Federal Communications Commission FCC 20-5 STATEMENT OF COMMISSIONER GEOFFREY STARKS APPROVING IN PART AND DISSENTING IN PART Re: Rural Digital Opportunity Fund, WC Docket No. 19-126; Connect America Fund, WC Docket No. 10-90. We have been talking about the “digital divide” for 25 years, but it is clear that we still have not finished the job of bringing quality affordable broadband access to all Americans. Across the United States, low-income people, people of color, and people in rural areas either are not getting online or are making great sacrifices to get connected. Solving the problem of internet inequality is a moral imperative, and it is essential to our country’s competitiveness. Other countries are making enormous investments to get their citizens connected to high-speed, quality broadband. China, for example, plans to deploy fiber-optic connections to 80 percent of the homes in that country. If we leave millions of our fellow Americans behind, our country will fall behind. The scale of this challenge is enormous and, at least for now, the resources available for meeting it are limited. In terms of just sheer magnitude, the $20 billion Rural Digital Opportunity Fund that we vote on here today is nearly ten times bigger than its predecessor, the Connect America Fund II, and will commit us to a certain course for more than a decade. There are no do-overs here with this money and precious time. We must, therefore, be good stewards of these funds and, perhaps more importantly, be clear-eyed and far-sighted about our connected future. When I talk to rural Americans—in the town library, the family farm, the community center—each has their own personal story, but if you listen closely, they often share a common theme: they need us to get this right. They are getting left behind. Over the last number of months, I have laid out a four-part plan for a data-driven and fiscally responsible approach to promoting rural broadband deployment. The FCC should: (1) provide funding based on accurate data and maps, (2) promote affordable broadband options, (3) incentivize providers to offer future-proof broadband, and (4) hold auction winners accountable. While there are parts of this item that are good advancements, today’s Report and Order falls short of satisfying my vision of how to best get broadband to our rural Americans. Accordingly, I will approve in part and dissent in part. First, as I laid out in my statement on the Notice of Proposed Rulemaking in this proceeding, I have zero tolerance for continuing to spend precious universal service funds based on bad data. There is bipartisan—and nearly universal—agreement that our existing broadband deployment data contains fundamental flaws. And yet today’s Report and Order presses ahead with funding decisions based on mapping data that doesn’t reflect reality, plowing the same mission-critical error into a newer, much larger program. We must do better. We know—from the tireless work of researchers, state agencies, and community activists—that our failure to get broadband mapping right has had serious consequences. The data generated from Form 477, on which we will double down today, fell so far short in Georgia that the state decided to stand up its own mapping initiative using a more rigorous approach. For each target county, state officials developed a database of all the premises in the county and then worked with providers to gather deployment information on a much more granular level than Form 477 provides. Last September, the Georgia Broadband Deployment Initiative released preliminary results for three counties. The results are striking. Source: Georgia Department of Community Affairs Exhibit A: Lumpkin County, which covers 283 square miles in the foothills of the Blue Ridge Mountains and looks mostly “served” by 25/3 Mbps broadband in our Form 477 data. Georgia’s analysis paints nearly the opposite picture: the vast majority of Lumpkin County is unserved. For the County’s 30,000 residents, that’s an enormous problem. This week, I spoke with Mayor Sam Norton of Dahlonega, a small city in Lumpkin County. He emphasized the real-world consequences of limited access to broadband. He told me that the lack of quality, affordable broadband is hurting the County’s ability to attract new businesses and address “the cycle of generational poverty.” For example, there’s a pharmacy in the community that doesn’t have high-speed broadband. That has meant they have a harder time getting doctors’ orders and using credit card machines. Mayors like Mr. Norton see these problems up close—even as our flawed maps and data obscure our vision here at the FCC. Make no mistake: without reservation, I fully support providing the resources needed to connect rural communities. And I understand the urgency of getting those resources into the field. But I remain seriously concerned that the Report and Order the FCC adopts today would make funding decisions for Phase I—which is budgeted