Federal Communications Commission DA 20-1361 DA 20-1361 Released: November 16, 2020 OFFICE OF ECONOMICS AND ANALYTICS AND WIRELINE COMPETITION BUREAU ADOPT ADJUSTMENT FACTOR VALUES FOR THE 5G FUND GN Docket No. 20-32 1. Today the Office of Economics and Analytics (Office) and the Wireline Competition Bureau (Bureau) adopt 5G Fund adjustment factor values to help direct more 5G Fund support to harder to serve areas. See Office of Economics and Analytics and Wireline Competition Bureau Seek Comment on Adjustment Factor Values for the 5G Fund, Public Notice, 35 FCC Rcd 5704, 5710, paras. 1, 12 & n.39 (OEA/WCB 2020) (5G Fund Adjustment Factor Public Notice). Specifically, the values we adopt will increase support levels for bids to serve areas where the terrain elevation variation raises the expected costs of deploying 5G networks, and/or where the business case for 5G otherwise is likely to be weaker, relative to the support for bids for easier to serve areas. Likewise, the adjustment factor values will also be used in the process of disaggregating legacy high-cost support to account for differences between recipients’ subsidized service areas. These adjustment factor values will help ensure that additional 5G Fund support goes to the areas that need it the most. 2. In the 5G Fund NPRM and Order, the Federal Communications Commission (Commission) proposed to distribute up to $9 billion in two phases using multi-round, descending clock auctions to assign support for the deployment of 5G service in rural areas. See Establishing a 5G Fund for Rural America et al., Notice of Proposed Rulemaking and Order, 35 FCC Rcd 3994 (2020) (5G Fund NPRM and Order). To account for differences in the cost of providing service and business case considerations across eligible areas, the Commission proposed incorporating an adjustment factor into the 5G Fund auctions that would assign a weight to each geographic area, which would be applied to bidding for support amounts to make the areas most difficult to serve more attractive to bidders and increase the support to such areas. See id. at 4015-16, para. 66 & n.97. In addition to incorporating an adjustment factor into the 5G Fund auctions, the Commission proposed to apply this adjustment factor to the methodology for disaggregating legacy high-cost support in the transition to 5G Fund support. Id. at 4015-16, para. 66. Legacy high-cost support is currently provided to a competitive eligible telecommunications carrier’s entire study area, with no attribution to particular sub-areas within that study area. Id. at 4018, para. 71 & n.104. To illustrate the role of the adjustment factor in the disaggregation of legacy support, consider a hypothetical carrier serving one mountainous census tract and one flat census tract of equal size in its subsidized service area. Such a carrier might require 75% of its support to serve the mountainous tract and 25% to serve the flat tract. Were an unsubsidized carrier to enter the flat tract, which may be more likely given the relatively lower costs in the flat tract, if we did not apply the adjustment factor in calculating disaggregated support, the carrier would lose 50% of its funding and would be unable to continue serving the mountainous tract. However, applying an adjustment factor of three to the mountainous area would result in the carrier retaining 75% of its original support amount and allow it to continue serving the mountainous tract. On June 5, 2020, the Office and Bureau released a public notice seeking comment on the proposed adjustment factor values, the three analyses that inform the values, and the application of the adjustment factor to the disaggregation of legacy support. See 5G Fund Adjustment Factor Public Notice, 35 FCC Rcd at 5704, 5709-10, paras. 1, 12-14 & n.39; see also 5G Fund NPRM and Order, 35 FCC Rcd at 4060-61, paras. 201-03 (directing the Office and Bureau “to propose and seek public comment on (1) adjustment factor values and the underlying methodologies that could be used to develop them; and (2) a process by which the adjustment factor could be applied to the disaggregation of legacy support consistent with the proposal in the 5G Fund NPRM”). 3. In the 5G Fund Report and Order, the Commission adopted its proposal to incorporate an adjustment factor into the 5G Fund auctions that will assign a weight to each geographic area and apply that adjustment factor to bidding for support amounts; this adjustment factor also will be applied to the methodology for disaggregating legacy high-cost support. Establishing a 5G Fund for Rural America, Report and Order, FCC 20-150, at 23, para. 54 (Oct. 29, 2020) (5G Fund Report and Order). For a 5G Fund auction, the Commission deferred the final determination of the precise manner in which the adjustment factor will be incorporated into the auction mechanism to the pre-auction process. Id. at 25, para. 58. Below, we provide the adjustment factor values, and we discuss the studies underlying our decision to adopt these values for use in a 5G Fund auction and in the methodology for the disaggregation of legacy high-cost support. 