On August 14, 2026, the Federal Communications Commission (FCC) released its 2026 Section 706 Report (also known as the Broadband Deployment Report), concluding that “advanced telecommunications capability is being deployed to all Americans in a reasonable and timely fashion.”

What Congress Requires of the FCC

Section 706 of the Telecommunications Act of 1996 directs the FCC to conduct an annual inquiry into whether broadband is reaching all Americans, and provides that if the determination is negative, the FCC “shall take immediate action to accelerate deployment of such capability by removing barriers to infrastructure investment and by promoting competition in the telecommunications market.” A positive finding triggers no statutory obligation at all. The Section 706 finding serves as both a legal switch and the FCC’s principal report to Congress on the state of the digital divide. Through this annual proceeding, the FCC also sets the speed benchmarks—currently 100 megabits per second (Mbps) download / 20 Mbps upload for fixed service—that other federal broadband programs and Congress routinely use.

Because the Section 706 Report turns almost entirely on statutory interpretation, let's explore the mandate in full before reading the FCC’s latest findings. Section 706 has four operative pieces:

§ 1302(a)—The standing duty. The obligation runs to the FCC and to "each State commission with regulatory jurisdiction over telecommunications services"—the state public utility commissions (called public service commissions or utility commissions in some states) that regulate telephone and telecommunications carriers within their borders. Together they "shall encourage the deployment on a reasonable and timely basis of advanced telecommunications capability to all Americans (including, in particular, elementary and secondary schools and classrooms)" by using price cap regulation, forbearance, "measures that promote competition in the local telecommunications market," or "other regulating methods that remove barriers to infrastructure investment."

§ 1302(b)—The annual inquiry. The FCC “shall, within 30 months after February 8, 1996, and annually thereafter, initiate a notice of inquiry concerning the availability of advanced telecommunications capability to all Americans … and shall complete the inquiry within 180 days after its initiation.” In the inquiry, the FCC “shall determine whether advanced telecommunications capability is being deployed to all Americans in a reasonable and timely fashion.”

§ 1302(c)—The unserved-areas list. As part of that inquiry, the FCC “shall compile a list of geographical areas that are not served by any provider of advanced telecommunications capability” and, to the extent Census Bureau data are available, determine for each unserved area the population, population density, and average per capita income.

§ 1302(d)(1)—The definition. “Advanced telecommunications capability” is defined “without regard to any transmission media or technology, as high-speed, switched, broadband telecommunications capability that enables users to originate and receive high-quality voice, data, graphics, and video telecommunications using any technology.”

A Term of Art, Not a Synonym

That last definition is worth unpacking before the analysis begins.

The FCC goes out of its way—in its first substantive footnote and again in the body of the report—to insist that “advanced telecommunications capability” and “broadband” are not interchangeable. The statutory term “has a definition that is more limited than the term ‘broadband,’” so that while every service delivering advanced telecommunications capability is broadband, not every broadband service delivers advanced telecommunications capability. Earlier reports, the FCC concedes, sometimes used the two terms loosely.

This distinction establishes that the FCC is measuring something narrower than “any internet connection.” The distinction also gives the FCC a textual hook for technological neutrality: because Congress defined the term “without regard to any transmission media or technology,” the FCC reads "advanced telecommunications capability" to require that satellite, fixed wireless, fiber, and cable be evaluated against the same yardstick rather than ranked against one another.

What the definition does not supply is a number. Section 706 contains no speed benchmark of any kind. The statute describes capability sufficient to “originate and receive high-quality voice, data, graphics, and video”—a functional standard, not a quantitative one. Every numerical threshold in this report—100/20 Mbps for fixed service, 35/3 Mbps for mobile, 1 gigabit per second (Gbps) per 1,000 students for schools—is an FCC construct, adopted in earlier proceedings and carried forward here. That matters when we reach what the FCC did with its long-term speed goal.

