******************************************************** NOTICE ******************************************************** This document was converted from WordPerfect to ASCII Text format. Content from the original version of the document such as headers, footers, footnotes, endnotes, graphics, and page numbers will not show up in this text version. All text attributes such as bold, italic, underlining, etc. from the original document will not show up in this text version. Features of the original document layout such as columns, tables, line and letter spacing, pagination, and margins will not be preserved in the text version. If you need the complete document, download the WordPerfect version or Adobe Acrobat version, if available. ***************************************************************** FCC 97-391 Before the FEDERAL COMMUNICATIONS COMMISSION Washington, D.C. 20554 In the Matter of ) ) Amendment of the Commission's ) ET Docket No. 95-183 Rules Regarding the 37.0-38.6 GHz and ) RM-8553 38.6-40.0 GHz Bands ) ) ) Implementation of Section 309(j) of the ) PP Docket No. 93-253 Communications Act -- Competitive ) Bidding, 37.0-38.6 GHz and 38.6-40.0 GHz) REPORT AND ORDER AND SECOND NOTICE OF PROPOSED RULE MAKING Adopted: October 24, 1997; Released: November 3, 1997 By the Commission: TABLE OF CONTENTS Paragraph I. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . .1 II. EXECUTIVE SUMMARY. . . . . . . . . . . . . . . . . . . . . . . . 3 III. BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . .4 IV. DECISION -- SERVICE RULES A. Service Areas. . . . . . . . . . . . . . . . . . 12 B. Permissible Operations in the 39 GHz Band. . . . 18 1. Point-to-Multipoint Operations . . . . . 20 2. Mobile Operations. . . . . . . . . . . . 21 3. The Balanced Budget Act Requirements for Flexible Use26 C. Channeling Plan. . . . . . . . . . . . . . . . . 27 D. Licensing Rules 1. Eligibility. . . . . . . . . . . . . . . 29 2. License Term . . . . . . . . . . . . . . 36 3. Performance Requirements: Renewal and Build-out38 4. Spectrum Aggregation Limit . . . . . . . 51 5. Technical Rules a. Frequency Tolerance and Efficiency Standard 58 b. Antenna Requirements . . . . . . 64 c. Frequency Coordination and Power Flux Density Limit. . . . . . . . . . . . . . 66 6. Partitioning and Disaggregation. . . . . 70 7. Regulatory Status. . . . . . . . . . . . 75 E. Treatment of Incumbent 39 GHz Licensees. . . . . 77 1. Reconciling Service Areas of 39 GHz Incumbents with BTA Service Areas of New Licensees . . . . . 78 2. Repacking. . . . . . . . . . . . . . . . 81 3. Disposition of Pending 39 GHz Band Applications a. Background . . . . . . . . . . . 83 b. Processing of Pending Applications87 i. Pending Mutually Exclusive 39 GHz Applications . . . . . . . . . . 88 ii. Applications Within the 60-day Public Notice Period on November 13, 1995. . . 92 iii. Modification Applications94 iv. Applications That Are Partially Mutually Exclusive . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 V. DECISION -- COMPETITIVE BIDDING ISSUES A. Auctionability of the 39 GHz Band. . . . . . . . 98 B. Competitive Bidding Design and Procedures 1. Competitive Bidding Design . . . . . . 105 2. Applicability of Part 1, Standardized Auction Rules 108 C. Bidding Issues 1. Grouping of Licenses . . . . . . . . . 109 2. Reserve Price or Minimum Opening Bids. .111 3. Bid Increments . . . . . . . . . . . . .112 4. Stopping Rules . . . . . . . . . . . . .114 5. Activity Rules . . . . . . . . . . . . .117 6. Duration of Bidding Rounds . . . . . . .121 D. Procedural and Payment Issues 1. Short Form Applications. . . . . . . . .123 2. Amendments and Modifications . . . . . .127 3. Upfront Payments . . . . . . . . . . . .129 4. Down Payment and Full Payment. . . . . .132 5. Bid Withdrawal, Default, and Disqualification136 6. Long-Form Applications and Petitions to Deny139 E. Regulatory Safeguards 1. Transfer Disclosure Requirements . . . .141 2. Anti-Collusion Rules . . . . . . . . . .143 F. Treatment of Designated Entities 1. Overview and Objectives. . . . . . . . .145 2. Eligibility for Bidding Credits. . . . .148 a. Small Business Definition. . . .149 b. Bidding Credits. . . . . . . . .152 c. Installment Payments . . . . . .155 3. Transfer Restrictions and Unjust Enrichment Provisions160 4. Entrepreneurs' Block . . . . . . . . . .163 VI. SECOND NOTICE OF PROPOSED RULE MAKING A. Background . . . . . . . . . . . . . . . . . . .165 B. Partitioning and Disaggregation For Small Business Licensees166 VII. PROCEDURAL MATTERS A. Regulatory Flexibility Act . . . . . . . . . . .171 B. Ex Parte Rules--Non-Restricted Proceeding. . . .173 C. Comment Dates. . . . . . . . . . . . . . . . . .174 D. Initial Paperwork Reduction Act of 1995 Analysis176 E. Ordering Clauses . . . . . . . . . . . . . . . .177 F. Further Information. . . . . . . . . . . . . . .179 APPENDIX A -- PARTIES FILING COMMENTS AND REPLY COMMENTS APPENDIX B -- REGULATORY FLEXIBILITY ACT APPENDIX C -- FINAL RULES I. INTRODUCTION 1. In the Notice of Proposed Rule Making and Order in the above-captioned proceeding, we proposed to amend the rules for fixed, point-to-point microwave service in the 38.6-40.0 GHz ("39 GHz") band, and to adopt a conforming set of new rules for the virtually unused 37.0-38.6 GHz ("37 GHz") band in order to allow for the expansion of 39 GHz-type service. Since the time we made these proposals, technological developments have sparked additional applications for the frequencies in the 36- 51 GHz band that had not been proposed when we issued the NPRM and Order. For example, some entities have submitted proposals for non-terrestrial systems -- such as Sky Station International's proposed use of platforms located in the stratosphere to build a global stratospheric telecommunications system, Motorola Satellite Systems' proposed 72-satellite NGSO/FSS M-Star system, and Hughes Communications, Inc.'s proposed satellite GSO/FSS Expressay system. While we seek to create a regulatory environment that will permit the construction of these projects, we also are interested in providing sufficient flexibility for terrestrial-based licensees to provide the public with innovative services. We believe that the public interest would be served by permitting the market to decide which entrepreneurial efforts will succeed. 2. In this Report and Order, we amend Parts 1, 2, and 101 of the Commission's Rules to facilitate more effective use of the 39 GHz band, by implementing a number of improvements such as licensing by Basic Trading Areas (BTAs) and employing competitive bidding procedures as a means for choosing among mutually exclusive license applicants. In addition, we conclude that our regulatory framework should be expanded to include service rules for mobile operations in the 39 GHz band. By facilitating implementation of mobile services, 39 GHz licensees will be able to modify their service offerings quickly and efficiently to provide the services that consumers demand and that technology makes possible. Thus, 39 GHz service providers will be better positioned to respond to the dictates of the marketplace. Moreover, such flexibility will promote competition by increasing both the diversity of potential service offerings and the number of providers that can offer any service. Finally, we address those 39 GHz applications held in abeyance pursuant to the processing freeze imposed in the NPRM and Order, as modified in our subsequent Memorandum Opinion and Order. In this Second Notice of Proposed Rule Making, we seek additional comments regarding the use of partitioning and disaggregation by parties utilizing bidding credits under our competitive bidding licensing rules. By these actions, we will foster the continued development of a variety of microwave operations in the 39 GHz band, which will facilitate provision of, inter alia, communications infrastructure for commercial and private mobile radio operations and competitive wireless local telephone service. II. EXECUTIVE SUMMARY 3. In our decision today, we take a number of steps to simplify and streamline the licensing process for the 39 GHz band. What follows is a synopsis of the major aspects of our decision. A. Licensing Rules We are allotting the 39 GHz spectrum for licensing throughout the United States by BTAs (constituting 487 service areas). We are authorizing an additional six BTA-like areas, covering the following U.S. territories: American Samoa; Guam; Northern Mariana Islands; San Juan, Puerto Rico; Mayagez/Aguadilla-Ponce, Puerto Rico; and, the United States Virgin Islands. Thus, a total of 493 authorizations will be issued for each channel block in the 39 GHz band. Incumbent 39 GHz licensees, however, will be able to retain their rectangular service areas, provided they meet the build-out requirements described infra. The existing 39 GHz channeling plan -- fourteen paired 50 MHz channel blocks, with a spacing of 700 MHz between the transmit and receive frequencies -- is retained. We also retain the existing framework of license terms for 39 GHz licensees; the licensees who received their authorizations prior to August 1, 1996, will retain the license term specified in their authorizations, while all licensees receiving a license after that date will have a ten-year license term from the date of grant. For each license held, 39 GHz licensees must show that they are providing "substantial service" when they file their renewal application. All 39 GHz band licensees will receive an explicit renewal expectancy if they satisfy the "substantial service" requirement. Any entity may apply for a 39 GHz license. In addition, we are not adopting a limit on the amount of 39 GHz spectrum that can be held by a single entity. 39 GHz licensees will be able to offer a variety of services including point-to-point, point-to- multipoint, and mobile operations (with implementation of mobile operations occuring after the Commission completes a rulemaking proceeding addressing inter-licensee and inter- service interference issues). All 39 GHz licensees are permitted to partition and/or disaggregate their licenses. B. Technical Rules We are eliminating the requirement that licensees meet the current standard for frequency tolerance. Protection against objectionable interference will be ensured by the existing emission limits. Licensees will not be required, as a general rule, to deploy Category A antennas. We are also eliminating the alternative Category B antenna option to permit use of other types of antennas. We note, however, that users of other than Category A antennas will be required to upgrade such antennas if they pose interference problems. C. Disposition of Pending 39 GHz Applications We dismiss without prejudice major amendments filed on or after November 13, 1995. We dismiss without prejudice all pending mutually exclusive applications, unless the mutual exclusivity was resolved by an amendment of right filed before December 15, 1995. We dismiss without prejudice all applications that had not been placed on public notice or completed the 60-day cut-off period as of November 13, 1995. D.Competitive Bidding We will award 39 GHz licenses through competitive bidding. We conclude that a series of auctions of several channels at a time is the fairest, fastest and most administratively efficient way of distributing these licenses. Simultaneous multiple round bidding and a simultaneous stopping rule will be used. We also adopt the Milgrom-Wilson activity rule. Applicants will apply for the 39 GHz auction by filing a short-form application (FCC Form 175) and paying an upfront payment. Upfront payments will be determined by the Wireless Telecommunications Bureau and announced by Public Notice prior to the auction. At the conclusion of the auction, winning bidders must supplement their upfront payments sufficient to bring the deposit up to 20 percent of their winning bid and file their long-form applications. Small businesses with revenues of not more than $40 million are eligible for a 25 percent bidding credit, and very small businesses with average annual gross revenues of not more than $15 million are eligible for a 35 percent bidding credit on all 39 GHz licenses. These bidding credits are not cumulative. E. Second Notice of Proposed Rule Making We request comments on the use of partitioning and disaggregation by parties taking advantage of bidding credits under our competitive bidding licensing rules. III. BACKGROUND 4. On September 9, 1994, the Telecommunications Industry Association ("TIA") filed a Petition for Rule Making seeking to increase the amount of spectrum available for operations contemplated in the 39 GHz band. Currently, the 39 GHz band is allocated for non-Government, fixed, point-to-point microwave communications. When we initiated this proceeding with the December 15, 1995, NPRM and Order, we acknowledged that the demand for use of 39 GHz spectrum was increasing dramatically due to the projected need for point-to-point spectrum by Personal Communications Services ("PCS") and cellular licensees, and by providers who require or furnish other types of point-to-point services. We proposed a regulatory framework to improve the 39 GHz band licensing process and to allow interested parties to expand their operations to the 37 GHz band. One of our main goals in initiating this proceeding was to facilitate operations that provide communications infrastructure, such as "backhaul" and "backbone" communications links. We received 34 comments and 17 reply comments in response to the NPRM and Order. 5. In the NPRM and Order, we also looked at permitting an array of fixed services in the 37 GHz band. Subsequently, Motorola and other satellite entities expressed their interest in this band as well, and similar interests were expressed for other high gigahertz bands. Accordingly, we decided to address the 36.0-51.0 GHz bands in a unified manner, and in a Notice of Proposed Rulemaking adopted earlier this year, we sought comment on our proposals for these frequency bands. However, because the 39 GHz band is significantly licensed and subject to additional applications for license, we believe that it is in the public interest to refine our rules at this time to allow existing and new licensees to maximize the array of services they can provide to the public. Indeed, the record in this proceeding demonstrates that our initial view of the potential uses for 39 GHz spectrum was too narrow. In addition to providing support for existing services (e.g., broadband PCS, cellular, and other commercial and private mobile radio operations), 39 GHz band providers plan to use this spectrum to satisfy needs for a host of other fixed services, such as: (1) wireless local loops, (2) call termination or origination services to long distance companies, (3) connection of the customers of a competitive access provider ("CAP") or a local exchange carrier ("LEC") to its fiber rings, (4) connection and interconnection services to private networks operated by business and government as well as other institutions, (5) Internet access, and (6) cable headend applications. In some cases, 39 GHz band licensees are already using the spectrum for such purposes. 