at more than $16 billion—using data we all know is wrong. That is a “ready, fire, aim” approach that favors speed of funding over the lasting results that Americans really need. The Report and Order asserts that we’ll clean up our mapping problem on that as-yet unknown day when we get to Phase II. But I have not seen any evidence that convinces me that, having already spent 75 percent of our budget, we can feel confident that there will be sufficient money left at Phase II. The Report and Order does not even attempt to estimate how many more people will be unserved when we finally get our maps in order. So, we do not know how many communities will need to be covered by Phase II, and we are not allowing a robust challenge process that would give them an opportunity to identify themselves and participate in Phase I. Without that information, how can we reassure the communities we will knowingly leave out of Phase I that we will be able to meet their needs when we finally get to them? Simply put, we are making a promise to people like the residents of Lumpkin County that we cannot know we will be able to keep. Without having better data and maps in place, we can’t even estimate how many areas similarly deemed “served” like most of Lumpkin County will be left out of RDOF Phase I. What we do know is that Lumpkin County is not alone. In Georgia, Exhibit B from Georgia’s mapping program is Tift County, which is where my Special Advisor, Alisa Valentin, grew up and where her folks still live—and they are always complaining about their lack of broadband. Similarly, I met a number unserved New Yorkers in Hudson, New York—the eighth most rural district in the United States—when I attended a Rural Broadband Field Hearing titled “Closing the Digital Divide: Connecting Rural Americans to Reliable Internet Service” with Congressman Antonio Delgado late last year. Many other states and organizations have begun working to provide an accurate picture of broadband availability because we have failed to do so, including the USTelecom maps that reported that nearly 40 percent of the total locations in Virginia and Missouri that are “served” according to our Form 477 data are actually unserved. Second, we have not done enough to ensure that once broadband is available, families can actually afford it. I share the excitement of many stakeholders about getting high-speed, next generation networks into communities previously denied access. But there’s nothing in our decision today that addresses the needs of low-income families. I had hoped that the subscribership target considered in the Notice of Proposed Rulemaking would incentivize winners to provide a range of broadband packages, including some lower-cost options. But the final Report and Order does not require winners to meet any subscribership target. Going forward, I’m encouraging all stakeholders involved in universal service to recommit to ensuring that cost is not a barrier. Third, to ensure universal service funds are put to the best use, we must envision the connectivity needs of the future—and build toward them. For too long, the FCC has subsidized networks that are obsolete by the time they are built. Less than 10 years ago, we awarded Connect America Fund Phase I support to price-cap carriers to provide service to approximately 524,000 locations. The CAF I rules only required the carriers to provide “broadband” at download speeds of just 4 Mbps and even slower uploads. Under our current rules, quite correctly, that is no longer broadband. That experience raises a critical question: Could we not have reasonably foreseen in 2013 and 2013 that 4/1 Mbps networks, and even 10/1 Mbps networks, would not stand the test of even a decade? That lack of foresight reverberates into the Report and Order before us today. At least 108,000 of those locations, more than 20 percent of locations served through CAF I, will presumptively be eligible for RDOF support. We must learn from that experience. Universal Service Fund dollars are too scarce and too badly needed to be spent building the networks of the past. Has the Commission learned its lesson? We built to 4/1 Mbps in 2010, and it didn’t last. Do I think that the 25/3 Mbps baseline that we set out today will last 10 years from today? I do not. No new lessons are learned from the second mule kick. We need to be building future-proof connections with this money. In my conversations with rural electric co-ops, they tell me that customers overwhelmingly want at least 100/100 Mbps networks. In the coming years, faster upload and download speeds will become increasingly essential to rural life, opening up options to work from home, upload farm data, access medical care, and participate in educational experiences not available locally. For that reason, I support adopting changes to the clearing round rule endorsed by rural electric co-ops and many other potential auction participants that will help ensure that faster and more upgradeable