `  i.Preliminary Issues   XH4?64. ` ` The synthesis model uses tables that show the cost per foot of copper cable, by pair size. In selecting input values for the cost of copper cables, we must first address a number of preliminary issues: the extent to which 24 and 26 gauge copper cable should be used in the synthesis model; whether cable installation costs should differ between feeder and distribution cable; and whether cable installation costs should vary for underground, buried, and aerial cable.  X 4@65. ` ` Use of 24 and 26Gauge Copper. The HAI default values assume that all copper cable below 400 pairs in size is 24gauge and all copper cable of 400 pairs and larger  Xy4is 26gauge.Ey, D6 yOV'ԍ HAI Inputs Portfolio at 20.E The BCPM default values include separate costs for 24 and 26gauge copper of all sizes. We tentatively reject the HAI sponsors' argument that 26gauge copper costs  XK4should be used for all larger pair sizes of copper cable. We tentatively conclude that the model should use both 24gauge and 26gauge copper in all available pairsizes. Based on a preliminary analysis of the results of the structure and cable cost survey, it appears that a significant amount of 24gauge copper cable in larger pair sizes currently is being deployed.  X4 We seek comment on these tentative conclusions.  X4A66. ` ` Distinguishing Feeder and Distribution Cable Costs. We reaffirm the  X4Commission's tentative conclusion in the 1997 Further Notice that the same input values should be used for copper cable whether it is used in feeder or in distribution plant. Although" 0*&&``"  X4the BCPM sponsors previously disagreed with this tentative conclusion,D6 yOy'ԍ BCPM Sept. 24, 1997 comments at 13. We also note that the BCPM default values now include the same costs for feeder and distribution copper cable. they have not provided persuasive data for this position. We seek comment on this tentative conclusion.  X'  X4 B67. ` ` Distinguishing Underground, Buried, and Aerial Installation Costs. The HAI and BCPM sponsors both claim that their proposed values for cable costs include the cost of  X4installation.  D6 yO^ 'ԍ The HAI sponsors assert that these costs include "engineering, installation, and delivery, as well as the cable material itself." HAI Inputs Portfolio at 20. The BCPM sponsors represent that their default values for cable costs include the material cost, supply cost, taxes, placing, splicing, and engineering costs. BCPM, Loop Inputs Documentation at 15.  The BCPM defaults provide separate cost estimates for aerial, buried, and underground cable. The HAI default cable costs do not vary by type of plant and, therefore, appear to assume that installation costs are the same for aerial, underground, and buried cable. For buried copper cable, the HAI defaults include a multiplier to estimate the additional cost  X14of the filling compound used in buried cable to protect the cable from moisture.E1D6 yO'ԍ HAI Inputs Portfolio at 23.E For  X 4underground cable, HAI adds a per foot material cost for the conduit material.E D6 yOc'ԍ HAI Inputs Portfolio at 24.E  X 4C68. ` ` We tentatively conclude that we should adopt separate input values for the cost of aerial, underground, and buried cable. Based on our analysis of cable cost data, we have found considerable differences in the per foot cost of cable, depending upon whether the cable was strung on poles, pulled through conduit, or buried. We seek comment on this tentative  X4conclusion. ` `  Xb'` `  ii.Cost Per Foot of Copper Cable  X44D69. ` ` We now turn to the cost per foot of 24 and 26gauge copper cable. Both the HAI and BCPM sponsors provide default input values for copper cable costs that are based upon the opinions of their respective experts, but without data that enable us to substantiate those opinions. In addition, the Commission received cable cost data from a number of LECs, including data received in response to the structure and cable cost survey developed by staff, which staff is continuing to analyze, as noted above.   X4E70. ` ` At the December 11, 1998 workshop, Commission staff described how they had  X|4estimated the preliminary copper cable costs, by pair size and by plant type (i.e., aerial, buried, or underground), that had been posted on the Commission's Web site prior to the  XP4workshop. For copper cable, the staff estimated high and low values for the cost of the"P( 0*&&``" smallest pair size of 26gauge copper cable based on an analysis of the HAI default values  X4and the values submitted by states filing cost models in this proceeding. These estimates were adjusted for larger pair sizes of 26gauge cable and different structure types using estimates in Gabel and Kennedy's analysis of RUS data, which was published by the National Regulatory  X4Research Institute (NRRI Study).CD6 {O'ԍ See supra note 128.C The cost of 24gauge copper cable was estimated by applying a multiplier to the 26gauge estimates based on the relative weight of the copper in  Xv4these two gauges.vZD6 yO 'ԍ An earlier version of HCPM calculated 24gauge copper cable by multiplying the values for 26gauge  {OI 'copper cable by 1.17. See HCPM Dec. 15, 1998 model description at 19.  XH4F71. ` ` While the HAI sponsors support using the publicly available RUS data in the  X14NRRI Study to estimate cable costs,1D6 yO'ԍ Letter from Chris Frentrup, MCI Worldcom, to Magalie Roman Salas, FCC, dated Feb. 9, 1999 (MCI  {O^'Feb. 9, 1999 ex parte). Sprint questions the reliability and suitability of this  X 4data, and urges us instead to use the cable cost data provided by incumbent LECs. D6 yO'ԍ Letter from Pete Sywenki, Sprint, to Magalie Roman Salas, FCC, dated Jan 29, 1999 (Sprint Jan. 29,  {O'1999 ex parte). As  X 4Sprint points out, the RUS data contain information from only the two lowest density zones.U h D6 {O'ԍ Sprint Jan. 29, 1999 ex parte at 89.U Because loops are longer in sparsely populated areas, lower gauge copper often is used.  X 4G72. ` ` We tentatively conclude that we should use, with certain modifications, the estimates in the NRRI Study for the per foot cost of aerial, underground, and buried 24gauge  X4copper cable. As described below,G D6 {O;'ԍ See infra paras. 8586.G we also tentatively conclude that we should estimate the cost of 26gauge copper cable by adjusting our 24gauge estimates with ratios derived from cost data submitted by several nonrural LECs. The proposed cost estimates for 24 and 26gauge aerial, underground, and buried copper cable in various pair sizes are shown in  X44Appendix A. We seek comment on these tentative conclusions and proposed values.   X4H73. ` ` Although the RUS data were collected from the two lowest density zones, we note that none of the models considered by the Commission has the capability of varying cable costs by density zones. Nor have parties proposed cable cost values that vary by density zone. We also believe that Sprint has mischaracterized the analysis of the RUS data in the NRRI Study. For example, Sprint challenges the validity of the study because some of the observations have zero values for labor or material, while failing to recognize that these" 0*&&``F"  X4values were excluded from Gabel and Kennedy's regression analysis.YD6 {Oy'ԍ Sprint Jan. 29 ex parte, attachment at 5.Y Similarly, Sprint's complaint that Gabel and Kennedy do not analyze the components of total cable costs, labor and material, separately overlooks that Gabel and Kennedy's regression analysis is designed to  X4explain the variation in total costs.YZD6 {O'ԍ Sprint Jan. 29 ex parte, attachment at 7.Y   X4I74. ` ` The NRRI Study provides estimates for outside plant structure and cable costs using cost data derived from construction contracts supplied by the RUS for a sample of  X_4companies that operate under various soil, weather, and population density conditions. _D6 yO 'ԍ To develop these estimates, Gabel and Kennedy first developed from the raw data reflected on these contracts a data base that contains outside plant structure and cable costs. The outside plant structure and cable cost data in this data base are derived from 171 contracts for 57 companies in 27 states, adjusted to 1997 dollars. NRRI Study at 2. In generating these estimates, Gabel and Kennedy used standard regression techniques to measure  X14the effect of geological and density conditions on cable and structure costs.z1D6 yO'ԍ In these regression analyses, Gabel and Kennedy used data from the HAI model on line counts and rock, soil, and water conditions for the geographic region in which each company in the data base operates. NRRI Study at 3436. Regression analysis is a standard method used to study the dependence of one variable, the dependent variable, on one or more other variables, the explanatory variables. It is used to predict or forecast the mean value of the dependent variable on the basis of known or expected values of the explanatory variables.  {O'For a discussion of regression analysis, see William H. Greene (1990), Econometric Analysis, New York: MacMillan Publishing Company.  In general, the econometric formulations that Gable and Kennedy developed to estimate cable costs measure the effect on these costs of cable size and the placement of two or more cables on the same route.  X 4J75. ` ` We tentatively conclude that one substantive change should be made to Gabel and Kennedy's analysis. Gabel and Kennedy used the ordinary least squares statistical technique to estimate the cost of structure and cables. The ordinary least squares technique fits a straight line to the data by minimizing the sum of squared prediction errors. The ordinary least squares technique is efficacious, however, only for a data set lacking statistical  XK4outliers.KD6 yO!'ԍ Statistical outliers are values that are much higher or lower than other data in the data set. Such outliers have an undue influence on regression results, since the residual associated with each outlier is squared in calculating the regression. In order to mitigate the influence of such outlier values, statisticians have developed socalled robust regression techniques for estimating regression equations. We tentatively conclude that a robust regression technique should be used for analyzing the RUS data. We seek comment on this tentative conclusion."!0*&&``{"Ԍ X4ԙK76. ` ` Specifically, we tentatively conclude that the robust regression technique proposed by Huber should be applied to the RUS data. Essentially, this algorithm uses a standard statistical criterion to determine the most extreme outliers, and excludes them. Thereafter, as suggested by Huber, it iteratively performs a regression, then for each observation calculates an observation weight based on the absolute value of the observation residual. Finally, the procedure performs a weighted least squares regression using the calculated weights. This process is repeated until the values of the weights effectively stop changing. We have used the robust regression parameter estimates for cable, conduit, and buried structure. The use of robust estimation did not improve the statistical properties of the estimators for pole costs, so we tentatively conclude that the ordinary least squares technique  X 4is appropriate for pole costs. D6 yO 'ԍ For the robust regression of the pole cost equation, the value of the Fstatistic was not statistically significant at the five percent level. We seek comment on these tentative conclusions and analysis.   X 4L77. ` ` 24Gauge Aerial Copper Cable. We tentatively conclude that we should use the regression equation in the NRRI Study, as modified by the Huber methodology described  X 4above, to estimate the cost of 24gauge aerial copper cable, with three adjustments. D6 yO'ԍ This modified regression equation is set forth in Appendix D, section I.A. The appendix also includes an example of how we propose to use this equation to estimate the cost of 24gauge aerial copper cable.  X4M78. ` ` First, we propose to adjust the equation to reflect the superior buying power that nonrural LECs may have in comparison to the LECs represented in the RUS data. We seek comment on whether an adjustment for superior bargaining power is necessary, and, if so, how such an adjustment should be made.  X4N79. ` ` Based on data entered into the record in a proceeding before the Maine Public Utilities Commission, Gabel and Kennedy determined that Bell Atlantic's material costs for aerial copper cable are approximately 15.2 percent less than these costs for the RUS  X4companies.;xD6 yO'ԍ NRRI Study at 47.; We tentatively conclude that this figure represents a reasonable estimate of the difference in the material costs that non-rural LECs pay in comparison to those that the RUS companies pay. To reflect this degree of buying power in the cable cost estimates that we derive for non-rural LECs, we propose to reduce the regression coefficient for the number of copper pairs by 15.2 percent for aerial copper cable. This coefficient measures the incremental or additional cable cost associated with one additional copper pair and therefore largely reflects the material cost of the cable. We seek comment on this proposed adjustment. We also invite parties to suggest alternative methods for capturing the impact of superior buying power. " "0*&&``"Ԍ X4O80.` ` Second, we propose to adjust the equation in the NRRI Study to account for LEC engineering costs, which were not included in the RUS cable data. The BCM2 default values include a loading of five percent for engineering. The HAI sponsors claim that engineering constitutes approximately 15 percent of the cost of installing outside plant cables. This percentage includes both contractor engineering and LEC engineering. The cost of contractor engineering already is reflected in the RUS cable cost data. Based on the record, we tentatively conclude that we should add a loading of 10 percent to the material and labor cost of the cable (net of LEC engineering and splicing costs) to approximate the cost of LEC engineering. We seek comment on this tentative conclusion and invite commenters to justify an alternative loading factor for LEC engineering.  X 4P81.` ` Third, we propose to adjust the equation to account for splicing costs, which also were not included in the RUS data. In the NRRI Study, Gabel and Kennedy determined that the ratio of splicing costs to copper cable costs (excluding splicing and LEC engineering  X 4costs) is 9.4 percent for RUS companies.; D6 yO7'ԍ NRRI Study at 29.; We tentatively conclude that we should adopt a loading of 9.4 percent for splicing costs. We seek comment on this tentative conclusion.  Xy4Q82. ` ` 24Gauge Underground Copper Cable. We tentatively conclude that we should use the regression equation in the NRRI Study, as modified by the Huber methodology described above, to estimate the cost of 24gauge underground copper cable. We also tentatively conclude that we should use the same three adjustments proposed for 24gauge  X4aerial copper cable, with one exception.LXD6 {O&'ԍ See Appendix D, section I.B.L We tentatively conclude that we should reduce the regression coefficient for the number of copper pairs by 16.3 percent, to reflect superior  X4buying power, based on the analysis in the NRRI study.hZD6 yO'ԍ Based on data entered into the record in the aforementioned proceeding before the Maine Commission, Gabel and Kennedy determined that Bell Atlantic's material costs for underground copper cable are  {O'approximately 16.3 percent less than these costs for the RUS companies. See NRRI Study at 47.h We seek comment on the use of this equation and the proposed adjustments.  X4R83. ` ` 24Gauge Buried Copper Cable. We tentatively conclude that it is necessary to modify the regression equation in the NRRI Study, as modified by the Huber methodology described above, to estimate the cost of a 24gauge buried copper cable, because the equation in the study includes labor and material costs for both buried cable and structure. Appendix  XN4D provides further detail on this proposed equation.LN D6 {O $'ԍ See Appendix D, section I.C.L We seek comment on this tentative conclusion and proposed equation. " #0*&&``<"Ԍ X4S84.` ` We propose to make the same three adjustments to this equation as we proposed for 24gauge aerial and underground cables, with the exception of the adjustment for superior buying power. Because the NRRI Study does not include a recommendation for such an adjustment for buried cable, we tentatively conclude we should use 15.2 percent, which is the lower of the reductions used for aerial and underground cable. We seek comment on the use of these adjustments for 24gauge buried cable.  X_4T85. ` ` 26Gauge Copper Cable. Because the NRRI Study did not provide estimates for 26gauge copper cable, we must either use another data source or find a method to derive these estimates from those for 24gauge. The HAI sponsors support the proposal presented by Commission staff at the workshop to use the relative weight of copper to adjust the 24gauge copper costs to derive 26gauge copper costs, although they would make further adjustments  X 4to reflect the cost of 26gauge copper for cable sizes of 400 pairs and larger.J D6 {Oe 'ԍ MCI Feb. 9, 1999 ex parte.J The BCPM sponsors challenge the assumption that the cost of copper cable is closely tied to the relative  X 4weight of the copper in the cable. ZD6 yO'ԍ Letter from Pete Sywenki, Sprint, to Magalie Roman Salas, FCC, dated Feb. 26, 1999 (Sprint Feb. 26,  {O'1999 ex parte). Both the HAI sponsors and the BCPM sponsors argue that the cost of splicing is not directly a function of investment, but rather is primarily a  X4function of the number of pairs to be spliced, and the distance between splices.D6 {O'ԍ MCI Feb. 9, 1999 ex parte at 67; Sprint Feb. 26, 1999 ex parte, attachment at 6. Although they agree that splicing costs should be estimated using the average cost per pairfoot, they disagree over what those costs should be.  X44U86. ` ` We tentatively conclude that we should derive cost estimates for 26gauge cable by adjusting our estimates for 24gauge cable. We agree with the BCPM sponsors that the cost of copper cable should not be estimated based solely on the relative weight of the cable. Instead, we propose to use the ordinary least squares regression technique to estimate  X4the ratio of the cost of 26gauge to 24gauge cable for each plant type (i.e., aerial, underground, buried). We propose to estimate these ratios using data on 26gauge and 24gauge cable costs submitted by Aliant and Sprint and the BCPM default values for these  X4costs.FD6 yO 'ԍ We are not able to use the HAI default values in addition to these data to estimate these ratios because the HAI defaults do not have separate values for 26gauge and 24gauge cable costs for each different cable size. While we would prefer to develop these ratios based on data from more than these three sources, we tentatively conclude that these are the best data available on the record for this purpose. We seek comment on these tentative conclusions and proposed analysis,  XP4including the regression techniques described in Appendix D.SPD6 {O%'ԍ See Appendix D, sections I.D, E, F.S We invite parties to propose"P$0 0*&&``" alternative methods of deriving cost estimates for 26gauge cable.  X'` ` c. Cost of Fiber Cable  X4V87. ` ` In selecting input values for fiber cable costs, we must determine values for the cost per foot of fiber for various strand sizes for aerial, underground, and buried cable. Both the HAI and BCPM sponsors provide default input values for fiber cable costs that are based upon the opinions of their respective experts, without data enabling us to substantiate those opinions. In addition, the Commission received cable cost data from a number of LECs, including data received in response to the structure and cable cost survey, which staff is continuing to analyze, as noted above.  X 4W88. ` ` At the December 11, 1998 workshop, Commission staff described how they had computed the preliminary fiber cable costs, by pair size and by plant type (aerial, buried,  X 4or underground) that had been posted on the Commission's Web site prior to the workshop.  X 4 Using a methodology similar to the one used for copper cable, staff estimated the cost of the smallest size fiber cable based on an analysis of proposed values and used the analysis in the NRRI Study to derive costs for larger sizes.  XK4X89. ` ` We tentatively conclude that we should use the RUS data and the analysis in the NRRI Study, with certain adjustments, to estimate fiber cable costs. For the reasons discussed above for copper cable, we also tentatively conclude that the cost of fiber cable will vary for aerial, underground, and buried plant. We tentatively select the input values for the per foot cost of aerial, underground, and fiber cable in various strand sizes, as shown in Appendix A. We seek comment on these tentative conclusions and proposed values.  X4Y90. ` ` Aerial Fiber Cable. We tentatively conclude that we should use the regression equation in the NRRI Study, as modified by the Huber methodology described above, to estimate the cost of aerial fiber cable, with three adjustments similar to those made for copper  Xe4cable.eD6 yO'ԍ This modified regression equation is set forth in Appendix D, section II.A. The appendix also includes an example of how we propose to use this equation to estimate the cost of aerial fiber cable. We seek comment on this tentative conclusion.  X74Z91. ` ` As noted, we propose three adjustments to the equation used in the NRRI Study to estimate the cost of aerial fiber cable. First, based on the NRRI Study, we propose to reduce by 33.8 percent the regression coefficient for the number of fiber strands, to reflect the  X4superior buying power of nonrural LECs.[Z D6 yO#'ԍ Based on data entered into the record in the aforementioned proceeding before the Maine Commission, Gabel and Kennedy determined that Bell Atlantic's material costs for fiber cable are approximately 33.8 percent  {OS%'less than these costs for the RUS companies. See NRRI Study at 47.[ Second, for the reasons described earlier,EBD6 {O'ԍ See supra para. 80.E we"%Z0*&&``" tentatively conclude that we should add a loading of 10 percent to the material and labor cost of the cable (net of LEC engineering and splicing costs) to approximate the cost of LEC engineering. Finally, we tentatively conclude that we should add a loading for splicing costs of 4.7 percent to the material and labor cost of the cable (net of LEC engineering and splicing  X4costs), based on the estimates in the NRRI Study.;ZD6 yO'ԍ NRRI Study at 29.; We seek comment on these tentative conclusions and proposed adjustments.  X_4[92. ` ` Underground Fiber Cable. We tentatively conclude that we should use the regression equation in the NRRI Study, as modified by the Huber methodology described above, to estimate the cost of underground fiber cable, with three adjustments similar to those  X 4made for aerial fiber cable.M D6 {O'ԍ See Appendix D, section II.B.M We seek comment on this tentative conclusion.  X 4\93. ` ` As noted, we propose three adjustments to the NRRI equation for the cost of underground fiber cable. First, based on the NRRI Study, we propose to adjust downward by 27.8 percent the regression coefficient for the number of fiber strands, to reflect the superior  X 4buying power of nonrural LECs.E |D6 {O'ԍ See NRRI Study at 47.E Second, for the reasons described earlier,C D6 {Of'ԍ See supra para. 80.C we tentatively conclude that we should add a loading of 10 percent to the material and labor cost of the cable (net of LEC engineering and splicing costs) to approximate the cost of LEC engineering. Finally, we tentatively conclude that we should add a loading for splicing costs of 4.7 percent to the material and labor cost of the cable (net of LEC engineering and splicing  X44costs), based on the estimates in the NRRI Study.;4D6 yO'ԍ NRRI Study at 29.; We seek comment on these tentative conclusions and proposed adjustments.  X4]94. ` ` Buried Fiber Cable. We tentatively conclude that it is necessary to modify the regression equation in the NRRI Study, as modified by the Huber methodology described above, to estimate the cost of a buried fiber cable, because the equation in the study includes labor and material costs for both buried fiber cable and structure. Appendix D provides  X4further detail on the proposed modifications to the equation used in the NRRI Study.M0 D6 {Ot$'ԍ See Appendix D, section II.C.M We seek comment on this tentative conclusion and proposed equation."|& 0*&&``"Ԍ X4ԙ^95.` ` We also propose three adjustments to the proposed equation. First, based on the NRRI Study, we propose to reduce by 27.8 percent the regression coefficient for the  X4number of fiber strands, to reflect the superior bargaining power of nonrural LECs.D6 yOK'ԍ Based on data entered into the record in the aforementioned proceeding before the Maine Commission, Gabel and Kennedy determined that Bell Atlantic's material costs for underground and aerial fiber cable are  {O'approximately 33.8 and 27.8 percent lower than the RUS values. See NRRI Study at 47. No data are available for buried fiber. We tentatively conclude that we should use the lower of these two numbers 27.8 percent for buried fiber cable.  X4Second, for the reasons described earlier,CzD6 {O 'ԍ See supra para. 80.C we tentatively conclude that we should add a loading of 10 percent to the material and labor cost of the cable (net of LEC engineering and splicing costs) to approximate the cost of LEC engineering. Finally, we tentatively conclude that we should add a loading for splicing costs of 4.7 percent to the material and labor cost of the cable (net of LEC engineering and splicing costs), based on the estimates in the NRRI  XH4Study.;H D6 yO'ԍ NRRI Study at 29.; We seek comment on these tentative conclusions and proposed adjustments.   X ' ` ` c. Cable Fill Factors  X 4_96. ` ` In determining appropriate cable sizes, network engineers include a certain amount of spare capacity to accommodate administrative functions, such as testing and repair, and some expected amount of growth. The percentage of the total usable capacity of cable  X 4that is expected to be used to meet anticipated demand is referred to as the cable fill factor.3 D6 yO'ԍ We note that the actual fill factor may be lower than the fill factor used to design the network (sometimes referred to as administrative fill), because cable and fiber are available only in certain sizes. For example, assume a neighborhood with 100 households has a current demand of 120 telephones. Dividing the 120 pair demand by an 80 percent administrative fill factor establishes a need for 150 pairs. However, cable is not sold in 150 pair units. The company will purchase the smallest cable that is sufficient to provide 150 pairs, which is a 200 pair cable. The fill factor that occurs and is measurable, known as the effective fill, will be the number of pairs needed to meet demand, 120 pairs, divided by the number of pairs installed, 200 pair, or 60 percent. 