Five takeaways: the mandate is annual; the proceeding includes a 180-day completion deadline; Congress singles out schools and classrooms; Congress names competition among the tools the FCC is expected to use; and the law requires a list of unserved areas, not merely national percentages.

The Interpretive Shift

The largest policy move in this report is not a number, but a reading of the statute.

In the 2024 Section 706 Report, the FCC assessed broadband against five universal service goals—deployment, affordability, adoption, availability, and equitable access—and concluded that broadband was not being deployed in a reasonable and timely fashion. The 2026 Report rejects that framework outright, calling the 2024 Report “a lone outlier” that “strayed from the statute’s plain language.”

The FCC’s reasoning:

• “Availability” means “deployment.” The statute “identifies ‘availability’ alone as the object of the Commission’s section 706 inquiry,” and availability “depends on whether broadband ‘is being deployed.’”
• Loper Bright compels a single answer. Citing Loper Bright Enterprises v. Raimondo (the Supreme Court ruling that overturned the 40-year-old "Chevron deference" doctrine), the FCC states that it must follow the best textual reading, “for if a statutory reading ‘is not the best, it is not permissible.’”
• The Infrastructure Investment and Jobs Act did not expand the inquiry. The Infrastructure Investment and Jobs Act’s reference to “the statutorily mandated goals of universal service for advanced telecommunications capability under section 706” is best read to refer only to goals already in section 706—reasonableness, timeliness, and the § 1302(a) toolkit. “Nothing in the Infrastructure Act suggests that Congress intended…to expand the Commission’s discretion.”
• Affordability is acknowledged, then set aside. The FCC writes that it does “not minimize the importance of factors like affordability,” and asserts that “evidence shows broadband prices are going down as the cost of living skyrockets along other dimensions.” (The supporting footnote cites a USTelecom pricing index and a Phoenix Center paper.) But affordability, adoption, and equitable access “are not found in the actual text of section 706,” and using them as proxies “would make the inquiry unduly subjective.”
• “Is being deployed,” not “has been deployed.” The current FCC finds that the 2024 Report’s binary reading “effectively read the ‘reasonable and timely’ language out of the statute.” The FCC will now assess incremental progress.

That last reasoning is the one that determines the outcome. Under a progress-based test, the question is no longer whether Americans have broadband; the question is whether the number who lack broadband is falling. On that test, the FCC, not surprisingly, finds the answer to be “Yes.”

Benchmarks

In this proceeding, the FCC sought comment on the 100/20 Mbps standard for fixed broadband adopted in the 2024 report and proposed to eliminate the FCC’s long-term goal of 1000/500 Mbps since it “risks skewing the market by unnecessarily potentially picking technological winners and losers.” The FCC said that “assuming a long-term goal of 1,000/500 Mbps may be unreasonably prejudicial to technologies such as satellite and fixed wireless that presently do not support such speeds.”

In the report, the FCC makes the following conclusions.

• Fixed: 100/20 Mbps is retained. The record “unanimously supports a benchmark of at least this speed.” Commenters seeking higher or symmetrical speeds rely on “market trends and aspirational standards rather than an analysis of the present use of broadband service.” The FCC agrees with USTelecom that raising the benchmark would be premature given the “still-nascent” Broadband Equity, Access, and Deployment (BEAD) Program.
• The 1,000/500 Mbps long-term goal is abolished without replacement. The FCC does not decline to retain the gigabit goal, or sunset it, or commit to revisiting it. The FCC adopts its proposal to “abolish without replacement” the long-term goal established in the 2024 Report—the strongest formulation available, and one that forecloses the incremental option of keeping a long-term goal while adjusting its level. The stated grounds are that a long-term goal “is not mentioned in section 706 and could appear to violate our obligation to conduct our analysis in a technologically neutral manner.” The FCC quotes the NCTA: technological advancement “will continue to drive increases in broadband speeds without regard to any artificial long-term goal.” The FCC concludes, “The private sector, driven by competition, does not need the Commission to centrally plan technological development.”
• Mobile: no single benchmark. The FCC analyzes 5G-NR (fifth-generation New Radio) at 35/3 Mbps outdoor stationary as its primary metric—the highest speed the FCC collects—and retains 7/1 Mbps 5G-NR and 5/1 Mbps 4G LTE (Long-Term Evolution) figures.
• Schools: 1 gigabit per second (Gbps) per 1,000 students retained; no long-term benchmark. A long-term goal “could ‘skew[] the market by unnecessarily potentially picking technological winners and losers.’”