4. Adjustment Factor Values. In the 5G Fund Adjustment Factor Public Notice, the Office and Bureau proposed values for an adjustment factor that operates along two dimensions: terrain elevation variation and demand, using median household income as a proxy. See 5G Fund Adjustment Factor Public Notice, 35 FCC Rcd at 5710, Fig. 1. These two dimensions were included to account for differences in network deployment costs and business case considerations that stem from the geographic and economic variations in the United States. Id. at 5706-08, paras. 7-11. We proposed in the 5G Fund Adjustment Factor Public Notice that areas be sorted into terrain elevation variation and demand factor groups according to their characteristics. See generally id. at 5714-15, Appx. A: Terrain Elevation, 5716-49, Appx. B: Economic Analyses Supporting the Proposed Adjustment Factor. Under the approach proposed in the 5G Fund Adjustment Factor Public Notice, an area’s terrain classification is determined by its average standard deviation of elevation. See id. at 5714. Areas are separated into one of three categories: 1) flat (standard deviation of 40 meters or less); 2) hilly (standard deviation between 40 and 115 meters); and 3) mountainous (standard deviation greater than 115 meters). Id. Similarly, areas’ demand classification is determined by the areas’ median household income. See id. at 5710, Fig. 1, para. 15. We note that the category thresholds for the medium- and high-income categories represent 2017 median household incomes that are 150% and 200% of the poverty line for a family of three, respectively. Annual Update of the HHS Poverty Guidelines, 82 Fed. Reg. 8831 (Jan. 31, 2017). Consistent with the adjustment factor values we adopt below, we will use the latest available data on terrain and median household income appropriate for such purposes to classify areas into the adjustment factor categories concurrent with the Commission’s release of the map of final areas eligible for 5G Fund Phase I support. The terrain elevation variation dimension is intended to address, in part, network cost differences across areas, while the demand factor is intended to address differences in expected revenues across areas. 5G Fund Adjustment Factor Public Notice, 35 FCC Rcd at 5706-08, paras. 6-11. 5. We adopt the following adjustment factor values, as proposed in the 5G Fund Adjustment Factor Public Notice. Id. at 5710, Fig. 1. We find that these adjustment factor values, informed by the three economic analyses laid out in the 5G Fund Adjustment Factor Public Notice, Id. at 5710-11, paras. 15-17, 5716-40, Appx. B, paras. 1-52. appropriately reflect the relative cost of serving areas with differing terrain characteristics, as well as the potential business case for each area, with less profitable areas receiving greater weight and therefore more support. 5G Fund Report and Order at 23-24, para. 54. Using these values to help distribute 5G Fund support to, and disaggregate legacy support in, a range of areas across the country that are geographically and economically diverse serves the public interest. Fig. 1: Adjustment Factor Values   Terrain Elevation Variation     Flat Hilly Mountainous Demand Factors Low 1.2 2.4 3.8 Medium 1.1 2.3 3.5 High 1.0 2.0 3.0 6. Use of An Adjustment Factor in Bidding. Commenters generally support the use of an adjustment factor to increase support in higher-cost, less-profitable areas, and no commenter suggests alternative adjustment factor values to those proposed in the 5G Fund Adjustment Factor Public Notice. See, e.g., AST&Science LLC 5G Fund NPRM Comments at 28. Although no commenter objects to the use of terrain elevation variation and median household income in the determination of the adjustment factor, several commenters suggest that the adjustment factor should consider other variables. For example, the Massachusetts Department of Telecommunications and Cable suggests that the adjustment factor should incorporate the “differences in the cost of labor and transportation to both deploy and operate 5G service, as well as differences in the cost of utility and other operating costs.” Massachusetts Department of Telecommunications and Cable 5G Fund NPRM Comments at 10. The Massachusetts Department of Telecommunications and Cable argues that it would be a departure from the Commission’s past practice in Mobility Fund Phase II, Connect America Fund Phase II (Auction 903), and the Rural Digital Opportunity Fund Phase I (Auction 904) if the 5G Fund adjustment factor does not incorporate the costs of labor, transportation, and the differences in the cost of utility and other operating costs. Id. We disagree. None of these programs adopted an adjustment factor that increased the support awarded per unit in the auction and thus references to such programs are inapposite. The Mobility Fund Phase II Order, which the Massachusetts Department of Telecommunications and Cable cites, rejected relying upon a model that included such factors and instead decided to use a reverse auction mechanism to award support. See Connect America Fund; Universal Service Reform – Mobility Fund, Report and Order and Further Notice of Proposed Rulemaking, 32 FCC Rcd 2152, 2158-59, paras. 