networks are deployed wherever possible. Fourth, consistent with our obligation to spend universal service funds responsibly, we must create real accountability for companies that receive subsidies. Looking at our recent efforts, I see warning signs that we should not ignore. For example, I am extremely frustrated more than a dozen winners from our last universal service auction have already defaulted. And some providers have recently announced that they will not make their CAF II milestones, which is concerning. Communities that have already waited too long for broadband should not be delayed by providers unable to fulfill their obligations. Letters of credit are one way we promote responsibility and protect the fund. I understand that the Commission is trying to strike a careful balance here. We don’t want to over-insure and make the program unaffordable for providers, and I support the changes to the draft Report and Order’s Letter of Credit mechanism because I expect that they will promote more participation in the auction. Moving forward, we must continuously evaluate how well our enforcement mechanisms work. I will be watching closely to see how we handle the problems that have arisen with some CAF II winners. Finally, I want to underscore my commitment to working with the states to close the digital divide. Because I value those partnerships, I cannot support provisions of the Report and Order that penalize the many states that have made their own investments in rural broadband deployment. Here is what I said at the Open Meeting on January 30, 2020 when the Commission voted on this Order: The version of the Order now before us excludes from RDOF any area that the Commission ‘know[s] to be awarded funding through the U.S. Department of Agriculture’s ReConnect Program or other similar federal or state broadband subsidy programs, or those subject to enforceable broadband deployment obligations.’ See Report & Order (January 30, 2020 version) ¶ 13 (emphasis added). Based on my initial research, that means that the nearly 30 states that fund rural broadband through their own programs may find their eligibility reduced or eliminated. These provisions discourage badly needed state-federal partnerships, risk unequal application of the rules between states, and create an unnecessary risk of litigation. We should not be surprised when state officials are confused and concerned upon seeing this decision. The FCC has long encouraged states to work with us toward our shared universal service goals. See, e.g., FCC, Universal Service, Federal-State Joint Board, https://www.fcc.gov/general/universal-service-federal-state-joint-board (last visited Jan. 30, 2020). Several members of Congress wrote to the Chairman just yesterday urging that we delay consideration of this Order until state and federal investments could be better coordinated. Letter from Rep. Anna G. Eshoo, et al., to Ajit Pai, Chairman, FCC (Jan. 29, 2020). Instead of taking time to work with various states, we’re blindsiding them with an exclusion they were never given an opportunity to weigh in on. That is unfair, unwise, and inconsistent with our obligation under the Administrative Procedure Act to give parties fair notice and an opportunity to comment. See 5 U.S.C. § 553. I see no evidence in this Order that we have attempted to understand the variety of state programs our decision will impact. Had we given state officials notice of this rule and worked collaboratively with them, I have no doubt we would have a better record on which to base this decision. Nor do I see any standards against which those programs will be evaluated or a plan to ensure we have canvassed all relevant state programs, creating a risk that our rules will be applied arbitrarily and capriciously. As I said before, I understand the urgency of getting RDOF funds to places that need them. But failing to coordinate with states will, in my view, risk undermining the effectiveness of this important effort. We should have taken the time to get it right. Instead, we have damaged our working relationship with important state partners and created litigation risk that jeopardizes the laudable aspects of this decision. *** On February 3, 2020, I received notice that the Report and Order we had voted had been revised by the Chairman in response to the concerns I raised in my statement at the meeting. The majority subsequently approved a revised version of the Report and Order. I have retained my original language above because I want a clear record of the chaotic process by which the majority adopted a revised item, because I stand by the full strength of the legal and policy issues stated there, and because my fundamental concerns remain. In relevant part, the post-adoption revision approved by the majority was changed to read: In addition, we will exclude those census blocks which have been identified as having been awarded funding through the U.S. Department of Agriculture’s ReConnect