3 If cable fill factors are set too high, the cable will have insufficient capacity to accommodate small increases in demand or service outages. In contrast, if cable fill factors are set too low, the network could have considerable excess capacity for many years. While carriers may choose to build excess capacity for a variety of reasons, we must determine the appropriate cable fill factors to use in the federal mechanism. If the fill factors are too low, the resulting excess capacity will increase the model's cost estimates to levels higher than an efficient firm's costs, potentially resulting in excessive universal service support payments. "'l0*&&``w"Ԍ X4`97. ` ` Variance Among Density Zones. In general, both the HAI and BCPM sponsors provide default fill factors for copper cable that vary by density zone, and they agree that fill  X4factors should be lower in the lowest density zones.D6 yOK'ԍ As explained below, default values in BCPM 3.1 for distribution cable do not vary by density zone. HAI sponsors claim that an outside plant engineer is more interested in providing a sufficient number of spares than in the ratio  X4of working pairs to spares, so the appropriate fill factor will vary with cable size.cXD6 yO'ԍ HAI Dec. 11, 1997 submission, Inputs Portfolio at 39, 63.c For example, 75 percent fill in a 2400 pair cable provides 600 spares, whereas a 50 percent fill in a six pair cable provides only three spares. Because smaller cables are used in lower density zones, HAI recommends that lower fill factors be used in the lowest density zones to ensure there will be enough spares available. The BCPM sponsors claim that less dense areas require lower fill ratios because the predominant plant type is buried and it is costly to add additional  X 4capacity after installation.v D6 yO'ԍ BCPM 3.1 May 26, 1998 (Preliminary Edition) Loop Inputs Documentation at 51.v We tentatively agree with the HAI and BCPM sponsors that fill factors for copper cable should be lower in the lowest density zones, which is reflected in the fill factors that we propose in this Notice. We seek comment of this tentative finding.   X 4a98. ` ` Distribution Fill Factors. The fill factors proposed by the HAI sponsors for  X 4distribution cable are somewhat lower than for copper feeder cable.  xD6 yO'ԍ HAI 5.0 default values range from 50 percent in the lowest density zone to 75 percent in the highest density zone for distribution cable sizing fill factors, and range from 65 percent in the lowest density zone to 75 percent in the highest density zone for copper feeder cable sizing fill factors. HAI Dec. 11, 1997 submission, Inputs Portfolio at 39, 63. The BCPM default fill factors for distribution cable, on the other hand, currently are set at 100 percent for all density  Xy4zones.9Zy` D6 yO'ԍ BCPM Dec. 11, 1997 submission. Earlier versions of BCPM, however, had lower fill factors for  {OR'distribution than for feeder. See, e.g., Further Notice at para. 118. Default values in BCPM 3.1 range from 75 to 85 percent for feeder cable. 9 This difference is related to the differences between certain assumptions that were made in the HAI and BCPM models. The HAI proponents claim that the level of spare capacity provided by their default values is sufficient to meet current demand plus some  X44amount of growth.c4 D6 yOg 'ԍ HAI Dec. 11, 1997 submission, Inputs Portfolio at 39, 63.c This is consistent with the HAI model's approach of designing plant to meet current demand, which on average is 1.2 lines per household. BCPM, on the other hand, designs outside plant with the assumption that every residential location has two lines, which is more than current demand. Because it is costly to add distribution plant at a later point in time, incumbent LECs typically build enough distribution plant to meet not only"(0*&&``"  X4current demand, but also anticipated future demand.(D6 {Oy'ԍ For example, in an ex parte meeting on March 24, 1999, Ameritech representatives said that Ameritech  {OC'designs distribution plant to meet "ultimate" demand and designs feeder plant that is "growable." See Letter  {O 'from Celia Nogales, Ameritech, to Magalie Roman Salas, FCC, dated March 25, 1999 (Ameritech March 25 ex  {O'parte). BCPM adopts this convention. Setting the fill factor at 100 percent in BCPM offsets BCPM's assumption that every household has two lines and the resulting estimation of appropriate cable sizes is sufficient to meet current  X4demand, rather than long term growth. D6 yO$ 'ԍ Commission staff requested that the BCPM sponsors make this change from earlier versions of BCPM to avoid double counting. That is, cable sizing in BCPM is a function of both the number of lines per customer  yO 'location and the fill factor. In HAI, cable sizing is a function of the number of customer locations and current demand of 1.2 telephones per household.  X4b99. ` ` In a meeting with Commission staff, Ameritech raised the issue of whether industry practice is the appropriate guideline for determining fill factors to use in estimating the forwardlooking economic cost of providing the services supported by the federal mechanism. Ameritech claims that forwardlooking fill factors should reflect enough capacity to provide service for new customers for a few years until new facilities are built, and should account for the excess capacity required for maintenance and testing, defective copper pairs,  X 4and churn.F\ D6 yOT'ԍ Ameritech filed data, subject to the protective order in this proceeding, showing how these considerations  {O'are used to calculate the actual and forwardlooking fill factors in Ameritech's territory. See Ameritech March  {O'25 ex parte.F  X 4c100. ` ` We tentatively conclude that the fill factors selected for use in the federal  X 4mechanism generally should reflect current demand, D6 yO3'ԍ We define "current demand" to include a reasonable amount of excess capacity to accommodate short term growth. and not reflect the industry practice of building distribution plant to meet "ultimate" demand. The fact that industry may build distribution plant sufficient to meet demand for ten or twenty years does not necessarily suggest that these costs should be supported by universal service support mechanisms. This also appears to reflect the assumptions underlying the HAI and BCPM default fill factors. Because the synthesis model designs outside plant to meet current demand in the same manner as the HAI model, we believe the fill factors should be set at less than 100 percent. We tentatively select the HAI defaults for distribution fill factors and tentatively conclude that they reflect the appropriate fill needed to meet current demand. We seek comment on these tentative conclusions.  X4d101. ` ` Feeder Fill Factors. In contrast to distribution plant, feeder plant typically is")0*&&``1"  X4designed to meet only current and short term capacity needs.]D6 {Oy'ԍ See, e.g., Ameritech March 25 ex parte.] The BCPM copper feeder default fill factors are slightly higher than HAI's, but both the HAI and BCPM default values appear to reflect current industry practice of sizing feeder cable to meet current, rather than long term, demand. Because both the HAI and BCPM default values assume that copper feeder fill reflects current demand, we tentatively select copper feeder fill factors that are the average of the HAI and BCPM default values. We seek comment on these tentative selections.   XH4e102. ` ` Fiber Fill Factors. Because of differences in technology, fiber fill factors typically are higher than copper feeder fill factors. Standard fiber optic multiplexers operate on four fiber strands: primary optical transmit, primary optical receive, redundant optical transmit, and redundant optical receive. In determining appropriate fiber cable sizes, network engineers take into account this 100 percent redundancy in determining whether excess  X 4capacity is needed that would warrant application of a fill factor. ZD6 yO'ԍ That is, fiber plant with a 100 percent fill factor has an actual utilization of 50 percent; whereas copper plant with a 50 percent fill factor has an actual utilization of 50 percent. Both the HAI and BCPM models use the standard practice of providing 100 percent redundancy for fiber and set the default fiber fill factors at 100 percent. We tentatively conclude that the input value for fiber fill in the federal mechanism should be 100 percent. We seek comment on this tentative conclusion.  XK' C.` ` Structure Costs   X' ` ` 1. Background   X4f103. ` ` In the 1997 Further Notice, the Commission sought comment and adopted tentative findings and conclusions on issues relating to the cost of outside plant. The Commission directed the HAI and BCPM proponents to justify fully their default values for  X4their mix of aerial, underground, and buried structure (i.e., plant mix) and sought comment on  X4the input values that will accurately reflect the impact of varying terrain conditions on costs.YD6 {O'ԍ 1997 Further Notice, 12 FCC Rcd at 18541.Y The Commission noted that "recent installations of outside structure may more closely meet  Xi4forwardlooking design criteria than do historical installations."[iDD6 {O^"'ԍ 1997 Further Notice, 12 FCC Rcd at 18541. [ The Commission found that an efficient carrier will vary its plant mix according to the population density of an area and tentatively concluded that the assignment of plant mix defined by the model should reflect";*0*&&``_"  X4both terrain factors and line density zones.YD6 {Oy'ԍ 1997 Further Notice, 12 FCC Rcd at 18541.Y Because burying cable in very rocky areas is costly, for example, the Commission tentatively concluded that relatively more feeder and distribution cable should be assigned to aerial installation for all density zones in wire centers characterized as "hard rock" conditions than in those wire centers with other terrain  X4conditions.[ZD6 {O'ԍ 1997 Further Notice, 12 FCC Rcd at 18541. [  Xv4g104. ` ` Outside plant structure refers to the set of facilities that support, house, guide, or otherwise protect distribution and feeder cable and varies by plant mix. Aerial structure  XH4consists of telephone poles, and associated hardware, such as anchors and guys. Buried  X14structure consists of trenches.w1D6 yO 'ԍ When a plow is used to place buried cable, a separate trench is not required.w Underground structure consists of trenches and conduit, and for feeder plant, manholes and pullboxes. As noted, underground cable is placed underground within conduits for added support and protection. Structure costs include the initial capital outlay for physical material associated with outside plant structure, including manholes; conduit, trenches, poles, anchors and guys, and other facilities; the capitalized cost for supplies, delivery, provisioning, right of way fees, taxes, and any other capitalized costs directly attributable to these assets; and the capitalized cost for the labor, engineering, and materials required to install these assets. For example, buried and underground structure costs include capitalized labor, engineering, and material costs for such activities as plowing or trenching, backfilling, boring cable, and cutting and restoring asphalt, concrete, or sod, or any  XK4combination of such activities.hhC  X4h105. ` ` Both the HAI and BCPM sponsors provide default input values for structure costs that are based upon the opinions of their respective experts, without backup data that allows us to substantiate these values. Although BCPM provides one nationwide set of default values for structure costs, the BCPM sponsors have argued that we should use companyspecific inputs and have proposed alternative values for companyspecific structure  X4costs in some study areas."|D6 {O'ԍ See Letter from Pete Sywenki, Sprint, to Magalie Roman Salas, FCC, dated June 11, 1998, attachment; BellSouth reply comments dated June 12, 1998, at 2 (arguing only state specific input values are appropriate); Letter from William W. Jordan, BellSouth, to Magalie Roman Salas, FCC, dated August 7, 1998, attachment,  yO1!'Responses to FCC Staff Questions of June 25, 1998, Question 4.  In addition, the Commission received other structure cost data from a number of LECs, including data received in response to the structure and cable cost survey developed by staff, which staff is continuing to analyze, as noted above.  XN' 2.` ` Issues for Comment  "7+f 0*&&``_"Ԍ X4i106. ` ` The synthesis model uses structure cost tables that identify the per foot cost of structure by type (aerial, buried, or underground), loop segment (distribution or feeder), and terrain conditions (normal, soft rock, or hard rock), for each of the nine density zones. For aerial structure, the cost per foot that is entered in the model is calculated by dividing the total installed cost per telephone pole by the distance between poles. As described below, we tentatively conclude that we should use, with certain modifications, the estimates in the NRRI Study for the per foot cost of aerial, underground, and buried structure. In general, these estimates are derived from regression equations that measure the effect on these costs of density, water, soil, and rock conditions.  X1'  X 'a.` ` Cost of Aerial Structure  X 4j107. ` ` We tentatively conclude that we should use the regression equation for aerial  X 4structure in the NRRI Study as a starting point. D6 {ON'ԍ See NRRI Study at 52, Table 212. This regression equation is set forth in Appendix D, section III.A. We propose to use this equation to develop proposed input values for the labor and material cost for a 40foot, class four telephone pole. We develop separate pole cost estimates for normal bedrock, soft bedrock, and hard  X4bedrock.NZD6 {O'ԍ See Appendix D, section III.A.N The regression coefficients estimate the combined cost of material and supplies. The NRRI Study reports that the average material price for a 40foot, class four pole is  Xb4$213.94.GbD6 yO'ԍ NRRI Study at 51, Table 211.G We note that this estimate is very close to results obtained from the data  XK4submitted in response to the 1997 Data Request. According to the Commission staff's  X64analysis of these data, the unweighted average material cost of a 40foot, class four pole is  X4$213.97, and the weighted average, by line count, is $228.22.|D6 yOL'ԍ This is slightly higher than the HAI default value of $201.00 for the same height and class pole, and considerably less than the BCPM default value of $368.17 We seek comment on this tentative conclusion and analysis.  X4k108.` ` We tentatively conclude that we should add to these estimates the cost of anchors, guys, and other materials that support the poles, because the RUS data from which this regression equation was derived do not include these costs. In the NRRI Study, Gabel and Kennedy used the RUS data to develop the following cost estimates for anchors, guys and other polerelated items: $32.98 in rural areas, $49.96 in suburban areas, and $60.47 in urban  Xg4areas.QgD6 {O#'ԍ See NRRI Study at 55, Table 214.Q We tentatively conclude that these are reasonable estimates for the cost of anchors, guys, and other polerelated items. We seek comment on these tentative conclusions and proposed values."9,f 0*&&``"Ԍ X4   X4l109.` ` We also tentatively add an estimate for the cost of LEC engineering, which is not reflected in the data from which Gabel and Kennedy derived cost estimates for poles and anchors, guys, and polerelated materials. For the reasons described above for copper and fiber cable, we tentatively conclude that we should add a loading of 10 percent to the material and labor cost (net of LEC engineering) for poles, anchors, guys, and other polerelated items. We seek comment on these tentative conclusions and invite proposals justifying an alternative loading factor for LEC engineering.  X14m110.` ` In order to obtain proposed input values that can be used in the model, we must convert the estimated pole costs into per foot costs for each of the nine density zones. For purposes of this computation, we propose to use for density zones 1 and 2 the per pole cost that we have estimated for rural areas, based on the NRRI Study; for density zones 3 through 7 the per pole cost for suburban areas; and for density zones 8 and 9 the per pole cost for urban areas. We then divide the estimated cost of a pole by the estimated distance between poles. We propose to use the following values for the distance between poles: 250 feet for density zones 1 and 2; 200 feet for zones 3 and 4; 175 feet for zones 5 and 6; and 150 feet for zones 7, 8, and 9. For the most part, these values are consistent with both the HAI and BCPM defaults. We seek comment on these proposals.  X4'` ` b. Cost of Underground Structure  X4n111. ` ` We tentatively conclude that we should adopt a similar methodology to estimate the cost of underground structure, as we proposed for the cost of aerial structure. We tentatively conclude that we should use the equation set forth in the appendix as a starting  X4point for this estimate.%ZD6 {O:'ԍ See Appendix D, section III.B. This regression equation is based on the RUS data, but was developed after the publication of that report. The NRRI Study does not set forth a regression equation for estimating the cost of underground structure.% We propose to use this equation to develop proposed input values for the labor and material cost for underground cable structure. We develop separate cost estimates for underground structure in normal bedrock, soft bedrock, and hard bedrock for  X|4density zones 1 and 2.RX|D6 yO'ԍ This regression equation was developed using underground cost data for density zones 1 and 2. The variable in this equation that represents the density zone of the geographic area in which the underground costs are incurred is not statistically significant at any standard level of significance.R As we did for aerial structure, we tentatively conclude that we should add a loading factor of 10 percent for LEC engineering. We seek comment on these tentative conclusions.  X 4o112.` ` We are able to develop directly from the regression equation cost estimates for underground structure only in density zones 1 and 2, because the RUS data is from companies" - 0*&&``" that operate only in those density zones. We tentatively conclude that we should derive cost estimates for density zones 3 through 9 by extrapolating from the estimates for density zone 2. We further tentatively conclude that we should perform such extrapolation based on the growth rate between density zones in the BCPM and HAI default values for underground and  X4buried structure.D6 yO'ԍ We propose to use this same extrapolation method for both underground and buried structure. Although we would prefer to rely on data specific to the density zone, rather than extrapolated, we tentatively conclude that, based on our current analysis, this is the  Xv4best data currently available for this purpose.vXD6 yO 'ԍ As noted, staff is continuing to analyze the data received in response to the survey on cable costs and structure. We seek comment on these tentative  X_4conclusions. In Appendix D, we describe the proposed method of extrapolation.N_D6 {O 'ԍ See Appendix D, section III.B.N We seek comment on this proposed method and invite parties to suggest alternative methods for estimating costs in density zones 3 through 9.  X '` ` c. Cost of Buried Structure (#  X 4p113. ` ` We tentatively conclude that we should use the modified equation for estimating the cost of 24gauge buried copper cable and structure to estimate the cost of  X 4buried structure.b BD6 yO'ԍ This equation is set forth in Appendix D, section III.C.b It is necessary to modify this equation because estimates derived from it include labor and material costs for both buried cable and structure. Appendix D provides  Xy4further detail on the modified equation.NyD6 {O'ԍ See Appendix D, section III.C.N We seek comment on this tentative conclusion.  XK4q114. ` ` For the reasons described above, we tentatively conclude that we should add a loading of 10 percent for LEC engineering to the estimates generated by the modified equation. We seek comment on this tentative conclusion.  X4r115. ` ` We are able to develop directly from the regression equation cost estimates for buried structure only in density zones 1 and 2, because the RUS data is from companies that operate only in those density zones. We tentatively conclude that we should derive cost estimates for density zones 3 through 9 by extrapolating from the estimates for density zone 2. We further tentatively conclude that we should perform such extrapolation based on the same method proposed for estimating the cost of underground structure. We seek comment  Xe4on these tentative conclusions.ned D6 yOz%'ԍ This extrapolation method is described in Appendix D, section III.B.n"e. 0*&&``"Ԍ X'ԙ` ` d. Plant Mix  X4s116. ` ` As discussed above, we have tentatively selected input values for the costs of cable and outside plant structure that differ for aerial, buried, and underground cable and structure. Because these cost differences can be significant, the relative amount of plant type  X4in any given area, i.e., the plant mix, plays a significant part in determining total outside plant investment. The synthesis model provides three separate plant mix tables, for distribution, copper feeder, and fiber feeder, which can accept different percentages for each of the nine density zones. Although we tentatively propose using nationwide input values for plant mix, as we have for other input values, we seek comment on an alternative to nationwide plant mix input values, as discussed below.  X 4t117. ` ` The BCPM sponsors claim that in low densities there generally is a greater percentage of buried plant than underground plant, and conversely, in higher densities there is  X 4more underground than buried plant.T D6 yO9'ԍ BCPM, Loop Inputs Documentation at 49, 54.T The BCPM default plant mix values reflect these assumptions. Although the HAI default plant mix values for feeder plant also reflect these assumptions, HAI's assumptions with respect to distribution plant mix are quite different than BCPM's, as discussed below. The HAI sponsors suggest that aerial plant is still the most prevalent plant type, but claim that their default plant mix values reflect an increasing trend  XM4toward the use of buried cable in new subdivisions._\MXD6 {OV'ԍ HAI Inputs Portfolio at 37. The HAI sponsors quote a 1994 edition of Bellcore's BOC Notes on the  {O 'LEC Networks: "The most common cable structure is still the pole line. Buried cable is now used whenever feasible, but pole lines remain an important structure in today's environment."_ The HAI default values generally  X64assume that there is more aerial plant than the BCPM default values. 6|D6 yOc'ԍ In the four lowest density zones, the HAI defaults assume slightly less aerial fiber feeder than the BCPM defaults, but in all other cases, assume more aerial plant than BCPM. In addition, the HAI model provides that a certain percentage of buried plant can be shifted to aerial (and vice versa) based on certain cost minimization routines. The BCPM defaults have separate values for plant mix in hard rock terrain, which generally assume there is slightly more aerial and less buried plant than the normal and soft rock terrain defaults.  X4u118. ` ` Distribution Plant. The BCPM default values for distribution plant assume that there is no underground plant in the lowest density zone and the percentage increases with each density zone to 90 percent underground distribution plant in the highest density zone. In contrast, the HAI default values for distribution plant mix place no underground structure in the six lowest density zones and assume that only 10 percent of the structure in the highest  Xg4density zone is underground.Egd D6 yO|%'ԍ HAI Inputs Portfolio at 36.E The BCPM default values assume there is no aerial plant in"g/ 0*&&``~" the highest density zone in normal and soft rock terrain, and 10 percent aerial plant in hard rock terrain. In contrast, the HAI default values assume that there is significantly more aerial cable, 85 percent, in the highest density zone, but notes that this includes riser cable within multistory buildings and "block cable" attached to buildings, rather than to poles.  X4v119. ` ` We tentatively select input values for distribution plant mix that more closely reflect the assumptions underlying BCPM's default values than HAI's default values for several reasons. The synthesis model does not design outside plant that contains either riser cable or block cable, so we do not believe it would be appropriate to assume that there is as high a percentage of aerial plant in densely populated areas as the HAI default values assume. Although our proposed plant mix values assume somewhat less underground structure in the lower density zones than the BCPM default values, we disagree with HAI's assumption that there is very little underground distribution plant and none in the six lowest density zones. We tentatively select the distribution plant mix values set forth in Appendix A, and seek comment on our tentative conclusions. We tentatively propose input values, for the lowest to the highest density zones, that range from zero percent to 90 percent for underground plant; 60 to zero percent for buried plant; and 40 to ten percent for aerial plant.  Xb4w120. ` ` Feeder Plant. The default plant mix percentages for feeder plant are generally similar in the BCPM and the HAI models. Although the BCPM default values vary between normal or soft rock terrain and hard rock terrain, as noted above, and the HAI default values  X4differ between copper and fiber feeder, the plant mix ratios across density zones are similar.kD6 yO'ԍ The BCPM default values for copper and fiber feeder are the same.k For example, both the BCPM default values and the HAI default values assume that there is  X4only five or ten percent of underground feeder plant in the lowest density zone.9XXD6 yO'ԍ HAI default values assume five percent underground feeder plant in the lowest density zone for both copper feeder and fiber. BCPM default values assume ten percent underground copper and fiber feeder in normal and soft rock terrain and five percent in hard rock terrain.9 The HAI defaults assume there is somewhat more aerial feeder cable than the BCPM defaults, except for fiber feeder cable in the four lowest density zones. The BCPM defaults assume there is no aerial feeder plant in the three highest density zones, except in hard rock terrain. Despite these differences, the relative amounts of aerial and buried plant across density zones are  X|4generally similar.|xD6 yO 'ԍ For aerial plant, in the lowest to the highest density zones, the BCPM defaults range from 40 to zero percent for normal and soft rock terrain, and from 50 to five percent for hard rock terrain; and the HAI defaults range from 50 to five percent for aerial copper feeder and 35 to five percent for fiber feeder. For buried plant, in the lowest to the highest density zones, the BCPM defaults range from 50 to zero percent for normal and soft rock terrain, and from 45 to zero percent for hard rock terrain; and the HAI defaults range from 45 to five percent for buried copper feeder and 60 to five percent for fiber feeder.  "e0 0*&&``~"Ԍ X4x121. ` ` We tentatively select input values for feeder plant mix, set forth in Appendix A, that generally reflect the assumptions underlying the BCPM and HAI default plant mix percentages, with certain modifications. We tentatively propose input values, for the lowest to the highest density zones, that range from five percent to 95 percent for underground plant; 50 to zero percent for buried plant; and 45 to five percent for aerial plant. Based on the  X4Commission staff's preliminary review of the structure and cable survey data,D6 {O'ԍ The survey is described above. See supra para. 53; see also App. C. Ā the proposed values, unlike the HAI and the BCPM (for normal and soft rock) default values, assume that there is no buried plant in the highest density zone. In contrast to the BCPM defaults, the proposed values assume there is some aerial plant in the three highest density zones. We tentatively find that it is reasonable to assume that there is some aerial feeder plant in all density zones, as HAI does, particularly in light of our assumption that there is no buried feeder in the highest density zone, where aerial placement would be the only alternative to underground plant. Although the HAI sponsors have proposed plant mix values that vary between copper feeder and fiber feeder, they have offered no convincing rationale for doing so. We tentatively conclude that, like the BCPM defaults, our proposed plant mix ratios should not vary between copper feeder and fiber feeder. We seek comment on our tentative conclusions.  