18-20 (2017). For Auction 903 and Auction 904, the Massachusetts Department of Telecommunications and Cable misconstrues the Commission’s adoption of cost thresholds for the purposes of establishing eligible locations as setting reserve prices based upon these factors. Moreover, the reverse auction mechanism used in Auction 903 and Auction 904 did not award support based upon the cost thresholds used for establishing eligible locations, but rather awarded support as determined by a formula that incorporates the clock percentage as well as the service tier and latency of the bid and the applicable reserve price. Rural Digital Opportunity Fund Phase I Auction Scheduled for October 29, 2020; Notice and Filing Requirements and Other Procedures for Auction 904, Public Notice, 35 FCC Rcd 6077, 6147-48, paras. 215-19. In any event, insofar as the Commission addressed this issue differently in other proceedings, we acknowledge that we are taking a different approach here, declining to incorporate additional costs into the adjustment factor for the reasons explained herein. As noted above, the 5G Fund adjustment factor is not meant to reflect actual costs and is meant to increase the availability of funds to areas that are more challenging to serve due to terrain and the business case. Further, the Commission specifically rejected the Massachusetts Department of Telecommunications and Cable’s proposal to incorporate additional factors into the 5G Fund adjustment factor. See 5G Fund Report and Order at 25-26, para. 60. Smith Bagley notes that the adjustment factor does not account for the existing infrastructure in an area. See Smith Bagley 5G Fund Adjustment Factor Public Notice Reply Comments at 3, 6-7. While it argues that failing to account for “deficient infrastructure” on Tribal lands will result in these areas receiving no support in the 5G Fund auctions, id. at 6-7, Smith Bagley does not identify an appropriate data source for us to determine whether the infrastructure in an area is deficient. The Commission’s decision to require legacy support recipients to submit network infrastructure data for their subsidized service areas in their initial report of current service offerings and subsequent annual reports, 5G Fund Report and Order at 40, 42, paras. 93, 97, likewise would not be sufficient to allow us to determine the infrastructure of areas outside of the areas that receive legacy support, nor would such data provide us with an indication of the state of the area’s electrical or plumbing infrastructure that Smith Bagley asserts should be considered. See Smith Bagley 5G Fund Adjustment Factor Public Notice Reply Comments at 6. Conversely, we do have reliable data on terrain variation and median household income. We conclude that these variables taken together can sufficiently account for the relative differences in costs and business case such that bidders in the 5G Fund auctions will be encouraged to bid on these high-cost Tribal lands. 7. We are not persuaded by these arguments and decline to increase the number of components or categories that make up the adjustment factor. We acknowledge that terrain elevation variation and median household income do not exhaust the list of potentially relevant variables. Likewise, we acknowledge that when we separate areas into categories, the areas near the midpoint of the category will have their relative costs and business cases more accurately represented by the adjustment factor values than areas at the margins. Nevertheless, as noted in the 5G Fund Report and Order, the adjustment factor adopted by the Commission is not intended to fully offset the differences inherent in providing service to different types of areas. See 5G Report and Order at 23-24, para. 54. Rather, it is intended to “make the most difficult areas to serve more attractive at auction in order to encourage more bidding for these areas.” 