The Rationale Proves Too Much

Each of the three reasons the FCC gives for abolishing its long-term goal applies with equal force to the benchmarks the FCC kept.

“Not mentioned in section 706.” Neither is 100/20 Mbps. Neither is 35/3 Mbps. Neither is 1 Gbps per 1,000 students. If a threshold’s absence from the statutory text were disqualifying, the FCC would have to abandon every benchmark it uses—which would leave the functional definition in § 1302(d)(1) unadministrable. The FCC does not explain why absence from the text disqualifies a future benchmark but not a present one.

Technological neutrality. A 100/20 Mbps benchmark also excludes technologies that cannot deliver 100/20 Mbps; that is what a benchmark is for. Commissioner Anna Gomez makes the point directly: “Lowering standards to ensure certain technologies can meet a benchmark is the opposite of technological neutrality.” In her reading, neutrality means evaluating all technologies against a single standard—not lowering the standard until every technology meets it.

Central planning. If setting a forward-looking speed target is central planning, so is setting a current one—and so is conditioning billions of dollars in universal service support on it, which the FCC does throughout its discussion of high-cost programs. Commissioner Gomez notes that the FCC’s own Rural Digital Opportunity Fund (RDOF) established 1,000/500 Mbps as its Gigabit performance tier, and that more than 85 percent of RDOF winners committed to gigabit service. RDOF bidders chose among performance tiers, and Commissioner Gomez notes that more than 85 percent of winners chose the Gigabit tier of 1,000/500 Mbps. Those commitments are enforceable, with letters of credit and buildout milestones behind them. So the gigabit standard is not speculative inside the FCC's own high-cost programs; it is an obligation the agency is enforcing against the great majority of RDOF winners today, in the same report that declares long-term speeds impossible to predict.

There is also a tension inside the report itself. The FCC justifies retaining 100/20 Mbps partly by pointing to BEAD, a program created by the Infrastructure Investment and Jobs Act and administered by the National Telecommunications and Information Administration (NTIA)—precisely the kind of reliance on an outside statute that the FCC condemns a few pages earlier when it criticizes the 2024 Report for reading the Infrastructure Act into Section 706. Extrinsic sources appear to be available when they support the benchmark the FCC wants to keep, and unavailable when they support the one it wants to drop.

None of this establishes that abolishing the gigabit goal was wrong as policy. But it does establish that the textual rationale offered is not the operative reason—because the FCC applies that rationale nowhere else in the same report.

What the Data Show

The 2026 Report uses the FCC’s June 2025 Broadband Data Collection (BDC) as its primary source, built on the Broadband Serviceable Location Fabric—a dataset of every location in the U.S. where fixed broadband can be installed. The FCC reports on three fixed categories: Fixed Wireline (copper, cable, fiber), Fixed Terrestrial (wireline plus terrestrial fixed wireless), and Any Fixed Service (terrestrial plus satellite).

Fixed availability at 100/20 Mbps, June 2025

	Wireline	Terrestrial	Any Fixed
United States	93.9%	96.9%	100.0%
Urban	98.7%	99.5%	100.0%
Rural	75.3%	86.7%	100.0%
Tribal lands	74.6%	87.6%	99.9%

Rural wireline availability at the FCC’s own benchmark is still 75.3 percent—meaning roughly one in four rural Americans has no wireline option at the benchmark speed. On rural Tribal lands, wireline availability is 57.1 percent; on rural federal reservations, 44.7 percent; in rural Alaskan villages, 40.6 percent.

