On April 29, 2026, the Federal Communications Commission (FCC) released a Notice of Proposed Rulemaking (NPRM)—a formal request for public comment on possible rule changes—titled Reforming the High-Cost Program for an All-IP Future. As of now, the document is circulated for tentative consideration at the FCC's May 20, 2026 open meeting and does not yet constitute an agency action. The NPRM proposes reassessing roughly $1.6 billion in annual federal broadband subsidies. That money currently flows to small, rural telephone companies serving some of the hardest-to-reach communities in the country. Depending on what the FCC ultimately decides, those carriers could face reduced support, new build-out obligations, or an entirely restructured funding relationship. The consequences of the FCC's decisions could impact rural residents, community institutions, and local economies that depend on these networks. State broadband officials managing their share of the $42.5 billion Broadband Equity, Access, and Deployment (BEAD) Program will have a direct stake: the NPRM explicitly raises the question of how the FCC's Universal Service Fund (USF) and BEAD support should be coordinated going forward.

What the High-Cost Program Does and Why It Is Under Review

The USF—grounded in Section 254 of the Communications Act of 1934, which requires the FCC to "preserve and advance universal service"—is meant to ensure that Americans in rural and high-cost areas have access to communications services comparable to those in urban areas, at reasonably comparable rates. The USF is funded by contributions from telecommunications providers and VoIP (Voice over Internet Protocol) service providers, assessed on their interstate and international end-user revenues; providers typically pass some portion of these costs to their customers.

The High-Cost Program, administered by the Universal Service Administrative Company, is the USF program that subsidizes the cost of building and maintaining networks in areas where the economics of private investment alone would not support adequate service. There are currently a dozen different mechanisms under the High Cost umbrella. This NPRM focuses on a specific subset—the mechanisms that serve rate-of-return carriers, a class of smaller rural telephone companies that historically have set their rates and received support based on their actual costs of service. (Price-cap carriers—generally the large incumbent providers like AT&T and Verizon—are regulated differently and are not the subject of this proceeding.)

What's Under Review

The three mechanisms examined in this NPRM account for approximately $1.6 billion of the High-Cost Program's $4.5 billion annual budget.

1. Connect America Fund

Connect America Fund Broadband Loop Support (CAF BLS) subsidizes carriers with high local loop—the physical lines connecting customers to the network—costs for both voice service and broadband-only lines. There are currently 249 rate-of-return carriers receiving CAF BLS. The FCC paid CAF BLS carriers approximately $995 million in support claims in 2025. Each CAF BLS carrier was required to deploy broadband service capable of at least 25/3 Mbps (25 megabits per second download, 3 megabits per second upload) to a defined number of locations in their service area by the end of 2023. The NPRM notes that more than 25 percent of CAF BLS—roughly $250 million— currently subsidizes traditional voice-only service. 

2. High Cost Loop Support

High Cost Loop Support (HCLS) provides support for traditional Public Switched Telephone Network (PSTN) voice lines in areas with average costs above a specified threshold. HCLS effectively subsidizes local voice service rates by shifting cost recovery from the intrastate to the interstate jurisdiction. (The federal USF is funded through assessments on interstate revenues, while intrastate cost recovery falls under state-level mechanisms.) Of the 249 CAF BLS carriers, 243 also received HCLS in 2025. HCLS support has declined sharply over time—from $741 million in 2015 to $202 million in 2025—as consumers have shifted from traditional voice lines to broadband-only service. There are no speed or deployment obligations attached to HCLS.

3. Alternative Connect America Cost Model

The Alternative Connect America Cost Model (A-CAM) was developed to give rate-of-return carriers a path to model-based—rather than cost-based—support, in exchange for defined broadband deployment obligations over a fixed term. (A cost model estimates what it should cost to serve an area, producing a fixed monthly support amount, rather than basing support on the carrier's reported costs.) Three versions are now approaching their end dates:

• A-CAM I was originally offered in 2016 with a 10-year term. Nine carriers at the holding company level (meaning a parent company with multiple subsidiaries is counted once, so the number of individual service areas receiving support may be higher) still receive this support. A-CAM I is set to sunset at the end of 2026. A-CAM I and Revised A-CAM I together account for approximately $174 million annually.
• Revised A-CAM I was adopted in 2018, extending support through 2028 with a higher funding cap and increased deployment obligations. Currently, 71 carriers at the holding company level still receive Revised A-CAM I support.
• A-CAM II was also adopted in 2018, with support also running through 2028. Currently, 71 carriers at the holding company level receive this support. A-CAM II provides approximately $218 million annually. 

A fourth mechanism, Enhanced A-CAM, is not under review in this proceeding. It was first offered in 2023 and provides support over a 15-year period through the end of 2038, with obligations to deploy at least 100/20 Mbps service. Three hundred and four (304) carriers accepted Enhanced A-CAM, voluntarily transitioning away from prior mechanisms. 

Three Paths Forward

The NPRM does not propose a specific rule. Instead, it lays out three possible paths and asks for public comment on each.

Option 1—Update existing mechanisms to align with the current broadband deployment and funding environment. This would involve modifying CAF BLS and HCLS, and determining next steps for sunsetting A-CAM mechanisms, without replacing them wholesale.

Option 2—Create a new, modernized fixed-support mechanism to replace the existing patchwork of legacy and model-based programs with a single structure. The NPRM asks whether such a mechanism should be model-based, what types of costs it should cover (capital expenditures, operating expenses, or both), and what deployment obligations should attach.