5G Fund Adjustment Factor Public Notice, 35 FCC Rcd at 5704, para. 1. Moreover, we selected terrain elevation variation and median household income as the two dimensions for the adjustment factor characteristics because they are important factors in characterizing deployment costs and business case considerations, respectively, and because there is more readily available and verifiable data with which to apply these two factors. See id. at 5747, Appx. B, Fig. B-12. As we discussed in the 5G Fund Adjustment Factor Public Notice, terrain elevation variation captures differences in network costs because “wireless network engineering principles indicate that greater variability of terrain in a given geographic area reduces the signal strength received by a mobile user, which requires wireless carriers to build more sites to provide the same quality of service.” See id. at 5726, Appx. B, para. 18. As a result, areas with higher terrain elevation variation generally have higher capital expenditures, operating expenditures, and leasing costs. See id. at 5706-07, para. 7. Similarly, we also discussed the importance of demand factors and the role that expected revenues play in carriers deployment decisions. See id. at 5716, Appx. B, para 2. The Entry Model Adjustment Factor study found that, all else equal, areas with higher median household incomes are more likely to be covered, a finding consistent with the basic assumption that higher income areas are more profitable. See id. at 5721, Appx. B, para 15. 8. Economic Analyses. To inform the proposed adjustment factor values, the Office and Bureau included three economic analyses. See generally id. at 5716-40, Appx. B, paras. 2-52. The first analysis (the Entry Model) used coverage data to estimate the effect that an area’s physical and demographic characteristics have on carriers’ network deployment decisions. See id. at 5716-26, Appx. B, paras. 2-17. The second analysis (the Cell Site Density Model) examined how cell site spacing changes as terrain roughness increases. See id. at 5726-35, Appx. B, paras. 18-38. The third analysis (the Auction Bidding Model) used Mobility Fund Phase I (Auction 901) bidding data to estimate how terrain roughness and other factors affected carriers’ bids. See id. at 5735-40, Appx. B, paras. 39-52. 9. Discussion of the economic analyses in the record is limited, and no party submitted an alternative economic analysis. T-Mobile commends “the Commission’s efforts to derive a reasonable adjustment factor for use in the 5G Fund,” T-Mobile 5G Fund NPRM Reply Comments at 5. and RWA notes that “the Entry Model and the Cell Site Density Model consider factors that are relevant to the terrain, demand for, and costs of constructing a 5G network.” RWA 5G Fund Adjustment Factor Public Notice Comments at 6. RWA and NTCA, on the other hand, argue that the Auction Bidding Model should not be used to determine the adjustment factor values. Id. at 6-8; NTCA 5G Fund Adjustment Factor Public Notice Comments at 4-5. RWA and NTCA contend that bidding data from the Mobility Fund Phase I auction is distorted and that the Mobility Fund Phase I auction is not an appropriate analogue because it provided one-time funding for capital expenditures, while the 5G Fund would be a ten-year continuing support mechanism that provides funding for capital expenditures and operational expenses. RWA 5G Fund Adjustment Factor Public Notice Comments at 7-8; NTCA 5G Fund Adjustment Factor Public Notice Comments at 4-5. In addition, RWA contends that the Mobility Fund Phase I auction data should be disregarded because bidding decisions were based on 2012 pricing, which is not comparable to today’s pricing. RWA 5G Fund Adjustment Factor Public Notice Comments at 8. RWA and NTCA also argue that the Commission should not rely on Mobility Fund Phase I data because, at the time of the auction, carriers could still use network equipment from low-cost equipment suppliers that have since been designated by the Commission as national security threats. Protecting Against National Security Threats to the Communications Supply Chain Through FCC Programs et al., Report and Order, Further Notice of Proposed Rulemaking, and Order, 34 FCC Rcd 11423, 11433, para. 26 (2019); see generally Protecting Against National Security Threats to the Communications Supply Chain Through FCC Programs – Huawei Designation, Order, 35 FCC Rcd 6604 (PSHSB 2020); Protecting Against National Security Threats to the Communications Supply Chain Through FCC Programs – ZTE Designation, Order, 35 FCC Rcd 6633 (PSHSB 2020). Thus, RWA and NTCA argue that the auction analysis should not be used in the determination of the 5G Fund adjustment factor. RWA 5G Fund Adjustment Factor Public Notice Comments at 6-7; NTCA 5G Fund Adjustment Factor Public Notice Comments at 4-5. 