Xb4y122. ` ` Alternatives to Nationwide Plant Mix Values. In the 1997 Further Notice, the Commission tentatively concluded that plant mix ratios should vary with terrain as well as density zones. Because the synthesis model does not provide separate plant mix tables for different terrain conditions, the proposed nationwide plant mix values do not vary by terrain. One method of varying plant mix by terrain would be to add separate plant mix tables, as there are in BCPM, to the synthesis model. We observe that, while the BCPM model provides separate plant mix tables, the BCPM default values reflect only slightly more aerial and less buried plant in hard rock terrain than in normal and soft rock terrain. Another method of varying plant mix would be to use company specific or state specific input values  X4for plant mix as advocated by the BCPM sponsors and other LECs.ZD6 yO'ԍ As noted above, although the BCPM sponsors have provided nationwide default values, they generally  yOh'advocate company specific input values.  Xg4z123. ` ` We generally have chosen not to use study area specific input values in the federal mechanism, and recognize that historical plant mix ratios may not reflect an efficient carrier's plant type choice today. On the other hand, historical plant mix also may reflect terrain conditions that will not change over time. For example, because it is costly to bury cable in hard rock, a carrier serving a very rocky area would tend to use more aerial than buried plant. The Commission staff's analysis of current ARMIS data reveals a great deal of variability in plant mix ratios among the states. In certain state proceedings, U S West has  X 4proposed an algorithm for adjusting plant mix to reflect its actual sheath miles as reported in" 10*&&``b" ARMIS. We seek comment on a modified version of this algorithm as an alternative method of determining plant mix percentages.  X4{124. ` ` The proposed algorithm uses ARMIS 4308 data on buried and aerial sheath distances and trench distances to allocate model determined structure distance between aerial,  X4buried, and underground structures. D6 yO'ԍ Structure distance, also known as route distance, measures the distance of the pole line or the trench. Sheath distance measures cable distance. If there is only one cable along a particular route then structure distance and sheath distance are equal. When, however, there is more than one cable along a route, sheath distance will be a multiple of the structure distance. The first step is to set the underground structure distance equal to the ARMIS trench distance and to allocate that distance among the density zones on the basis of the nationwide plant mix defaults. Then an initial estimate of aerial plant is calculated as the sum of the synthesis model structure distances by density zone multiplied by the nationwide aerial plant mix defaults. A second estimate of aerial plant is calculated by multiplying structure distance less trench miles by the aerial percentage of total ARMIS sheath miles. Then an adjustment ratio is calculated by dividing the second estimate by the initial estimate. This adjustment ratio is then applied to each density zone to adjust the nationwide default so that the final synthesis model plant mix reflects the study area specific plant mix. The buried plant mix percentage is determined as a residual equal to one minus sum of the underground and aerial percentages. We seek comment on this alternative to nationwide plant mix values. We also invite parties to suggest other alternatives to determine plant mix in the synthesis model.  XK4|125. ` ` We also seek comment on whether we should allow the synthesis model to choose the plant mix on the basis of minimum annual cost. We note that this optimization would be constrained to reflect the embedded underground plant percentage, because underground plant is typically deployed in relatively dense areas for reasons of public safety. Embedded percentages of aerial and buried plant, on the other hand, may reflect zoning ordinances but we note that these ordinances in turn may reflect purely aesthetic concerns rather than public safety. If we were to determine that we should use study area specific plant mix input values, we seek comment on whether the synthesis model should be permitted to use its optimization feature for percentages of aerial and buried plant.  Xe' D.Structure Sharing   X7' 1.` ` Background   X 4}126. ` ` Outside plant structures are generally shared by LECs, cable operators, electric utilities, and others, including competitive access providers and interexchange carriers. To the extent that several utilities may place cables in common trenches, or on common poles, it is appropriate to share the costs of these structures among the various users and assign a portion" 20*&&``;" of the cost of these structures to the LEC.  X4~127. ` ` In the 1997 Further Notice, the Commission tentatively concluded that 100 percent of the cost of cable buried with a cable plow should be assigned to the telephone  X4company.D6 {O'#X\  P6G;ɒP#э 1997 Further Notice, 12 FCC Rcd at 18547, para. 80. In addition, the Commission also tentatively concluded that Sprint's suggested value of 66 percent is an acceptable aggregate default input value for the percent of costs  Xx4assigned to the LEC for all other shared facilities.xZD6 {O '#X\  P6G;ɒP#э 1997 Further Notice, 12 FCC Rcd at 18548, para. 81. Several commenters disagreed with these  Xa4tentative conclusions.aD6 {O 'ԍ See AT&T/MCI Sept. 24 comments at 1213; Florida PSC Sept. 24 comments at 67; GTE Sept. 24 comments at 9. The Commission also sought comment on AT&T's contention that changes to the regulatory climate will increase the extent to which carriers are required or  X34willing to share structures.c3FD6 {O*'ԍ 1997 Further Notice, 12 FCC Rcd at 18548, para. 82.c  X 4128. ` ` Several comments relating to structure sharing values were filed in response to  X 4the 1997 Further Notice. D6 {Ow'ԍ See, e.g., AT&T/MCI Sept. 24 comments; Bell Atlantic Sept. 24 comments; GTE Sept. 24 comments. Both the BCPM and HAI models vary the percentage of costs they assume will be shared depending on the type of structure (aerial, buried, or underground) and line density. The model proponents differ significantly, however, on their assumptions as  X 4to the extent of sharing and, therefore, assignment of structure cost to the LEC." j D6 {O'ԍ See HAI Dec. 11, 1997 submission, App. B at 57; BCPM Jan. 31, 1997 submission, Att. 9. The BCPM sponsors assume that an efficient telephone company will benefit only marginally from sharing. The HAI sponsors assume that utilities will engage in substantial sharing with telephone companies, and generally assigns between 25% and 50% of the cost of shared facilities to the LEC.  X4   X}' 2.` ` Issues for Comment   XO4129. ` ` We tentatively adopt the following structure sharing percentages that represent the percentage of structure costs to be assigned to the LEC. For aerial structure, we tentatively assign 50 percent of structure cost in density zones 16 and 35 percent of the costs in density zones 79 to the LEC. For underground and buried structure, we tentatively assign 90 percent of the cost in density zones 12, 85 percent of the cost in density zone 3, 65 percent of the cost in density zones 46, and 55 percent of the cost in density zones 79 to the  X4LEC.T D6 {O%'ԍ See Appendix A for a complete list of the input values that we tentatively adopt in this Further Notice."30*&&``"Ԍ X4ԙ130. ` ` We believe that the structure sharing percentages that we tentatively adopt reflect a reasonable percentage of the structure costs that should be assigned to the LEC. We note that our tentative conclusions reflect the general consensus among commenters that structure sharing varies by structure type and density. While disagreeing on the extent of sharing, the majority of commenters agree that sharing occurs most frequently with aerial  X4structure and in higher density zones.D6 {O'ԍ See, e.g., HAI Dec. 11, 1997 submission, App. B at 57; BCPM Jan. 31, 1997 submission, Att. 9; Montana State Cost Study at 4647. For example, no commenter attributes more than 50 percent of the cost of aerial structure to the LEC. The sharing values that we tentatively adopt reflect these guidelines. In addition, we note that the Washington Utilities and Transportation Commission has adopted structure sharing values that are similar to those that  X14we tentatively adopt.o1"D6 {O 'ԍ See Washington USF Proceeding, Docket No. UT980311(a), App. D.o We also note that the sharing values that we tentatively adopt fall within the range of values proposed by HAI and BCPM.  X 4131. ` ` In addition, we agree with the Nebraska Public Service Commission that there  X 4are some opportunities for sharing even in the lowest density zones. D6 yO:'ԍ Letter from Frank E. Landis, Nebraska Public Service Commission, to Magalie Roman Salas, FCC, dated May 22, 1998 (Nebraska State Cost Study) at 5. As noted by the Nebraska Commission, "[e]ven in these more remote regions of the state, there will be some  X 4opportunities for sharing as new homes and businesses are constructed."I D6 yOd'ԍ Nebraska State Cost Study at 5.I We therefore do not assign 100 percent of the cost of buried or underground structure to the LEC in the lowest density areas, as suggested by the BCPM proponents.  XK4132. ` ` We seek comment on the tentative conclusions set forth in this section. In addition, we seek comment on AT&T's contention that the structure sharing percentages  X4should reflect the potential for sharing, rather than the LEC's embedded sharing practice.D6 yOj'ԍ AT&T Sept. 24 comments at 1213. For example, AT&T contends changes in the regulatory climate have increased the extent to which carriers are required or are willing to share structures.  X4 E.Serving Area Interfaces    X' 1.` ` Background   X4  X4133. ` ` A serving area interface (SAI) is a centrally located piece of network equipment that acts as a physical interface between a copper feeder cable connecting a wire"|4 0*&&``"  X4center and neighborhood distribution copper cables.4XD6 yOy'ԍ Generally, when a neighborhood is located near a wire center, copper feeder cable, using analog transmission, is deployed to connect the wire center to the SAI. From the SAI, copper cables of varying gauge extend to all of the customer premises in the neighborhood.4 The model includes copper cable and  X4SAI investment only when there are no DLC or T1 terminals.D6 yO'ԍ Both indoor and outdoor SAI investments are a function of the total number of pairs, both feeder and distribution, that the SAI terminates. The model input table lists  X4prices for indoor and outdoor SAIs of various sizes.X@D6 yO 'ԍ The current version of the model supports eighteen SAI sizes. SAI capacities currently supported are 7200, 5400, 4200, 3600, 3000, 2400, 2100, 1800, 1200, 900, 600, 400, 300, 200, 100, 50, 25 and 1 line facilities. An indoor SAI is generally used in multiunit buildings housing business establishments or residential accommodations. The construction of an outdoor SAI involves the additional cost of metal cabinets for housing protection and connection materials. Thus, the cost of constructing an outdoor SAI tends to be somewhat higher than the cost of constructing an indoor SAI. Consequently, an outdoor SAI is generally used only when there is no place to house an indoor SAI.  X14134. ` ` Both the sponsors of BCPM and HAI have submitted default input values for indoor and outdoor SAI costs. In addition, Sprint submitted cost estimates for a 7200 pair  X 4indoor SAI. ` D6 {O'ԍ Indoor SAI Cost Analysis, submitted by Sprint Local Telecommunications Division, July 30, 1998. Because the cost of a SAI depends on the cost of its components, we tentatively conclude that, in the absence of contract data between the LECs and suppliers, it is necessary to evaluate the cost of these components. Our analysis therefore begins with a review of the data and justifications submitted by the HAI sponsors and Sprint regarding the  X 4cost of the components that comprise a 7200 pair indoor SAI. D6 yOJ'ԍ We note that the BCPM defaults do not specify estimates for the cost of SAI components.  Xy4135. ` ` On or around November 25, 1998, Commission staff posted preliminary ranges of SAI input values on the Commission's Web site to elicit comment and empirical data from  XK4interested parties on the cost of SAIs.=K D6 {O~'ԍ Workshop Public Notice at 2. Commission staff used BCPM default inputs as the low end of the ranges for both indoor and outdoor SAIs, and Sprint's cost estimates as the high end of the range for indoor SAIs. The high end of the range for outdoor SAIs represented staff's analysis of stateapproved SAI parameters. Staff's preliminary ranges for SAI costs did not include HAI inputs because staff concluded that HAI had not included all of the materials and splicing required to install this equipment. = Commission staff also conducted a workshop on"K540*&&``*"  X4December 11, 1998, to discuss the posted preliminary inputs.(D6 {Oy'ԍ See Common Carrier Bureau Releases Preliminary Common Input Values to Facilitate Selection of Final  {OC'Input Values for the ForwardLooking Cost Model for Universal Service, Public Notice, CC Docket Nos. 9645,  {O '97160, DA 99295 (rel. Feb. 5, 1999) (Preliminary Input Values Public Notice); Workshop Public Notice. See  {O'also, Preliminary Input Values Handouts, dated December 11, 1998.  X' 2.` ` Issues for Comment   X'` ` a. Cost of a 7200 Pair SAI  Xv4136. ` ` Our proposed approach takes into account the cost of the following SAI components for a 7200 pair indoor SAI: building entrance splicing and distribution splicing; protectors; tie cables; placement of feeder blocks; placement of crossconnect jumpers/punch down; and placement of distribution blocks. Of these, we tentatively conclude that protector and splicing costs are the main drivers of SAI costs, and crossconnect costs and feeder block and distribution block installation costs greatly contribute to the difference in Sprint's and the  X 4HAI proponents' indoor SAI costs. D6 {OU'ԍ  See Appendix D, section IV for a breakdown of costs for each component calculated to derive the proposed cost of a 7200 pair DLC. Based upon the following analysis of the record  X 4regarding these costs, we propose a total cost of $21,708 for the 7200 pair indoor SAI. D6 yO'ԍXThe following is a breakdown of costs for a 7200 SAI size:(#  yO('Building Entrance Splicing hhCqpp  *$1,014.00  yO'Protectors  hhCqpp  *$12,520.00  yO'Placement of Feeder Blocks hhCqpp  *$930.00  yO'Placement of Cross Connects hhCqpp  *$4,067.00  yOH'Placement of Distribution Blocks hhCqpp  *$2,363.00  yO'Placement of 41 each 100 Pair Distribution Tie Cables pp  *$187.00  yO'Distribution Splicing hhCqpp  *$627.00  yO'` `  hhCqpp  *______xxX  yOh'Total` `  hhCqpp  *$21,708.00  We seek comment on this tentative analysis.  X4137. ` ` Protector Costs. The cost of the protector is the single greatest contributor to the difference in Sprint's and HAI's indoor SAI costs. HAI proposes a cost of $2.00 per pair for protector material, and Sprint initially proposed a $6.62 cost per pair for protector material. In its review of Sprint's proposed cost, staff concluded that all of the parts identified in Sprint's proposal may not be necessary for SAI construction. Staff also believed, however, that HAI's proposal was for less than a fully functional SAI, and found HAI's  X4proposed cost to be too low. Having analyzed the ex parte submissions, staff proposed a cost"6r0*&&``s"  X4of $4.00 per pair for protector material.eD6 yOy'ԍ Preliminary Input Values Handouts, dated December 11, 1998.e In its February 4, 1999, ex parte submission, Sprint agreed that $4.00 is a reasonable estimate of the cost. We tentatively adopt this  X4proposed value and seek comment.XD6 yO'ԍ Appendix D shows how we use this value to estimate the total cost of protectors in a 7200 pair indoor  {O'SAI. See Appendix D, section IV.  X4138. ` ` Splicing and Labor Rates. HAI and Sprint propose different splicing rates, but do not dispute splice setup time. The HAI sponsors propose a splicing rate of 300 pairs per  Xx4hour, while Sprint argues for a splicing rate of 100 pairs per hour."xD6 {O 'ԍ See Letter from Chris Frentrup, MCI WorldCom, to Magalie Roman Salas, FCC, dated January 21, 1999. On January 20, 1999, the sponsors of HAI provided a demonstration of splicing, in support of their splicing rate. Letter from Kenneth Cartmell, U S West, dated February 8, 1999, to Magalie Roman Salas, FCC; Letter from Pete Sywenki, Sprint, to Magalie Roman Salas, FCC, dated February 4, 1999.  We believe that HAI's proposed rate is a reasonable splicing rate under optimal conditions, and therefore, we tentatively conclude that Sprint's proposed rate is too low. We note that the HAI sponsors have submitted a letter from AMP Corporation, a leading manufacturer of wire connectors, in  X 4support of the HAI rate. D6 {Oi'ԍ See attachment to letter from Chris Frentrup, Senior Economist, MCI WorldCom, to Magalie Roman Salas, FCC, dated January 21, 1999. We recognize, however, that splicing under average conditions does not always offer the same achievable level of productivity as suggested by the HAI sponsors. For example, splicing is not typically accomplished under controlled lighting or on a worktable. Having accounted for such variables, we propose to adjust the splicing rate to 250 pairs per hour. We also propose a $60.00 per hour labor rate for splicing, which is  X 4within the range of filings on the record.{ D6 yOP'ԍ The $60.00 per hour rate is the prevalent labor rate for mechanical apprentices. { We seek comment on these proposed values. D6 yO'ԍ Appendix D, shows how the proposed splicing and labor rates are used in calculating the cost of a 7200  {O'pair indoor SAI. See Appendix D, section IV.  X{4139. ` ` CrossConnect Costs. The crossconnect is the physical wire in the SAI that connects the feeder and distribution cable. Sprint asserts that the "jumper" method generally will be employed to crossconnect in a SAI. In contrast, HAI suggests that the "punch down" method is generally used to crossconnect. We tentatively conclude that neither the jumper method nor the punch down method is used exclusively in SAIs. In buildings with high churn rates, such as commercial buildings, carriers may be more likely to use the jumper method. On the other hand, in residential buildings, where changes in service are less likely, carriers may be more likely to use the less expensive punch down method. Based on the record, it appears that both methods are commonly used, and that neither is used substantially"70*&&``X" more than the other. Therefore, we tentatively conclude that we should assume that each  X4method will be used half the time. We seek comment on this tentative conclusion.D6 yOb'ԍ Appendix D shows how this tentative conclusion is used to determine proposed costs for a 7200 pair  {O*'SAI. See Appendix D, section IV. In particular, we invite parties to justify a particular allocation between the jumper and punch down methods.  X4140. ` ` Feeder Block and Distribution Block Installation Rates. Sprint proposes an installation rate of 60 pairs per hour, while the HAI sponsors propose 400 pairs per hour. Because neither feeder block installation nor distribution block installation is a complicated procedure, we tentatively conclude that Sprint's rate of 60 pairs per hour is too low. We recognize, however, that installation conditions are not always ideal. Like splicing, feeder block and distribution block installations are not typically accomplished under controlled lighting or on a worktable. Having accounted for such variables, we propose a rate of 200  X 4pairs per hour. We seek comment on this proposed value. "D6 yO'ԍ Appendix D show how this proposed value is used in the calculation of a 7200 pair SAI.  X ' ` ` b. Cost of Other SAI Sizes  X4141. ` ` Because we currently do not have similar componentbycomponent data for other SAI sizes, we propose to determine the costs of the other SAI sizes by extrapolating from the cost of the 7200 pair indoor SAI. We believe that this is a reasonable approach because there is a linear relationship between splicing and protection costs, which are the main drivers of cost, and the number of pairs in the SAI. We look to the HAI data to determine the relationship in cost among the various sizes of SAI. Specifically, we develop a ratio of our proposed cost for a 7200 pair indoor SAI to the cost proposed by HAI. We then propose to apply this ratio, 2.25, to the values submitted by the HAI sponsors for other sizes of indoor and outdoor SAIs. Applying this factor, we tentatively adopt the cost estimates for indoor and outdoor SAIs contained in Appendix A. We propose to use the HAI, rather than BCPM data, in this manner because BCPM has not submitted estimates for all of the SAI sizes used in the model. We note that using the BCPM data in this way would result in roughly the same estimates. We seek comment on these tentative conclusions and proposed values.  XN4  X7' F.Digital Loop Carriers  X 4  X ' 1.` ` Background   X4142. ` ` A digital loop carrier (DLC) is a piece of network equipment that converts a digital signal carried on optical fiber cable to an analog, electrical signal that is carried on" 80*&&``b"  X4copper cable and is compatible with customers' telephones.D6 yOy'ԍ Optical fiber cable carries a digital signal that is incompatible with most customers' telephone equipment, but the quality of the signal degrades less with distance compared to a signal carried on copper wire. Generally, when a neighborhood is located too far from the wire center to be served by copper cables alone, an optical fiber cable will be deployed to a point within the neighborhood, where a DLC will be placed to convert incoming digital signals to analog signals and outgoing analog signals to digital. From the DLC, copper cables of varying gauge extend to all of the customer premises in the neighborhood. Because of the high cost of DLCs, a single DLC is shared among a number of customers. The model uses fiber cable and DLCs whenever it calculates that this configuration is cheaper than using copper cable or when the distance exceeds the maximum copper loop length. When using DLCs, the model determines the size and number of DLCs that should be installed at a location, based on cost minimization and engineering constraints. In designing outside plant, the model uses five  Xv4different sizes of DLCs.v@D6 yOg 'ԍ The current version of the model supports a fifth DLC size in addition to those already supported. DLC  {O/'capacities currently supported are 2016, 1344, 672, 96, and 24 line facilities.   In order to run the model, a user must input the fixed and perline cost for each of these DLC sizes. The total cost of a particular DLC is determined by multiplying the number of lines connected to the DLC times the perline cost of the DLCs, and then adding the fixed cost of the DLC.  X ' 2.` ` Issues for Comment   X 4143. ` ` Both the sponsors of BCPM and HAI have submitted default values for DLC costs. Because these values are based on the opinions of experts without data to enable us to substantiate these opinions, however, we tentatively conclude that we should not rely on these data. We also tentatively conclude that the most reliable data on DLC costs available to the Commission at this time are the contract data submitted to the Commission in response to the  Xb41997 Data Request, and in ex parte submissions following the December 11, 1998 workshop.  XM4We seek comment on these tentative conclusions.  X64  X4144. ` ` Following their submission of DLC data to the Commission in response to the  X41997 Data Request, US West, Bell South, and ATU resubmitted their data on the record in  X4this proceeding.D6 {O>'č In response to the 1997 Data Request, Ameritech, Bell Atlantic (including NYNEX), Bell South, SBC, US West, GTE, Sprint, ATU, and PRTC originally submitted data to the Commission on DLC costs in 1997.  {O 'Bell South and US West resubmitted their data on the record of this proceeding subject to the Protective Order. Letter from William W. Jordan, Bell South, to Magalie Roman Salas, FCC, dated March 15, 1999; Letter from Robert B. McKenna, US West, to Magalie Roman Salas, FCC, dated March 8, 1999. Letter from Alane C.  {O*#'Weixel, counsel for ATU, to Magalie Roman Salas, FCC, dated May 6, 1999 (ATU May 6, 1999 ex parte). At the December 11, 1998 workshop, staff of the Common Carrier Bureau"90*&&``)"  X4discussed the DLC costs data on the record in this proceeding.D6 {Oy'ԍ See Material Cost Workshop Digital Loop Carrier Equipment "Template for Determining DLC Cost"  {OC'handout.  In an effort to elicit further discussion of DLC input values, staff presented a template of the components of a typical  X4DLC. The HAI sponsors, GTE, and Aliant submitted data using the template of DLC costs.$D6 yO'ԍ Letter from W. Scott Randolph, GTE, to Magalie Roman Salas, FCC, dated February 11, 1998; Letter from Robert A. Mazer & Albert Shuldiner, Counsel for Aliant, to Magalie Roman Salas, FCC, dated February 8,  {O7'1998. The HAI sponsors filed an ex parte letter with DLC cost information using a different template. Letter from Chris Frentrup, MCI WorldCom, to Magalie Roman Salas, FCC, Docket No. 9645, 97160, dated January  yO '21, 1999 (following their January 20, 1999 presentation). US West and Sprint filed responses to the HAI presentation on February 8, 1999 and February 4, 1999, respectively.  Staff found the data submitted by the HAI sponsors to be significantly lower than the contract data on the record, and staff concluded that it would be inappropriate to use it, especially as no support was provided in justification. Because the data submitted by the companies are based on actual costs incurred in purchasing DLCs, we tentatively conclude that they are more reliable than the opinions proffered, and, therefore, should be used to estimate the cost of DLCs. Although we would prefer to have a larger sampling of data, we note that the data represent the costs incurred by several of the largest nonrural carriers, as well as two of the smallest nonrural carriers. We also note that, throughout this proceeding, the Commission  X 4has repeatedly requested cost data on DLCs.^ D6 {OR'ԍ In addition to the data submitted in response to the 1997 Data Request, and following the December 11,  {O'1998 workshop, the Bureau requested further data on DLC costs in the 1997 Further Notice and in the Inputs  {O'Public Notice. See also Preliminary Input Values Public Notice. We believe that we are using the best data  X 4available on the record to determine the cost of DLCs. & D6 {Oa'ԍ Only US West, Bell South, and ATU presented their contract data from the 1997 Data Request in a  {O+'format that staff could use. Some of the data and comments that were submitted in response to the 1997 Data  {O'Request, but not refiled on the record under the Protective Order, could not be used because the data were either inadequate or presented in a format from which staff could not extract relevant information.   