Program, or awarded funding through other similar federal or state broadband subsidy programs to provide 25/3 Mbps or better service. This is consistent with our overarching goal of ensuring that finite universal service support is awarded in an efficient and cost-effective manner and does not go toward overbuilding areas that already have service. Report & Order ¶ 13. The new version also adds, among other changes, a footnoted explanation that “[o]ur intent is to exclude areas where 25/3 Mbps or better service has been or will be deployed without Rural Digital Opportunity Fund support, not to prevent winning bidders from accessing other funding sources, including from states.” Report & Order ¶ 13 n.32. In light of my concerns that the Commission failed to satisfy our obligations under the Administrative Procedure Act, the majority attempted to establish in the revised Report and Order that states should have been on notice that we might exclude from RDOF the census blocks that have been awarded funding through state subsidy programs. They failed in my mind. On this point, the only limited language they can point to in the NPRM is language explaining that the Commission seeks to “ensure that [the Commission’s] limited universal service support is awarded in an efficient and cost-effective manner, without overbuilding to areas that already have service” and proposing that the Rural Digital Opportunity Fund framework be guided by, among other goals, “reducing waste and inefficiency in the high-cost program.” Id. That is not sufficient notice. Notably, the only state explicitly and specifically mentioned in the NPRM as potentially being excluded based upon its state funded broadband program was New York. Rural Digital Opportunity Fund; Connect America Fund, GN Docket Nos. 19-126, 10-90, Declaratory Ruling and Notice of Proposed Rule Making, 34 FCC Rcd 6778, 6795, ¶ 48 n.91 (2019). The comments now cited in footnote 32 do not convince me that we have met our responsibility to “adequately frame the subjects for discussion.” Connecticut Light & Power Co. v. Nuclear Regulatory Comm’n, 673 F.2d 525, 533 (D.C. Cir. 1982). On the contrary, our framing of this issue in the NPRM and in the draft Report and Order as only excluding the state of New York may have misled commenters into believing that a nationwide exclusion was not on the table. For example, the Verizon comments cited in and relied upon by the majority in footnote 32 refer to “specified pre-existing federal or state broadband programs,” particularly the CAF auction and “matching CAF support for New York”—two programs actually named in the NPRM and the draft Report and Order. Verizon Comments 8-9. Indeed, we received significant engagement on the New York issue, including communications from numerous members of Congress. See Letter from Antonio Delgado et al., Members of Congress, New York State Congressional Delegation, to Ajit Pai, Chairman, FCC at 41 (Jan. 17, 2020) (signed by 22 members of Congress); Letter from Antonio Delgado, Member of Congress, New York State Congressional Delegation, to Ajit Pai, Chairman, FCC at 58 (Jan. 28, 2020); Letter from Charles E. Schumer & Kirsten Gillibrand, Senators, United States, to Ajit Pai, Chairman, FCC at 33 (Jan. 17, 2020). I have no doubt that, if we had provided adequate notice, we would have heard from many more affected states—and they would have articulated and illustrated why punishing states that have made broadband deployment investments and deterring future state efforts is counterproductive. Nor do the revisions to the Report and Order reassure me that we have gathered the necessary information to understand how the majority’s changes will impact RDOF’s overall success and structure. There is a lot at stake here. We failed to engage with the states, and, accordingly, the Report and Order contains no real analysis about how these state broadband deployment programs work. While much is unknown about how RDOF’s state-subsidy exclusion will impact our federal effort and those state funded broadband programs, we do know this: there are a lot of them, and they distribute significant funding. More than half the states have some kind of state broadband program, and together these programs distribute well over $1 billion in funding. NTIA has given us a head start on this by cataloguing state programs on its BroadbandUSA website. See, e.g., State Broadband Programs, NTIA, https://broadbandusa.ntia.doc.gov/ (last visited Feb. 5, 2020). It will take significant analysis to understand how these programs will interact with RDOF, but their sheer size and scope heightens my grave concern that we do not have even a reasonable estimate—let alone a precise understanding—of how many census blocks are being funded in full or in part by a state broadband subsidy program. It is possible, for example, that some states have spread their funding around scattered locations such that they have made small (perhaps even