Mobile: ~95 percent of Americans live in areas with 5G-NR at 35/3 Mbps, including ~79 percent in rural and ~86 percent in Tribal areas.

Both fixed wireline and mobile: ~91 percent nationally, but ~98 percent urban versus ~61 percent rural.

Demographics (§ 1302(c)): Areas without both fixed and mobile service have statistically significantly lower population, density, and per capita income than served areas—for the Fixed Wireline/mobile comparison, $39,939 versus $43,521 average per capita income (2024 dollars).

“Virtually Universally Available"

The report’s opening summary states that with satellite included, broadband deployment to homes or businesses “appears to be virtually universally available to all Americans.” The Section 706 finding puts a number on it: counting satellite, only about 25,000 Americans lack 100/20 Mbps fixed service—one in roughly every 14,000.

That claim rests on a modeled availability figure. The FCC estimates that 99.7 percent of the U.S. population had access to a 100/20 Mbps satellite service as of June 30, 2025. In the same paragraph, the FCC reports that satellite accounts for 2.3 percent of all fixed broadband connections in Form 477 subscription data, and acknowledges that service reliability “may decline during periods of congestion” and that actual speeds “could depend on location, time of day, and type of service plan.”

The FCC gives a reason for counting satellite at all, noting that in 2026 it updated its rules governing spectrum sharing between geostationary and non-geostationary satellite systems. The change was intended to ease capacity constraints and encourage a more competitive satellite marketplace. The FCC says that "[g]iven this development," the Report presents fixed availability estimates both with and without satellite. Whether a 2026 rule change justifies counting satellite in a June 2025 dataset is a fair question—the rule postdates the data—but the FCC did explain itself.

The FCC does distinguish between availability and usability. A footnote to the "virtually universally available" sentence notes that while satellite is widely deployed, its usability "may be limited by certain factors or during certain periods of time, resulting in slower speeds and higher latency in remote areas"—citing X-Lab's Starlink capacity analysis. The finding paragraph repeats the caveat, listing line-of-sight obstructions, capacity constraints in high-demand areas, and network congestion. The FCC also preserves the distinction structurally, presenting every fixed estimate both with and without satellite.

What the report does is subordinate the caveat. The claim runs in the body text; the qualification runs in a footnote at the first instance and in a trailing sentence at the second. The summary bullets at the top of the report and FCC Chairman Brendan Carr's statement carry the satellite-inclusive figure without the caveat attached. And the caveat is qualitative throughout—"may be limited," "could result"—while the availability figure is precise to a tenth of a percent. Nothing in the majority report estimates how many of the 99.7 percent can actually obtain 100/20 Mbps satellite service in practice. Commissioner Gomez's statement, discussed below, supplies the numbers the majority leaves qualitative.

Update: Four days after release, the FCC issued an erratum correcting the finding paragraph itself. The original text described Starlink as "a worldwide network already with over 7 million American subscribers." The correction removes the American figure and replaces it with a global one—a subscriber base growing from 6 million to 12 million between June 2025 and June 2026. The FCC's statement of its statutory conclusion had characterized worldwide subscribership as domestic. After the correction, the finding offers no estimate of how many Americans actually subscribe to satellite broadband.

Competition: The Number the Report Does Not Headline

Section 1302(a) names “measures that promote competition in the local telecommunications market” as one of the four tools Congress expected the FCC to use. Competition is not a side issue in Section 706; it is one of the statutory instruments. That makes the report’s competition data directly relevant to the mandate—and the report tells two different stories about it.

The story in the summary

The FCC’s opening bullets state that the number of competitive options available to Americans “continues to grow,” citing a roughly 56 percent increase between June 2023 and June 2025 in the share of households with three or more Fixed Terrestrial options at 100/20 Mbps. By June 2025, ~46 percent of households had no more than one Fixed Wireline provider offering service at 100/20 Mbps, down from ~51 percent a year earlier.