X 4  X 4145. ` ` We note that ATU asserts that material handling and shipping costs should be added to the DLC prices reflected in the contract it submitted. ATU suggests that these costs  X4could represent up to 10 percent of the material cost of a DLC..ZD6 {O'ԍATU May 6, 1999 ex parte. ATU also suggests that costs for placement, installation, and testing should be added to the DLC material costs it submitted. We note that these site preparation costs are already separately accounted for in the model.. It is unclear whether the DLC data submitted by other parties include these costs. We seek comment on the extent, if any, to which we should increase our proposed estimates for DLCs to reflect material handling and shipping costs. "4:0*&&``"Ԍ X4146. ` ` We recognize that the cost of purchasing and installing a DLC changes over time. Such changes occur because of improvements in the methods and components used to produce DLCs, changes in both capital and labor costs, and changes in the functionality requirements of DLCs. Thus, we believe it is appropriate to adjust the contract data to reflect 1999 prices. In order to capture changes in the cost of purchasing and installing DLCs over time, we propose a 2.6 percent annual reduction in both fixed DLC cost and per line DLC cost. This proposed rate is based on the change in cost calculated for electronic digital switches over a four year period. We believe that the change in the cost of these switches over time is a reasonable proxy for changes in DLC cost, because they are both types of digital telecommunications equipment. We also note that the 2.6 percent figure is a conservative estimate, based on the change in cost of remote switches. Our analysis suggests that the change in cost of host switches over the past four years is much higher. Finally, we  X 4note that use of the current consumer price index results in a similar figure over four years. J D6 {Oe 'ԍ See infra para. 226.J The indexed amount is based on the effective date of the contracts. Based upon an average of the contract data submitted on the record, adjusted for cost changes over time, we tentatively  X 4adopt the cost estimates for DLCs contained in Appendix A. We seek comment on this proposed analysis and the proposed values.  Xb' VI. SWITCHING AND INTEROFFICE FACILITIES ׃  X4' A. Background   X4147. ` ` The central office switch provides the connection between a subscriber's local loop and the outside world. Modern digital switches provide voice, data, and video signals  X4connecting telephones, fax machines, and computers on the public switched network.k XZD6 yO'ԍ The functions performed by the switch for local service include: line termination; line monitoring; usage call processing, routing, and completion; interconnection to other carriers; billing and maintenance; and vertical services and features. We note that not all of these functions are supported by universal service.k In order to accomplish this, a telephone network must connect customer premises to a switching facility, ensure that adequate capacity exists in that switching facility to process all calls, and interconnect the switching facility with other switching facilities to route calls to their  X|4destination. A wire center is the location of the switching facility. |zD6 yO 'ԍ The wire center boundaries define the area in which all customers are connected to a given wire center. The infrastructure to  Xe4interconnect the wire centers is known as the "interoffice" network, and the carriage of traffic  XN4among wire centers is known as "transport."  X 4148. ` ` In the Universal Service Order, the Commission stated that "[a]ny network function or element, such as . . . switching, transport or signaling, necessary to provide" ;  0*&&``"  X4supported services must have an associated cost."w D6 {Oy'ԍ Universal Service Order, 12 FCC Rcd at 8913, para. 250 (criterion two).w In the 1997 Further Notice, the Commission sought comment on issues that affect the input values relating to the forward X4looking economic cost of switching and interoffice transport.k ZD6 {O'ԍ 1997 Further Notice, 12 FCC Rcd at 1856066, paras. 12138.k The Switching and Transport  X4Public Notice established several guidelines relating to switching, the design of the interoffice  X4network, and interoffice cost attributable to providing supported services.< |D6 {OG 'ԍ Switching and Transport Public Notice at 26. The Bureau guidelines established that: (1) the models permit individual switches to be identified as host, remote, or standalone; (2) switching investment costs should be separately estimated for host, remote, and standalone switches; (3) models should include switch capacity constraints; (4) all of the lineside port costs and a percentage of usage costs should be assigned to the cost of providing the supported service; and (5) models should accommodate an interoffice network that is capable of connecting switches designated as hosts and remotes in a way that is compatible with capabilities of equipment  {O 'and technology that are available today and current engineering practices. Id.< In the Platform  X4Order, the Commission concluded that the federal mechanism should incorporate, with certain  X4modifications, the HAI 5.0a switching and interoffice facilities module.^0 D6 {Oa'ԍ Platform Order, 13 FCC Rcd at 21354, para. 75.^  XR4149. ` ` Both HAI and BCPM have provided default input values for estimating the  X;4forwardlooking economic cost of switching and interoffice network.; D6 {O'ԍ See Letter from Richard N. Clarke, AT&T, to Magalie Roman Salas, FCC, dated February 3, 1998 (HAI Feb. 3 submission) App. B; BCPM April 30, 1998 submission, Switch Model Inputs. On November 25,  X$ 41998, the Bureau announced the release of preliminary input values on its Web site.P$ D6 {O'ԍ See Workshop Public Notice at 2.P On December 1, 1998, the Bureau held a public workshop designed to elicit comment on the  X 4switching inputs values to be used in the federal mechanism. D6 {OU'ԍ See Workshop Public Notice. The December 1, 1998 workshop addressed issues relating to switching and expenses. On February 5, 1999, the Bureau released a set of revised preliminary input values that included switching and  X 4interoffice transport values to facilitate the review and selection of final input values.d D6 {O 'ԍ Preliminary Input Values Public Notice. d  X4150. ` ` In this section, we tentatively adopt and seek comment on the inputs associated with the installation and purchase of new switches, the use of the LERG to identify hostremote switch relationships, and other switching and interoffice input issues that have been raised by commenters. The remaining switching and interofficerelated input values that we"U<0*&&``"  X4tentatively adopt in this Further Notice are provided in Appendix A.D6 yOy'ԍ These inputs values are generally agreed upon by the parties or have not been disputed.  X' B. Issues for Comment   X' 1. ` ` Switch Costs   Xv4151. ` ` We now examine the inputs associated with the purchase and installation of new switches. Specifically, we must select values for the fixed and perline cost of host and remote switches, respectively.  X 4152. ` ` Switch Cost Data. Both the sponsors of BCPM and HAI have submitted default values for switch costs. To a large extent, however, these values are based on nonpublic information or opinions of their experts, but without data that enable us adequately to substantiate those opinions. Consistent with the recommendation of the Joint Board and  X 4criterion eight in the Universal Service Order, we tentatively conclude that we should not rely on these submissions because the underlying data are not sufficiently open and available to the public. We also tentatively conclude that it is not necessary to rely on this information,  X{4because the Commission, in conjunction with the work of Gabel and Kennedy,{XD6 {O'ԍ David Gabel and Scott Kennedy, Estimating the Cost of Switching and Cables Based on Publicly  {ON'Available Data, The National Regulatory Research Institute, NRRI 9809, April 1998 (NRRI Study). the Bureau of Economic Analysis (BEA) of the Department of Commerce, and the U.S. Department of Agriculture Rural Utility Service (RUS), has compiled publicly available data on the cost of purchasing and installing switches. This information was gathered from depreciation reports filed by LECs at the Commission and from reports made by LECs to RUS.  X4153. ` ` The depreciation data contains, for each switch reported: the model designation of the switch; the year the switch was first installed; and the lines of capacity and bookvalue cost of purchasing and installing each switch at the time the depreciation report was filed with  X4the Commission. D6 yO'ԍ Until 1996 large incumbent LECs were required to file depreciation rate reports with the Commission pursuant to 47 C.F.R.  43.43. Prior to filing these reports, companies generally submit depreciation rate studies that include data for each digital switch in operation. See Appendix E of this Further Notice for a description of the data set and an explanation of adjustments made to the data. The RUS data contains, for each switch reported: the switch type (i.e.,  X4host or remote); the number of equipped lines; cost at installation; and year of installation."XD6 yO"'ԍ Many small telephone companies receive financial assistance from RUS, which requires these companies to report the payments made for new switches. See Appendix E for a description of the RUS data and an explanation of adjustments made to the data." "~= 0*&&``"Ԍ X4154. ` ` The sample that we propose to use to estimate switch costs includes 1,060 observations. The sample contains 921 observations selected from the depreciation data, which provide information on the costs of purchasing and installing switches gathered from 20 states. The sample also contains 139 observations selected from the RUS data, which provide information from across the nation on the costs of small switches purchased and installed by rural carriers. The combined sample represents purchases of both host and remote switches, with information on 468 host switches and 592 remote switches, and covers switches installed between 1989 and 1996. This set of data represents the most complete public information available to the Commission on the costs of purchasing and installing new switches.  X 4155. ` ` In response to the 1997 Data Request, the Commission received a second set of information pertaining to 1,486 switches. Upon analysis, however, Commission staff identified one or more problems with most of the data submitted: missing switch costs; zero or negative installation costs; zero or blank line counts; unidentifiable switches; or missing or inconsistent Common Language Local Identification (CLLI) codes. After excluding these corrupted observations, 302 observations remained. The remaining observations represented switches purchased by only four companies. We tentatively conclude that the data set we  X{4propose to use is superior to the data set obtained in response to the 1997 Data Request, both in terms of the number of usable observations and the number of companies represented in the data set. We seek comment on this tentative conclusion.  X!4156. ` ` Following the December 1, 1999 workshop, three companies voluntarily  X 4submitted further data regarding the cost of purchasing and installing switches. D6 {O'ԍ BellSouth January 29, 1999 ex parte, Sprint February 5, 1999 ex parte, and GTE February 22, 1999 ex  {OM'parte. Because these submissions were received late in the process, Commission staff has not had sufficient time to analyze the quality and content of the information. We seek comment on the use of  X4this data set as a substitute or complement to the data set we propose.   *  X4157. ` ` Adjustments to the Data. The cost figures reported in the depreciation information reflect the costs of purchasing and installing new switches. While the RUS cost data also contain information on purchasing and installing new switches, they do not include: (1) the cost associated with purchasing and installing the main distribution frame (MDF); (2) the cost associated with purchasing and installing power equipment; (3) the cost of connecting  X$4each remote switch to its respective host switch; and (4) LEC engineering costs."&$$D6 yO!'ԍ Letter from W. Scott Randolph, GTE, to Magalie Roman Salas, FCC dated December 18, 1998 (GTE  {O"'Dec. 18 ex parte) at 5 and 6; #C\  P6QɒP#NRRI Study at 97 and 102; Letter from Pete Sywenki, Sprint, to Magalie Roman  {O#'Salas, FCC, dated December 22, 1998 (Sprint Dec. 22 ex parte) at 1321; Letter from Richard Clarke, AT&T, to  {OU$'Magalie Roman Salas, FCC, dated January 7, 1999 (AT&T Jan. 7 ex parte) at 1. " In order to make the depreciation and RUS information comparable, we propose to add estimates of these" >0*&&``" four components to the switch costs reported in the RUS information. These additions are discussed below. We seek comment on this proposed approach.  X4158. ` ` In order to account for the cost of MDF equipment omitted from the RUS information, AT&T recommends using the HAI 5.0a default value of $12 per line for MDF. We tentatively conclude that $12 per line is a reasonable cost for purchasing and installing  Xv4MDF equipment.KvD6 {O'ԍ AT&T Jan. 7 ex parte at 1. K No party contests this value. We seek comment on this tentative conclusion and invite commenters to submit alternative values.  X14159. ` ` In order to account for the cost of central office power equipment omitted from the RUS information, AT&T recommends using the HAI 5.0a default values for these inputs.  X 4We tentatively use the following input values for power equipment: $12,000 for switches with 0999 lines; $40,000 for switches with 1,0004,999 lines; and $74,500 for switches with 5,00025,000 lines. These values are derived from a range of values on the record in this  X 4proceeding, including state cost studies.^ ZD6 {O'ԍ See, e.g., Commonwealth of Kentucky, An Inquiry Into Universal Service Funding Issues, Administrative  {O'Case No. 360 (1998) App. F at 14 (Kentucky Cost Study); Compliance Proceeding for Implementation of the  {O]'Texas High Cost Universal Service Plan, Order No. 30, Project No. 18515 (August 24, 1998). We seek comment on the values we tentatively adopt and invite commenters to submit alternative values.  Xy4160. ` ` Gabel and Kennedy estimate that the average cost of terminating a remote on a  Xb4host switch is $27,598.@bD6 yO'ԍ NRRI Study at 102104.@ Relying on this estimate, we tentatively conclude that $27,598 should be added to the cost of each remote switch reported in the RUS data. We seek comment on this tentative conclusion and invite commenters to submit alternative values.  X4161. ` ` Gabel and Kennedy also recommend, based on a data analysis undertaken by RUS, that the cost of switches reported in the RUS data should be increased by 8 percent in  X4order to account for the cost of LEC engineering.3D6 {O'ԍ Id.3 Relying on those estimates, we tentatively conclude that 8 percent should be added to the total cost, including MDF, power, and remote connection costs, of each switch reported in the RUS data. We note that the proposed value is based on the only information on the record on this issue. We seek comment on this tentative conclusion and invite commenters to submit alternative values.  XN4162. ` ` We tentatively conclude that switch costs should be estimated based on a sample of public data that includes both RUS and depreciation data. As noted, this information represents the broadest range of data publicly available for both small and large" ?0*&&``<" switches. We seek comment on the appropriateness of merging the two data sets.  X4163. ` ` Methodology. In order to determine the reasonable forwardlooking cost of switches, based on the selected data set, we propose to employ regression analysis. In the process of estimation, we propose, where appropriate, to make adjustments to the information compiled by the above parties. These proposed modifications to the data and estimation techniques used by the Commission are discussed below.  XH4164. ` ` We tentatively conclude that the cost of a switch should be estimated as a linear function of the number of lines connected to the switch, the type of switch installed (i.e., host or remote), and the date of installation. We adopt a linear function based on examination of the data and statistical evidence. Sprint recommends using a nonlinear function, such as the loglog function, to take into account the declining marginal cost of a  X 4switch as the number of lines connected to it increases. D6 {ON'#C\  P6QɒP#э Sprint Dec. 22 ex parte at 12. Sprint criticized the Commission's preliminary switch regression presented in the December 1998 workshop based on the "Rsquared" statistical goodness of fit criterion. However, after adjusting for data transformations associated with moving to a loglog specification, the R yO'squared of a loglog regression (0.54) suggested by Sprint is lower than the Rsquared in the linear regression#XP\  P6QynXP#  yOp'#C\  P6QɒP#(0.78). Specifically, we note that the Rsquared measure resulting from a regression employing a loglog functional form is not directly comparable to the Rsquared measure from a linear regression. In order for the two measures to be comparable, the Rsquared measure computed from the loglog regression must be computed using observed and predicted cost measures, not the logs of these measures. We also note that the loglog  yO'regression we employed is of the form:#XP\  P6QynXP#  XA4 #5\  P6Q مP#Ln(Cost) = a1 + a2*Ln(Lines) + a3*Host + a4*Ln(Time) + a5*Ln(Lines)*Ln(Time) + a6*Host*Ln(Time) + e# C\  P6QɒP# where Ln(x) denotes the natural log of x. Because Sprint did not make these necessary adjustments, we believe that their criticism of the use of a linear function is misplaced. For a discussion of the "Rsquared" statistical  {O'goodness of fit criterion and a discussion of loglog specifications, see William H. Greene, Econometric Analysis, 192193 and 251 (1990). We tentatively conclude that the linear function we adopt provides a better fit with the data than the loglog function. A discussion of the effect of time and type of switch on switch cost is presented below. We  X4seek comment on these tentative conclusions.#XP\  P6QynXP#  Xb4165. ` ` Based upon an analysis of the data and the record, we tentatively conclude that the fixed cost (i.e., the base getting started cost of a switch, excluding costs associated with connecting lines to the switch) of host switches and remote switches differ, but the perline variable cost (i.e., the costs associated with connecting additional lines to the switch) of host  X4and remote switches are approximately the same. This is consistent with statistical evidenceVD6 {O $'#C\  P6QɒP#э#C\  P6QɒP# See General Wald Test for omitted variables in Ramu Ramanathan, Introductory Econometrics with  {O$'Applications 170 (1989)."@0*&&``%"  X4and the comments of the HAI sponsors.#XP\  P6QynXP#D6 yOy'#C\  P6QɒP#э Letter from Richard Clarke, AT&T, to Magalie Roman Salas, FCC, dated January 7, 1999 (AT&T Jan. 7  {OA'ex parte) at 1. "The primary difference between a host switch and remote switch is in the extent and complexity of the `getting started equipment,' associated with each type of switch (e.g., switch central processor functions, SS7 nonscaleable equipment, maintenance and testing, call recording for billing purposes, etc.). Because most of these functions for lines terminating a remote switch are performed at that switch's host, very little of this type of `getting started' equipment is required at the remote. In contrast, the scaleable equipment used to terminate lines and trunks and to perform basic call processing is essentially the same at the host and remote. In fact, the line units used by Lucent 5E Remote Switching Modules are identical to those used by 5E host or standalone switches. Similarly, the line cards used in Nortel DMS 100 host or standalone switches are the  {O 'same as those used in DMS 100 remotes, or in DMS 10 host or remote switches." Id.ķ#XP\  P6QynXP# We seek comment on this tentative conclusion.  X4166. ` ` Accounting for Changes in Cost Over Time. We recognize that the cost of purchasing and installing switching equipment changes over time. Such changes result, for example, from improvements in the methods used to produce switching equipment, changes in both capital and labor costs, and changes in the functional requirements that switches must meet for basic dial tone service. In order to capture changes in the cost of purchasing and installing switching equipment over time, we propose to modify the data to adjust for the effects of inflation, and explicitly incorporate variables in the regression analysis that capture cost changes unique to the purchase and installation of digital switches. We describe this process below.  X 4167. ` ` To the extent that the general level of prices in the economy change over time, the purchasing power of a dollar, in terms of the volume of goods and services it can purchase, will change. In order to account for such economywide inflationary effects, we propose to multiply the cost of purchasing and installing each switch in the data set by the  X4grossdomesticproduct chaintype price indexO Xd D6 yO'#C\  P6QɒP#э#C\  P6QɒP# The grossdomesticproduct chaintype price index, which tracks economywide inflation, is published monthly by the Bureau of Economic Analysis of the U.S. Department of Commerce in the Survey of Current Business. O for 1997 and then divide by the grossdomesticproduct chaintype price index for the year in which the switch was installed,  Xb4thereby converting all costs to 1997 values.!b D6 yO'ԍ Figures are adjusted after estimation for both realized and expected inflation between 1997 and 1999.  {O_'See Appendix E for an explanation of these adjustments.  X44168. ` ` In order to account for cost changes unique to switching equipment, we  X4propose to enter time terms directly into the regression equation." D6 yO#'ԍ Time was added to the regression in reciprocal form as an independent variable to measure fixed cost changes unique to remote switches. Then, a time term was added in conjunction with the host identifier variable to measure the fixed cost changes unique to host switches. A time term was also added in conjunction with the line variable, in order to measure cost changes unique to line additions on switches. GTE expresses concern"A"0*&&``" that, under certain specifications of time, the regression equation produces investments for remote switch "getting started" costs that are negative and that such specifications overstate  X4the decline in switch costs.#D6 {OK'#C\  P6QɒP#э GTE Dec. 18 ex parte at 4.#C\  P6QɒP#ј The HAI sponsors also caution that the historical large  X4percentage price declines seen in recent years may not continue.$ZD6 {O'#C\  P6QɒP#э AT&T Jan. 7 ex parte at 4.#C\  P6QɒP#ј We tentatively conclude that the reciprocal form of time in the regression equation proposed would satisfy these concerns by yielding projections of switch purchase and installation costs that are positive yet declining over time.  XH4169. ` ` Ameritech and GTE advocate the use of the Turner Price Index,%HD6 yO '#C\  P6QɒP#э The index is published semiannually by AUS consultants. #C\  P6QɒP#ѱ which is an index designed to measure the changing cost of telecommunications plant, to convert the embedded cost information contained in the depreciation data to costs measured in current  X 4dollars.& |D6 {O0'#C\  P6QɒP#э See Ameritech Dec. 16, 1998 comments at 5; #C\  P6QɒP#GTE Dec. 18, 1998 ex parte at 4.#C\  P6QɒP# We note, however, that this index and the data underlying it are not on the public record. We prefer to rely on public data when available. Moreover, we tentatively conclude it is not necessary to rely on this index to convert switch costs to current dollars. As described in the preceding paragraph, the Commission has proposed to account for costs explicitly in the estimation process, rather than adopt a surrogate such as the Turner Price Index. We seek comment on this proposed approach. In addition, we seek comment on the potential impact of increased use of packet switches, including the possibility that manufacturers will reduce the price of circuit switches to maintain market share.  X44170. ` ` Treatment of Switch Upgrades. The bookvalue costs recorded in the depreciation data include both the cost of purchasing and installing new equipment and the cost associated with installing and purchasing subsequent upgrades to the equipment over time. Upgrades costs will be a larger fraction of reported bookvalue costs in instances where the bookvalue costs of purchasing and installing switching equipment are reported well after the initial installation date of the switch. In order to estimate the costs associated with the purchase and installation of new switches, and exclude the costs associated with upgrading switches, we propose to remove from the data set those switches installed more than three years prior to the reporting of their associated bookvalue costs. We believe that this restriction would eliminate switches whose book values contain a significant amount of upgrade costs, and recognizes that, when ordering new switches, carriers typically order equipment designed to meet shortrun demand.  X 4171. ` ` We tentatively conclude that we should reject the suggestion of Ameritech," B&0*&&``" GTE, and Sprint that the costs associated with purchasing and installing switching equipment  X4upgrades should be included in our cost estimates.'D6 {Ob'#C\  P6QɒP#э Ameritech Dec. 16, 1998 comments at 45; GTE Dec. 18, 1998 ex parte at 45; Sprint Dec. 22, 1998 ex  {O,'parte at 57.#C\  P6QɒP# The model platform we adopted is intended to use the most costeffective forwardlooking technology available at a particular period of time. The installation costs of switches, as configured by us above, reflect the most costeffective forwardlooking technology for meeting industry performance requirements. Switches, augmented by upgrades, may provide carriers the ability to meet performance requirements, but do so at greater costs. Therefore, such augmented switches do not constitute costeffective forwardlooking technology. In addition, as industry performance requirements change over time, so will the costs of purchasing and installing new switches. The historical cost data employed in this proposed analysis reflect such changes over time, as do the timetrended cost estimates. We seek comment on this tentative conclusion.  X 4172. ` ` Additional Variables. Several parties contend that additional independent variables should be included in our regression equation. Some of the recommended variables include minutes of use, calls, digital line connections, vertical features, and regional, state, and  X 4vendorspecific identifiers. ( $D6 {O|'#C\  P6QɒP#э GTE Dec. 18, 1998 ex parte at 5; Sprint Dec. 22, 1998 ex parte at 13; Ameritech Dec. 16, 1998  yOF'comments at 6.#x\  P6QdP#  For the purposes of this analysis, our proposed model specification is limited to include information that is in both the RUS and depreciation data sets. Neither data set includes information on minutes of use, calls, digital line connections, vertical features, or differences between host and standalone switches. Nor do they contain detail sufficient to allow us to obtain such information from other sources. State and regional identifiers are not included in the proposed regression because we only have depreciation data on switches from 20 states. Thus, we could not accurately estimate regionwide or statewide differences in the cost of switching. Our proposed model specification also does not include vendorspecific variables or variables distinguishing host switches from standalone switches because the model platform does not distinguish between different types of switches.  X4173. ` ` Switch Cost Estimates. Using the regression analysis discussed above, we tentatively adopt the fixed cost (in 1999 dollars) of a remote switch as $186,400 and the fixed cost (in 1999 dollars) of both host and standalone switches as $447,000. We tentatively adopt the additional cost per line (in 1999 dollars) for remote, host, and standalone switches  XN4as $83.)N~D6 yO}"'ԍ See Appendix E for regression results, and an explanation of how cost estimates are derived from these results. We seek comment on these tentative conclusions.  