6. Several satellite entities commenting in the 36-51 GHz proceeding contend that we should delay taking final action on the 39 GHz band until after the World Radio Conference - 97 (WRC-97). For example, in its comments in the 36-51 GHz proceeding, Lockheed Martin Corporation (Lockheed Martin) states that our proposed band plan for spectrum between 36 - 51.4 GHz is fraught with risk of rejection through the WRC-97 process. As a result, Lockheed Martin argues that it would be unreasonable for us to take further action in this proceeding without the assurance that our entire plan will receive the necessary international endorsement. Any action now, it maintains, will adversely affect the interests of those services (particularly, satellite) that rely on international allocations. In addition, some satellite commenters argue that because high density fixed services are deployed only in the 38.5 - 39.5 GHz band in other parts of the world, we should designate the 39.5 - 40.0 portion of the 39 GHz band for satellite services. Such a designation, they maintain, would be consistent with international and domestic allocations. 7. We are not persuaded by these commenters that a delay in concluding this proceeding or changing the service designation for the 39.5 - 40.0 GHz band would be in the public interest. Current allocations for this segment of the 39 GHz band contain both fixed and satellite services. The actions we take here today do not alter those allocations. We further note that our actions here do not constrain our ability to later modify the Table of Allocations with respect to this segment of the band, or our overall band segmentation plan proposed in the 36-51 GHz proceeding, should future events (e.g., WRC-97 decisions) require a different result. 8. Moreover, we note there is wide support for the premise that the types of fixed and satellite services likely to be offered in spectrum above 36 GHz will not be able to share the same spectrum blocks. There have been numerous presentations by various terrestrial fixed service entities supporting this notion, and this conclusion has been reiterated in the records of both this and the 36-51 GHz proceeding. Similarly, various satellite entities have indirectly conceded that sharing between terrestrial and satellite is not likely in bands above 36 GHz, even though they recommend that the sharing option continue to be pursued. For example, many of the comments in the 36-51 GHz proceeding express doubt about the feasibility of our proposal to establish an "underlay" license for terrestrial services in those bands that would be designated for satellite services. Underlying this concern is the recognition of the potential for interference between the two types of operations. Against this backdrop, we conclude that some form of band segmentation will be required to accommodate planned services in the spectrum above 36 GHz. The current use and allocation of the 39 GHz band is consistent with this result, and therefore, we see no basis for delaying this proceeding. 9. Further, of the bands comprising our 36-51 GHz segmentation plan, the 39 GHz band is the only one involving current licensees. Indeed, we continue to authorize additional operations in the band. Over the last four years, we have licensed 55 entities to render a variety of fixed point-to-point services in more than 200 metropolitan areas throughout the country. As a result, in some of these areas all 39 GHz spectrum has been assigned. In fact, many of these authorized stations operate in the 39.5 - 40.0 GHz portion of the 39 GHz band. 10. Given the significant level of licensing in the 39 GHz band, we are presented with the challenging question of how to accommodate commercial satellite operations in the 39.5 - 40.0 segment of the band. We are not persuaded that redesignation of that portion of the 39 GHz band for satellite services only, as recommended by satellite proponents, is the most prudent course of action at this time. In light of the near unanimous concern about the feasibility of terrestrial-satellite sharing, it would appear that grandfathering existing 39 GHz terrestrial licensees would not be a viable option. While relocation or repacking of existing licensees might be possible, we believe such an alternative would be extremely burdensome to terrestrial licensees presently operating within that portion of the band. For example, re- packing the existing licensees in the 39.5 - 40.0 GHz portion to some other portion of the 39 GHz band could require existing licensees to change frequencies, purchase new equipment and/or perform a major retrofit. In addition, a new terrestrial frequency plan would be required -- one based on a different transmit/receive frequency separation. Such a change would impose significant costs on equipment manufacturers and licensees. Furthermore, a change in the frequency plan would require further rulemaking, which would result in additional delay in the deployment of new services to the public. 11. In addition, this repacking alternative could impair the ability of existing licensees to provide continued service to their customers. According to several 39 GHz licensees, a broad base of customers have been established and a variety of services are being offered. In addition, the 39 GHz companies are making major strides toward becoming effective competitors to incumbent local exchange carriers. Given the likelihood of inter-service interference and the rapid implementation of service by 39 GHz licensees, the satellite industry's request for delayed action in this proceeding and a spectrum designation change is not persuasive. Again, should future events dictate that a different course of action with respect to the 39 GHz band is warranted, nothing that we have done here will prevent us from taking the appropriate action at that time. IV. DECISION -- SERVICE RULES A. Service Areas 12. Background. The current licensing process in the 39 GHz band allows each licensee to define its own service area. In the NPRM and Order, we proposed to license prospectively all channel blocks in the 39 GHz band using BTAs. Alternatively, we asked whether some or all of the channel blocks should be made available for licensing over significantly larger geographic areas, or whether smaller geographic areas should be used to meet the needs of those who might desire individual links. 13. Discussion. After careful consideration of the record, we will adopt our proposal to license new 39 GHz licenses based on pre-defined geographic areas rather than the applicant-defined rectangular areas currently authorized in the 39 GHz band. Use of pre-determined service areas will provide a more orderly structure for the licensing process. Moreover, Commission-defined service areas will foster efficient utilization of 39 GHz spectrum in an expeditious manner. Our experience in the 39 GHz band has shown that while applicant-defined service areas may give entities the opportunity to apply only for that area which they intend to serve, this opportunity does not result in expeditious licensing of the spectrum because the mutually exclusive situations are complex and often overlapping. In contrast, the use of Commission-defined service areas should facilitate rapid delivery of services to the public. For these aforementioned reasons, we therefore reject the suggestion by some commenters that we continue licensing the 39 GHz band by permitting applicants to define their own service areas. For those interested in tailoring a service area to other smaller or larger markets, we note that today we also are proposing service rules to allow partitioning and disaggregation by 39 GHz licensees. 14. In choosing the most appropriate definition for 39 GHz service areas, we observe that our conclusion that this band is auctionable (explained below in Section V-A) requires us to apply the criteria of Section 309(j)(4)(C) of the Communications Act of 1934, as amended, ("Act" or "Communications Act"). This Section mandates that we consider certain factors when establishing service areas for auctionable services. The first of these criteria is that the service area promote an equitable distribution of licenses and services among geographic areas. We believe that use of BTAs fulfills this objective because they are intended to represent the natural flow of commerce, comprising areas within which consumers have a community of interest. As a result, we believe that BTAs are representative of the geographic areas in which the types of services envisioned for the 39 GHz band are likely to be provided. The second criterion we are required to consider is whether the service area is appropriate to provide economic opportunity for a wide variety of applicants, including small businesses, rural telephone companies, and businesses owned by members of minority groups and women. We believe that BTAs are sufficiently large to accommodate the array of services proposed for the 39 GHz band in a manner which provides opportunities for a variety of licensees. For example, broadband PCS licensees use BTAs or Major Trading Areas ("MTAs," which are regional aggregations of BTAs), as their primary service areas, and may seek to use 39 GHz band spectrum for backbone and backhaul. Thus, the BTA-sized service areas for support spectrum will be compatible with the primary service areas defined for broadband PCS providers. We also believe that other services, such as telephony, would find sufficient population within BTAs to support the pursuit of various business opportunities. In addition, we believe that other services anticipated for 39 GHz spectrum, such as wireless local loop, competitive access, local exchange, and Internet access, are of a local nature for which use of BTAs also would be appropriate. Moreover, we believe that use of BTAs as the service area definition for the 39 GHz band will also satisfy the third criterion of Section 309(j)(4)(C), which requires that we establish service areas in a manner which will promote investment in and rapid deployment of new technologies and services. Accordingly, we agree with the commenters who advocate the use of BTAs for licensing the 39 GHz band. 15. We disagree with those commenters who contend that the service areas for the 39 GHz band should be based on larger geographic areas. We believe that BTAs offer a sufficiently large service area to allow applicants flexibility in designing a system to maximize population coverage and to take advantage of economies of scale necessary to support a successful operation. Moreover, to the extent that 39 GHz licensees desire to provide service over a larger geographic region, the rules we adopt today will allow them to aggregate BTAs. We do not believe, however, nor does the record indicate, that the majority of licensees will seek to provide service over vast geographic regions. Thus, we believe that larger service areas would be inappropriate for the 39 GHz band. 16. Finally, although GTE expressed some concern that any Rand McNally licensing agreement should be reasonable, we do not believe that the existence of Rand McNally's copyright interest in the BTA listings will present an impediment to use of these areas by 39 GHz band licensees. We expect that potential licensees and Rand McNally will execute a licensing agreement similar to those already undertaken in other contexts. In particular, Rand McNally has already licensed the use of its copyrighted MTA/BTA listing and maps for a number of services, such as PCS, 800 MHz Special Mobile Radio (SMR) service, and Local Multipoint Distribution Service ("LMDS"), and the company has also reached an agreement with the American Mobile Telecommunications Association ("AMTA") for a blanket copyright license for the conditional use of copyrighted material in the 900 MHz SMR service. These agreements authorize the conditional use of Rand McNally's copyrighted material in connection with these particular services, require interested persons using the material to include a legend on reproductions (as specified in the license agreement) indicating Rand McNally's ownership, and provide for a payment of a license fee to Rand McNally. 17. While the services to be provided in the 39 GHz band do not appear to be covered by any blanket copyright license agreement, we will take the approach we used in MM Docket No. 94-131 and leave it to the parties to negotiate an arrangement with Rand McNally to make use of its listings. The 39 GHz licensees and other parties interested in using the copyrighted materials may, of course, negotiate their own licensing arrangement with Rand McNally, but we encourage interested parties and Rand McNally to explore the possibility of entering into blanket license agreements, similar to those referenced above, to cover the 39 GHz band. We note that a 39 GHz BTA authorization grantee who does not obtain a copyright license through a blanket license agreement (or some other arrangement) with Rand McNally for use of the copyrighted material may not rely on the grant of a BTA-based authorization from the Commission as a defense to any claim of copyright infringement brought by Rand McNally against such grantee. The MTA/BTA Listings, the MTA/BTA Map and the license agreements noted above are available for public inspection at the Wireless Telecommunications Bureau, Reference Room, Room 5322, 2025 M Street, N.W., Washington, D.C., 20554. B. Permissible Operations in the 39 GHz Band 18. Background. In the NPRM and Order, we raised questions about expanding the array of services provided in the 39 GHz band to include point-to-multipoint and mobile operations. These services are permitted under the Table of Allocations for this spectrum band, however, we have not previously promulgated rules which would govern point-to-multipoint and mobile operations. The only type of service authorized under our current service rules is point-to-point operations. The 39 GHz band is currently being licensed and used for non-Government, terrestrial-based, fixed, point-to-point microwave service. In addition, there are no satellite operations in the 39 GHz band. Accordingly, our efforts to improve the licensing and service rules for non-Government service in this band are not affected by any existing assignments under different allocations. We take note of the fact that the 39 GHz band contains the following allocations: Domestically, the 38.6-39.5 GHz portion of the band is allocated for non-Government use to provide fixed and mobile services and FSS (space-to-Earth) on a primary basis. In addition to these primary allocations, the 39.5-40.0 GHz portion of the band is allocated on a shared basis between Government and non-Government users on a primary basis for FSS (space-to-Earth) and Mobile-Satellite Service ("MSS") (space-to-Earth). Government use of 39.5-40.0 GHz is limited to military systems. Internationally, the 39 GHz band is allocated on a co-primary basis for fixed and mobile services and FSS (space-to-Earth), and on a secondary basis for use by the Earth-Exploration Satellite service (space-to-Earth). The 39.5-40.0 GHz portion of the band is also allocated on a primary basis for MSS (space-to-Earth). 19. Accordingly, in the NPRM and Order, we requested public comment on whether we should also establish service rules which would permit point-to-multipoint and mobile services. Many parties commenting in this proceeding have encouraged us to allow them flexibility to determine the best uses of the 39 GHz band; in particular, they have requested authority to provide point-to-multipoint and mobile service, as the technology to provide these services becomes available. We have considered these comments in connection with the recent amendment to Section 303 of the Communications Act concerning criteria we must consider when permitting flexible use of the electromagnetic spectrum, which was enacted after the NPRM and Order and the comment period had been completed in this proceeding. 