The story further into the report

As of June 2025, 2,082 entities reported making fixed broadband available to residential consumers at speeds exceeding 200 kbps in at least one direction. The total number of fixed providers fell by approximately 4 percent from June 2024 to June 2025, and the decline was steeper in rural areas—approximately 4 percent in rural areas versus approximately 2 percent in urban areas. Market structure is also highly concentrated: the overwhelming majority of providers cover less than 1 percent of the U.S. population, and in both June 2023 and June 2025, only 13 providers—three of them satellite—reported making fixed broadband available to at least 5 percent of the U.S. population.

Different measures, different outcomes

Both service availability figures are accurate, because they count different things. The household figures measure the share of households with a given number of options, weighted by where people live. The provider figure counts firms nationally, and a firm registers once whether it serves one county or forty states. The national provider count is dominated by very small companies, and a 4 percent decline in that count can coexist with rising household options: the firms that move the household numbers are the large ones, and the firms disappearing are mostly not. That is a reconciliation, not a reassurance. Small fixed providers are concentrated in rural markets, where the report finds the decline is steepest and where households have the fewest alternatives to begin with.

What's missing

The report states the decline in national provider count and offers no explanation for it: no discussion of consolidation, of exit by small rural providers, or of whether the decline reflects mergers, wind-downs, or changes in reporting. There is no discussion of what a shrinking provider base implies for the § 1302(a) directive to promote competition. And the FCC’s affirmative finding does not mention the provider decline at all.

Implications for rural deployment

For state broadband offices, the practical point is this: the number of firms capable of bidding on a rural buildout is declining, and the number of firms operating at a national scale has not changed in two years. That bears directly on the competitiveness of BEAD subgrantee selection and on what happens to service quality in rural markets once construction obligations lapse.

The Funding Gap—and the Qualifier Attached to It

About 3.1 percent of Americans (10,581,094 people) lack access to Fixed Terrestrial service at 100/20 Mbps. Of those, about 32 percent (3,394,105) live in areas with an enforceable federal funding commitment; about 68 percent (7,186,989) do not. Excluding BEAD, the Rural Digital Opportunity Fund accounts for nearly half of committed locations; U.S. Treasury’s Capital Projects Fund accounts for about 21 percent. The report notes that the FCC’s Broadband Funding Map does not yet include BEAD commitments; the snapshot underlying these figures is dated November 23, 2025.

The FCC then discounts its own figure. The unfunded, unserved share is “likely a meaningful over-statement of locations that lack both funding and service,” the report says, because it does not reflect “the general expansion of broadband coverage since June 2025” or federally supported buildouts reported more recently, including BEAD data.

The caveat is fair on its face—BEAD commitments are real and are not included on the Broadband Funding Map. But the FCC is discounting the single number in the report that quantifies the remaining unfunded gap by pointing to data the report does not contain and does not attempt to estimate. No corresponding qualifier is attached to the deployment figures, which are drawn from the same June 2025 baseline and are subject to the same staleness. The reader is told to treat the bad news as overstated and the good news at face value.

Schools and Classrooms: Three Measures, and the Least Flattering One Is the FCC’s Own

Section 706 names elementary and secondary schools and classrooms twice. The FCC's unit of analysis changes twice, and the resulting numbers are not comparable.

Measure one—school districts. The FCC previously relied on Connect K-12, a Connected Nation project that reported the share of school districts meeting the 1 Gbps per 1,000 students benchmark using publicly available E-Rate data. Connect K-12 stopped publishing after 2023, having found that 74 percent of school districts met the benchmark that year.

Measure two—“entities,” approximating Connect K-12. FCC staff reviewed Connect K-12’s methodology, approximated it, and report that approximately 78 percent of entities met the 1 Gbps per 1,000 student threshold in 2024, “up from 74% in 2023.” While the 74 percent baseline is a district figure produced by Connect K-12, the 78 percent is an entity figure produced by FCC staff. The report presents these two figures as a year-over-year comparison without reconciling the units of measurement.