X ' 2. ` ` Use of the Local Exchange Routing Guide (LERG)  " C)0*&&``"Ԍ X4174. ` ` We tentatively conclude that the Local Exchange Routing Guide (LERG) database should be used to determine hostremote switch relationships in the federal universal  X4service mechanism.*"D6 yOK'Ѝ The LERG is a database of switching information maintained by Bellcore that includes the existing hostremote relationships. The HAI proponents have placed on the record the portion of the LERG that identifies the hostremote relationships. Letter from Chris Frentrup, MCI, to Magalie Roman Salas, FCC, dated September 14,  {O'1998 (MCI Sept. 14 ex parte). In the 1997 Further Notice, the Commission requested "engineering and cost data to demonstrate the most costeffective deployment of switches in general and  X4hostremote switching arrangements in particular."g+D6 {O 'ԍ 1997 Further Notice, 12 FCC Rcd at 1856061, para. 122.g In the Switching and Transport Public  X4Notice, the Bureau concluded that the model should permit individual switches to be identified  X|4as host, remote, or standalone switches.,"|DD6 {Oq 'ԍ Switching and Transport Public Notice at 2. Switches can be designated as either host, remote, or standalone switches. Both a host and a standalone switch can provide a full complement of switching services without relying on another switch. A remote switch relies on a host switch to supply a complete array of switching functions and to interconnect with other switches. The Bureau noted that, although standalone switches are a standard component of networks in many areas, current deployment patterns suggest that hostremote arrangements are more costeffective than standalone switches in  X74certain cases.]-7. D6 {O'ԍ Switching and Transport Public Notice at 23.] No party has placed on the record in this proceeding an algorithm that will  X 4determine whether a wire center should house a standalone, host, or remote switch.^. D6 {O'ԍ Platform Order, 13 FCC Rcd at 21355, para. 76.^  X 4175. ` ` In the Platform Order, we concluded that the federal mechanism should incorporate, with certain modifications, the HAI 5.0a switching and interoffice facilities  X 4module.a/ R D6 {O'ԍ Platform Order, 13 FCC Rcd at 2135455, para. 75.a In its default mode, HAI assumes a blended configuration of switch technologies  X 4to develop switching cost curves.a0 D6 yOD'ԍ HAI Feb. 3, 1998 submission, Model Description at 58. a HAI also allows the user the option of designating, in an input table, specific wire center locations that house host, remote, and standalone switches. When the hostremote option is selected, switching curves that correspond to host, remote, and standalone switches are used to determine the appropriate switching investment. The LERG database could be used as a source to identify the hostremote switch relationships. In  X<4the Platform Order, we stated that "[i]n the inputs stage of this proceeding we will weigh the benefits and costs of using the LERG database to determine switch type and will consider alternative approaches by which the selected model can incorporate the efficiencies gained "Dt00*&&``"  X4through the deployment of hostremote configurations."^1D6 {Oy'ԍ Platform Order, 13 FCC Rcd at 21355, para. 76.^  X4176. ` ` The majority of commenters support the use of the LERG database as a means  X4of determining the deployment of host and remote switches.2^ZD6 {O'ԍ See, e.g., Aliant Switching and Transport Public Notice comments at 2; BellSouth et al. Switching and  {O'Transport Public Notice comments, att. 1 at 12; GTE Switching and Transport Public Notice comments at 11;  {OZ'Bell Atlantic Switching and Transport Public Notice reply comments at 2. These commenters contend that the use of the LERG to determine hostremote relationships will incorporate the accumulated knowledge and efficiencies of many LECs and engineering experts in deploying  Xv4the existing switch configurations.3vD6 {O 'ԍ Bell Atlantic Switching and Transport Public Notice reply comments, att. 1 at 2; BellSouth et al.  {Oq 'Switching and Transport Public Notice reply comments, att. 1 at 23. Commenters also contend that an algorithm that realistically predicts this deployment pattern is not feasible using publicly available data and  XH4would be "massive and complex."4HD6 {O'ԍ See, e.g., AT&T/MCI Switching and Transport Public Notice comments at 6; BellSouth et al. Switching  {O'and Transport Public Notice reply comments, att. 1 at 2. The HAI proponents argue, however, that use of the LERG to identify hostremote relationships may reflect the use of embedded technology,  X 4pricing, and engineering practices.p5Z 8 D6 yO'ԍ Letter from Richard N. Clarke, AT&T, to Magalie Roman Salas, FCC, dated January 6, 1998 at 11  {O'(AT&T Jan. 6 ex parte). For example, the HAI sponsors contend that factors influencing the placement of switches, such as line demand, switch prices, and life cycle costs, may have changed over time.p Although the HAI proponents oppose the use of the LERG, they have taken steps to ensure that the LERG database is compatible with use in the  X 4switching module in the synthesis model.6 Z D6 {O'ԍ See MCI Sept. 14 ex parte; Letter from Richard N. Clarke, AT&T, to Magalie Roman Salas, FCC, dated  {O'September 16, 1998 (AT&T Sept. 16 ex parte).  X 4177. ` ` We tentatively conclude that the LERG database is the best source currently available to determine hostremote switch relationships in the federal universal service mechanism. As noted above, no algorithm has been placed on the record to determine whether a wire center should house a standalone, host, or remote switch. In addition, a majority of commenters agree that development of such an algorithm would be difficult using  XK4publicly available data.z7^KD6 {O"'ԍ See, e.g., Ameritech Switching and Transport Public Notice comments at 3; AT&T/MCI Switching and  {O|#'Transport Public Notice comments at 6; BellSouth et al. Switching and Transport Public Notice comments att. 1  {OF$'at 12; GTE Switching and Transport Public Notice at 1112.z We tentatively conclude that the use of the LERG to identify the hostremote switch relationships is superior to HAI's averaging methodology which may not,"4E70*&&``U" for example, accurately reflect the fact that remote switches are more likely to be located in rural rather than urban areas. We therefore tentatively agree with the BCPM proponents and other commenters that use of the LERG is the most feasible alternative currently available to incorporate the efficiencies of hostremote relationships in the federal universal service mechanism. We seek comment on these tentative conclusions. In particular, we encourage parties to comment on any alternative source or methodology that will identify hostremote switch relationships on a forwardlooking basis.  XH' 3. ` ` Other Switching and Interoffice Transport Inputs   X 4178. ` ` General. Several commenters assert that the depreciation studies on which the Commission relied to develop switch costs include all investments necessary to make a switch  X 4operational.8 D6 {Oe 'ԍ AT&T Jan. 7 ex parte; GTE Dec. 18 ex parte at 57; Sprint Dec. 22 ex parte at 9. These investments include telephone company engineering and installation, the main distribution frame (MDF), the protector frame (often included in the MDF), and power  X 4costs.i9 ZD6 {O'ԍ AT&T Jan. 7 ex parte; Sprint Dec. 22 ex parte at 9.i To avoid double counting these investments, both as part of the switch and as separate input values, the model proponents agree that the MDF/Protector investment per line  X4and power input values should be set at zero.:D6 {O-'ԍ AT&T Jan. 7 ex parte; GTE Dec. 18 ex parte at 56; Sprint Dec. 22 ex parte at 9. In addition, commenters agree that the  Xy4Switch Installation Multiplier should be set at 1.0.w;y~D6 {O'ԍ See, e.g., AT&T Jan. 7 ex parte; GTE Dec. 18 ex parte at 6.w We agree that including these investments both as part of the switch cost and as separate investments would lead to double counting of these costs. We therefore tentatively conclude that the MDF/Protector investment per line and power input values should be set at zero. We further tentatively conclude that the Switch Installation Multiplier should be set at 1.0. We seek comment on these tentative conclusions.  X4179. ` ` Analog Line Offset. We tentatively conclude that the "Analog Line Circuit Offset for Digital Lines" input should be set at zero. The HAI proponents contend that the switch investment in the model should be adjusted downward to reflect the cost savings  X4associated with terminating digital, rather than analog, lines.<D6 {OT!'ԍ AT&T Jan. 7 ex parte. The HAI proponents contend that the cost of terminating digital lines is significantly less expensive than terminating analog lines. The HAI proponents assert that this cost savings is due primarily to: (1) the elimination of a MDF and protector frame termination; and (2) the economic efficiencies of terminating multiple lines on a DS1 trunk termination instead of individual analog line terminations. Further, HAI contends that the depreciation data on which the Commission relied in developing switch investments do not"7Fj <0*&&``" reflect adequately the cost savings that would be realized if "60+% of lines are terminated on  X4DLC as occurs in the TELRIC models."F=D6 {Ob'ԍ AT&T Jan. 7 ex parte. F HAI contends that the depreciation data used to  X4determine costs reflect the use of only approximately 15 percent digital lines.E>ZD6 {O'ԍ AT&T Jan. 7 ex parte.E  X4180. ` ` The HAI proponents suggest that the analog line offset input should be set to $15.00 per line to reflect additional savings in switch investment for terminating digital lines  Xv4in the model.? vD6 {O 'ԍ AT&T Jan. 7 ex parte. The HAI proponents reach this value by first concluding that the total switch cost savings for digital lines should be approximately $20.00 per line. HAI calculates this cost savings as follows: (1) $12.00 for the MDF/Protectors that are not needed in the switch for incoming lines terminated on DLC; and (2) $8.00 perline "efficiencies" (e.g., greater usage concentration, eliminated line cards, and lower cost to terminate trunks carrying DLC lines). Next, HAI concludes that the average fraction of digital lines included in the "historical" data is roughly 15% (or one quarter of the 60% digital line usage engineered in the model). The HAI proponents then deduct from the $20.00 value, one quarter of the digital line usage savings that is already taken into consideration in the historical data, to arrive at the analog line offset value of $15.00 per line.  The BCPM proponents and GTE recommend setting the analog line offset to  X_4zero.@_ D6 {O'ԍ GTE Dec. 18 ex parte at 6; Sprint Dec. 22 ex parte at 12; Sprint Jan. 8 ex parte at 15. Sprint contends that the analog line offset is inherent in the switching curve in the  XH4model, thus making this input unnecessary.NAHP D6 {OI'ԍ Sprint Dec. 22 ex parte at 12.N Sprint argues that an unknown mixture of  X14analog and digital lines are taken into consideration in developing the switch curve.NB1D6 {O'ԍ Sprint Dec. 22 ex parte at 12.N GTE  X 4asserts that the analog offset must be set to zero to "track with the switching inputs."JC tD6 {O?'ԍ GTE Dec. 18 ex parte at 6.J  X 4181. ` ` We note that the record contains no basis on which to quantify savings beyond those taken into consideration in developing the switch cost. We also note that the depreciation data used to determine the switch costs reflect the use of digital lines. The switch investment value will therefore reflect savings associated with digital lines. We also note that HAI's proposed analog line offset of $15.00 per line is based on assumptions that are neither supported by the record nor easily verified. For example, it is not possible to determine from the depreciation data the percentage of lines that are served by digital connections. It is therefore not possible to verify HAI's estimate of the digital line usage in the "historical" data. In addition, HAI provides little support for its conclusion that there is a $20.00 per line cost savings using digital lines. HAI merely attributes a portion of this"GC0*&&``2" estimate to certain "efficiencies" realized from terminating digital rather than analog lines. In the absence of more explicit support of HAI's position, we tentatively conclude that the Analog Line Circuit Offset for Digital Lines should be set at zero. We seek comment on this tentative conclusion.  X4182. ` ` Switch Capacity Constraints. We tentatively adopt the HAI default switch  Xv4capacity constraint inputs as proposed in the HAI 5.0a model documentation.WDvD6 yO'ԍ HAI Feb. 3, 1998 submission, App. B at 3839.W The forwardlooking cost mechanism contains switch capacity constraints based on the maximum line and traffic capabilities of the switch. The HAI proponents now recommend increasing the switch  X14line and traffic capacity constraints above the HAI input default values for those inputs.E1XD6 {O: 'ԍ AT&T Jan. 7 ex parte. The HAI proponents included the updated switch capacity constraints in a table  {O 'attached to the Jan. 7 ex parte. HAI contends that the default input values no longer reflect the use of the most current  X 4technology.GF D6 {Oh'ԍ AT&T Jan. 7 ex parte. G For example, HAI contends that the maximum equipped line size per switch  X 4should be increased from 80,000 to 100,000 lines.HG FD6 {O'ԍ AT&T Jan. 7 ex parte. H  X 4183. ` ` We tentatively conclude that the original HAI switch capacity constraint default values are reasonable for use in the federal mechanism. We note that commenters have  X4reviewed these values and are in general agreement with the HAI default values.HD6 {O'ԍ See, e.g., BellSouth Inputs Public Notice reply comments at Exhibit 213; BCPM Inputs Portfolio at 3436. For example, we note that the HAI and BCPM default values for maximum equipped lines per  Xb4switch are identical at 80,000 lines per switch.Ib2 D6 {OE'ԍ See HAI Dec. 11 submission, Model Inputs at 80; BCPM April 30, 1998 submission, Switching Model Inputs at 34. We also note that the HAI model documentation indicates that the 80,000 line assumption was based on a conservative estimate "recognizing that planners will not typically assume the full capacity of the switch can be  X4used."_J D6 {OZ!'ԍ See HAI Dec. 11 submission, Model Inputs at 80._ The HAI proponents therefore selected the 80,000 line limitation as the maximum equipped line size value with the knowledge that the full capacity of the switch may be"HJ0*&&``"  X4higher.0KD6 yOy'ԍ In addition, we note that a decision to adopt the revised HAI values for maximum equipped lines per switch would have only a minimal impact on the overall forwardlooking cost estimation because less than 2 percent of wire centers have more than 80,000 lines. A review of the data indicates that, of the 12,506 wire centers served by nonrural LECs, only 189 (1.5 percent) have more than 80,000 lines and 57 (0.5 percent) have  {O'more than 100,000 lines. See HAI Feb. 3, 1998 model submission.0 We seek comment on our tentative conclusion.  X4184. ` ` Switch Port Administrative Fill. We tentatively adopt a switch port administrative fill factor of 94 percent. HAI defines the switch port administrative fill as "the percent of lines in a switch that are assigned to subscribers compared to the total equipped  X4lines in a switch."_LzD6 yO 'ԍ HAI Dec. 11, 1997 submission, Inputs Portfolio at 80._ HAI assigns a switch port administrative fill factor of 98 percent in its  Xv4default input values._Mv D6 yO1'ԍ HAI Dec. 11, 1997 submission, Inputs Portfolio at 80._ The BCPM default value for the switch percent line fill is 88  X_4percent.NX_D6 yO'ԍ BCPM April 30, 1998 submission, Switch Model Inputs at 2021. BCPM defines Switch Percent Line Fill as the ratio between the number of working lines on the switch and the total number of lines for which the switch is engineered.  X14185. ` ` The BCPM proponents contend that switches have significant unassigned capacity due to the fact that equipment is installed at intervals to handle one to three years'  X 4growth.NO D6 {On'ԍ Sprint Dec. 22 ex parte at 17.N BCPM most recently contends that U S WEST and BellSouth have companywide  X 4average fills in the range of 76 percent.NP L D6 {O'ԍ Sprint Dec. 22 ex parte at 18.N Sprint, on behalf of the BCPM proponents, now  X 4recommends an average fill factor of 80 percent.NQ D6 {Od'ԍ Sprint Dec. 22 ex parte at 18.N  X 4186. ` ` We note that the switch port administrative fill factor of 94 percent has been adopted in several state universal service proceedings and is supported by the Georgetown  Xy4Consulting Group, a consultant of BellSouth.YR^ypD6 {O!'ԍ BellSouth Inputs Public Notice reply comments at Exhibit 213; Kentucky Cost Study App. F at 13;  {Od"'Louisiana Public Service Commission, State ForwardLooking Cost Studies for Federal Universal Service  {O.#'Support (May 19, 1998) (Louisiana Cost Study).Y We also note that this value falls within the range established by the HAI and BCPM default input values. The BCPM model documentation established a switch line fill default value of 88 percent that included"KIR0*&&``"  X4"allowances for growth over an engineering time horizon of several years."gSD6 yOy'ԍ BCPM April 30, 1998 submission, Switch Model Inputs at 2021.g BCPM has provided no additional evidence to support its revised value of 80 percent. We therefore tentatively adopt a switch port administrative fill factor of 94 percent. We seek comment on this tentative value.  X4187. ` ` Trunking. We tentatively conclude that the switch module should be modified to disable the computation that reduces the end office investment by the difference in the interoffice trunks and the 6:1 line to trunk ratio. In addition, we tentatively adopt the HAI suggested input value of $100.00 for the trunk port investment, per end.  X 4188. ` ` The HAI switching and interoffice module developed switching cost curves using the Northern Business Information (NBI) publication, "U.S. Central Office Equipment  X 4Market: 1995 Database."bT XD6 yO'ԍ HAI Dec. 11, 1997 submission, Model Description at 52. b These investment figures were then reduced per line to remove  X 4trunk port investment based on NBI's implicit line to trunk ratio of 6:1.`U D6 yOn'ԍ HAI Dec. 11, 1997 submission, Model Description at 53.` The actual number of trunks per wire center is calculated in the transport calculation, and port investment for these trunks is then added back into the switching investments.  Xy4189. ` ` The BCPM proponents contend that, under the HAI trunk investment approach, raising the pertrunk investment leads to a decrease in the switch investment per line under  XK4the HAI approach, "despite a reasonable and expected increase" in the investment per line.NVKxD6 {Ot'ԍ Sprint Dec. 22 ex parte at 10.N The BCPM proponents argue that the trunk port input value should be set at zero to avoid  X4producing "contradictory" results.NW D6 {O'ԍ Sprint Dec. 22 ex parte at 10.N GTE also notes that the selection of the trunk port input value creates a dilemma in that it is used to reduce the end office investment, as noted above,  X4and to develop a tandem switch investment.JXD6 {O<'ԍ GTE Dec. 18 ex parte at 6.J GTE recommends that the switch module be modified by disabling the computation that reduces the end office investment by the  X4difference in the computed interoffice trunks and the 6:1 line to trunk ratio.JY. D6 {O"'ԍ GTE Dec. 18 ex parte at 6.J The HAI sponsors agree that the trunk port calculation should be deactivated in the switching"J Y0*&&``"  X4module.ZD6 {Oy'ԍ Letter from Chris Frentrup, MCI, to Magalie Roman Salas, FCC, dated Feb. 9, 1999 (MCI Feb. 9 ex  {OC'parte) at 24.  X4190. ` ` We agree with commenters that the trunk port input creates inconsistencies in reducing the end office investment. We do not, however, agree with the suggestion of the BCPM sponsors to simply set this input value at zero. As noted by GTE, this input value is  X4also used to calculate the tandem switch investment.J[$D6 {Ob 'ԍ GTE Dec. 18 ex parte at 6.J Consistent with the suggestions by GTE and the HAI sponsors, we tentatively conclude that the switch module should be modified to disable the computation that reduces the end office investment by the difference in the computed interoffice trunks and the 6:1 line to trunk ratio.  X 4191. ` ` Because the trunk port input value is also used to determine the tandem switch  X 4investment, we must determine the trunk port, per end investment.\ D6 yOj'ԍ HAI defines this input as the "per trunk equivalent investment in switch trunk port at each end of a trunk." HAI Dec. 11, 1997 submission, Appendix B (HM 5.0 Inputs, Assumptions, and Default Values) at 46. The HAI input value for  X 4trunk port investment per end is $100.00.`] D6 yO'ԍ HAI Dec. 11, 1997 submission, Inputs Portfolio at 102.` GTE and Sprint contend that this value should be  X 4much higher ranging from $200.00 to $500.00.{^ D6 {O$'ԍ GTE Dec. 18 ex parte at 6; Sprint Dec. 22 ex parte at 10.{ BellSouth notes that four states have  X 4issued orders addressing the cost of the trunk port for universal service.I_Z 0 D6 yO'ԍ Letter from William W. Jordan, BellSouth, to Magalie Roman Salas, FCC, dated August 7, 1998  {Og'(BellSouth Aug. 7 ex parte) Attachment to Question 1 at 5, 9, 13, 17 (dated July 15, 1998). The four states are Kentucky, Louisiana, North Carolina, and South Carolina.I These states  X 4estimate the cost of the trunk port ranging from $62.73 to $110.77.;` R D6 {O'ԍ Id. ; We tentatively conclude that the record supports the adoption of a trunk port investment per end of $100.00, as suggested by the HAI sponsors. As noted above, this value is consistent with the findings of several states and BellSouth. In addition, GTE and Sprint provide no data to support their proposed trunk port investment value. We therefore tentatively adopt the HAI suggested input value of $100.00 for the trunk port investment, per end. We seek comment on our tentative conclusions.  X4J VII. EXPENSES ׃  X4   X4192. ` ` In this section, we address the inputs in the model related to expenses,"K`0*&&``" including general support facilities (GSF) expenses. In light of the criteria identified in the  X4Universal Service Order, the Commission intends to select inputs that will result in a reasonable allocation of joint and common costs for nonnetworked related costs such as GSF, plant specific and nonspecific expenses, and corporate and customer operations. The Commission seeks to develop an appropriate methodology for estimating these types of expenses to "ensure that the forwardlooking economic cost [calculated by the federal mechanism] does not include an unreasonable share of the joint and common costs for non Xa4supported services."aaD6 {O'#X\  P6G;ɒP#э Platform Order, 13 FCC Rcd at 21357, para. 81.  X3' A.Background   X 4193. ` ` GSF costs and expenses include the investment and expenses related to vehicles, land, buildings, and general purpose computers. Other expenses (that are not  X 4associated with GSF) include: plant specific expenses,b ZD6 yO'ԍ #X\  P6G;ɒP#Plant specific expenses include the cost of maintaining telecommunications plant and equipment. plant nonspecific expenses,c D6 yOr'ԍ #X\  P6G;ɒP#Plant nonspecific expenses include the cost of engineering, network operations, and power expenses.  X 4corporate operations expenses,d zD6 yO'ԍ #X\  P6G;ɒP#Corporate operations expenses include the cost of administration, human resources, legal, and accounting expenses. and customer services expenses.e D6 yOC'ԍ #X\  P6G;ɒP#Customer service expenses include the cost of marketing, billing, and directory listing expenses. For purposes of this  X 4Further Notice, costs associated with common support services (often called overhead expenses) refer to plant nonspecific expenses, corporate operations expenses, and customer service expenses.  XO4194. ` ` In the 1997 Further Notice, the Commission sought comment on how to remove costs for nonregulated activities from costs for regulated activities in order to incorporate the appropriate amount of GSF investment and expenses in estimating the costs of  X 4providing the supported services.f b D6 {O'ԍ #X\  P6G;ɒP#1997 Further Notice, 12 FCC Rcd at 18569, para. 148. The Commission tentatively concluded that GSF expenses should vary by state with respect to land values because a large share of GSF expenses is attributable to the cost of land.  X4195. ` ` In the Further Notice, the Commission also sought comment on how to  X4establish forwardlooking expenses in the selected federal mechanism.g D6 {O@%'ԍ #X\  P6G;ɒP#1997 Further Notice, 12 FCC Rcd at 1857273, para. 157. The Commission"L g0*&&``" specifically sought comment on which expenses should be calculated on a perline basis and which should be calculated as a percentage of investment. The Commission also sought comment on whether there are measures other than lines and investment to which specific  X4expenses should be tied.hD6 {O4'ԍ #X\  P6G;ɒP#1997 Further Notice, 12 FCC Rcd at 18572, 1857477, paras. 157, 162, 165, 168, 171. With respect to plant specific expenses, the Commission sought comment on whether maintenance expense estimates should depend upon plant mix and, in particular, whether an increase in the use of aerial cable also increases maintenance expenses, and whether plant specific expenses should vary with such characteristics as climate or soil  X_4type.i_ZD6 {Oj 'ԍ #X\  P6G;ɒP#1997 Further Notice, 12 FCC Rcd at 18574, para. 162. In addition, the Commission asked commenters to identify the complete set of  XH4forwardlooking expenses for which universal service support should be available.jHD6 {O 'ԍ #X\  P6G;ɒP#1997 Further Notice, 12 FCC Rcd at 1857477, paras. 162, 165, 168, 171.  X 4196. ` ` In the Platform Order, we adopted HAI's algorithm for calculating expenses and GSF costs, as modified to provide some additional flexibility in calculating expenses  X 4offered by the BCPM sponsors.k ~D6 {O'ԍ #X\  P6G;ɒP#Platform Order, 13 FCC Rcd at 21357, para. 81. With this added flexibility, the model allows the user to estimate expenses as either a perline amount or as a percentage of investment. We noted that many of the questions regarding how best to calculate expenses will be resolved in the input  X 4selection phase of this proceeding.l D6 {Oj'ԍ #X\  P6G;ɒP#Platform Order, 13 FCC Rcd at 21360, para. 87.  X{' B.Issues for Comment   Xd'  XM'1.` ` Plant Specific Operations Expenses  X6'  X4197. ` ` We first address the inputs related to plant specific operations.mD6 yOr'ԍ #X\  P6G;ɒP#Note that plantspecific operations expenses include general support facilities (GSF) expenses. Plant specific operations expenses are the expense costs related to the maintenance of specific kinds of  X4telecommunications plant.nxH 2 D6 yO 'ԍ #X\  P6G;ɒP#Plant specific operations expenses correspond to the following ARMIS 4303 report accounts: 6110 Network Support Expense 6120 General Support Expense 6210 COE Switch XX` ` 6212 COE Digital Electronic Switch only (#` 6220 Operator Systems"%m0*&&%"Ԍ6230 COE Transmission XX` ` 6231 Radio Systems (#` XX` ` 6232 COE Circuit DDS(#` XX` ` 6232 COE Circuit Other than DDS(#` X6310 Information Origination/Termination(# XX` ` 6311 Station Apparatus (only)(#` X6341 Large PBX(# X6351 Public Telephone(# X6362 Other Terminal Equipment(# 6411 Poles 6421.1 Aerial Cable Metallic (Copper) 6421.2 Aerial Cable Fiber 6422.1 Underground Cable Metallic (Copper) 6422.2 Underground Cable Fiber 6423.1 Buried Cable Metallic (Copper) 6423.2 Buried Cable Fiber 6441 Conduit Systems  "Mn0*&&``"Ԍ  X4198.