1. Point-to-Multipoint Operations 20. Discussion. Given the fact that the 39 GHz service is still in its early stages of development, we believe that it is imperative that we not take any regulatory actions that would hamper the service's continued development and growth potential. We note, as a general matter, that the type of services proposed for the 39 GHz band by the commenters can be offered on both a point-to-point and point-to- multipoint basis. Although a few commenters contend that we should defer allowing point-to-multipoint operations in this band until specific technical rules are adopted to protect against interference to point-to- point users (such as equipment specifications), there is no evidence in the record that point-to-point and point-to-multipoint operations are inherently incompatible in the same band or licensing area. Therefore, we will adopt 39 GHz rules for point-to-multipoint operations. 2. Mobile Operations 21. Discussion. We have considered the comments of several parties requesting that we establish rules to permit mobile operations in this band. WinStar argues that such flexibility would give licensees the opportunity to make use of technological advances, and would confer the benefits of these advances to subscribers. Milliwave believes that making the 39 GHz band available for a wide array of services, including mobile, will foster innovation and competition in a changing telecommunications market, stimulate infrastructure investment, job creation, and efficient spectrum use. ART suggests that although there does not appear to be an immediate demand for mobile services in the 39 GHz band, such use should not be precluded. To ensure adequate interference protection in a mobile (and point-to- multipoint) environment, ART urges the Commission to license 39 GHz spectrum under the General Wireless Communications Service ("GWCS") rules until rules are adopted for the proposed Licensed Millimeter Wave Service. 22. Parties opposing authorization of mobile services in the 39 GHz band argue that there are no technical parameters to protect both fixed and mobile operations from mutual interference. In particular, TIA argues that mobile equipment now available in the marketplace is designed such that it would receive interference from fixed stations, that coordination is difficult between fixed stations and mobile facilities, that international spectrum harmony would be disrupted, and that manufacturing economies of scale would be disrupted. TIA also argues that advocates for mobile services fail to present documentation that mobile systems would work in the band. ANS and PCIA argue that fixed and mobile operations cannot co-exist because there is significant threat of interference. Harris argues that co-location of fixed and mobile service systems, and the expected increased density of 39 GHz transmitters, combined with their expected evolution toward point-to-multipoint configuration, makes sharing with mobile services unrealistic. BizTel, while promoting flexible service concepts, nevertheless argues that it is questionable whether mobile services could exist on a co-primary basis with fixed uses. It further argues that any mobile service use should be authorized on a secondary basis only. 23. After careful review of the record evidence, we have decided to permit implementation of mobile operations in the 39 GHz band. Permitting such flexibility will enable providers to modify their offerings quickly and efficiently to provide the services that consumers demand and that technology makes possible. Thus, providers will be better positioned to respond to the dictates of the marketplace. Moreover, such flexibility will promote competition by increasing both the diversity of potential service offerings and the number of providers that can offer any service. Thus, the requirements of Section 303(y) are fulfilled because both technological development and investment therein will be stimulated. Moreover, this broad view of the character of 39 GHz service comports with the development of the industry thus far because parties are developing a wide variety of fixed services and, as discussed earlier in this section, some parties may be developing, or planning to develop, mobile services technology capable of operating without interference to fixed facilities in this band. Accordingly, we are convinced that establishing rules for mobile operations will best serve the public interest. In addition, we observe that in a number of other contexts we have authorized licensees to provide both mobile and fixed operations within the same service -- e.g., GWCS, the Commercial Mobile Radio Services ("CMRS"), and the Interactive Video and Data Service ("IVDS"). 24. For the most part, the objections that have been raised to mobile operations in this proceeding are misplaced. Since the service is licensed on an exclusive, area-wide basis, (whether by incumbents' rectangular service areas or by new licensees' BTAs), the issue of technical compatibility of fixed and mobile operations within a service area is one that can and should be resolved by the licensee. To the extent that a licensee has the technological wherewithal to provide one or the other, or both, types of services, the licensee will do so in a manner that the market directs. Governmental direction in this service is unnecessary except to the extent that the operations of one licensee may interfere with that of another. Even if mobile operations are not now compatible with fixed operations within a licensee's service area, if adequate protections against inter-licensee interference are in place, a failure to authorize mobile use in this spectrum might delay implementation of a dual (mobile and fixed) operation when it does become feasible. Accordingly, we agree that 39 GHz licensees should have the flexibility to provide mobile services. 25. We recognize that inter-licensee interference issues are magnified under this approach. For example, a mobile unit operating in a fixed microwave environment on the same frequency calls for a different interference analysis and a more difficult resolution than the operation of two or more fixed microwave systems on the identical frequency in the same vicinity. In addition, the Department of Defense has stated that it has plans to implement satellite downlinks at 39.5-40.5 GHz in the future. NASA has also identified 39.5 - 40.0 GHz as a possible space research band to accommodate future earth-to-space wideband data requirements. Such plans, however, should not affect the continued development of the 39 GHz band for non-Government use. We believe that it is likely that military satellite systems will be able to share with non-Government terrestrial and/or fixed satellite systems, provided that the Government receiving Earth stations are limited in number. We intend to address these interference issues in a future, separate proceeding that will focus on developing inter-licensee and inter- service standards and criteria. Until these standards and criteria are adopted we will not permit mobile operations in the 39 GHz band. 3. The Balanced Budget Act Requirements for Flexible Use 26. The Balanced Budget Act authorizes us to allocate spectrum so as to provide flexible use, if such use is consistent with international agreements to which the United States is a party and we find that: (1) such an allocation would be in the public interest; (2) such use would not deter investment in communications services and systems, or technical development; and (3) such use would not result in harmful interference among users. In the NPRM and Order, we sought comment on whether we should allow point-to-multipoint and mobile operations in addition to the traditional point-to-point services authorized in the 39 GHz band. As discussed supra, we find that the flexible use approach adopted herein is consistent with the new statute. Accordingly, we will permit point-to-point, point-to-multipoint and mobile operations on the 39 GHz band. However, as explained supra, we will defer mobile use until a future rulemaking proceeding can establish interference criteria. Accordingly, we find, as required by Section 303(y) of the Communications Act, as amended by the Balanced Budget Act, that no harmful interference will be caused by allowing both point-to-point and point-to-multipoint operations in the 39 GHz band. We conclude further, based on the above-mentioned comments in the record, that point-to-multipoint use will not deter investment in communications services and systems, or in technology development. To the contrary, permitting point-to-multipoint use will stimulate creative technology development and facilitate investment therein. It is in the public interest to afford 39 GHz licensees flexibility in the design of their systems to respond readily to consumer demand for their services, thus allowing the marketplace to dictate the best uses for this band. Accordingly, we find that the requirements of Section 303(y) of the Communications Act, as amended, are fulfilled to justify point-to-multipoint use of the 39 GHz band as part of a flexible use approach. While at this time, we are not determining the specific provisions for interference protection with regard to mobile use, we will adopt such requirements before permitting mobile operations in this band. C. Channeling Plan 27. Background. The existing 39 GHz channeling plan consists of fourteen paired 50 MHz channel blocks, with a spacing of 700 MHz between the transmit and receive frequencies. Within this framework, 39 GHz licensees have the flexibility to subdivide their channels in the manner they deem most appropriate to meet service demands. As discussed in the NPRM and Order, TIA, however, has proposed that licensees who subchannelize their 50 MHz channel blocks be required to conform to an underlying grid of 1.25 MHz subchannels. TIA argued that this restriction would ease frequency coordination at channel edges and at geographic boundaries. 28. Discussion. We will retain our current channel plan and we decline to adopt TIA's proposal regarding subchannelization. Adopting a standard subchannelization plan at this early stage in the development of the 39 GHz service would potentially hamper licensees' efforts to meet their customer demands and could unnecessarily impose technical and economic costs on equipment users and limit the range of services potentially available. Moreover, given the short propagation transmission characteristics at these frequencies, lack of a subchannelization plan is not likely to cause any significant coordination problems in the 39 GHz band. Furthermore, because we anticipate that one of the uses for the 39 GHz band is provision of CMRS infrastructure, we are concerned that adoption of a subchannelization plan may frustrate such use if it is inconsistent with the channeling plan for particular CMRS providers. Thus, we believe that the existing approach that allows 39 GHz licensees to freely subdivide their channel blocks will not only avoid this unintended result but also facilitate the most flexible and efficient use of 39 GHz spectrum. As we observed in the NPRM and Order, however, our decision not to adopt a standard subchannelization plan does not preclude the industry from developing its own voluntary standards in this area. D. Licensing Rules 1. Eligibility 29. Background. The issue of eligibility restrictions was first raised by TIA's original proposal that applicants receive a license for a single channel only after demonstrating their need for multiple paths within the service area. Additional channels would be authorized only if the existing channels were operating at or near expected capacity. Thus, our primary focus in the NPRM and Order was on the question of whether eligibility should be restricted to those entities who could demonstrate need for 39 GHz spectrum and the means for meeting that need. While we recognized that such restrictions are designed to weed out applicants who are financially unqualified or engaging in speculation, we tentatively concluded that use of competitive bidding would operate more effectively and efficiently in ensuring that this spectrum was put to its highest valued use. Accordingly, we declined to recommend any eligibility restrictions. 30. Discussion. Two commenters argue that eligibility restrictions should be imposed for somewhat different reasons than those suggested by the NPRM and Order. Specifically, ALTS and BizTel contend that eligibility restrictions should be imposed as a safeguard against potential anticompetitive abuses by LECs. ALTS states that we should "establish safeguards to prevent incumbent LECs from obtaining all of the desirable channel blocks in a given market and to ensure an opportunity for CLECs to obtain licenses." BizTel states that it is contrary to the public interest, and possibly an antitrust violation, for the Commission to allow a LEC with monopoly power to obtain a 39 GHz license covering any portion of its home operating territory. BizTel argues that allowing such LEC participation in bidding would "frustrate the most viable alternatives available today for deployment of competitive local telecommunications services." BizTel asserts that, at a minimum, any LEC with monopoly power should be required to certify full compliance with the "Competitive Checklist" set forth at Section 271(c)(2)(B) of the Communications Act, as a prior condition to participating in the 39 GHz auction for licenses covering any portion of its home territory. Other commenters propose that the Commission substitute its own assessment of the appropriate array of uses and users of 39 GHz spectrum for that of the marketplace. 31. In opposition, a number of other commenters contend that there is no reason to restrict eligibility of the LECs. U S West, for example, argues that neither ALTS nor BizTel provides evidence to support their assertion that LECs will impede competition. According to U S West, the result of eliminating LECs from bidding for spectrum within their respective home operating territories could be that there would be no incentive for quick and economical deployment of wireless local loop in the rural areas of their service region. Further, Pacific argues that the "Competitive Checklist" is associated with the ability of a LEC to offer inter-LATA services and has no relevance to eligibility for 39 GHz licenses. Pacific also states that a safeguard against the warehousing of spectrum by a LEC is to apply the same construction requirement on a LEC that applies to other 39 GHz licensees. 32. In addressing this eligibility issue, we inquire whether open eligibility poses a significant likelihood of substantial competitive harm in specific markets, and, if so, whether eligibility restrictions are an effective way to address that harm. This approach results in reliance on competitive market forces to guide license assignment absent a compelling showing that regulatory intervention to exclude potential participants is necessary. Such an approach is appropriate here because it best comports with our statutory guidance. When granting the Commission authority in Section 309(j)(3) to auction spectrum for the licensing of wireless services, Congress acknowledged our authority "to [specify] eligibility and other characteristics of such licenses." However, Congress specifically directed that we exercise that authority so as to "promot[e] . . . economic opportunity and competition." Congress also emphasized this pro-competitive policy in Section 257, where it articulated a "national policy" in favor of "vigorous economic competition" and the elimination of barriers to market entry by a new generation of telecommunications providers. This approach is also consistent with our analysis in the LMDS R&O. Finally, implementation of this approach is consistent with the court's treatment of eligibility issues in Cincinnati Bell. In that decision, the Court looked to statistical data and general economic theory as support for predictive judgments by the Commission such as that eligibility restrictions are required. 