Measure three—the FCC’s own, by student. FCC staff then built a separate measure from E-Rate Form 471 data. Here the unit of analysis is the “final recipient entity, which is generally a single school or a single building within a school district.” For each entity, staff divided total download bandwidth by the number of full-time students; an entity that exceeded 1 Mbps per student was considered to meet the benchmark. Staff then summed the students at qualifying entities and divided by the total number of students. District bandwidth is allocated across buildings in proportion to student share, and the measure counts only connections between applicants and internet service providers or state and regional networks—excluding dark fiber and connections to libraries and non-instructional facilities.

On that measure, 52 percent of students were at schools meeting the benchmark in 2025, up from 32 percent in 2021. In 32 states, at least half of students were in schools that met it.

Why the change in unit matters 

A district-level measure treats a 40,000-student district and a 300-student district as a single observation each, and the measure can pass a district-wide aggregate purchase of bandwidth even when individual buildings fall short. Moving to the building level and weighting by enrollment addresses a different, harder question: how many actual students are in a building with adequate connectivity? The FCC says so itself. In a footnote, the FCC explains that Connect K-12’s measure was at the district level while its own measure examines bandwidth per student, and that the FCC approach “better reflects the impact of the E-Rate program by matching students to the actual services they receive.”

So the FCC built a more demanding measure, called it better, ran it—and got 52 percent. Nearly half of American students attend schools that do not meet the FCC’s own short-term benchmark. That figure appears once and is never mentioned again.

The FCC also declined to use alternative sources suggested by commenters—including the State Educational Technology Directors Association’s Universal Connectivity Imperative, the 2024 National Educational Technology Plan, Funds For Learning’s E-Rate affordability analysis, and New America’s broadband equity research—as “not sufficiently comprehensive for measuring progress toward the short-term benchmark.”

The Finding

The FCC’s affirmative finding rests on rates of change: 23 percent fewer Americans lacked Fixed Terrestrial service at 100/20 Mbps in June 2025 than in June 2024, and almost 43 percent fewer than in June 2023. The number lacking 5G-NR at 35/3 Mbps fell nearly 16 percent year over year and over 30 percent over two years.

The 2026 Report also credits the FCC’s own policymaking, and the window it chooses is worth noting. The finding attributes progress to FCC efforts since the 2024 Report, “particularly those from the last 16 months”—a period beginning around April 2025, shortly after the current Chairman took office. But the deployment data being explained run from June 2024 to June 2025, and network construction reported in June 2025 reflects planning, permitting, and capital commitments made well before that. The report offers no analysis connecting specific FCC actions to specific deployment outcomes; the attribution is asserted rather than demonstrated. That is a familiar feature of agency self-reporting under both parties, but it is worth naming here because this passage does the work of showing that the FCC discharged its § 1302(a) duty to encourage deployment.

The 2026 Report is candid about what remains: 10.5 million people still lack fixed terrestrial service at 100/20 Mbps, including more than 9 million in rural areas and roughly 500,000 on Tribal lands. The Report states the FCC “may [not] now rest on its laurels and declare the job well done.”

Did the FCC Fulfill the Mandate?

Three elements of § 1302 deserve close scrutiny here.

1. Timeliness: The FCC missed the statutory deadline by more than six months, never says so, and used the extra time to analyze data it already had.

Section 1302(b) requires the FCC to “complete the inquiry within 180 days after its initiation.” The FCC initiated this inquiry with the Nineteenth Section 706 Report Notice of Inquiry (NOI) (FCC 25-46, adopted August 4, 2025 and released August 8, 2025). At the time, the FCC’s own fact sheet for that item stated: “The Commission’s section 706 report must be issued within 180 days after release of the Notice.” That deadline fell on February 4, 2026. The report arrived 371 days after initiation.