` ` Nationwide Estimates. We tentatively conclude that we should adopt input values that reflect the average expenses that will be incurred by nonrural carriers, rather than a set of companyspecific maintenance expense estimates. We make this tentative conclusion for a number of reasons. First, we note that this tentative conclusion is consistent with a  X4recommendation of the state Joint Board members.oD6 {O'ԍ #X\  P6G;ɒP# See State High Cost Report at 21; 1997 Further Notice, 12 FCC Rcd at 18574, para. 161. Second, we have not been able to obtain current costtobook cost ratios for each ARMIS reporting firm, which would be necessary to calculate company or study area specific expensetoinvestment ratios in the proposed methodology described below. Further, we tentatively conclude that the use of national or regional averages for input factors is more consistent with the forwardlooking nature of the high cost model because it mitigates the rewards to less efficient companies. We seek comment on these tentative conclusions. Parties advocating the use of companyspecific values or other alternatives to nationwide or regional estimates should identify the method and data readily available to firms that would be used to estimate plantspecific expenses. Commenters should also indicate how their proposal is consistent with the goal of estimating forwardlooking costs. We note that the proposed expense estimates in Appendix A are nationwide averages.  Xb4199.` ` In support of the use of companyspecific factors, a number of commenters and workshop participants argue that maintenance expenses vary widely by geographic area and the type of plant installed. Others contend that plantspecific expenses are highly dependent on regional wage rate differentials. At this time, we have been unable to verify significant regional differences among study areas or between companies based solely on labor rate variations using the publicly available ARMIS expense account data for plantspecific"Njo0*&&``" maintenance costs. Nonetheless, we believe that expenses vary by the type of plant installed. The synthesis model takes this variance into account because, as investment in a particular type of plant varies, the associated expense cost also varies. We seek comment on the degree to which regional wage rate differentials exist and are significant. We ask parties to suggest independent data sources on variations of wage rates between regions. We seek comment on a methodology that permits such distinctions without resorting to selfreported information from companies.  XH4200.` ` One possible approach would be to use indexes calculated by the President's  X14Pay Agent for calculating locality pay differentials for Federal employees.p1D6 {O 'ԍ #X\  P6G;ɒP#Report on Localitybased Comparability Payments for the General Schedule, Annual Report of the  yOt 'President's Pay Agent, Appendix II, 1995.#x6X@`7X@# Under this methodology, we would first calculate a baseline expense factor for the laborrelated portion of each plantspecific expense account according to a formula which is based on the sum of an expense factor for that category by study area, a weight representing the total investment in a study area, and the regional wage differential deflator calculated in the Pay Agent's report  X 4applicable to the study area.q( "D6 {ONԍ #X\  P6G;ɒP#The baseline expense factor is calculated according to the formula Ei =  j eijwjrj where Ei is the baseline  {O[Nexpense factor for category i, eij is the expense factor for category i and for study area j, wj is a weight based on  {O%Nthe number of switched lines in study area j, and rj is the regional deflator from the Pay Agent's report for the  {O'region most applicable to study area j. The baseline expense would then be disaggregated to each wire center or study area using the deflator. We seek comment both on the validity of this approach as well as on the specific implementation.   Xb4201. ` ` We also tentatively conclude that we should not adopt different expense estimates for small, medium, and large nonrural companies on a per line basis. In order to determine if economies of scale should be a factor in plantspecific expenses, Commission staff tested whether significant differences in maintenance expenses per line could be discerned from segmenting companies into small carriers with less than 500,000 access lines, medium carriers with between 500,000 and 5,000,000 access lines, and those large carriers  X4with over 5,000,000 access lines.rD6 yO'ԍ #C\  P6QɒP#We note that the data used for this analysis is available on the Commission's Web site at http://www.fcc.gov. We have found no significant differences in the expense factor perline or perinvestment estimates based on these criteria. Therefore, to estimate costs associated with an efficient network as determined by the forwardlooking mechanism, we tentatively conclude that plantspecific maintenance factors should be estimated on a national basis. We seek comment on these tentative conclusions.  XN4202. ` ` Methodology. Commenters advocate two methods of estimating plant specific operations expenses. The BCPM sponsors contend that all expenses should be calculated on a"7Oj r0*&&``8" perline basis. The BCPM default estimates for these accounts are based on a survey of companies. The HAI sponsors argue that expenses should be calculated as a percentage of investment. Specifically, the HAI sponsors assert that plant specific operations expenses should be calculated as a fixed percentage of investment. ` ` .  X4203. ` ` Although we agree with the HAI sponsors that plant specific operations expenses should be estimated as a percentage of investment, we tentatively decline to adopt the flat percentages they advocate. By using ARMIS investment values that are not converted to current levels, the flatrate method proposed by the HAI sponsors does not attempt to use forwardlooking estimates. We also tentatively decline to adopt the perline BCPM default estimates. Based on a private survey of companies, the BCPM values fail to comply with  X 4criterion eight identified in the Universal Service Order, because the underlying data for these  X 4values are not open to and verifiable by the public nor made available under the Protective  X 4Order.s D6 yOR'ԍ#C\  P6QɒP# Protective Order, 13 FCC Rcd at 13910.#Xx6X@DQX@#ў In contrast to the BCPM proposal, the methodology that we tentatively adopt here is primarily based on readily identifiable and publicly available ARMIS data. Although ARMIS data reflect the embedded costs incurred by incumbent LECs, we take steps in our proposed methodology to convert these costs to forwardlooking estimates, as described below. We note that this methodology was proposed by Commission staff in the public workshop on maintenance expenses on December 10, 1998.  X:4204. ` ` In order to estimate forwardlooking plant specific operations expenses, we  X#4have considered the requirements set forth in the Platform Order, and information provided in  X4workshops, comments and expartes. We tentatively conclude that the input values for each plant specific operations expense account should be calculated as the ratio of booked expense to current investment. These expensetoinvestment ratios would then be multiplied in the model by the modelderived investment for each investment account or group of accounts, to produce an estimate of the plant specific operations expenses.  X4205. ` ` Our proposed methodology for estimating expense to investment ratios consists of four steps. First, staff obtained from some of the ARMISfiling companies, accountspecific current cost to book cost (currenttobook) ratios for the related investment accounts. The currenttobook ratio is a tool that is used to restate the historic, financial account balance on a company's books, which reflects investment decisions made over many years, to present  X4day replacement cost.UtXXD6 yO"'ԍ #X\  P6G;ɒP#An example of a currenttobook ratio for a facility would be defined as the ratio of the current cost of a facility divided by its embedded cost. Therefore, if a pole cost $200 to install in 1980, and $400 today, the currenttobook ratio is $400/$200 = 2.0.U For each account or subaccount, a currenttobook ratio is developed by first revaluing each type of equipment at its current replacement cost. The sum of these current costs are then divided by the total, embedded cost account balance. The"Pxt0*&&``" resulting currenttobook ratio will be greater than one if current costs are rising relative to the historic costs and less than one if current costs are declining. Currenttobook ratios for the years ending 1995 and 1996 were provided by the following five holding companies:  X4Ameritech, Bell Atlantic, Bell South, GTE, and Southwestern Bell.uD6 yO4'ԍ #X\  P6G;ɒP#The currenttobook ratios submitted by these companies are proprietary information subject to provisions  {O'in the Protective Order and therefore are not reproduced here.  Although we would prefer to have data from more companies, the other ARMISfiling carriers informed us that, they either no longer maintain this type of information, or never used currenttobook ratios  Xv4for accounting purposes.vv"D6 yOI 'ԍ #X\  P6G;ɒP#We note that the five carriers that provided currenttobook ratios account for 66 percent of plant specific expenses and 70 percent of the investment reflected in ARMIS.  XH4206.` ` Second, staff calculated composite currenttobook ratios for each account. For each study area of the five holding companies that provided currenttobook ratios, we obtained yearend 1995 and 1996 investment balances from ARMIS for the plant accounts  X 4consistent with the aforementioned plantspecific expense accounts.iw zD6 yO.'ԍ #X\  P6G;ɒP#Investment balances were obtained from the following plant accounts: X` ` 2112 2116 Network Support Investment (#  yO'X` ` 2121 2124 General Support Investment(#  yON'X` ` 2210 COE Switch hhC(# ` `  2212 COE Digital Electronic Switch only(#`  yO'X` ` 2230 COE Transmission (# XX` `  2231 Radio Systems (#` XX` `  2232 COE Circuit DDS(#` XX` `  2232 COE Circuit Other than DDS(#`  yO'X` ` 2310 Information Origination/Termination (# XX` `  2311 Station Apparatus (#` X` `  2341 Large PBX(# ` `  2351 Public Telephone X` `  362 Other Terminal Equipment(# ` ` 2410 Cable and Wire Facilities ` `  2411 Poles  yOv' ` `  2421.1 Aerial Cable Metallic (Copper)(#` ` `  2421.2 Aerial Cable Fiber ` `  2422.1 Underground Cable Metallic (Copper) ` `  2422.2 Underground Cable Fiber ` `  2423.1 Buried Cable Metallic (Copper) ` `  2423.2 Buried Cable Fiber ` `  2241 Conduit Systems i Study areaspecific currenttobook ratios for the two periods were multiplied by the 1995 and 1996 ARMIS investments in each account to derive the forwardlooking, "current," yearend 1995 and 1996 " Qw0*&&`` " investment levels by account and by study area. The ARMIS and current investments were then summed separately, by year and by account, for all study areas of the five holding companies. The resulting total current investment (by year and by account for the sum of all study areas) was then divided by the total ARMIS investment (by year and by account for the sum of all study areas) producing two sets of composite currenttobook ratios (year end 1995 and 1996).  X_4207. ` ` Third, to calculate the expensetoinvestment ratios for the plantspecific operations expense accounts, staff obtained total, yearend 1995 and 1996 investment account balances from the ARMIS 4303 reports for all ARMISfiling companies. To make these embedded account balances forwardlooking, staff next multiplied each investment account balance for each year by the currenttobook ratios for the same year developed earlier. The 1995 and 1996 "current" balances for each account were then averaged by adding the two years together and dividing by two.  X 4208.` ` Finally, from the 1996 ARMIS 4303 report, staff obtained the 1996 balances for each plantspecific operations expense account for all ARMISfiling companies. The expense account balances were divided by their respective average "current" investment to obtain expensetoinvestment ratios. We tentatively conclude that these expensetoinvestment ratios should be applied in the mechanism to the modelderived investment balances to obtain forwardlooking plantspecific operations expense estimates. The industrywide expensetoinvestment ratios are listed in Appendix A. We seek comment on these proposed input values, tentative conclusions, and the proposed methodology outlined above.  X4209. ` ` Converting Expense Estimates to Current Values. We recognize that plant specific expenses will change over time. Because we initially used data from 1996 in the methodology described above, we tentatively conclude that it is appropriate to adjust this data to account for inflation and changes in productivity by obtaining revised 1997 currenttobook ratios from those companies providing data. In addition, we tentatively conclude that we should use the most current ARMIS data available necessary for the maintenance factor methodology. Because expense and investment balances for 1998 are not available from ARMIS at this time, we have also not been able to include them in calculating the plantspecific maintenance factors. We tentatively conclude that we should use these data in the final computation of expense estimates. We seek comment on these tentative conclusions.  X4210.` ` GSF Investment. GSF investment includes buildings, motor vehicles, and general purpose computers. The synthesis model uses a threestep algorithm to estimate GSF for each study area. First, the model calculates a GSF investment ratio for each GSF account by dividing the ARMIS investment for the account by the ARMIS total plant in service (TPIS). Second, the model calculates a preliminary estimate GSF investment for each account"Rw0*&&``!" by multiplying the GSF investment ratio for that account times the model's estimate of  X4TPIS.xD6 yOy'ԍ #C\  P6QɒP#As calculated by the model, TPIS excludes GSF investment, while ARMIS TPIS includes GSF investment. HAI Dec. 11, 1997 submission. Finally, the model reduces each of the preliminary GSF investment estimates by  X4multiplying by one of two factors, which are the same as those used in the HAI model.y 7D6 yO'ԍ #C\  P6QɒP#These two factors are one minus either the Total Operations General Support Allocator (Total Operations Allocator) or the Office Worker General Support Allocator (Office Worker Allocator). The Total Operations Allocator is applied to the Motor Vehicles, Garage Work Equipment, and Other Work Equipment accounts, while the Office Worker Allocator is applied to the Furniture, Office Equipment, Buildings and General Purpose  {O'Computer accounts. See HAI Dec. 11, 1997 submission. Each of these allocators is a fraction. The Total Operations Allocator is the ratio of the sum of customer operations expenses and corporate operations expenses to total operating expenses. The Office Worker Allocator is the ratio of the sum of corporate operations expenses and network operations expenses to the sum of customer operations expenses, corporate operations expenses and network operations expenses.  X4211.` ` We tentatively conclude that the model's preliminary estimate of GSF investment should be reduced, because only a portion of GSF investment is related to the cost of providing the services supported by the federal mechanism. We also tentatively conclude that the synthesis model should not use the same factors as those used in the HAI model. The HAI sponsors, who developed the expense module in the synthesis model, have not shown why these particular factors should be used for this purpose. Instead, we tentatively conclude that total GSF investment should be reduced by factors that reflect the percentage of customer operations, network operations, and corporate operations used to provide the supported  X 4services.gz@ D6 yO'ԍ#C\  P6QɒP# We tentatively conclude that the Office Worker Allocator should equal the ratio of the sum of customer operations expenses, network operations expenses, and corporate operations expenses assigned to supported services, to the sum of those expenses calculated on a total regulated basis. In principle the Total Operations Allocator should equal the Office Worker Allocator. Due to equations embedded in the HAI expense module, however, the total operations general support allocator is set equal to one minus the office worker general support allocator. We tentatively conclude that GSF investment should be calculated as the product of the Office Worker Allocator, calculated on a nationwide basis, and the preliminary GSF investment, which is calculated on a study area specific basis.g We seek comment on these tentative conclusions.  X ' 2.` ` Common Support Service Expenses   X4212. ` ` We next address common support service expenses, which are comprised of corporate operations, customer service expenses, and plant nonspecific expenses. Corporate operations expenses are those costs associated with general administrative, executive planning, human resources, legal, and accounting expenses for total company operations. Customer service expenses include marketing, billing, operator services, directory listing, and directory"4Sz0*&&``"  X4assistance costs.'{0D6 yOy'ԍ#X\  P6G;ɒP# Corporate operations and customer services include the following ARMIS accounts and their subaccounts: X` ` 6610 Marketing Total(# ` `  6611 Product Management ` `  6612 Sales ` `  6613 Product Advertising ` ` 6620 Service Expense Total ` `  6621 Call Completion (Operator Service Expense) ` `  6622 Number Services (Directory Publishing Expense) ` `  6623 Customer Services ` ` 6710 Executive and Planning Total ` `  6711 Executive ` `  6712 Planning ` ` 6720 General and Administrative ` `  6721 Accounting and Finance ` `  6722 External Relations ` `  6723 Human Resources ` `  6724 Information Management ` `  6725 Legal ` `  6726 Procurement ` `  6727 Research and Development  yO'` `  6728 Other General and Administrative#x6X@`7X@#' Plant nonspecific expenses are common network operations and maintenance type of expenses, including engineering, network operations, power and testing  X4expenses, that are considered general or administrative overhead to plant operations.^|D6 yOC'ԍ #X\  P6G;ɒP# Nonspecific plant expenses include the following ARMIS expense accounts: ` ` 6510 Other Property Plant and Equipment Expense ` ` 6530 Network Operations ` `  6531 Power ` `  6532 Testing^ Commission staff held public workshops where they sought comment on various paradigms and econometric estimation techniques used to calculate these factors. Commission staff also discussed possible methods for subtracting nonrecurring costs from expense estimates and for  Xv4adjusting estimates for inflation and potential wage differentials.}v8D6 {O_'ԍ#C\  P6QɒP# See 1997 Further Notice, 12 FCC Rcd at 18578, para 173.  XH4213.` ` PerLine Basis. Common support services are costs that cannot readily be associated with any particular maintenance expense or investment account. As a result, we tentatively conclude that these expenses (unlike plantspecific expenses) should be estimated" T}0*&&`` "  X4on a perline basis, as advocated by the BCPM sponsors.k~D6 yOy'ԍ #X\  P6G;ɒP#BCPM model default values on a perline permonth basis are the following:p  *(#   yO '` ` Aggregate USOA AccountqXppBCPM Default Values xxX(#  yO'` ` Other PP & E (6510) hhCqpp$ 0.03  *XxxX (#x  yOa'` ` Network Operations (6530)qX pp  1.33  * (#   yO)'` ` Marketing (6610) hhCXqX pp  0.35(#  yO'` ` Service Exp./Customer Operations (6620)h pp 2.42 (#  yO'` ` xec., Planning, G&A (6700)qXpp 2.29(#  yOI '` ` Total PerLine PerMonth Expensespp$ 6.42p  * XxxXX (# k We tentatively conclude that the HAI sponsors have failed to justify their proposal that expense estimates for certain accounts be based on a percentage of ARMISreported expenses or a percentage of total capital costs  X4and operations expenses."( D6 yO'ԍ #X\  P6G;ɒP#For example, the HAI sponsors propose that network operations expense be estimated at 50 percent of ARMISreported network operations expense. Corporate overhead expense was estimated to be 10.4 percent of  {O$'the total of capital costs and operations expenses as a default value. See 1997 Further Notice, 12 FCC Rcd at 18572, 18577, paras. 164, 170. We seek comment on these tentative conclusions.  X4214. ` ` Nationwide Estimates. Commenters such as Aliant, Sprint, GTE, and Bell South have argued for the inclusion of all accounts, and have argued further that these types of corporations and customer service expenses are inherently company specific in nature and should be evaluated in this manner. We tentatively conclude that inputs for corporate operations, customer services, and plant nonspecific expenses should also be estimated on a nationwide basis rather than a more disaggregated basis. We seek comment on this tentative conclusion.  X 4215.` ` Costs associated with plant nonspecific expenses used to supply and run network operations by definition cannot be directly allocated to individual maintenance or investment accounts. Commenters have suggested that these types of expenses may vary among carriers and between study areas. They argue that these differences may be a result of company specific plant configurations, geographic and labor demographic variables, onetime exogenous costs, and nonrecurring adjustments such as reengineering expenses. They further argue that administrative support expense differences are also a function of regional wage differentials and plant specifications. As stated earlier, we cannot at this time distinguish significant differences in regional wage differentials for administrative services based solely on ARMIS expense data for these accounts. Further, costs associated with corporate overhead and customer services accounts are not directly linked to specific company investment levels. We tentatively conclude that, for forwardlooking cost estimates, these types of administrative and service expenses are less dependent on carrier physical plant or geographic differentials than those that also correlate to company size (number of lines) and"U0*&&``" demand (minutes of use), which were used as estimation variables to develop the model inputs. We seek further comment on this analysis.  X4216. ` ` We also tentatively conclude that we should not adopt different estimates for small, medium, and large high cost nonrural companies for common support service expenses. As with plant specific expenses, Commission staff tested whether statistically significant differences in common support service expenses per line could be determined from segmenting companies into small carriers with less than 500,000 access lines, medium carriers with between 500,000 and 5,000,000 access lines, and those large carriers with over 5,000,000 access lines. We have further reviewed whether expense estimates varied due to the total number of Dial Equipment Minutes (DEMs) reported by companies in addition to the number of lines. As with the plantspecific accounts, we could find no significant differences in the expense factor perline based on these criteria. Therefore, consistent with the forwardlooking costs associated with an efficient network as determined by the federal mechanism, we tentatively conclude that we should estimate these nonspecific network operations expenses  X 4on a nationwide, perline basis. We seek comment on this tentative conclusion.   Xy4217.` ` Data Source. Following standard economic analysis and forecasting methods,  Xb4we propose to use publicly available 1996 ARMIS expense data bD6 yO'ԍ #C\  P6QɒP#Data was taken from 1996 ARMIS 4301, Subject to Separations (Column F) for Accounts 6610, 6620, 6710 and 6720. Data was taken from 1996 ARMIS 4303, Subject to Separations (Column M) for Accounts 6510 and 6530. Line counts were taken from 1996 ARMIS 4308, Table III, Total Switched Lines (Column DJ) and Total Access Lines (Column DM).  and minutes of use  XK4information from NECA, KD6 yO'ԍ #C\  P6QɒP#Dial Equipment Minutes of Use (DEMS) for 1996 were taken from NECA, available on the Commission's Web site at http://www.fcc.gov/Bureaus/Common_Carrier/Reports/FCCState_Link/neca.html.  by study area, to estimate the portion of these companywide expenses to be covered by universal service support. We believe that consolidation of this  X4data produces a sufficient number of observations by study area for each of these accounts.1D6 yO'ԍ#C\  P6QɒP# See Appendix F for further explanation of the data sources used and the method for consolidating study areas between data sets to reconcile the number of observations for the variables used in the regression equations.1 Public data for 1996 was used in this analysis in order to compare the estimates obtained with proprietary information received from a previous data request. We note that this methodology was proposed by Commission staff in a public workshop on December 1, 1998. We seek comment on this proposal.  X4218. ` ` Regression Methodology. Using standard multi-variate regression analysis, we developed two different specifications to determine the portion of corporate and customer operations and plant nonspecific expenses subject to universal service support. Each equation estimates total expenses per total lines as a function of switched lines per total lines, special lines per total lines, and toll minutes per total lines, either in combination (Specification 1) or"7V` 0*&&``8"  X4separated between intrastate toll and interstate toll minutes per total lines (Specification 2).QD6 yOy[ԍ #C\  P6QɒP#Specification 1 used the following regression equation: Expense/Total Lines = 1 (Switched Lines/Total  yOA[Lines)+ 2 (Special Lines/Total Lines)+ 3 (Toll Minutes/Total Lines). Specification 2 used the following  yO [equation: Expense/Total Lines = 1 (Switched Lines/Total Lines)+ 2 (Special Lines/Total Lines)+ 3 (State Toll  yO[Minutes/Total Lines)+ 4 (Interstate Toll Minutes/Total Lines) Q  X4219.` ` Each specification has been chosen to separate the portion of expenses that could be estimated as attributable to special access lines and toll usage, which are not supported by the high cost mechanism, rather than switched lines and local usage. Commission staff found from an earlier formulation that, when the model included both a switched line component and a local usage component, the number of switched lines and local DEMs were so highly correlated that it did not increase the explanatory power of the model to include both variables. As a result, we tentatively conclude that we should not include local dial equipment minutes per total lines as an explanatory variable, despite suggestions by a number of workshop participants and commenters. Because both regression equations produce reasonable estimates, and in order to prevent any potential advantage to firms which might have a different mix of toll minutes, we propose to use the average of the estimates from the two specifications. We seek further comment on this proposed regression methodology.  X 4220. ` ` Removal of OneTime and NonSupported Expenses. In order to eliminate the impact of onetime nonrecurring expenses on forwardlooking estimates, we have sought verifiable public information on exogenous costs and those that are recovered through nonrecurring charges and tariffs. These include specific one time charges for the cost of mergers, acquisitions, and process reengineering. We also sought to estimate the cost of providing permanent number portability, network and interexchange carrier connection, disconnection, and reconnection (i.e., churn) costs. Other recurring functions that we have attempted to identify include vertical features expenses, billing and collection expense not related to supported services, operational support systems and other expenses associated with providing unbundled network elements and wholesale services to competitive local exchange carriers, collocation expenses, and costs associated with SS7 services.  