10. We acknowledge the contextual differences between the Mobility Fund Phase I auction and the upcoming 5G Fund auctions, but do not find that such differences unduly undermine the analysis. While the timing and one-time funding nature of the Mobility Fund Phase I auction and the presence of Huawei and ZTE as low-cost equipment options for Mobility Fund Phase I support recipients may have influenced the absolute bid amounts, RWA and NTCA fail to explain why the relative bid amounts would differ significantly compared with a more recent long-term funding auction where bidders could not use Huawei and ZTE equipment. The absolute level of the bids does not necessarily affect the relative differences across areas. For example, if all bids were 20% lower in absolute level due to factors related to the auction’s context, the ratio of bids across areas would be unaffected. We find it more likely that the calculated adjustment factor should be largely invariant to differences in funding type and radio equipment costs. There are two cases to consider. In the case where the costs to build and operate towers are the same across terrain types and more towers are needed to cover rougher terrain, the cost of radio equipment would have no effect on the calculated adjustment factors. In the case where towers cost more to build and operate in rougher terrain, the absolute cost of radio equipment could affect the adjustment factor. However, given that radio equipment costs are a very low percentage of the overall costs to build and operate a network, the change in the calculated adjustment factor would be negligible. See generally Omnibus Broadband Initiative, A Broadband Network Cost Model: A basis for public funding essential to bringing nationwide interoperable communications to America’s first responders, OBI Technical Paper No. 2 (Apr. 2010), https://transition.fcc.gov/national-broadband-plan/broadband-network-cost-model-paper.pdf. Similarly, RWA and NCTA’s arguments that the Mobility Fund Phase I auction is not an appropriate point of comparison because that auction provided only funding for capital expenditures whereas the 5G Fund will provide funding for both capital and operational expenditures likewise does not undermine our analysis here because the adjustment factor values we adopt are meant to capture the relative differences in cost and business case for different areas. That is, reliance upon bid amounts in an auction that did not award operational expenses should not affect the relative differences in costs because bidders in the 5G Fund auctions will be able to consider the entirety of costs (including both capital and operational expenditures). Thus, any additional operational expenses will be reflected in higher bidding values in the auction but the relative differences between areas is likely to remain the same. Moreover, our conclusions about the appropriateness of using Mobility Fund Phase I auction data are also consistent with all three models producing comparable adjustment factor estimates. We find that the information regarding the relative bidding incentives across areas produced by the Auction Bidding Model outweighs any concerns with the absolute levels of the bidding data. 11. Use of an Adjustment Factor for Disaggregation of Legacy High-Cost Support. In the 5G Fund Adjustment Factor Public Notice, the Office and Bureau sought comment on the appropriate adjustment factor values for the disaggregation of legacy high-cost support to account for differences in costs across areas and the underlying methodologies that could be used to develop the values. See 5G Fund Adjustment Factor Public Notice, 35 FCC Rcd at 5706, 5709, paras. 5, 12 & n.39; 5G Fund NPRM and Order, 35 FCC Rcd at 4015-16, 4060-61, paras. 66, 201-03. In cases where the transition of legacy support occurs across areas of different types, such as eligible areas and ineligible areas, the adjustment factor would be used to scale the actual square kilometers associated with each disaggregated area. See 5G Fund NPRM and Order, 35 FCC Rcd at 4017-18, para. 71. In the 5G Fund Report and Order, the Commission concluded that the adjustment factor values that are adopted by the Office and Bureau for a 5G Fund auction also would be used for the disaggregation of legacy high-cost support. 5G Fund Report and Order at 23-24, paras. 54-55. Accordingly, we adopt the adjustment factor values proposed in the 5G Fund Adjustment Factor Public Notice, as set forth in Fig. 1 above, for use in the process of disaggregating legacy support. 5G Fund Adjustment Factor Public Notice, 35 FCC Rcd at 5706, para. 5. 