The annual cadence has also lapsed. The prior report was released March 18, 2024, a gap of roughly 29 months. And that report was itself late: then-Commissioner Carr’s March 2024 dissent opened by noting it was “the FCC’s first in three years.” This is a failure of long standing, and not one confined to a single administration.

The Report does not mention the 180-day deadline anywhere. The phrase does not appear in the document. A report that devotes nine paragraphs to the discipline of following Section 706’s “plain language” is silent on the one deadline in that section that the FCC plainly missed.

The delay would matter less if the additional year had bought better data. It did not. The report’s central deployment findings rest on BDC data as of June 30, 2025—roughly 13.5 months old at the time of release. The FCC opened its eighth BDC filing window on January 2, 2026, for availability data as of December 31, 2025, with a filing deadline of March 2, 2026. That collection closed more than five months before this report was adopted. The FCC also demonstrates in this Report that it can work with December datasets. The Report's Figure 3, which tracks the number of fixed providers over time, draws on five semiannual BDC snapshots: June 2023, December 2023, June 2024, December 2024, and June 2025. The agency was willing to build a trend line on December collections; it simply did not extend that line to December 2025. [Update: The erratum makes the point again. Correcting the finding paragraph, the FCC reached for satellite subscriber data through June 30, 2026, drawn from a corporate filing made ten days before the report was released. The agency was willing to use figures a full year newer than the deployment baseline on which the finding rests.]

The Report works around the gap rather than closing it. Its Broadband Funding Map figures come from a November 23, 2025 snapshot—more recent than the availability data they are matched against. Its Form 477 subscription figures are as of June 30, 2025 but reflect filings submitted through December 30, 2025. And the report discounts its own unserved-and-unfunded estimate by invoking “the general expansion of broadband coverage since June 2025”—an acknowledgment, in the FCC’s own words, that the analytical baseline is stale.

There is a pointed precedent here. Chairman Carr’s March 2024 dissent from the prior report faulted it for relying on “broadband deployment statistics that are 15 months old when newer information is available.” The 2026 Report relies on statistics roughly 13.5 months old. Whatever one concludes about the 2024 Report, that objection has not been cured; it has been narrowed by about six weeks.

The practical consequence for readers is that the report’s headline conclusion—that the unserved population is falling quickly—describes a moment more than a year past, during which BEAD subgrantee selection, the end of the Affordable Connectivity Program, and a substantial reworking of federal broadband funding were all in motion. A December 2025 baseline would have told policymakers something about the present.

2. Schools and classrooms: Measured, then omitted from the finding.

Section 1302(a) and (b) both single out “elementary and secondary schools and classrooms.” The Report does collect the data. But the Section 706 Finding contains no mention of schools or classrooms at all.

That omission matters because the school data are the weakest in the 2026 Report. If 52 percent of students attend schools meeting the FCC’s own short-term benchmark, then roughly half do not—and the FCC’s finding does not weigh that fact one way or the other. A statute that names a constituency twice arguably requires the ultimate determination to say something about it.

3. The § 1302(c) list: Satisfied in substance, not in form.

The statute requires the FCC to “compile a list of geographical areas that are not served by any provider” and report population, density, and per capita income “for each such unserved area.” The Report aggregates served and unserved populations at the Census block group level, because that is the smallest geography for which per capita income is available, and publishes county-level availability with demographics. The FCC publishes statistical comparisons of served versus unserved areas.

What it does not publish is a discrete list of unserved areas. What the Report offers is a table of all counties with percentages—not an enumeration of areas served by no provider. Whether that satisfies § 1302(c) is a fair question for Congress. Note also a definitional gap the report leaves open. Section 1302(c) asks for areas not served by any provider of advanced telecommunications capability, "as defined by subsection (d)(1)"—so the list depends entirely on which services qualify. The FCC never says. The Commission sorts the Broadband Data Collection's technology codes into three tiers—Fixed Wireline, Fixed Terrestrial, and Any Fixed Service—and reports estimates both including and excluding satellite, without identifying which tier corresponds to the statutory category. That choice has enormous consequences. Counted at the Any Fixed Service tier, the unserved population is roughly 25,000 people nationwide, about one in every 14,000 Americans. Counted at Fixed Terrestrial, it is 10.5 million. The FCC performs its demographic analysis at the Fixed Terrestrial line—the more useful choice—but never explains why that line, and not the other, defines the areas Congress asked it to list.  The demographic analysis the FCC actually performs uses Fixed Terrestrial service as the dividing line—a more useful choice, but not the one the statutory text on its face describes.