X4221. ` ` Without obtaining proprietary information from carriers, we have been unable to find an objective public data source or discern a systematic method for excluding many of these costs from the expense data used to calculate the input factors. AT&T and MCI WorldCom presented an analysis to Commission staff on January 14, 1999, proposing a method to estimate, nonsupported, nonrecurring, or onetime expenses for customer, network, and corporate operations expenses. Averaging data for five years (19931997) of corporate Security and Exchange Commission (SEC) 10K and 10Q filings, a percentage of corporate and network operations identified as onetime charges were estimated for the BOCs and all Tier One companies. Because the SEC reports do not specifically indicate whether the"Wx0*&&``^" onetime expenses were actually made during the year(s) indicated, we tentatively conclude that we should not use these figures to adjust the 1996 ARMIS data used in estimating the expense input values. The analysis does indicate, however, that onetime expenses for  X4corporate operations can be significant and should be estimated, if possible.)D6 yO4'ԍ #C\  P6QɒP# According to the analysis, on average, approximately 20 percent of yearly corporate operations expenses consisted of nonrecurring charges for costs relating to merger and acquisitions and workforce restructuring over the five year period. Network Operations onetime charges for things such as process reengineering were also calculated using the same methodology. On average, 2.6 percent of yearly Network Operations expenses were attributed to nonrecurring charges.) Because this type of data detail is not publicly available from ARMIS or easily reconcilable from other public company financial reports to individual account expenses for a specific year, we invite comment on how to identify and estimate these expenses.  XH4222.` ` We tentatively conclude that, if it is determined that expense estimates to be used as inputs in the highcost mechanism are to be revised annually, as suggested by various  X 4parties, xD6 {OC'ԍ#C\  P6QɒP# See 1997 Further Notice, 12 FCC Rcd at 18578, para. 173. onetime nonrecurring costs should be systematically excluded. We further recommend that, to the extent possible, efforts be made to use current information supplied and verified by the companies, if none can be found independently, to more accurately reflect forwardlooking expenses. We seek comment on this tentative conclusion and recommendation.   X4223. ` ` Removal of NonSupported Expenses. Cost reductions were made for continuous nonsupportable services which could be identified and estimated from publicly available (ARMIS) expense data. Expense adjustments were made to calculated input values for marketing expenses. Though the HAI sponsors and state Joint Board members suggested  X44that marketing expenses be excluded entirely,Z4 D6 yO'ԍ #X\  P6G;ɒP#Both the HAI sponsors as well as the State Joint Board indicated that all marketing expenses be  {O'excluded. See 1997 Further Notice, 12 FCC Rcd at 18576, paras. 166, 167. See also, Letter from Chris  yO'Fentrup, MCIWorldcom to Magalie Roman Salas, FCC, dated January 15, 1999.Į commenters and workshop participants noted  X4that Section 214 of the Communications Act requires eligible telecommunications carriers to advertise the availability of residential local exchange and universal service supported  X4services. , D6 yO 'ԍ #X\  P6G;ɒP#Section 214 (e)(1)(B) of the Communications Act states that eligible telecommunications carriers shall, throughout the service area for which the designation of universal service support is received, "advertise the availability of such services and the charges therefor using media of general distribution." 47 USC  214(e)(1)(B).  X4224. ` ` We tentatively conclude that an analysis made by Economics and Technology, Inc., regarding the disaggregation of marketing and advertising expenses made by companies"X0*&&``" for basic telephone service, is the most accurate method on the record for apportioning marketing expenses between supported and nonsupported services. This analysis attributes an average of 95.6 percent of company marketing costs to nonsupported customers or activities,  X4such as vertical and new services."D6 {O4'ԍ #X\  P6G;ɒP#See Further Comments of the National Cable Television Association Inc., CC Docket 9645, Appendix 3A. The adjustment for supported local loop Marketing Expenses was made by deducting 95.6 percent of estimated expenses, maintaining 4.4 percent of Account 6610. See further discussion in Appendix F for calculations of expense reductions.  We seek comment on this proposed analysis for estimating marketing expenses.  Xv4225.` ` We also propose adjustments for nonsupported service costs related to coin operations and collection, published directory, access billing, interexchange carrier office  XH4operation, and service order processing, HD6 yO 'ԍ #C\  P6QɒP#A deduction was made for average expenses relating to Coin Collection and Administration, Carrier Access Billing, and Other Customer Services Total at 12.46 percent from the Customer Services Subaccount 6623, maintaining 87.54 percent of the costs associated with this account. See discussion in Appendix F for the calculations and methodology of expense reductions.  which are associated with specific expense accounts used in the regression analysis. Under this methodology, percentage reductions would be made to the estimated coefficients for those accounts using calculations based on a time trend  X 4analysis of average ARMIS 4304 expense data D6 yON' #C\  P6QɒP#э ARMIS 4304 Accounts Subject to Separations, Basic Local Loop and Other Related Services for Total Customer Operations Expense. for five years (19931997). We seek comment on this proposed methodology.   X 4226.` ` Converting Expenses to 1999 Values. In order to bring forward the 1996 data relied upon for estimating common support service expenses, we propose to use a 6.0 percent productivity factor for each year (1997 and 1998) to reduce the estimated input values for each account. The 6.0 percent productivity factor is based on the 6.5 percent "Xfactor" used  Xb4in the Commission's price cap methodology.Qb D6 yO'ԍ #X\  P6G;ɒP# Under price cap regulation, the productivity (X) factor includes a .50 percent (.005) Consumer Productivity Dividend (CPD), which is used as a supplement to calculated productivity measures to assure that consumers additionally benefit from price cap rules through lower access charge rates. Thus, for universal service calculations, this dividend was subtracted from expected productivity increases or cost savings that can be experienced each year (1997 and 1998) by firms that may provide universal service. Q We note that the D.C. Circuit Court of Appeals recently reversed and remanded for further explanation the Commission's decision to  X44select 6.0 percent as the first component of the Xfactor.4D6 {O#'ԍ #C\  P6QɒP#United States Telephone Association v. FCC, No. 971469 (D.C. Cir. May 21, 1999). In light of that remand, we seek comment on whether we should continue to adjust our expense input values to reflect productivity gains. If we determine that such adjustment is appropriate, we may want to use"Y40*&&``" an alternative method of estimating productivity. We seek comment on what other measures we could use to adjust our expense data for gains in productivity. We further propose to add an inflation factor for each year based on the fixed weighted Gross Domestic Product Price  X4Index (GDPPI) for 1997 (2.1120 percent) and for 1998 (2.1429 percent).aXD6 yO4'ԍ #X\  P6G;ɒP# These inflation factors provided by the Department of Commerce were used by price cap companies in their annual tariff filings to the Commission to allow an increase in revenues for the years 1997 and 1998 under the price cap regulatory mechanism and access reform rules. a Thus, we propose a net reduction of 3.888 percent for 1997 and 3.8571 percent for 1998 when using the 6.0 percent productivity factor. We seek comment on this method for converting expenses to 1999 values.  XH4227.` ` Estimates of Corporate Operations, Customer Operations, and Plant Non X14Specific Expenses. Appendix A contains a summary of the proposed perline, permonth input figures for both plant nonspecific expenses, corporate operations, and customer operations adjusted expenses as calculated using the aforementioned methodology. We seek  X 4comment on these proposed values.H D6 yO'ԍ#C\  P6QɒP# Input values for common support service expenses are reproduced here: ` `  XXhhCqppAdjusted (#h  yO'` ` Aggregate USOA AccountqXppExpense Inputs(#  yOm'` ` Other PP & E (6510) hhCq$pp0.07(#h  yO5'` ` Network Operations (6530)qpp1.35  yO'` ` Marketing (6610) hhCXqX pp0.02(#  yO'` ` Service Exp./Customer Operations (6620)h pp1.07 (#  yO'` ` Exec., Planning, G&A (6700)qXpp2.60(#  yO'` ` Totals PerLine PerMonthqX $ pp5.11(#  ?4#Xx6X@DQX@# H   X 4X\\(#- VIII. CAPITAL COSTS ׃  X4228. ` ` In this section, we address the inputs in the model related to capital costs:  Xy4depreciation, cost of capital, and annual charge factors.  XK'A.Depreciation    X' 1. ` ` Background    X4229. ` ` #Xj\  P6G;ynXP#We now consider the inputs related to the calculation of depreciation expenses. "Z0*&&``6"  X4The model uses "adjusted projected lives" to recover the current costs of the assets.D6 {Oy'ԍ 1997 Further Notice, 12 FCC Rcd at 18570, para. 149. The projected life of an asset is the asset's expected service life at installation, reflecting not only the physical life of the equipment, but also the obsolescence associated with the replacement of older equipment with equipment that uses new technologies and forecasts of future replacements. The adjusted projected life of an asset is its projected life adjusted by its future net salvage value. Future net salvage is the percentage of the asset's value that the owner expects to obtain when  {Oc'selling the asset at the end of its useful life. Id.ĩ Under this approach, the annual depreciation charges associated with an asset are computed by  X4dividing the asset's current cost by its adjusted projected life.DD6 yO 'ԍ Depreciation charges are computed in this manner for the first year. In subsequent years, depreciation charges are computed using reserve. A shorter life will increase the annual depreciation expense.  X4230. ` ` In the Universal Service Order, the Commission concluded that "economic lives and future net salvage percentages used in calculating depreciation expense should be within  Xa4the FCCauthorized range" and use currently authorized depreciation lives.~aD6 {O'ԍ Universal Service Order, 12 FCC Rcd at 891314, para. 250 (criterion 5).~ In the 1997  XL4Further Notice, the Commission tentatively concluded that it should adopt depreciation expenses that reflect a weighted average of the rates authorized for carriers that are required  X 4to submit their rates to us.e . D6 {O'ԍ 1997 Further Notice, 12 FCC Rcd at 18571, para. 152.e The Commission also sought comment on whether adjusted projected asset lives should reflect the lives of facilities and equipment dedicated to providing only the services supported by universal service or whether the asset lives should reflect a  X 4decision to replace existing plant with plant that can provide broadband services.3 D6 {OL'ԍ Id.3 The May  X 44 Public Notice requested further information on these issues.Q R D6 {O'ԍ See Inputs Public Notice. Q    X' 2. ` ` Issues for Comment   Xl'` ` a. Method of Depreciation (#   X>4231. ` ` Before selecting values for projected life and future net salvage value, we first tentatively adopt the method of depreciation that should be used in the model, that is, how depreciation allowances should be allocated over the life of an asset. The Commission's"[0*&&``" depreciation accounting rules require carriers to use straightline equallife group  X4depreciation.D6 yOy'ԍ 47 C.F.R.  32.2000(g). Straightline depreciation is an accounting technique in which an asset's value is divided into equal parts over its useful life. The equallife group procedure subdivides assets according to age.  {O 'See Amendment of Part 31 (Uniform System of Accounts for Class A and B Companies) so as to Permit Depreciable Property to be Placed in Groups Comprised of Units with Expected Equal Life for Depreciation  {O'Under the Straight-Line Method, Report and Order, 83 FCC2d 267 (1980), recon., 87 FCC2d 916 (1981), supplemental opinion, 87 FCC2d 1112 (1981).  Both the HAI and BCPM proponents advocate the use of straightline  X4depreciation in calculating depreciation expenses.k[D6 yO'ԍ HAI June 1, 1998 comments at 14; BCPM June 1, 1998 comments at 8.k Ameritech suggests that the depreciation method used for a specific geographic area should be consistent with any studies that underlie  X4the development of economic lives or net salvage values for that same area.YD6 yO@ 'ԍ Ameritech June 12, 1998 reply comments at 23. Y GTE proposes that incumbent LECs be allowed to use depreciation lives based on the expected economic life  Xv4of the asset. v{D6 yO'ԍ GTE June 12, 1998 reply comments at 19. GTE urges the Commission to allow incumbent LECs to use the same depreciation rates and salvage values as they use for financial reporting or, in the alternative, to establish a range based on the depreciation rates and salvage values used by interexchange carriers and competitive LECs for their financial reporting. Because the Commission's rules require the use of straightline depreciation, rather than a more accelerated depreciation method, we tentatively conclude that this method, which is used for all Commissionproposed depreciation, is also appropriate for use in the  X14high cost support mechanism. We seek comment on this tentative conclusion.  X 4` ` b. Depreciation Lives and Future Net Salvage Percentages  X 4232. ` ` In estimating depreciation expenses, the model uses the projected lives and  X 4future net salvage percentages for the asset accounts in Part 32 of the Commission's rules.M c D6 {O'ԍ See 47 C.F.R.  32.2000(j)M Traditionally, the projected lives and future net salvage values used in setting a carrier's rates have been determined in a triennial review process involving the state commission, the Commission, and the carrier. In order to simplify this process, the Commission has prescribed ranges of acceptable values for projected lives and future net salvage  XK4percentages._K D6 {O 'ԍ See 47 C.F.R.  32.2000(g)(iii). _ The Commission's prescribed ranges reflect the weighted average asset life for  X44regulated telecommunications providers. These ranges are treated as safe harbors, such that carriers that incorporate values within the ranges into their depreciation filings will not be challenged by the Commission. Carriers that submit life and salvage values outside of the"\0*&&``"  X4prescribed range must justify their submissions with additional documentation and support.(D6 {Oy'ԍ The Commission has proposed streamlining the depreciation prescription process by, inter alia, expanding the prescribed range for the digital switching plant account and eliminating salvage from the  {O 'depreciation process. See 1998 Biennial Regulatory Review Review of Depreciation Requirements for  {O'Incumbent Local Exchange Carriers, Notice of Proposed Rulemaking, CC Docket No. 98137, 13 FCC Rcd  yO'20542 (1998). ( Commission authorized depreciation lives are not only estimates of the physical lives of assets, but also reflect the impact of technological obsolescence and forecasts of equipment  X4replacement. We believe that this process of combining statistical analysis of historical information with forecasts of equipment replacement generates forwardlooking projected lives that are reasonable estimates of economic lives and, therefore, are appropriate measures of depreciation.  XH4233. ` ` In the 1997 Further Notice, the Commission tentatively concluded that it should adopt depreciation expenses that reflect a weighted average of the rates authorized for  X 4carriers that are required to submit their rates to us.p ~D6 {OK'ԍ 1997 Further Notice, 12 FCC Rcd at 18571, para. 152.p The values submitted by the HAI sponsors essentially reflect such a weighted average. The HAI values represent the weighted  X 4average depreciation lives and net salvage percentages from 76 study areas.V D6 yO'ԍ HAI June 1, 1998 comments at 10.V According to the HAI sponsors, these depreciation lives and salvage values reflect the experience of the incumbent LEC in each of these study areas in retiring plant, and its projected plans for future  X 4retirements.3 D6 {O'ԍ Id.3  X{4234. ` ` We tentatively conclude that HAI's values represent the best forwardlooking  Xd4estimates of depreciation lives and net salvage percentages.hd2 D6 yOG'ԍ The proposed values for these inputs are listed in Appendix A.h We seek comment on this tentative conclusion. Generally, these values fall within the ranges prescribed by the Commission for projected lives and net salvage percentages. Although the HAI values for  X4four account categories fall outside of the Commission's prescribed ranges, D6 yO 'ԍ HAI's lives and salvage values fall within the Commission's prescribed ranges with the exception of values for four accounts: Digital Circuit Equipment; Garage Work Equipment; Operator Systems; and Poles. these values still reflect the weighted average of projected lives and net salvage percentages that were approved  X4by the Commission and therefore are consistent with the approach proposed in the 1997  X4Further Notice. As noted above, the fact that an approved value falls outside of the prescribed range simply means that the carrier that proposed the value was required to provide additional justification to the Commission for this value. We are satisfied that HAI calculated"]0*&&``\" its proposed rates using the proper underlying depreciation factors and that HAI's  X4documentation supports the selection of these values.  X4235. ` ` We disagree with the BCPM sponsors and other incumbent LECs that the Commission's prescribed ranges are not appropriate for determining depreciation rates in a  X4competitive environment.D6 yO'ԍ Aliant June 1, 1998 comments at 34; Ameritech June 1, 1998 comments at 4; BCPM June 1, 1998 comments at 1113; GTE June 1, 1998 comments at 1516; Southwestern June 1, 1998 comments at 910. These parties argue that rapid changes in technology and the opening of local telecommunications markets to competition shorten asset lives significantly  X_4beyond what the Commission has prescribed._ D6 yO0 'ԍ BCPM June 1, 1998 comments at 12; Southwestern June 1, 1998 comments at 17; GTE June 1, 1998 comments at 16; Ameritech June 1, 1998 comments at 4. The BCPM sponsors claim that these factors cause existing equipment to become obsolete at a faster pace, thus reducing the overall  X14economic value of the assets more quickly.M1xD6 yOZ'ԍ BCPM June 1, 1998 comments at 910.M We agree with the HAI sponsors that there is no evidence to support the claim that increased competition or advances in technology require the use of shorter depreciation lives in the model than are currently prescribed by the  X 4Commission.J D6 yO'ԍ HAI June 1, 1998 comments at 13.J The Commission's prescribed lives are not based solely on the engineered life of an asset, but also consider the impacts of technological change and obsolescence. We note that the depreciation values we tentatively adopt are generally at the lower end of the prescribed range. We further note that although the average depreciation rate for an incumbent LEC's Total Plant in Service is approximately seven percent, incumbent LECs are retiring plant at a four percent rate. This difference has allowed depreciation reserves to increase so that the depreciation reserveratio is greater than 50 percent. We tentatively conclude that the existence of this difference implies that the prescribed lives are shorter than the engineered lives of these assets. In addition, this difference provides a buffer against technological change and competitive risk for the immediate future. We therefore tentatively conclude that the Commission's prescribed ranges are appropriate to determine depreciation rates for the model. We seek comment on these tentative conclusions.   X4236. ` ` We tentatively decline to adopt the values for projected lives and net salvage percentages submitted by the BCPM proponents. The BCPM proponents based their default values for projected lives and salvage on a LEC industry data survey requesting forward X|4looking values.H|D6 yO#'ԍ BCPM Dec. 11 submission at 80.H With regard to projected lives, the BCPM values generally fall outside of "|^( 0*&&``"  X4the Commission's prescribed ranges.D6 yOy'ԍ The eight categories in which BCPM's values fall outside required ranges for projected lives were: Digital Circuit Equipment; Digital Switching; Aerial CableMetallic; Aerial CableNonMetallic; Underground CableMetallic; Underground CableNonMetallic; Buried CableMetallic; and Buried CableNonMetallic. The two categories in which BCPM's values fall outside required ranges for net salvage percentage were Digital Circuit Equipment and Poles. Because the BCPM sponsors fail to introduce sufficient evidence supporting their values, we tentatively decline to accept their approach.  X4 The BCPM proponents submitted values for projected life that are significantly shorter than  X4the already shortened Commission's prescribed life ranges.HxD6 yO 'ԍ BCPM Dec. 11 submission at 80.H This is significant because BCPM's values that fall outside of the prescribed ranges represent accounts that reflect the overwhelming majority of plant investment, thus potentially triggering a dramatic increase in support. We seek comment on this assessment.  XH' B.Cost of Capital   X 4237. ` ` The cost of capital represents the annual percentage rate of returnX D6 yO'ԍ Rate of return is the percentage which a telephone carrier is authorized to earn on its rate base. For example, if the rate of return is 11.25% and the rate base is $1 million, the carrier is authorized to earn $112,500. that a company's debtholders and equity holders require as compensation for providing the debt and  X 4equity capital that a company uses to finance its assets.R\ ( D6 {O'ԍ See Local Exchange Carriers' Rates, Terms, and Conditions for Expanded Interconnection Through  {O'Physical Collocation for Special Access and Switched Transport, Second Report and Order, CC Docket No. 93 yOY'316212 FCC Rcd 18370, 18765 (1997). R In the Universal Service Order, the Commission concluded that the current federal rate of return of 11.25 percent is a reasonable  X 4rate of return by which to determine forwardlooking costs.g L D6 {O'ԍ Universal Service Order, 12 FCC Rcd at 8913, para. 250.g  X4238. ` ` The HAI proponents have submitted data indicating that the incumbent LEC's  X{4cost of capital is 10.01 percent, not the current 11.25 percent federal rate of return.J{D6 yO 'ԍ HAI June 1, 1998 comments at 13.J The HAI proponents also contend that certain state commissions have determined that even lower  XM4costs of capital are appropriate.:MnD6 {Ol#'ԍ Id. at 13.: The BCPM proponents advocate a cost o f capital rate of  X6411.36 percent.B6D6 yO%'ԍ BCPM Dec. 11 submission.B"6_0*&&``U"Ԍ X4ԙ239. ` ` We find that both BCPM and HAI proponents have failed to make an adequate showing to justify rates that differ from the current 11.25 percent federal rate of return. We tentatively conclude, therefore, that the current rate is reasonable for determining the cost of universal service. If the Commission, in a rate represcription order, adopts a different rate of return, we tentatively conclude the model should use the more recently determined rate of return. We seek comment on these tentative conclusions.  X_' C.Annual Charge Factors   X14240. ` ` Incumbent LECs develop cost factors, called "annual charge factors," to determine the dollar amount of recurring costs associated with acquiring and using particular pieces of investment for a period of one year. Incumbent LECs develop these annual charge factors for each category of investment required. The annual charge factor is the sum of depreciation, cost of capital, adjustments to include taxes on equity, and maintenance costs.  X 4241. ` ` To develop annual charge factors, the BCPM proponents propose a model with  X4useradjustable inputs to calculate the depreciation and cost of capital rates for each account.HD6 yO 'ԍ BCPM Dec. 11 submission at 80.H The BCPM proponents state that this accountbyaccount process was designed to recognize  Xb4that all of the major accounts have, inter alia, differing economic lives and salvage values that  XM4lead to distinct capital costs.$MXD6 {OV'ԍ Id. BCPM's model includes all of the methodologies that are in practice today, including: Deferred taxes; Midyear, Beginning Year, and End Year placing conventions; GompertzMakeham Survival Curves; Future Net Salvage Values; Equal Life Group Methods; and others. The model also incorporates separate Cost  {O'of Debt and Equity rates, along with the Debt to Equity ratio. Id. HAI's model is also user adjustable and reflects the sum for  X64the three inputs: depreciation, cost of capital, and maintenance costs.G6DD6 yO+'ԍ HAI Dec. 11 submission at 41.G   X4242. ` ` Because the synthesis model uses HAI's expense module, with modifications,  X4we tentatively conclude that HAI's annual charge factor should be used.D6 yOv'ԍ The expense module contains the expense values including, plant specific maintenance ratios, and the algorithms that determine monthly cost perline given the results of all other modules. We believe that HAI's annual charge factor is consistent with other inputs used in the model adopted by the Commission, and therefore easier to implement. We seek comment on this analysis and our tentative decision to use HAI's annual charge factor.  X~4 IX. OTHER ISSUES RELATED TO THE HIGH COST MECHANISM ׃  XP' A. Alternatives to the ForwardLooking Cost Model "P`, 0*&&``"Ԍ X4ԙ243. ` ` It is our expectation that the model outputs will be fully verified in time for implementation on January 1, 2000, and we remain firmly committed to the idea that support based on forwardlooking costs will provide the best assurance of predictable, specific, and sufficient support as competition develops. In the unlikely event that the model is not ready for timely implementation, however, we seek comment on how the Commission might determine support levels without resort to a forwardlooking cost model. Commenters addressing this issue should specifically describe how their proposal will generate sufficient support to meet the goals of section 254, even as competition develops in the local exchange.  X1' B. Proposed Modification to Procedures for Distinguishing Rural and NonRural  X 'Companies   X ' 1. ` ` Background   X 4244. ` ` In the Universal Service Order, the Commission determined that rural and nonrural carriers will receive federal universal service support determined by separate  X4mechanisms, at least until January 1, 2001.sD6 {O 'ԍ Universal Service Order, 12 FCC Rcd at 8927, para. 273.s The Commission stated that it would define rural carriers as those carriers that meet the statutory definition of a rural telephone company  Xd4in section 153(37) of the Communications Act.dZD6 {Oo'ԍ See 47 U.S.C.  153(37); Universal Service Order, 12 FCC Rcd at 8944, para. 310. Under this definition, a "local exchange carrier operating entity" is deemed a "rural telephone company" to the extent that such entity X` ` (A) provides common carrier service to any local exchange carrier study area that does not include either  XX` `  (i) any incorporated place of 10,000 inhabitants or more, or any part thereof, based on the most recently available population statistics of the Bureau of the Census; or ` XX` `  (ii) any territory, incorporated or unincorporated, included in an urbanized area, as defined by the Bureau of the Census as of August 10, 1993; ` X` ` (B) provides telephone exchange service, including exchange access, to fewer than 50,000 access lines;  X` ` (C) provides telephone exchange service to any local exchange carrier study area with fewer than 100,000 access lines; or  X` ` (D) has less than 15 percent of its access lines in communities of more than 50,000 on the date of enactment of the Telecommunications Act of 1996.   