single-location) investments in many census blocks. Under the approach outlined in the Report and Order, census blocks where even one home has received a state subsidy for 25/3 Mbps deployment will be left out of Phase I. That could create a disproportionate exclusion from RDOF that has ramifications for our budget and auction mechanics. Notably, the post-adoption revision appears to attempt to address this issue with respect to the U.S. Department of Agriculture’s Rural Development Broadband Reconnect Program. The Report and Order now directs the Bureau to exclude only “portions of any census blocks” “that are substantially overlapped by a ReConnect awardee” rather than the entire census blocks. See Report & Order ¶ 13 n.31 (emphasis added). This inconsistency with the treatment of state programs is not explained. Had we proposed this idea in the NPRM, I am confident that we would have received informative and necessary information from states about how their subsidy programs work. As it stands, the households in these census blocks will be cut out of Phase I, and I don’t see any evidence or analysis that we have any idea how many will be impacted. We didn’t ask, and as a result, we don’t know. That deficit is not harmless. See Sprint Corp. v. FCC, 315 F.3d 369, 376 (D.C. Cir. 2003) (explaining that “an utter failure to comply with notice and comment cannot be considered harmless if there is any uncertainty at all as to the effect of that failure” (quoting Sugar Cane Growers Cooperative v. Veneman, 289 F.3d 89, 96 (D.C. Cir. 2002)). Without that information and analysis, this exercise can hardly be called “reasoned decision-making.” Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 52 (1983). Even with the post-adoption changes, the Report and Order remains confusing and confused. The majority has attempted to clarify that, with respect to state programs, the Commission will only exclude “census blocks which have been identified as having been awarded funding . . . through similar state subsidy programs to provide 25/3 Mbps or better service.” Report & Order ¶ 13. On one level, that provides some clarity to states that subsidized slower speeds. But on another level, what about state programs that provided a shorter subsidy period or other terms different from those provided by RDOF, or state programs that plan to fund 25/3 Mbps deployment, but not on a timeline consistent with RDOF? Based on this text, the states will not know, and it is not clear how the FCC will proceed. Given the high stakes here—including a large commitment of our universal service funds, life-changing access to broadband for unserved Americans, and significant investments by the states—I agree with the D.C. Circuit that “elementary fairness compels clarity.” Gen. Elec. Co. v. EPA, 53 F.3d 1324, 1329 (D.C. Cir. 1995), as corrected (June 19, 1995) (quoting Radio Athens, Inc. v. FCC, 401 F.2d 398, 404 (D.C. Cir. 1968)). Additionally, I remain concerned that we have opened the door to arbitrary and capricious application of the state-subsidy exclusion among different states. The voluntary reporting process described in paragraph 14 does not allay my concerns. Report & Order ¶ 14. Rather than gathering information from state programs in advance to inform this Order, as I believe we should have, the Report and Order “direct[s] the Bureau to provide an opportunity to identify census blocks that have been awarded support by a federal or state broadband subsidy program to provide 25/3 Mbps or better service.” Id. This is a recipe for incomplete information and unfairness. Finally, I raise again that this state-subsidy exclusion utterly upends our long-standing policy to work with states to meet our universal service goals. In a confusing about-face, the majority’s action today effectively pits federal broadband dollars against state broadband dollars and penalizes states for their self-help in closing the digital divide. That is bad policy that should alarm our state partners. I outline the practical and legal problems with this aspect of the revised Report and Order here in some detail because I want parties to understand that these issues could have been handled thoughtfully. The RDOF program will distribute a tremendous amount of money and, if well managed, could do enormous good. We could have asked states to partner with us to achieve RDOF’s goals. For reasons unclear to me, we rushed this item and skipped important steps. The Commission could have taken the time to gather the necessary facts, hear from affected states and other parties, and incorporate that information into our decision. I regret that we did not. *** Here’s the final point: we must avoid waking up 10 years from now, with another $20 billion spent, still failing to understand with precision which communities remain unserved and how we can effectively and accurately finish the job. I thank the staff of the Wireline Competition Bureau for their work on this very important issue. 7