Views of the Commissioners

Chairman Brendan Carr frames the Report as validation:

Commissioner Anna Gomez concurs; she does not dissent. Her central objection: “Omitting affordability and usability from the analysis was a policy decision with which I disagree.” Congress’s deployment requirement, she writes, “is a floor, not a ceiling.”

Her most substantive contribution is on satellite, and she notes she worked with the Chairman’s office and the FCC’s Wireline Competition Bureau to add usability caveats to the Report itself. She calls it “misleading to state that 99.7% of the U.S. population has access to satellite broadband at 100/20 Mbps,” citing that:

• in 2025 fewer than half of satellite broadband users actually experienced 100/20 Mbps speeds;
• in areas exceeding 6.66 households per square mile, satellite may not deliver 100/20 Mbps;
• one-time network demand fees of up to $1,500 have been imposed in high-density areas; and
• typical latency in remote areas including Alaska exceeds 100 milliseconds, against a 30-millisecond benchmark identified as optimal for emerging artificial intelligence applications.

On the abolished gigabit goal, she is direct, saying, “Lowering standards to ensure certain technologies can meet a benchmark is the opposite of technological neutrality.” She notes the Rural Digital Opportunity Fund set 1,000/500 Mbps as its Gigabit performance tier and that more than 85 percent of RDOF winners committed to gigabit service.

Commissioner Olivia Trusty supports the Report while narrowing its significance. “While the Section 706 review once was the centerpiece of the FCC’s broadband analysis, today Congress has situated it as one part of a broader marketplace inquiry.” She notes that Broadband DATA Act mapping and the Broadband Funding Map “play a direct role in policymaking without being filtered through the lens of the section 706 inquiry,” and cautions that the FCC “must be mindful of its limited scope.” It is a notable framing from a member of the majority: the Section 706 Report matters less than it used to.

The Bottom Line

The FCC did what Section 706 asks in one narrow sense: conducting an inquiry, compiling deployment data, and making a determination. Whether the FCC fulfilled the mandate is a harder question. The FCC completed the inquiry 371 days after initiating it, against a 180-day statutory deadline it never acknowledges—and used the extra year to analyze a June 2025 baseline rather than the December 2025 collection it closed in March. The FCC gathered data on schools and classrooms—which the statute names specifically—and then made its finding without reference to them. It documented a shrinking base of broadband providers, shrinking fastest in rural markets, without connecting that to the statutory instruction to promote competition. And it eliminated its long-term speed goal on a textual rationale it applies to no other benchmark in the same report.

Ten and a half million Americans still lack fixed terrestrial broadband at the FCC’s own benchmark. Roughly seven million of them live in areas where no federal dollars have yet been committed to deploy broadband networks. Under the reading adopted here, that is not a negative finding—and so no immediate action is required.

Notes

1. Actually, Figure 4 in the report shows 76.9% of households, not Americans, and that figure is for Any Fixed Service—i.e., counting satellite.

The Benton Institute for Broadband & Society is a non-profit organization dedicated to ensuring that all people in the U.S. have access to competitive, High-Performance Broadband regardless of where they live or who they are. We believe communication policy - rooted in the values of access, equity, and diversity - has the power to deliver new opportunities and strengthen communities.

---

© Benton Institute for Broadband & Society 2026. Redistribution of this email publication - both internally and externally - is encouraged if it includes this copyright statement.

---

For subscribe/unsubscribe info, please email headlinesATbentonDOTorg