33. In the case of the 39 GHz band, it is unlikely that substantial anticompetitive effects would result from LEC eligibility for two primary reasons. First, increased LEC provision of services other than those provided in local exchange markets, such as point-to-point backhaul and backbone transmission, will not diminish the generally competitive environment in which those services are now available. Second, even presuming that 39 GHz licenses will enable effective provision of services that can compete with local exchange service, such as wireless local loop, incumbent LECs should have little or no incentive to acquire those licenses with the anticompetitive intent of foreclosing entry by other firms and preserving market power. An incumbent strategy of preserving expected future profits by buying 39 GHz licenses cannot succeed because there are numerous other sources of actual and potential competition. As explained above, there are many non-LEC license holders in the 39 GHz band currently, and these licensees will be able to provide services that compete with wireline local exchange. In addition, our overall 36-51 GHz band plan contemplates making available considerable additional spectrum, including substantial unencumbered spectrum, for flexible terrestrial use at frequencies close to those covered by this Order. These future licenses should enable provision of whatever competitive services can be provided with the 39 GHz licenses. Further, entry by other wireless licensees is possible as well, such as CMRS firms now authorized to provide fixed services. Moreover, the 1996 Act has set the stage for new facilities-based, wireline entrants such as interexchange carriers and competitive LECs, and non-facilities-based wireline entrants utilizing the new local competition provisions. Finally, we have now provided for one additional potential competitive option in every region of the country in the form of the 1150 MHz LMDS licensee. We have imposed an eligibility restriction preventing in-region LECs (and cable television companies) from acquiring these large LMDS licenses for three years, guaranteeing that each license will be acquired by a firm new to provision of local exchange in the service area. Therefore, these licensees also constitute potential competition for incumbent LECs providing local exchange services. Given all these competitive possibilities, it is implausible that incumbent LECs would pursue a strategy of buying 39 GHz licenses in the hope of foreclosing or delaying competition, and implausible that they would succeed if that strategy were attempted. Therefore, we find that LEC eligibility for these licenses poses no likelihood of substantial competitive harm. 34. Note that several factors, taken together, explain the distinction between our resolution of the eligibility issue here and in the case of the 1150 MHz LMDS licenses. The 1150 MHz LMDS license blocks are unusually large, making possible the provision of voice, video, data, or some combination of these services. With the possibility of providing voice cheaply as part of a set of services, the 1150 MHz LMDS license is a particularly attractive competitive option, and incumbents are particularly likely to attempt acquisition in order to prevent entry by new competitors using the LMDS license. In addition, with only one large LMDS license available per geographic area, anticompetitive preemption is quite feasible and thus the risk of such acquisition is increased. Moreover, the 39 GHz licenses being made available within the near future (i.e., within a similar time frame as the LMDS spectrum), are encumbered, while LMDS licenses are largely unencumbered. Thus, we believe, 39 GHz licenses are less likely to be acquired by incumbent LECs for anticompetitive motives. Most importantly, as noted above, given the fact that we have now provided for an additional competitive option by imposing the 1150 MHz LMDS eligibility restriction, the competitive circumstances we face in this proceeding differ from those we faced in the LMDS proceeding. Our eligibility analysis and conclusion here, in fact, are consistent with our treatment of eligibility for the small, 150 MHz, LMDS licenses. 35. Because we see no likely and substantial competitive harm flowing from LEC eligibility, we reject the argument that LECs should be required to certify compliance with the "Competitive Checklist" as a precondition to participation in the 39 GHz auction. We also note as a general matter that LEC eligibility can be expected to yield efficiency benefits if there are complementarities between the ultimate use(s) of 39 GHz spectrum and the existing LEC services when offered in the same service area. For example, LECs might be able to achieve savings not available to new entrants by taking advantage of their current infrastructure, and imposition of restrictions would prevent realization of such savings. Restrictions might also prevent incumbent LECs from experimenting with certain technology and market combinations, and preclude or delay desirable entry by incumbents into new markets. 2. License Term 36. Background. Under our previous rules, all common carrier 39 GHz licensees who were licensed before August 1, 1996 (i.e., those licensed previously under Part 21 of our Rules) were subject to a fixed license term ending February 1, 2001, regardless of the grant date of their individual licenses. Private carrier 39 GHz licensees authorized before August 1, 1996 (i.e., those licensed previously under Part 94 of our Rules), received a five-year license which would run from the date of license grant. However, both private and common carrier licenses granted on or after August 1, 1996, the effective date of the Part 101 Report and Order, have a license term not to exceed ten years. In addition, neither the former fixed microwave rules in Parts 21 and 94, nor the current ones in the new Part 101, expressly provide for a renewal expectancy for common carrier or private carrier 39 GHz licensees. 37. Discussion. Two parties argue that we should increase the term to ten years for incumbents who have received a shorter period under the rules that predated those adopted in the Part 101 Report and Order. We decline to take this action. When we adopted the Part 101 rules, we decided to conform the license terms of common carrier and private carrier 39 GHz licensees on a going forward basis. We did not, therefore, alter the conditions under which incumbent licensees had taken their licenses, and we left in place a bifurcated approach toward renewal that would exist until the incumbents' current licensing cycle runs its course. We are unpersuaded that this approach, adopted only a year ago, should be altered. 3. Performance Requirements: Renewal and Build-out 38. Background--Renewal. We noted in the NPRM and Order that both cellular and PCS licensees receive a renewal expectancy, and we proposed adopting a similar standard in this proceeding. Under the PCS standard, a licensee receives a renewal expectancy upon demonstration that substantial service has been rendered during the license term and that there has been compliance with applicable Commission rules and policies and the Communications Act. In the broadband PCS context, we observed that a renewal expectancy will provide the PCS community with a stable regulatory environment that is conducive to investment, thereby fostering the rapid development of that service. Commenters support adopting a renewal expectancy for the 39 GHz service for similar reasons, as they recognize the benefits that such a presumption offers. 39. Background--Build-out requirements. Incumbent 39 GHz licensees are currently subject to the build-out requirements of Part 101 of our Rules, which require that at least one link be constructed in a licensee's geographic service area within eighteen months of the date of license grant. In the NPRM and Order, we proposed new build-out requirements for incumbent 39 GHz licensees in order to ensure that the spectrum was being used to provide service to the public. Because of our concern that such licenses be used to provide service to the public, we solicited comment on our proposal to allow incumbent 39 GHz licensees to retain their licenses only by meeting specific construction and loading requirements. We suggested three basic construction build-out options, each of which depended upon a specific number of fixed stations to be built within the licensees' geographic service area. The build- out options were each intended to ensure a minimum level of service. While the proposals represented a significant departure from the current build-out rules applicable to these licensees, in the NPRM and Order we stated that the purpose of these proposed measures was to minimize speculation without harming existing 39 GHz licensees who are responsibly developing the spectrum they have been assigned. 40. We also requested comment on build-out requirements for new licensees authorized pursuant to the competitive bidding rules promulgated herein. In the NPRM and Order, we observed that the Communications Act requires that any regulations implementing a competitive bidding system include performance requirements -- such as appropriate deadlines and penalties for performance failures -- to ensure prompt delivery of service to rural areas, to prevent stockpiling or warehousing of spectrum by licensees, and to promote investment in and rapid deployment of new technologies and services. The build-out requirements that apply to other fixed, microwave services licensed on a link-by-link basis, as well as those requirements that apply to mobiles services, did not appear appropriate for a fixed, geographically licensed service like 39 GHz. Accordingly, we asked for comment on what other methods we might employ to ensure that licensees are using their spectrum, servicing rural areas, and enabling the provision of new services to the public. We suggested that these goals might be accomplished if we required licensees to demonstrate substantial service in their service areas. As we noted in the NPRM and Order, the use of a substantial service standard has precedent in our Rules -- for example, Section 24.203(b) gives certain PCS licensees the option of meeting their build-out requirement by making a substantial service showing. 41. Discussion. The performance rules we are adopting for the 39 GHz band require each licensee to prove substantial service in order to achieve license renewal. We arrived at this approach based on two factors. First, the approach satisfies the dictates of Section 309(j)(4)(B) of the Communications Act, which requires the Commission to adopt effective safeguards and performance requirements for licensees in connection with any competitive bidding system. We believe that the requirements we establish herein will fulfill this obligation, because a license will be assigned in the first instance through competitive bidding, with the result that it will be assigned efficiently to an entity that has shown, by its willingness to pay market value, its willingness to put the license to its best use. 42. Second, the approach we are taking with regard to performance rules is also based on the record in this proceeding, which strongly supports giving 39 GHz licensees a significant degree of flexibility in meeting their performance requirement. As described above, the types of service available from 39 GHz providers is tremendously varied, and the service promises to develop in ways we cannot predict at this time. Thus, an inflexible performance requirement might impair innovation and unnecessarily limit the types of service offerings 39 GHz licensees can provide. Permitting licensees to demonstrate that they are meeting the goals of a performance requirement with a showing tailored to their particular type of operation avoids this pitfall. Moreover, our examples of presumed substantial service, based on a specific number of links per population standard, provides licensees with a degree of certainty of regarding their license requirements. Accordingly, we believe that the performance requirements we establish herein will permit flexibility in system design and market development, yet provide a clear and expeditious accounting of spectrum use by licensees to ensure that service is indeed being provided to the public. 43. We decline to adopt any of the build-out proposals we made for incumbent 39 GHz licensees in the NPRM and Order. The first option would have required licensees to meet a specific build-out benchmark. We have considered a number of possibilities for such a benchmark, and we have rejected those that appear infeasible. Our principal proposal fell into this category. We had proposed to require any licensee to construct and put in operation at least four links per 100 square kilometers of their service area within 18 months of adoption of a Report and Order in this proceeding. We are persuaded by several commenters' arguments that such a build-out requirement would be unduly restrictive and burdensome, thus unnecessarily limiting licensees' service options. For the same reasons, we reject a variant of our principal proposal, which would have combined the alternatives discussed below with an 18-month requirement to construct a certain number of links per 100 kilometers. 44. The other two alternatives we had proposed for providing licensees with specific build-out benchmarks are also problematic. One alternative provided for a specific number of links, increasing over time, per geographic area served by each licensee. This alternative does not adequately take into account the differences among licensees. Under this requirement, a licensee in a sparsely populated BTA would have to build an operation that could provide the same level of service as a licensee of a metropolitan BTA. Such an approach would result in either an overly burdensome requirement for the licensee of the smaller market or a very lenient and almost meaningless requirement for the licensee of the metropolitan BTA. Moreover, since market size is a reasonable proxy for gauging the appropriate comparative levels of spectrum use, we agree with the consensus of the commenters that any build-out standard should therefore be based on market population or population density. This approach is, in fact, an underpinning of standards that have been adopted for CMRS services such as PCS and SMR. 45. The second alternative would have required licensees to construct a specific number of link installations based on the market's population. In the case of 39 GHz, however, the services to be offered generally will be customized for each subscriber, and, for the most part, each subscriber will have equipment dedicated to its location. Moreover, 39 GHz licensees are not likely to install equipment until they receive an order. We further note that some commenters argue that adoption of a concrete standard would discourage growth, stymie new development, and deter investment in the 39 GHz arena. Accordingly, we are concerned that a requirement for a fixed number of links may interfere with the market decisions of a particular licensee and its customers. 