12. We note that some commenters oppose using the adjustment factor in the disaggregation process. See, e.g., RWA 5G Fund Adjustment Factor Public Notice Comments at 1-5; NTCA 5G Fund Adjustment Factor Public Notice Comments at 2. These commenters generally argue that, because the adjustment factor does not capture all of the characteristics of the particular service areas for which legacy support is provided, it is not appropriate to apply the factor when disaggregating legacy support. See, e.g., RWA 5G Fund Adjustment Factor Public Notice Comments at 2-5. For example, RWA argues that the adjustment factor should account for foliage. In its comments, RWA notes the “substantial adverse effect on signals in the 850 MHz cellular band” caused by pine trees and raises concerns that some areas classified as evergreen forest by the National Land Cover Database contain many pine trees while others contain other tree species and few or no pine trees. RWA 5G Fund Adjustment Factor Public Notice Comments at 4-5. While we acknowledge that there may be limitations in the precision of the terrain and income data upon which we rely, we note that each of the models already implicitly account for foliage, and thus it is unnecessary for us to add foliage as a third variable when determining the adjustment factor values. See 5G Fund Adjustment Factor Public Notice, 35 FCC Rcd at 5716-17, Appx. B, para. 3 & n.8 (using carrier-reported 4G LTE mobile broadband coverage data, which inherently incorporate clutter data including foliage into the carriers’ propagation modeling, in the Entry Model); id. at 5729-30, Appx. B, paras. 27-28 (using a simplified propagation model which incorporates antenna height and terrain, “where terrain reflects not only the variation in elevation, but also other factors that affect propagation such as buildings and foliage” in the Cell Site Density Model); id. at 5737-38, Appx. B, paras. 47-48 (using “measure[s] of terrain roughness, population density, tract median household income, road miles, and percent forested land” with “[t]errain roughness . . . captur[ing] the effect of terrain on increased construction costs and reduced signal propagation distances” in the Auction Bidding Model). RWA also asserts that the terrain categories are too broad and, as a result, areas near the margins will be disadvantaged. RWA 5G Fund Adjustment Factor Public Notice Comments at 4. They propose instead that the Commission rely on service providers’ knowledge of their subsidized areas to estimate the costs of deploying in those areas. See, e.g., RWA 5G Fund Adjustment Factor Public Notice Comments at 3-4. 13. The Commission rejected the argument that the adjustment factor should not be applied to the disaggregation of legacy support. 5G Fund Report and Order at 25, para. 59. In the 5G Fund Report and Order, the Commission found that “[u]sing an adjustment factor is appropriate because it will alleviate potential concerns over a carrier losing a disproportionate amount of its legacy support resulting from a disaggregation methodology in which more costly areas would be treated the same as less costly areas with respect to subsidies received.” Id. As the Commission indicated, this approach will help ensure that legacy high-cost support is available for harder-to-serve areas. Id. at 25, para. 59 n.149. 14. We also note that there are other reasons to apply the adjustment factor to the disaggregation of legacy high-cost support. Using an adjustment factor to disaggregate legacy support is preferable to the administrative burdens that would arise from requiring service providers to disaggregate their costs, and furthermore, it avoids the potential incentive issues associated with service providers self-reporting their own costs. For example, where part of a legacy support recipient’s service area would be served by a 5G Fund winner while its remaining area would continue to receive legacy support, the legacy support recipient would have the incentive to overestimate the amount of high-cost support flowing to the area that would continue to receive legacy support, thus maximizing the funds it would receive through preservation of service support. In addition, while we acknowledge that the adjustment factor does not account for all factors that affect network costs, the Commission indicated that the adjustment factor is meant to give an estimate of how a carrier may allocate legacy high-cost support within the area for which it receives such support. 5G Fund Report and Order at 24, para. 54; see supra para. 7. It is not meant to reflect the actual cost of deployment in that area. We maintain that applying an adjustment factor in the disaggregation process will lead to a more equitable distribution of legacy funding. Applying the adjustment factor will better reflect the distribution of high-cost support by accounting for cost differences arising from terrain elevation variation and business case differences arising from income disparities within a service area. Thus, we will use the adjustment factor values in Fig. 1 above for the disaggregation of legacy high-cost support. 15. For further information, contact Kate Matraves, Economic Analysis Division, Office of Economics and Analytics, at Catherine.Matraves@fcc.gov, or Nicholas Copeland, Economic Analysis Division, Office of Economics and Analytics, at Nicholas.Copeland@fcc.gov. -FCC- 2