X 4245. ` ` In addition, the Commission determined that LECs should selfcertify their" a0*&&``b"  X4status as a rural company each year to the Commission and their state commission.jD6 {Oy'ԍ Universal Service Order, 12 FCC Rcd at 894344, para. 310.j On September 23, 1997, the Common Carrier Bureau (Bureau) released a Public Notice requiring carriers seeking to be classified as rural telephone companies to file a letter with the  X4Commission by April 30 of each year certifying that they meet the statutory definition.ZD6 {O'ԍ SelfCertification as a Rural Telephone Company, Public Notice, DA 971748 (rel. Sept. 23, 1997) (Self {O'Certification Public Notice). The  X4SelfCertification Public Notice requires a LEC certifying as a rural carrier to explain how it  X4meets at least one of the four criteria set forth in the statutory definition.\D6 {O 'ԍ See 47 U.S.C.  153(37).` (#\ On March 16, 1999, the Bureau released a Public Notice revising the annual deadline for LECs seeking to be classified as rural carriers to July 1 of each year.  X3' 2. ` ` Issues for Comment   X 4246. ` ` On June 22, 1998, the Accounting Policy Division released a Public Notice with a list of the approximately 1,400 carriers that had certified as rural carriers as of April  X 430, 1998.jZ HD6 yO'ԍ Commission Acknowledges Receipt of Letters SelfCertifying LECs as Rural Telephone Companies,  {O'Public Notice, DA 981205 (rel. June 22, 1998). Under current procedures, each of these carriers would have to  yOb'submit another rural certification by July 1, 1999.q(#j Because a vast majority of the carriers certifying as rural serve under 100,000 access lines, we tentatively conclude that we should adopt new filing requirements for carriers filing rural selfcertification letters. We propose that carriers who serve under 100,000 access  X4lines should not have to file the annual rural certification letter unless their status has changed  X{4since their last filing. {j D6 yO'ԍ The National Exchange Carrier Association, Inc. (NECA) has requested that the Commission eliminate the annual rural certification process. NECA states that the majority of carriers that meet the rural definition are small LECs with limited resources, and whose status is not likely to change. Letter from Richard A. Askoff,  yO'NECA to Irene Flannery, Chief, Accounting Policy Division, FCC, dated April 9, 1999. xxX(#x We believe that this is a better approach because the overwhelming majority of the companies that filed rural certification letters qualified as rural telephone companies because they provide service to fewer access lines than either the 50,000 or 100,000 line thresholds identified in the statute. Access line counts can be verified easily with publiclyavailable data. Further, this relaxation in filing requirements would lessen the burden on many rural carriers and Commission staff. We estimate that this change will  X4eliminate the filing requirement for approximately 1,380 of the carriers that filed this year. We seek comment on this proposal.  X4247. ` ` As noted above, the Commission can easily determine whether a carrier"bR 0*&&``B"  X4satisfies criteria (B) or (C) of the rural telephone company definition,ID6 yOy'ԍ 47 U.S.C.  153(37)(B), (C).I because these criteria are based on information that can be verified easily with publicly available data the number of access lines served by a carrier. In contrast, criteria (A) and (D) require additional information and analysis to verify a carrier's selfcertification as a rural company. Specifically, under criterion (A) a carrier is rural if its study area does not include "any incorporated place of 10,000 inhabitants or more" or "any territory ... in an urbanized area,"  Xv4based upon Census Bureau statistics and definitions.DvXD6 yO 'ԍ 47 U.S.C.  153(37)(A).D Under criterion (D) a carrier is rural if it had "less than 15 percent of its access lines in communities of more than 50,000 on the date  XH4of enactment of the [1996 Act]."DHD6 yO 'ԍ 47 U.S.C.  153(37)(D).D  X 4248. ` ` We tentatively conclude that, once we have clarified the meaning of "local exchange operating entity" and "communities of more than 50,000" in section 153(37), we should require carriers with more than 100,000 access lines that seek rural status to file certifications for the period beginning January 1, 2000, consistent with the Commission's interpretation of the rural telephone company definition. We seek comment on this tentative conclusion. We also seek comment on whether we should require these carriers to recertify each year (after the filing for January 1, 2000) or, in the alternative, whether they should be required to recertify only if their status has changed.  XK4249. ` ` Most of the carriers asserting rural status under criterion (A) or (D) also claim  X44rural status under the access line thresholds in criterion (B) or (C).4xD6 yO]'ԍ Many carriers claim they meet three or four of the criteria. We also note that many carriers that only cite one criterion may qualify under several criteria. In these cases, the Commission does not need additional information to verify the carrier's rural status. If a carrier serves a local exchange study area with more than 100,000 access lines, however, the Commission needs additional information about the study area to determine whether criterion (A) or (D) is met. Based on the certifications we have received, we believe that carriers have adopted differing interpretations of criterion D. We tentatively conclude that criterion A, on the other hand, by referencing Census Bureau sources, can be applied consistently without  X4further interpretation by the Commission. We seek comment on this tentative conclusion.(#(#  Xe4250. ` ` We have identified at least two issues in the rural telephone company definition for which carriers have adopted different interpretations that affect the determination of whether a carrier satisfies the requirements of criterion D. Specifically, carriers differ on whether criterion (D) should be applied on a holding company or study areabystudy area basis. For example, while most carriers have asserted that they meet the 15 percent/50,000" c0*&&``" test in criterion (D) for a particular study area because less than 15 percent of its access lines within that study area are in communities of more than 50,000, at least one carrier claims it meets this criterion for all of its study areas, because less than 15 percent of its access lines nationwide are in such communities. In order to resolve these differences, we must interpret the phrase "local exchange operating entity" in the introductory text of section 153(37).  Xv4251. ` ` We therefore seek comment on how we should interpret the phrase "local exchange operating entity" in section 153(37) of the Communications Act. Specifically, we seek comment on whether that term refers to an entity operating at the study area level or at the holding company level. Although most of the carriers certifying under subparagraph (D) have construed the term to refer to an entity at the study area level, we note that at least one state commission, in denying a carrier's request for an exemption under section 251(f)(1) of the Communications Act, viewed the exemption claim from the perspective of the national  X 4operating entity. D6 {ON'ԍ See 47 U.S.C.  251(f)(1); Order Denying Motion, Docket No. M263, Iowa, Department of Commerce Utilities Board (Dec. 11, 1996). We also request information on how states have construed the rural telephone company definition in exercising their authority under section 251(f)(1) and section  X 4214(e)(2) of the Act.R "D6 yOz'ԍ 47 U.S.C.  214(e)(2), 251(f)(1).R  Xy4252. ` ` Carriers also have used different interpretations of the phrase "communities of  Xb4more than 50,000" in criteria (D) of the rural telephone company definition.DbD6 yO'ԍ 47 U.S.C.  153(37)(D).D Some carriers have used Census Bureau statistics for legally incorporated localities, consolidated cities, and censusdesignated places, to identify communities of more than 50,000. Other carriers have provided lists of communities without identifying the source of the designation or the population information. Some carriers have attempted to distinguish between rural communities and communities that may be characterized as urban or suburban. One carrier, for example, based its analysis of its service territories on the Commission's definition of  X4"rural area" in section 54.5 of the Commission's rules.BD6 yO'ԍ Section 54.5 provides the following definition of rural area: XA "rural area" is a nonmetropolitan county or county equivalent, as defined in the Office of Management and Budget's (OMB) Revised Standards for Defining Metropolitan Areas in the 1990s and identifiable from the most recent Metropolitan Statistical Area (MSA) list released by OMB, or any contiguous nonurban Census Tract or Block Numbered Area within an MSAlisted metropolitan county identified in the most recent Goldsmith Modification published by the Office of Rural Health Policy of the U.S. Department of Health and Human Services."  47 C.F.R.  54.5.  The carrier calculated its percentage"d0*&&`` " of rural/nonrural lines by determining whether each of its wire centers is associated with a metropolitan statistical area (MSA). If so, these lines were considered to be urban, unless the  X4wire center has rural pockets, as defined by the most recent Goldsmith Modification.HD6 {OK'ԍ See 47 C.F.R.  54.5.H  X4253. ` ` We seek comment on how we should interpret the phrase "communities of more than 50,000" in section 153(37) of the Act. We seek comment on whether we should define communities of more than 50,000 by using Census Bureau statistics for legally incorporated localities, consolidated cities, and censusdesignated places. In the alternative, we seek comment on whether we should distinguish between rural and nonrural communities in applying criterion D of section 153(37). Specifically, we seek comment on whether we should use the methodology in section 54.5 of the Commission's rules to determine whether a community is in a rural area. We also seek comment on other methods of defining communities with populations greater than 50,000 for purposes of applying criterion D.  X 4254. ` ` As noted above, states apply the definition of rural telephone company in determining whether a rural telephone company is entitled to an exemption under section 251(f)(1) of the Act and in determining, under section 214(e)(2) of the Act, whether to designate more than one carrier as an eligible telecommunications carrier in an area served by  Xb4a rural telephone company.RbZD6 yOm'ԍ 47 U.S.C.  214(e)(2), 251(f)(1).R Although the Commission used the rural telephone company definition to distinguish between rural and nonrural carriers for purposes of calculating universal service support, there is no statutory requirement that it do so. The Commission adopted the Joint Board's recommendation to allow rural carriers to receive support based on embedded cost for at least three years, because, as compared to large LECs, rural carriers generally serve fewer subscribers, serve more sparsely populated areas, and do not generally  X4benefit as much from economies of scale and scope.gD6 {Os'ԍ Universal Service Order, 12 FCC Rcd at 8936, para. 294.g The Commission also noted that for many rural carriers, universal service support provides a large share of the carriers' revenues, and thus, any sudden change in the support mechanisms may disproportionately affect rural  X4carriers' operations.g|D6 {O'ԍ Universal Service Order, 12 FCC Rcd at 8936, para. 294.g We seek comment on whether the Commission should reconsider its decision to use the rural telephone company definition to distinguish between rural and nonrural carriers for purposes of calculating universal service support. That is, we seek comment on whether there are differences between our universal service policies and the competitive policies underlying sections 251(f)(1) and 214(e)(2) that would justify definitions of "rural telephone company" and "rural carrier" that differ. " e0*&&``"Ԍ X4255. ` ` Finally, we address a necessary procedural matter. Currently, carriers are required to file rural certifications by July 1, 1999 to be classified as rural for January 1, 2000. Given our tentative conclusions above that we should modify the current filing requirements for rural certification, including eliminating the filing requirement for most carriers that have filed previously, we move the July 1, 1999 filing deadline to October 15, 1999.  X_4  X. PROCEDURAL MATTERS AND ORDERING CLAUSE ׃  X14 A. Ex Parte Presentations   X 4256. ` ` This is a permitbutdisclose noticeandcomment rulemaking proceeding. Ex  X 4parte presentations are permitted, except during the Sunshine Agenda period, provided that  X 4they are disclosed as provided in Commission's rules.h D6 {OS'ԍ See generally 47 C.F.R.  1.1202, 1.1203, 1.1206.h  X ' B. Initial Regulatory Flexibility Act   X~4257. ` ` As required by the Regulatory Flexibility Act (RFA),XZ~ZD6 {O'ԍ See 5 U.S.C.  603. The RFA, see 5 U.S.C.  601 et seq., has been amended by the Contract with America Advancement Act of 1996, Pub. L. No. 104121, 110 Stat. 847 (1996) (CWAAA). Title II of the CWAAA is the Small Business Regulatory Enforcement Fairness Act of 1996 (SBREFA).X the Commission has prepared this Initial Regulatory Flexibility Analysis (IRFA) of the possible significant economic impact on small entities by the proposals in this Further Notice. Written public comments are requested on the IRFA. These comments must be filed in accordance with the same filing deadlines as comments on the rest of this Further Notice, and should have a separate and distinct heading designating them as responses to the IRFA. The Commission will send a copy of this Further Notice, including the IRFA, to the Chief Counsel for  X4Advocacy of the Small Business Administration (SBA) in accordance with the RFA.I|D6 {O 'ԍ See 5 U.S.C.  603(a).I In addition, the Further Notice and IRFA (or summaries thereof) will be published in the Federal Register.  X4258. ` ` Need for and Objectives of Proposed Rules. In the Universal Service Order, the Commission adopted a plan for universal service support for rural, insular, and high cost areas to replace longstanding federal subsidies to incumbent local telephone companies with explicit, competitively neutral federal universal service mechanisms. In doing so, the Commission adopted the recommendation of the Joint Board that an eligible carrier's support should be based upon the forwardlooking economic cost of constructing and operating the"f0*&&`` " networks facilities and functions used to provide the services supported by the federal universal service mechanism.  X4259. ` ` Our plan to adopt a mechanism to estimate forwardlooking cost has proceeded in two stages. On October 28, 1998, the Commission completed the first stage of this proceeding: the selection of the model platform. The platform encompasses the aspects of the model that are essentially fixed, primarily assumptions about the design of the network and network engineering. In this Further Notice we move toward completion of the second stage of this proceeding, by proposing input values for the cost model, such as the cost of cables, switches and other network components, in addition to various capital cost parameters. In addition, we propose adoption of a road surrogate algorithm to determine the location of customers and a data set of customer locations. This Further Notice also seeks comment on other issues related to the federal high cost mechanism, including alternatives to the forwardlooking cost model and modifications to the procedures for distinguishing rural and nonrural companies.  X4260. ` ` Legal Basis: The proposed action is supported by sections 4(i), 4(j), 201205, 254, and 403 of the Communications Act of 1934, as amended, 47 U.S.C.  154(i), 154(j), 201205, 254, and 403.  X64261. ` ` Description and Estimate of the Number of Small Entities to which the Further  X!4Notice will Apply.  X4262. ` ` The RFA generally defines "small entity" as having the same meaning as the  X4term "small business," "small organization," and "small government jurisdiction."?D6 yOW'ԍ 5 U.S.C.  601(6).? In addition, the term "small business" has the same meaning as the term "small business concern" under the Small Business Act, unless the Commission has developed one or more definitions  X4that are appropriate to its activities.#XD6 yO'ԍ 5 U.S.C.  601(3) (incorporating by reference the definition of "small business concern" in 5 U.S.C.  632). Pursuant to 5 U.S.C.  601(3), the statutory definition of a small business applies "unless an agency after consultation with the Office of Advocacy of the Small Business Administration and after opportunity for public comment, establishes one or more definitions of such term which are appropriate to the activities of the agency and publishes such definition in the Federal Register."# Under the Small Business Act, a "small business concern" is one that: (1) is independently owned and operated; (2) is not dominant in its field  Xk4of operation; and (3) meets any additional criteria established by the SBA.kD6 {O$#'ԍ 15 U.S.C.  632. See, e.g., Brown Transport Truckload, Inc. v. Southern Wipers, Inc., 176 B.R. 82 (N.D. Ga. 1994). The SBA has defined a small business for Standard Industrial Classification (SIC) category 4813 (Telephone Communications Except Radiotelephone) to be small entities when they have no more than"=gb 0*&&``8"  X41,500 employees.AD6 yOy'ԍ 13 C.F.R.  121.201.A  X4263. ` ` The most reliable source of information regarding the total number of certain  X4common carriers appears to be data the Commission publishes annually in its Carrier Locator report, derived from filings made in connection with the Telecommunications Relay Service  X4(TRS).XD6 {O'ԍ Carrier Locator: Interstate Service Providers, Figure 1 (Jan. 1999) (Carrier Locator). See also 47  {Ob 'C.F.R.  64.601 et seq.  Xa4264. ` ` Although some affected incumbent LECs may have 1,500 or fewer employees, we do not believe that such entities should be considered small entities within the meaning of the RFA because they are either dominant in their field of operations or are not independently owned and operated, and therefore by definition not "small entities" or "small business concerns" under the RFA. Accordingly, our use of the terms, "small entities" and "small businesses" does not encompass incumbent LECs. Out of an abundance of caution, however, for regulatory flexibility analysis purposes, we will separately consider small incumbent LECs within this analysis and use the term "small incumbent LECs" to refer to any incumbent LEC  X 4that arguably might be defined by the SBA as "small business concerns."& D6 {O'ԍ See 13 C.F.R.  121.201, Standard Industrial Classification (SIC) 4813. Since the time of the  {O'Commission's 1996 decision, Implementation of the Local Competition Provisions in the Telecommunications Act  {O'of 1996, First Report and Order, 11 FCC Rcd 15499, 1614445 (1996), the Commission has consistently addressed in its regulatory flexibility analyses the impact of its rules on such incumbent LECs.  X{4265. ` ` Local Exchange Carriers. Neither the Commission nor SBA has developed a definition of small local exchange carriers. The closest applicable definition for these carriertypes under SBA rules is for telephone communications companies other than radiotelephone  X64(wireless) companies.P6D6 yO'ԍ 13 C.F.R.  121.210, SIC Code 4813.P The most reliable source of information regarding the number of these carriers nationwide of which we are aware appears to be data that we collect annually in  X4connection with the TRS.J2 D6 {O'ԍ Carrier Locator at Fig. 1.J According to our most recent data, there are 1,410 LECs.J D6 {O}!'ԍ Carrier Locator at Fig. 1.J Although it seems certain that some of these carriers are not independently owned and operated, or have more than 1,500 employees, we are unable at this time to estimate with greater precision the number of these carriers that would qualify as small business concerns under SBA's definition. Consequently, we estimate that there are fewer than 1,410 small"hV 0*&&``5"  X4entity LECs that may be affected by the proposals adopted in this Further Notice. We also note that, with the exception of a modification in reporting requirements, the proposals &Fadopted in this  X4Further Notice apply only to larger &F#  "nonrural" LECs.BD6 {OK'ԍ See supra para. 3.B  X4 266. ` ` Description of Projected Reporting, Recordkeeping, and Other Compliance  X4Requirements.  Xc4 267. ` ` On June 22, 1998, the Accounting Policy Division released a Public Notice with a list of the approximately 1,400 carriers that had certified as rural carriers as of April  X5430, 1998.jZ5ZD6 yO@ 'ԍ Commission Acknowledges Receipt of Letters SelfCertifying LECs as Rural Telephone Companies,  {O 'Public Notice, DA 981205 (rel. June 22, 1998). Under current procedures, each of these carriers would have to  yO 'submit another rural certification by July 1, 1999.q(#j Because a vast majority of the carriers certifying as rural serve under 100,000 access lines, we tentatively conclude that we should adopt new filing requirements for carriers filing rural selfcertification letters. We propose that carriers who serve under 100,000 access  X 4lines should not have to file the annual rural certification letter unless their status has changed  X 4since their last filing.  |D6 yO'ԍ The National Exchange Carrier Association, Inc. (NECA) has requested that the Commission eliminate the annual rural certification process. NECA states that the majority of carriers that meet the rural definition are small LECs with limited resources, and whose status is not likely to change. Letter from Richard A. Askoff,  yO^'NECA to Irene Flannery, Chief, Accounting Policy Division, FCC, dated April 9, 1999. xxX(#x We believe that this is a better approach because the overwhelming majority of the companies that filed rural certification letters qualified as rural telephone companies because they provide service to fewer access lines than either the 50,000 or 100,000 line thresholds identified in the statute. Access line counts can be verified easily with publiclyavailable data. Further, this relaxation in filing requirements would lessen the burden on many rural carriers and Commission staff. We estimate that this change will  XO4eliminate the filing requirement for approximately 1,380 of the carriers that filed this year.  X!4 268. ` ` We tentatively conclude that, once we have clarified the meaning of "local exchange operating entity" and "communities of more than 50,000" in section 153(37), we should require carriers with more than 100,000 access lines that seek rural status to file certifications for the period beginning January 1, 2000, consistent with the Commission's interpretation of the rural telephone company definition. We also seek comment on whether we should require these carriers to recertify each year (after the filing for January 1, 2000) or, in the alternative, whether they should be required to recertify only if their status has changed.  XR4 269. ` ` In addition, we address a necessary procedural matter. Currently, carriers are required to file rural certifications by July 1, 1999 to be classified as rural for January 1,";id 0*&&``E" 2000. Given our tentative conclusions above that we should modify the current filing requirements for rural certification, including eliminating the filing requirement for most" i0*&&``" carriers that have filed previously, we propose moving the July 1, 1999 filing deadline to" i0*&&``" October 15, 1999." i0*&&``"Ԍ X4ԙ 270. ` ` Steps Taken to Minimize Significant Economic Impact on Small Entities and  X4Significant Alternatives Considered. Throughout the Further Notice, we seek comment on the tentative conclusions that we propose. In addition, we believe that the reporting modifications that are proposed above will reduce the burden on rural LECs. As noted, we propose that carriers serving fewer access lines than either the 50,000 or 100,000 line thresholds should not be required to file annual rural certification letters unless their status has changed since their last filing.  XL4271. ` ` Federal Rules That May Overlap, Duplicate or Conflict with the Proposed  X74Rule. None.  X ' C. Initial Paperwork Reduction Act Analysis   X 4272. ` ` This Further Notice contains a proposed information collection. As part of its continuing effort to reduce paperwork burdens, we invite the general public and the Office of Management and Budget (OMB) to take this opportunity to comment on the information collections contained in this Further Notice, as required by the Paperwork Reduction Act of  X41995, Pub. L. No. 10413.<XD6 yO'ԍ A supporting statement, prepared in accordance with the Paperwork Reduction Act, that details the Commission's estimates with respect to the burdens imposed by the proposals in this Further Notice is available from the Commission or from the Office of Management and Budget.< Public and agency comments are due at the same time as other comments on this Further Notice; OMB comments are due 60 days from date of publication of this Further Notice in the Federal Register. Comments should address: (a) whether the proposed collection of information is necessary for the proper performance of the functions of the Commission, including whether the information shall have practical utility; (b) the accuracy of the Commission's burden estimates; (c) ways to enhance the quality, utility, and clarity of the information collected; and (d) ways to minimize the burden of the collection of information on the respondents, including the use of automated collection techniques or other form of information technology.  X' D. Deadlines and Instructions for Filing Comments   Xm4273. ` ` Pursuant to Sections 1.415 and 1.419 of the Commission's rules, 47 C.F.R.  1.415, 1.419, interested parties may file comments on or before July 2, 1999 and reply comments on or before July 16, 1999. Comments may be filed using the Commission's  X(4Electronic Comment Filing System (ECFS) or by filing paper copies. See Electronic Filing  X4of Documents in Rulemaking Proceedings, 63 Fed. Reg. 24,121 (1998).  X4274. ` ` Comments filed through the ECFS can be sent as an electronic file via the Internet to . Generally, only one copy of an electronic submission must be filed. If multiple docket or rulemaking numbers appear in the caption of"!j0*&&`` " this proceeding, however, commenters must transmit one electronic copy of the comments to each docket or rulemaking number referenced in the caption. In completing the transmittal screen, commenters should include their full name, Postal Service mailing address, and the applicable docket or rulemaking number. Parties may also submit an electronic comment by Internet email. To get filing instructions for email comments, commenters should send an email to ecfs@fcc.gov, and should include the following words in the body of the message, "get form