46. We conclude that a showing of substantial service, the approach we proposed for new 39 GHz licensees, should be applied to both incumbent and new licensees in the band. This approach will permit flexibility in system design and market development, while ensuring that service is being provided to the public. Although a finding of substantial service will depend upon the particular type of service offered by the licensee, one example of a substantial service showing for a traditional point-to-point licensee might consist of four links per million population within a service area. This revised performance standard should ensure that meaningful service will be provided without unduly restricting service offerings. 47. One of the principal problems that commenters identified with our build-out proposals was that they required too much too soon. We recognize that licensees must be given a reasonable amount of time to meet a performance requirement. Parties, particularly incumbent licensees, also argued that different build-out standards were unfair and would place an unreasonable burden on their ability to respond to market demands. Accordingly, we have decided that in order to impose the least regulatory burden on licensees as possible, but to remain consistent with our statutory responsibilities, we will combine the showing traditionally required for build-out and the showing required to acquire a renewal expectancy into one showing at the time of renewal. We believe this will give licensees a sufficient opportunity to construct their systems. We believe that applying a similar performance requirement to all licensees at the license renewal point will help establish a level playing field without compromising the goals of ensuring efficient spectrum use and expeditious provision of service to the public. 48. We believe that the deadline for compliance that we are adopting should negate concerns about a performance requirement being imposed too early in the license term. To establish a viable operation, we recognize that licensees must have sufficient time in which to develop market plans, secure necessary financing, develop and incorporate new technology in their systems, accommodate equipment manufacturers' production schedules, and build a customer base. Our approach takes these practicalities into account. We recognize that existing licensees who obtained their licenses before August 1, 1996, will receive a somewhat shorter period from the date of this decision to meet the construction threshold (i.e., about four years). Extending the build-out deadline past renewal, however, would not be prudent nor would it appear to be consistent with the objectives of Section 309(j) of the Communications Act. Moreover, these incumbents already have had at least a year, and in some cases more than two years, in which to set in motion their business plans. Thus, we do not believe this approach will adversely affect incumbent 39 GHz licensees. 49. We concur with those commenters who advocate adopting a renewal expectancy for all licensees in the 39 GHz band. As with cellular and broadband PCS licensees, affording 39 GHz providers the opportunity to earn a renewal expectancy will facilitate investment for their industry, provide stability over the long run, and better serve the public by reducing the possibility that proven operators will be replaced with less effective licensees. Like broadband PCS, we anticipate that such benefits to the 39 GHz community will promote the rapid development of the service. For such benefits to flow to the public in the most effective manner possible, the opportunity for a renewal expectancy should be available to all 39 GHz licensees, not just those licensed under the rules amended by this decision. Thus, we are not limiting this opportunity to newly licensed 39 GHz providers. The build-out/renewal requirements established herein will, if met, serve to give the incumbent licensee a renewal expectancy as well. 50. We are not persuaded by the arguments of some commenters that a build-out requirement should not be imposed because potential users of the 39 GHz band, such as broadband PCS licensees, are subject to other construction requirements. As we discussed supra, we do not believe that use of the 39 GHz spectrum will be limited to such uses. Moreover, our decision herein to adopt a requirement of substantial service by renewal will ensure that our 39 GHz rules do not work at cross purposes with build-out requirements to which broadband PCS licensees and others already are subject. 3. Spectrum Aggregation Limit 51. Background. In the NPRM and Order, we sought general comment on whether there should be a limit on the aggregation of 39 GHz channels within a single BTA. We also requested comment on whether the 39 GHz service represents a discrete market. In the event that we concluded that this service did constitute a discrete market, we indicated that a spectrum aggregation limit might be advisable to ensure that there would be an adequate number of licenses available to meet the needs of broadband PCS licensees and other competitors in the wireless marketplace. 52. Discussion. We agree with those commenters who oppose a 39 GHz spectrum aggregation limit. The record strongly supports the conclusion that 39 GHz licensees will participate in a number of broad markets, consisting of a host of short-range fixed communications provided by many operators who employ a range of different, but substitutable, technologies (both radio and wire). Therefore, we are not concerned with guaranteeing a particular number of 39 GHz competitors or with creating competition within the 39 GHz band. Moreover, as we noted above, there is no evidence that the 1400 megahertz of spectrum in the 39 GHz band is particularly important for, or unusually suited for, the creation of competition in two markets where market power still exists -- local telecommunications services and multi-channel video program delivery. Therefore, an aggregation limit is not needed in order to foster competition in these two markets. Indeed, a 39 GHz spectrum aggregation limit that was applicable to 39 GHz licensees might limit the ability of a licensee to bring efficient competition to these markets. 53. Although we believe that some of the 39 GHz spectrum will be used to satisfy CMRS and private mobile radio infrastructure needs, we are persuaded by the commenters that a great portion of this spectrum likely will be used to provide other wireless services, e.g., local area network ("LAN")-to- LAN, local access for long distance providers, wireless augmentations to CAPs' networks, and other high capacity data transmission networks. This is evidenced by current 39 GHz operations, which are not supporting CMRS communications infrastructure but generally tend to be local private line and local bypass services. Since this arena is already being served by multiple providers using a variety of technologies, it is clear that disaggregated ownership of 39 GHz spectrum is not necessary for the competitive provision of those services. 54. We also note that even the current users of the 39 GHz band are still in the early stages of developing their services, and that the particular uses of this spectrum are still being defined by the marketplace. As indicated above, 39 GHz spectrum can be used for almost any fixed, short-range communication -- the internal parts of almost any communications system (mobile or fixed) -- or the "last mile" of any fixed system, whether for voice, data, video, or more than one of the foregoing. At this time, we believe that it would be inappropriate for us to view the output of 39 GHz spectrum as falling into any one of these categories or to find that some limit on spectrum aggregation in order to foster competition in that category is necessary. Accordingly, we do not believe that it is appropriate to restrict the amount of 39 GHz spectrum that may be licensed to any one service or entity. 55. Moreover, we conclude that there may be benefits to the public in terms of efficiencies and types of services provided if we permit aggregation of 39 GHz spectrum. For example, spectrum aggregation would allow a licensee to expand its operation and thereby lower the per unit cost of equipment and its per capita cost of providing service to subscribers. Furthermore, a 39 GHz licensee with substantial spectrum can better compete with established service providers who have large transmission capacity. In addition, we conclude that it is not likely that aggregation of 39 GHz spectrum by a single entity would lead to undue market power. We note that other service providers, such as LECs and CAPs, have some significant competitive advantages over a competitor using only 39 GHz spectrum, such as an established customer base and transmission facilities that carry much more traffic than would be possible by a 39 GHz-based facility using only, for example, 700 MHz of spectrum. In addition, other service providers are not precluded from adding fiber or radio transmission facilities to their existing networks. Moreover, we have proposed to make available additional spectrum enabling more parties to compete in many of the types of services proposed by potential 39 GHz service providers, and we plan to consider these proceedings in connection with our global upper-gigahertz band plan proceeding. Therefore, we believe that even if a single licensee controls a significant part of the 39 GHz band in a single BTA, it could not control service prices or limit competition, given the number of providers of similar or substitutable services and the variety of transmission media at their disposal. 56. We also observe that 39 GHz licensees would be unable to overcome the competitive disadvantages of operating under a spectrum aggregation limit simply by improving engineering efficiency. While an entity with limited technical capacity may strive to use its facilities in the most efficient manner possible, those same engineering techniques and procedures may be utilized by other parties to similarly increase their efficiencies. For example, one of the most discussed means of increasing transmission capacity is the use of digital compression technology. For the most part, this technology is transferable from one transmission medium to another. Therefore, while a 39 GHz service provider might be able to gain a significant increase in engineering efficiency by employing such technology, this increase in efficiency will not give it any competitive advantage, because its competitors will have the same opportunities to deploy this technology. 57. We also do not believe that a spectrum aggregation limit is warranted to ensure that there is adequate support spectrum available for broadband PCS, cellular radio, and other commercial and private mobile radio operations. While the use of the 39 GHz band may help meet these needs, such backhaul and backbone support can also be provided by using wire-based technologies and over-the-air spectrum outside the 39 GHz band (e.g., at 6, 11, 18 and 23 GHz). Given this availability of substitutable spectrum for backhaul and backbone support, coupled with the aforementioned competition that exists to 39 GHz providers of alternative types of services, we find that imposing a spectrum aggregation limit for the 39 GHz band would be contrary to the public interest. 5. Technical Rules a. Frequency Tolerance and Efficiency Standard 58. Background. In the NPRM and Order, we tentatively concluded that only those technical rules required to minimize interference between channel blocks and between service areas are needed. Thus, as a mitigating interference factor, we proposed to adopt a 0.001% frequency tolerance for equipment operating in the 39 GHz band, instead of the 0.03% tolerance standard currently required by Section 101.107 of the Rules. In order to promote more efficient use of the spectrum, we also requested comment on adding an efficiency standard to our Part 101 rules, of 1 bit per second per hertz ("bps/Hz") for new assignments in this band. 59. Discussion. Initially, we believed that this spectrum principally would be used to provide support facilities for various mobile services. As a result, we proposed technical standards intended to ensure a certain level of equipment efficiency and performance. The record, however, indicates that much wider uses are anticipated. For example, a number of commenters stated that 39 GHz facilities will be employed to provide wireless equal access, LAN-to-LAN communications, and other high capacity data transmission services. In order to accommodate these varied services and to provide 39 GHz licensees the necessary technical flexibility to meet these demands, we have determined that any benefits to be gained by adoption of the proposed standards are outweighed by the limitations they would place on the development of 39 GHz service. For these same reasons, we have reevaluated our existing frequency tolerance standard and determined that it is unnecessary, particularly in light of other interference safeguards in our rules. We note that in our 220-222 MHz proceeding we concluded that interim spectral efficiency standards were warranted. This decision stemmed from one of our specific objectives in establishing the band, i.e., to encourage the development of spectrally efficient technologies. Here, however, there is sufficient evidence that 39 GHz licensees and manufacturers are proceeding with the improvement of spectrally efficient equipment. For example, one manufacturer, [P-Com], has off-the- shelf equipment which operates at an efficiency rate of 1.25 bits per hertz, a rate which exceeds the one bit per hertz rate proposed in the NPRM and Order. Given the advancements that are already made in this area, and that more are likely to follow, we believe that a spectral efficiency standard for 39 GHz equipment is unnecessary. 60. With respect to setting a spectrum efficiency standard -- which is principally designed to ensure that the licensee's technical quality of service to its end users meets a certain level -- setting a mandatory standard could be harmful to the continued development and growth of the 39 GHz service. If we set the standard at or below what licensees would voluntarily adopt, then the standard would have no effect. If we set it above the voluntary level, then we would be imposing a cost in excess of any benefit. Moreover, consistent with our actions in other proceedings, we believe it unwise to adopt technical rules that will require updating as technological advances are made because we believe 39 GHz licensees need maximum flexibility to respond to market forces. As commenter Columbia notes, "[t]he trend toward spectrum flexibility is one of the great achievements of . . . [the FCC] and is perhaps the single most important development of the decade in encouraging innovation and imaginative service to the public." In contrast, TIA contends this "hands-off" approach is premature because it would, among other things, "unleash large numbers of incompatible operators in individual markets without adequate safeguards against harmful interference . . . and create uncertainty over potential market demand and related production and performance requirements because specific uses for the [band] are not prescribed." 61. As a general matter, whenever spectrum is exclusively assigned and licensees cannot expect to obtain additional spectrum at a price significantly below its market value, we believe that a mandatory efficiency standard is unnecessary. Under these conditions, licensees can be expected to invest voluntarily in efficient technology up to the socially optimal level, and a mandatory standard would either have no effect (if it is at or below the voluntary level) or impose unjustified costs that exceed any resulting gain. We believe that mandatory standards are beneficial if they correct for under-investment in efficiency by licensees. A licensee with a shared assignment may under-invest in efficiency because much of the gain from that investment would accrue to others. But even if a licensee has an exclusive assignment, it may choose to under-invest in efficiency if it can expand capacity by obtaining spectrum at less than the market value. 62. In the 39 GHz band, however, neither of these conditions exists; thus, we find that a mandatory efficiency standard is not necessary. Given that the 39 GHz assignments will continue to be exclusive, other licensees will be denied any "free ride" from a gain in increased efficiency. In other words, the benefits gained by an increase in efficiency (e.g., more available spectrum) are not shared by other licensees who did not contribute, as would be the case in a shared environment. There is also little likelihood that 39 GHz licensees will be able to obtain additional 39 GHz spectrum below its market value because we expect that the remaining 39 GHz band will be subject to competing interests and that the competitive bidding process will be used to assign this spectrum. Thus, competitive forces of the marketplace should cause licensees to maximize the use of their assigned channels. While 39 GHz licensees may be able to obtain additional spectrum in other bands in the future, our use of auctions to select between future mutually exclusive applications for 39 GHz spectrum should ensure that these licensees are subject to full marketplace incentives to operate efficiently. Consequently, the use of competitive bidding procedures provides additional support for our finding that an efficiency standard is unnecessary. 63. As noted in paragraph 59, we have determined that a frequency tolerance standard is unnecessary. Our basis for this view stems from our desire to provide 39 GHz licensees flexibility in the operation of their facilities and to avoid imposing unnecessary regulations. In addition, we believe such a standard could inhibit technological advances, for equipment performance is likely to be influenced by customer demand. For those that might be concerned that elimination of this standard may lead to inter-system interference, we point to our existing out of band emission requirements (emission mask) contained in Sections 101.111 of the Rules. That rule requires frequencies removed in various percentage from the center frequency to be attenuated below the mean power of the transmitter. This means that the frequencies at the outer edges of an assigned 50 MHz channel or the edge of an aggregated group of 50 MHz channels power levels will be significantly reduced such that interference to an adjacent channel licensee is unlikely. Thus, we believe that strict adherence to Section 101.111 will be as effective in controlling inter-system interference as the imposition of a frequency tolerance standard. As observed in the NPRM and Order, "the effect of requiring operations to stay within the emission mask at all time would . . . reduce the frequency tolerance to levels more restrictive than the recommended [frequency tolerance]. In addition, concerns for inter-system interference should be further eased, as we are requiring neighboring and adjacent channel licensees to engage in frequency coordination before implementation of their planned operations. b. Antenna Requirements 64. Background. In the NPRM and Order, we proposed that for any new assignments in the 39 GHz band not acquired through competitive bidding, we would restrict licensees to the use of Category A antennas, which provide a more focused antenna pattern than Category B antennas, thus allowing for greater frequency reuse. Additionally, in the event that a BTA licensee was prevented from providing communications in its service area because an incumbent licensee of a grandfathered link is using a Category B antenna, we proposed to require the incumbent licensee to replace that antenna with one meeting the Category A antenna standard or cease transmission on the interfering link. We also proposed that in the case of licenses for grandfathered links in the 39 GHz band, all rule changes would only apply to facilities that are constructed after January 1, 1998, and to replacement equipment which is installed after that date. We believed that the January 1, 1998 date for implementing these requirements would allow manufacturers adequate time to make any necessary changes to their equipment production lines and to deplete inventory. 65. Discussion. There is evidence in the record that our proposal to require 39 GHz licensees to employ only Category A antennas is too restrictive because parties are contemplating a variety of system configurations that would require different types of antennas, e.g., sectorized or wide beam units, characteristics of which would be incompatible with the standards of a Category A antenna. These models represent a more cost-effective and technically suitable alternative to traditional narrowbeam Category A antennas when deployed in a point-to-multipoint configuration. As the deployment of 39 GHz facilities increases, we expect other system configurations to be developed in which narrowbeam antennas may not be the optimal solution. While DCT argues that Category A antennas should be required because they are inherently more efficient and less prone to cause interference (DCT Comments at 29), we conclude that the need to provide 39 GHz licensees the technical flexibility to meet service demands outweighs any benefits that would ensue by adopting the requirement. Therefore, we decline to require licensees in the 39 GHz band to use Category A antennas initially. We conclude that 39 GHz licensees should be given the flexibility to employ antennas other than Category A types, provided they do not cause interference problems. Should the use of an antenna other than a Category A become the source of an interference problem, however, we will require that the licensee immediately resolve such interference by replacing the antenna with a Category A model or one with better performance characteristics. c. Frequency Coordination and Power Flux Density ("PFD") Limit 66. Background. In the NPRM and Order, we noted that existing 39 GHz licensees are using the frequency coordination procedures of former Section 21.100(d) (now Section 101.103(d)) of our Rules to avoid interference between operations in the band. To further facilitate coordination between licensees in adjoining areas, we proposed to establish a maximum field strength limit that would apply at the boundaries of each service area. Under this proposal, licensees' operations not exceeding this limit would avoid the need to complete the formal coordination process. Also, licensees could negotiate higher or lower limits or enter into other mutually beneficial agreements to facilitate efficient spectrum use near their common boundaries. Due to our lack of technical data in the 39 GHz band, we did not propose a specific PFD or field strength limit. We therefore requested industry recommendations on a reasonable limit. We also sought comment on what effect, if any, our adoption of a PFD or field strength limit would have on the appropriateness of removing the existing EIRP limit. 67. Discussion. As an initial matter, we note that the National Spectrum Management Association (NSMA) stated in its initial comments that it was evaluating processes and technical criteria necessary to formalize a frequency coordination process for the 39 GHz band. On September 4, 1996, NSMA filed Supplemental Comments providing a report on the progress made in developing frequency coordination policies and procedures for precluding harmful interference among co-channel operators in the band. According to NSMA, however, further studies must be concluded to complete formal recommendations relating to its overall 39 GHz frequency coordination process, including issues related to harmful interference that may result from adjacent channel operations. Despite the incomplete state of NSMA's evaluations, it recommends that the Commission delegate to it the principal responsibility for promulgating recommendations regarding technical procedures and criteria for 39 GHz Fixed Service frequency coordination. 68. NSMA's Supplemental Comments indicate considerable progress toward developing a process that will minimize interference in the 39 GHz band. However, there is additional work to be done which we believe should be completed before taking final action on NSMA recommendations and considering revisions to our rules. As to measures we will take in the interim, we are persuaded by the record that adoption of a PFD limit or field strength limit now would not further our goal of facilitating the growth and development of the 39 GHz spectrum. In this connection, we note that there is a lack of consensus regarding the parameters necessary to establish a reasonable and practical PFD or field strength limit. As a result, we are concerned that establishing a service area boundary PFD or field strength limit without such information may stifle the development of advanced 39 GHz technology. Thus, we decline to adopt such a standard at this time, and consequently, we need not reevaluate the current EIRP at this time. As NSMA continues to evaluate means to control inter-licensee interference, we will also be exploring this issue in a future, separate proceeding. Meanwhile, we conclude that it is in the public interest to continue to use the frequency coordination procedures outlined in Section 101.103(d) of our Rules. We describe these procedures, infra, as modified to implement certain improvements supported by the record of this proceeding. Despite the fact that licensees will not be able to rely on PFD or field strength limits to avoid the formal coordination process, we believe that our modified coordination procedures will provide licensees substantial flexibility in system design while ensuring that inter-system interference will be kept to a minimum. Our experience with other services employing frequency coordination procedures shows that those services have been successfully implemented with little delay and rarely result in unresolved frequency interference cases. For example, this process has been in use in the common carrier point-to-point microwave industry for over 20 years with few interference complaints. Given the support in the record and the past success of the process in other services, we believe 39 GHz licensees will continue to benefit from this program. 69. Under our frequency coordination procedures, 39 GHz licensees will be subject to the requirements of Section 101.103(d) of our Rules, with certain modifications. As a result, they must provide values for the appropriate parameters listed in that subsection to each neighboring BTA licensee authorized to use adjacent and co-channel frequencies. Likewise, they must provide the same information to each potentially-affected, adjacent-channel licensee in the same BTA. Coordinating parties also must supply technical information related to their subchannelization plan and system geometry. Based on the propagation characteristics of this spectrum, coordination between neighboring systems need only encompass operations located within 16 kilometers of BTA boundaries. Currently, Section 101.103(d) of our Rules gives each party that receives a coordination notification 30 days in which to respond. The record in this proceeding indicates that 30 days is an inappropriate time frame for operations in the 39 GHz band because licensees often offer service that requires much shorter installation deadlines. In order to facilitate such rapid service installation schedules, we will require that recipients of coordination notifications respond within 10 days. Each licensee must complete this coordination process prior to initiating service within its service area. Finally, participating parties should resolve any problems that develop during this process. Only unresolved frequency conflicts should be reported to the Commission. In such cases we will resolve the conflicts. We believe that the coordination approach we are adopting does not preclude licensees from entering into private agreements that mitigate interference problems. These agreements may include an arrangement to conduct a one-time blanket coordination as opposed to coordinating each individual link as they are planned for activation, or arrangements for one party to compensate another financially for modifying its operation to accommodate new installations. 5. Partitioning and Disaggregation 70. Background. In the NPRM and Order, we proposed a partitioning scheme (similar to that adopted in broadband PCS), which we believed would encourage participation by rural telephone companies. In addition to seeking comment on partitioning for rural telephone companies, we also sought comment on whether the scope of partitioning should be broadened to include all applicants seeking to utilize the 39 GHz band, similar to what we offered in the Multipoint Distribution Service (MDS) context. In particular, we sought comment on methods available to meet the needs of those who might desire individual links, smaller geographic service areas, or smaller spectrum blocks. We presented the question of whether we should allow some form of partitioning or spectrum disaggregation to facilitate market entry by entities with these specialized needs. 71. Discussion. We conclude that partitioning and disaggregation should be permitted in the 39 GHz band. We further conclude that the option of partitioning should not be limited to rural telephone companies but should be made available to all entities eligible to be licensees in the 39 GHz band, including incumbent 39 GHz licensees. We thus concur with commenters who support partitioning, and note that no parties opposed this proposal. We believe that the availability of these options will enhance 39 GHz licensees' flexibility with respect to system design and service offerings. We also believe that partitioning and disaggregation opportunities further the objectives of Section 309(j) of the Communications Act by facilitating the development of niche markets and the arrival of new entrants, including small businesses, rural telephone companies and businesses owned by members of minority groups and women. In addition, these tools will promote efficient use of 39 GHz spectrum. 72. As a result, 39 GHz licensees acquiring their licenses under the new rules established herein will be permitted to acquire partitioned and/or disaggregated licenses in either of two ways: (1) they may form bidding consortia to participate in auctions, and then partition or disaggregate the licenses won among consortia participants after grant; or (2) they may acquire partitioned or disaggregated 39 GHz licenses from other licensees through private negotiation and agreement either before or after the auction. A licensee planning to partition or disaggregate its license must first be granted the license, and the licensee and partitionee and/or disaggregatee will be required to file an assignment application. We will require that a licensee disaggregate by frequency pairs. This requirement is necessary for administrative purposes: the database necessary to track authorizations could otherwise become too cumbersome and complex and processing could become delayed or prone to error. 73. Overall, we believe that partitioning and disaggregation will promote competition in the 39 GHz service and expedite the delivery of service to the public, particularly in rural areas. Moreover, partitioning and disaggregation will help to eliminate market entry barriers pursuant to Section 257 of the Communications Act by creating smaller, less capital intensive service areas that may be more accessible to small entities. We consider partitioning and disaggregation effectively to be types of assignments, which will, therefore, require prior approval by the Commission. In authorizing partitioning and disaggregation, we will follow existing assignment procedures. The licensee must file FCC Form 702 Assignment of License signed by both the licensee and qualifying entity. The qualifying entity will also be required to file an FCC Form 430 Licensee Ownership unless a current FCC Form 430 is already on file with the Commission. In addition, any 39 GHz BTA licensees taking advantage of bidding credits and seeking to utilize these options may be subject to the restrictions on assignments or transfer of control for such entities, delineated infra. We conclude that this approach is necessary in order to ensure that partitioning and disaggregation are not used as means to circumvent such restrictions. 74. We will require the entity acquiring a license by partitioning or disaggregation to satisfy the same construction requirements as the initial licensee, regardless of when its license was acquired. Should a licensee fail to meet the construction requirements, the license will cancel automatically. The cancelled license will, if it was partitioned from a rectangular service area, revert to the BTA licensee for that channel (unless the forfeiting entity is the BTA licensee for that channel). If the forfeited license was partitioned from a BTA, the license will be auctioned. In addition, parties must comply with our current technical rules with respect to service area boundary limits and protections. Coordination and negotiation among licensees must be maintained and applied in licensing involving partitioned areas and disaggregated spectrum. Finally, under partitioning or spectrum disaggregation, an entity will be authorized to hold its license for the disaggregated spectrum or partitioned area for the remainder of the original license term. We conclude that this approach is appropriate because we should not bestow greater rights to a licensee receiving its authorization pursuant to partitioning or spectrum disaggregation than we awarded under the terms of the original license grant. 7. Regulatory Status 75. Background. In the NPRM and Order we requested comment on whether a new licensee in the 39 GHz band should be allowed to use the spectrum for private use and also to provide a common carrier service. 76. Discussion. We conclude that 39 GHz band licensees should be permitted to serve as a common carrier or as a private licensee. Further, those licensees who select common carrier regulatory status will be able to provide private service, and those licensees who select private service provider regulatory status may share the use of their facilities on a non-profit basis or may offer service on a for- profit, private carrier basis subject to Section 101.135 of the Commission's Rules. Under this scenario, licensees will elect the status of the services they wish to offer and be governed by the rules applicable to their status. Although no commenters addressed this issue, we believe our approach will promote economic efficiencies by reducing construction and operating costs associated with having to provide separate facilities. This result also is consistent with Section 101.133(a) of our Rules. E. Treatment of Incumbent 39 GHz Licensees 77. Incumbent 39 GHz licensees are those who have been licensed under the current fixed microwave rules in 47 C.F.R. Part 101, or its predecessors, Parts 21 (for common carriers) or 94 (for private carriers). Their service areas are self-defined and generally are restricted to point-to-point operations. Many of these licensees have participated as commenters in this proceeding, and include WinStar, ART, BizTel, Columbia, and a number of PCS licensees. 1. Reconciling Service Areas of 39 GHz Incumbents with BTA Service Areas of New Licensees 78. While we have decided that BTAs are appropriate for the new licensing system in the 39 GHz band, we recognize that many of the newly-licensed BTA service areas will be encumbered by incumbent 39 GHz band licensees. These incumbents are authorized in various locations throughout the country, and their rectangular service areas will occupy portions of BTAs or cross BTA boundaries. Our licensing approach toward these encumbered areas will necessarily differ depending on whether the incumbent licensee's authorization covers all or a portion of a BTA. We believe that resolution of this issue is an essential element of our goal to adopt a rational licensing approach for the 39 GHz band. After careful consideration of the concerns expressed by various commenters, we conclude that the following approaches are appropriate. 79. Where an incumbent licensee's rectangular service area occupies only a portion of a BTA, the licensee's channels will be available for application under the new competitive bidding rules, but the incumbent will retain the exclusive right to use those channels within its rectangular service area. The holder of the BTA authorization thus will be required to design its system to protect against harmful interference to the incumbent by complying with the Commission's interference protection standards. Specifically, the BTA authorization holder will be required to coordinate with the rectangular service area licensee to ensure that interference protection is provided. Such a licensing policy enables incumbents and new licensees to operate concurrently and maximizes the provision of service to the public. We note that should such an incumbent lose its authority to operate, the BTA license holder will be entitled to operate within the portion of the forfeited rectangular service areas located within its BTA, without being subject to competitive bidding. This approach best serves the public because it gives the service providers an incentive to make efficient use of available spectrum, and it ensures that any disruption of service will be remedied as quickly as possible. This licensing design is similar to that used in the MDS service. When we were amending the MDS rules, we were faced with an analogous situation arising from our decision to change the method for licensing from one that provided 35-mile zone of protection around the licensee's transmitter site to one that provided exclusive rights within a BTA. We maintained the status quo for incumbents, by continuing to recognize the sanctity of their 35-mile zone, but we provided that the holders of the new BTA authorizations would receive contingent rights to encumbered MDS spectrum within the BTA. Accordingly, if an MDS incumbent lost its authorization (by, e.g., failing to construct), the forfeited channels would revert and become part of the BTA licensee's authorization. 80. Where an authorized incumbent licensee has a rectangular service area covering an entire BTA, we will not make those channels available for "overlay" licensing in that BTA. Unlike the scenario described above, in this situation a BTA will not have areas that are currently unassigned. Since incumbents will be required to construct and operate pursuant to Commission Rules, the public should be assured of receiving service throughout the BTA without the need to license an alternative provider. 2. Repacking 81. Background. In the NPRM and Order, we asked for comment on whether incumbent facilities should be relicensed on their current frequency or whether incumbent links should be "repacked" into a different portion of the band than initially occupied. We noted that under a repacking approach, most grandfathered links would be switched to one designated channel pair, provided that mutual interference would not result. 82. Discussion. There was very little discussion by commenters on the issue of repacking. WinStar addressed this issue within its discussion of fair treatment to incumbents, by pointing out that the Commission generally does not single out incumbent licensees for treatment harsher than that given to new licensees. Specifically, WinStar stated that the Commission chose not to repack incumbents when we established a mechanism for exclusive licensing of private carrier paging systems. We agree that our general approach up to this point has been to refrain from repacking, if possible. For example, in a proceeding to provide for spectrum sharing between private land mobile services and the UHF television broadcast service, we chose not to repack existing broadcast stations because we found that the relocation of existing UHF-TV stations into the remaining portion of the UHF-TV spectrum would be costly and cause a major disruption in existing television service. Similarly, we find that repacking the 39 GHz band would also cause a significant disruption of incumbent 39 operations. As noted throughout this proceeding, we do not intend to alter or restrict significantly the operations of incumbents. Moreover, we believe that we can coordinate with the extant licenses of 39 GHz incumbents so that they will not impair our new licensing system using BTAs and 50-MHz channel blocks. Accordingly, we do not believe that repacking is necessary under these circumstances. 3. Disposition of Pending 39 GHz Band Applications a. Background 83. On November 13, 1995, the Wireless Telecommunications Bureau ("Bureau"), pursuant to delegated authority, adopted and released an Order ("Freeze Order") announcing that the Commission would no longer accept for filing any new applications for 39 GHz licenses in the Common Carrier or Operational Fixed Point-to-Point Microwave Radio Services, pending Commission action on the TIA Petition. The Freeze Order was made effective upon its release. 84. The NPRM and Order extended the freeze, providing that pending applications would be processed only if (1) they were not mutually exclusive with other applications at the time of the Bureau's November 13, 1995, Freeze Order, and (2) the 60-day period for filing mutually exclusive applications had expired prior to November 13, 1995 (i.e., the applications were "ripe"). The NPRM and Order further provided that those applications that were mutually exclusive with others as of November 13, 1995, or within the 60-day period for filing competing applications on or after November 13, 1995, would be held in abeyance for processing and disposition. In addition, amendments to these frozen applications received on or after November 13, 1995, were also held in abeyance. Moreover, applications for modification of existing 39 GHz licenses (e.g., applications to modify existing licenses for the purpose of changing the height of an antenna) filed on or after November 13, 1995, were held in abeyance, as well as amendments thereto that were filed on or after November 13, 1995. Finally, no new applications to modify existing licenses, or amendments to pending modification applications, were to be accepted for filing on or after December 15, 1995, unless they (1) did not involve any enlargement of any portion of the proposed area of operation, and (2) did not change frequency blocks, other than to delete one or more. 85. On January 16, 1996, Commco filed a Petition for Reconsideration and an Emergency Request for Stay, asking the Commission to vacate that portion of the NPRM and Order imposing an interim freeze on the processing of mutually exclusive applications to establish new facilities in the 39 GHz band, including amendments thereto, pending as of November 13, 1995. BizTel, GHZ Equipment Company, Inc. ("GEC"), and TIA filed comments in support of the Stay Request. Additionally, on January 16, 1996, DCT Communications, Inc., filed a Petition for Partial Reconsideration, requesting that the Commission process (a) minor amendments, at least those that eliminate mutual exclusivity, and (b) as-yet uncontested applications for which the 60-day period for filing mutually exclusive applications had not expired prior to the November 13, 1995, Freeze Order. 86. On January 17, 1997, we reconsidered certain aspects of our processing freeze and decided to lift the processing freeze on amendments of right filed before December 15, 1995. Thus, all applications that were amended to resolve mutual exclusivity before that date were to be processed, provided they had completed their 60-day public notice period as of November 13, 1995. In addition, we clarified that applications to modify existing 39 GHz licenses and amendments thereto were to be processed regardless of when filed, provided they neither enlarge the service area nor change the assigned frequency blocks (except to delete them). In all other respects, our decisions regarding the filing and processing of 39 GHz applications and amendments were unaffected by the reconsideration decision. A summary of other main points of the decision follows: We decided to process those amendments of right filed on or after November 13, 1995, but before December 15, 1995. We noted that all other amendments filed on or after November 13, 1995, would continue to be held in abeyance. We affirmed our decision to continue to hold in abeyance all pending mutually exclusive applications, unless the mutual exclusivity was resolved by an amendment of right filed before December 15, 1995. Where the mutual exclusivity was resolved, we expressly stated that we would process the application provided it was "ripe" as of November 13, 1995 -- i.e., it had been placed on public notice and completed the 60-day cut-off period for filing of competing applications as of November 13, 1995. We affirmed our decision to hold in abeyance all applications that had not been placed on public notice or completed the 60-day cut-off period as of November 13, 1995. b. Processing of Pending Applications 87. In view of the goals of this proceeding, e.g., to foster competition among different service providers, to promote maximum efficient use of the spectrum, and to provide efficient service to customers by improving the licensing procedure, we conclude that what follows is the best approach for processing currently pending 39 GHz license applications that were affected by the November 13, 1995, Freeze Order and the December 15, 1995, freeze. The Commission has processed: (1) those 39 GHz applications that were not mutually exclusive as of December 15, 1995, and that, as of November 13, 1995, had passed the 60-day cut-off period for filing competing applications, (2) applications to modify existing licenses ("modification applications"), or amendments to modification applications, which do not enlarge the service area or change frequency blocks, except to delete them. For the reasons that follow, we have decided to dismiss, without prejudice, all other applications that have remained subject to the freeze, i.e., (1) applications that are mutually exclusive, (2) applications that were not yet on public notice, or for which the 60-day cut-off period had not been completed prior to November 13, 1995, and (3) modification applications or amendments thereto that do not meet the criteria set out infra, in paragraph 95. These applicants may reapply under the new geographic area licensing rules established in this proceeding. i. Pending Mutually Exclusive 39 GHz Applications 88. PCS and other CMRS licensees, equipment manufacturers, and TIA ask that we process 39 GHz applications that are pending and mutually exclusive. GTE, however, urges us either to (1) dismiss the pending 39 GHz applications that we are holding in abeyance and open a new application filing window for such frequencies and licensing areas under the new rules that we adopt in this proceeding; or (2) retain those applications on file and permit other interested parties to file competing applications that will be processed pursuant to adopted competitive bidding procedures and corresponding rules for 39 GHz authorizations. Some commenters recommend a specific time frame for allowing 39 GHz license applicants to resolve mutual exclusivity, i.e., between 60 days and six months after a Report and Order is issued in this proceeding. Bachow asks that the Commission dismiss, without prejudice, any mutually exclusive applications that remain after the time for resolving mutual exclusivity passes. 89. Some commenters further ask that the Commission dismiss as defective any applications which did not limit themselves to only one specified 39 GHz channel as of November 13, 1995, or which otherwise failed to satisfy a 1994 Public Notice that described the processing procedures and rules applicable to the 39 GHz band. Under this approach, any remaining applicants that are still subject to mutual exclusivity would be allowed to file amendments to reduce their proposed service area contours or otherwise enter into settlement agreements to resolve their conflicts. 90. We have determined that the best approach for processing pending mutually exclusive applications is to dismiss them without prejudice, and to allow these applicants to submit new applications under the competitive bidding rules established in this proceeding. We take this action because we find that it will optimize the public interest by promoting fair and efficient licensing practices. As we explain below in Section V-A ("Auctionability of the 39 GHz Band"), the use of a competitive bidding system for licensing the 39 GHz band constitutes the best method for choosing among mutually exclusive applicants. Competitive bidding allows spectrum to be acquired by the parties who value it most highly and increases the likelihood that innovative, competitive services will be offered to consumers. These benefits will be lost, in part, if we were to process pending mutually exclusive applications under our old rules. Moreover, under such an approach, those pending mutually exclusive applications that cannot be accommodated by the availability of alternative frequencies would be subject to comparative hearing (either formal or informal). While these rules may be useful in other bands to address the rare situation in which two point-to-point links cannot be coordinated to avoid interference, in the 39 GHz band, applicants seek to serve geographic areas rather than to provide service on a single point-to-point link basis. This, coupled with the exponential growth in demand for 39 GHz spectrum, results in a significant number of mutually exclusive applications, including "daisy-chain" situations, among entities seeking to acquire spectrum. Resolving these mutually exclusive applications through comparative hearings would be much slower and possibly more costly, both to the government and applicants, than competitive bidding. 91. We also find that those who believe that they should be afforded the opportunity to amend their pending applications to avoid mutual exclusivity had ample opportunity to file such amendments prior to the commencement of this rule making. We are not convinced that parties who have not already entered such agreements will successfully accomplish such agreements now. Moreover, even if such agreements are possible, the parties will have the opportunity to accomplish similar results through the partitioning and disaggregation rules we are adopting today. Similarly, parties may resolve existing conflicts by forming joint ventures or similar arrangements to apply for BTA licenses. If, however, we permitted pending mutually exclusive applicants to resolve their conflicts outside the structure of the competitive bidding process, other entities would be foreclosed from an opportunity to apply for 39 GHz spectrum under the flexible rules we adopt herein. This would have the result of limiting the pool of potential applicants to those who have already filed under the current, more restrictive rules, and may inhibit the development of new and innovative services in this spectrum. Accordingly, we find that existing applicants have a reasonable avenue of relief for their concerns in the procedures we adopt herein, and we deny their requests. ii. Applications Within the 60-day Public Notice Period on November 13, 1995 92. Some petitioners and commenters argue that we should process the "unripe" applications -- those that had not passed the 60-day public notice period as of the date of the November 13, 1995, Freeze Order. According to DCT, for example, all applications that have been or should have been placed on public notice announcing their susceptibility to petitions to deny as required by Section 309 of the Communications Act meet the processing requirements of the Communications Act. DCT contends that the disparate treatment of these applications and those we have decided to process would only make sense if there were no vacant channel pairs available for a second applicant in the same service area. DCT and WinStar argue that under the rules, if there were a vacant channel pair, a second applicant would have to yield ultimately to the first-in-time applicant with respect to the frequencies specified by the first-in-time applicant. 93. In the January 17, 1997, Memorandum Opinion and Order, supra, we held that unripe applications would continue to be held in abeyance because, until we had completed our consideration of the record, we were not in a position to state whether further applications may be filed, or how the applications presently held in abeyance would have been treated. Having concluded here that the 39 GHz band should be subject to significantly different rules than the ones used previously, we believe that the most fair and reasonable approach with regard to pending unripe applications is to dismiss them and allow these applicants to reapply under the new rules set forth in this proceeding. Taking into account our conclusion that these new rules further the public interest, we believe that applying the new 39 GHZ rules to those applications that were still subject to the possibility of competing applications under the former rules adequately balances the expectations of applicants with the public need for a better system for licensing use of the 39 GHz band. We further believe that we have crafted a fair approach because such applicants will be permitted to apply for spectrum under the new rules. iii. Modification Applications 94. In the NPRM and Order, we stated that we would hold in abeyance modification applications, and any amendments thereto, that were filed on or after November 13, 1995, the date of the Freeze Order. We stated that no new applications to modify existing licenses would be accepted after December 15, 1995, unless they did not involve any enlargement in any portion of the service area and did not change frequency blocks (unless to delete one). 95. In the January 17, 1997, Memorandum Opinion and Order, we clarified that any pending modification application or amendment thereto filed prior to November 13, 1995, was to be processed. Modification applications or amendments to such applications, filed between November 13 and December 15, 1995, which meet the criteria of Section 101.59 of our Rules and which do not enlarge the applicant licensee's service area, were to be accepted for filing and processed. Any modification application, or amendment thereto, which meets the criteria of Section 101.61 of our Rules were likewise to be accepted for filing and processed. All other modification applications and amendments thereto were to be held in abeyance. 96. For the same reasons that we dismiss without prejudice the pending mutually exclusive and unripe applications as discussed supra, we also dismiss without prejudice any modification application held in abeyance pursuant to the freeze. Such applications, if granted under the previous rules, would frustrate the goals underlying this proceeding by continuing the licensing scheme which we are abandoning today. As discussed supra, we must choose a point from which our new rules will apply, taking into account our conclusion that these new rules are in the best interest of the public for the development of new services in the 39 GHz band. We believe that it is fair to dismiss major modification applications because such applicants will be permitted to apply for additional spectrum, without disadvantaging potential new entrants, under the new rules. iv. Applications That Are Partially Mutually Exclusive 97. There are seven applications that are partially mutually exclusive. That is, these applications request more than one frequency pair, some of which are mutually exclusive with frequencies requested in other applications and some of which are not mutually exclusive. Although the non-mutually exclusive portion of these applications was subject to processing under our December 15, 1995, NPRM and Order, the mutually exclusive portion of each of the applications was required to be held in abeyance. The divided status of these applications has presented a unique processing issue. Our electronic process for addressing these applications does not permit partial grants because there is no capability for allowing an application to remain in pending status if final action has been taken on a portion of it. As a result, we have not been able to process the non-mutually exclusive portion of these applications until we had reached a decision regarding the disposition of pending mutually-exclusive applications in general. As we have now made this determination, we will process these applications as follows. Specifically, we will process to completion that portion of each of these applications that is non-mutually exclusive with other applications. However, we will dismiss the remainder of the application which cannot be granted due to mutual exclusivity, consistent with our order herein. V. DECISION -- COMPETITIVE BIDDING ISSUES A. Auctionability of the 39 GHz Band 98. Background. In the NPRM and Order, we proposed to use competitive bidding to select among mutually exclusive applications for initial licenses in the 39 GHz band. We reconsidered our previous decision not to license intermediate links by competitive bidding and the various factors that influenced our decision. First, we noted that point-to-point microwave channels used as part of end-to- end subscriber-based service offerings meet the "principal use" requirement of the Communications Act. Second, because BTAs are large areas, we stated that defining service areas by BTAs likely will result in the filing of mutually exclusive applications. Third, we noted that based upon our experience with auctions in other services, an auction for intermediate links within a well-defined service area will neither significantly delay the provision of other services, such as PCS, to the public nor impose significant administrative costs on the applicants or the Commission. Fourth, we noted that by placing licenses in the hands of those who value this spectrum most highly, competitive bidding will likely promote the development and rapid deployment of new technologies and ensure that new and innovative technologies are readily accessible to the American people. Finally, we noted that some of the licensees in the 39 GHz band have offered to sell or lease their licenses and may never have intended to directly serve the public, but rather to hold their own auctions and thereby deprive the public of the aforementioned benefits. 99. Discussion. Upon consideration of the record in this proceeding, we conclude that auctioning the 39 GHz band meets the new criteria set forth in Section 309(j) of the Communications Act, as amended by the Balanced Budget Act of 1997. During the pendency of this proceeding and after comments were received in this proceeding, Congress enacted the Budget Act which extended and expanded the Commission's auction authority. Many commenters support the award of unallocated spectrum through auctions for the 39 GHz band. Using the pre-Budget Act criteria for auctionability of spectrum, some commenters argued that the 39 GHz band did not meet such criteria because: (1) the band is being used for providing intermediate links and, therefore, is not principally being used to garner compensation from subscribers as required under the former "principal use" criteria of the Act; (2) an auction of the 39 GHz band does not promote the objectives contained in the Act; and (3) an auction of intermediate links could significantly delay the development and deployment of new products and services and impose significant costs on licensees and the Commission. As discussed below, as a result of the Budget Act provisions, the "principal use" criteria of 309(j)(2)(A) and "promote the objectives" criteria of 309(j)(2)(B) and 309(j)(3) of the Act no longer govern the auctionability of electromagnetic spectrum. Thus, we do not find these arguments to be compelling reasons not to employ competitive bidding procedures for 39 GHz spectrum. 100. Under the Budget Act, the Commission's auction authority covers all mutually exclusive applications for initial licenses or construction permits, with three limited exceptions which are not applicable in this proceeding. The Budget Act replaced language in Section 309(j)(2), formerly called "Uses to Which Bidding May Apply," which stated the requirements for spectrum to be auctionable, i.e., a determination that the principle use of the spectrum will be on a subscription basis and that competitive bidding will promote the objectives stated in Section 309(j)(3) with the new paragraph that expands the Commission's auction authority. Accordingly, under the amended Section 309(j), the Commission has the authority to auction the 39 GHz band. 101. We reject DCT's contentions that using competitive bidding procedures for this band violates Sections 309(j)(1) and 309(j)(6)(E), because the Commission is required to use various means to avoid mutual exclusivity, including the use of engineering solutions, negotiate threshold qualifications and service regulations, and licensing proceedings, before turning to auctions. DCT argues that because the NPRM and Order finds that current point-to-point rules are structured to avoid mutual exclusivity through frequency coordination, changing the rules to license by BTAs is tantamount to adopting a licensing system designed to encourage mutual exclusivity. However, the 39 GHz band has been the subject of significantly increased requests for large rectangular service areas and multiple channels. Frequency coordination techniques, suitable for the level of point-to-point spectrum demand existing prior to the existence of emerging technologies, are no longer adequate. Accordingly, our use of pre-defined geographic areas rather than the applicant-defined rectangular areas currently used as service areas furthers our public interest goals, as we concluded above in Section IV(C)(1). As we noted there, predetermined service areas will provide a more orderly structure for the licensing process and will