Option 3—Do nothing. The final option is to maintain the status quo for legacy mechanisms and allow the A-CAM mechanisms to sunset.

The NPRM asks extensive questions under each option, including:

• What speeds should be required?
• What latency thresholds are needed to support AI applications in rural areas? (Latency is the delay between sending and receiving data—lower latency means faster response times, which matters for real-time applications like video calls, remote medicine, and AI-driven tools.)
• Should cybersecurity plan requirements apply?
• Should support be conditioned on broadband adoption rates—for instance, requiring that 70 percent of locations subscribe before full support is paid?
• Should there be graduated penalty structures for carriers that fail to meet performance obligations?

The NPRM does not include an estimate of how many carriers might exit the market or how many locations would lose service under this option, information that stakeholders—especially rural residents and communities—may want to address in the record.

A-CAM I Extension

The NPRM does offer one concrete, near-term proposal: extending A-CAM I support by two years—from its current sunset at the end of 2026 through the end of 2028—to align it with the Revised A-CAM I and A-CAM II sunset dates. During the extension, carriers would continue receiving their previously authorized support amounts but would be required to maintain voice and broadband service and serve additional locations upon reasonable request. 

The NPRM proposes tiered support recovery for carriers that fail network testing during the extension period: no recovery for full compliance; 25 percent recovery for Level 1 non-compliance; 50 percent for Level 2; 75 percent for Level 3; and 100 percent for Level 4. 

Wait, Do We Still Need a High Cost Program?

A significant portion of the NPRM examines whether high-cost support is justified in areas where competitive alternatives already exist or where other federal funding has been committed to provide service.

The FCC estimates there are approximately 3.1 million broadband serviceable locations (BSLs)¹—locations that could be served with broadband—in the areas covered by the relevant rate-of-return carriers. Of those, 80 percent are already served by the incumbent carrier (the rate-of-return carrier receiving the subsidy) at speeds of at least 100/20 Mbps, 58 percent are served by an unsubsidized competitor (not counting satellite) at 100/20 Mbps, and only about 267,000 locations—roughly 9 percent—still lack 100/20 Mbps service from any terrestrial provider. For the 9 percent, more than 99 percent are covered by a low Earth orbit (LEO) satellite provider at speeds of at least 100/20 Mbps, based on the National Broadband Map as of June 30, 2025. 

For voice service, the FCC estimates that at least one mobile provider serves approximately 99 percent of the 3.1 million BSLs in these areas, and about 43 percent have fixed voice service from an unsubsidized interconnected VoIP competitor. 

The FCC invokes a longstanding principle that providing support in areas where an unsubsidized competitor already offers high-quality service is "an inefficient use of limited universal service funds," and asks whether ongoing support should be limited to areas where providers would not otherwise operate without a subsidy.

Oh, LEO

The NPRM devotes substantial attention to the role of LEO satellite service—specifically citing SpaceX's Starlink and Amazon's Leo—and notes that Starlink's "Residential" plan is offered for $120 per month with typical download speeds of up to 305 Mbps and upload speeds of up to 40 Mbps. The FCC currently provides up to $200 per month per location in USF support to some legacy carriers for voice and 25/3 Mbps service. 

The FCC asks whether locations covered by LEO satellite service should be removed from high-cost carriers' service obligations—and whether carriers should relinquish a corresponding portion of their support if satellite covers areas they are not serving. The FCC also raises, but does not resolve, the question of whether satellite providers would raise their prices significantly if terrestrial competitors lost subsidy support.

The FCC has a statutory obligation under Section 254 to periodically reassess what counts as adequate service and where public subsidy is actually needed. Some will argue the numbers above make that case plainly: if LEO satellite now provides 100/20 Mbps service to 99 percent of locations that lack terrestrial service at that speed, continuing to pay up to $200 per location per month for legacy voice and 25/3 Mbps service may not be prudent stewardship of public funds. Stakeholders who believe satellite-delivered service is an inadequate substitute will need to make that case in the record—with data on actual performance (speed and latency), affordability, reliability, and adoption rates in affected communities.

The IP Transition

The FCC asks what role USF funding should play in encouraging the transition from traditional Time-Division Multiplexing (TDM) networks—the technology underlying the legacy PSTN—to all-Internet Protocol (IP) networks. The NPRM asks about potential cost savings (reduced maintenance, electricity, and real estate costs) and the challenges such a transition could pose for rural areas, including the ability of rural 911 systems to function in an all-IP environment. 

The NPRM also includes a section titled "Delete, Delete, Delete," inviting comment on which existing High-Cost Program rules have become unnecessary and should be eliminated. 

Issues to Keep an Eye On

The NPRM raises a number of issues that stakeholders might want to track.

Where is the FCC headed?

Several passages in the NPRM strongly signal the FCC's current thinking. The juxtaposition of $120/month Starlink service against $200/month USF support per location, the observation that 99 percent of currently unserved locations are covered by LEO satellite, and repeated invocation of the "inefficiency" of subsidizing competition collectively suggest the FCC is, at a minimum, skeptical that current support levels are warranted across all areas.

Conspicuously absent: affordability.

The NPRM addresses coverage—whether a service is technically available—but does not ask whether satellite or other alternatives are affordable to rural residents, including low-income households. The Starlink pricing cited ($80–$120/month) may be within reach for some rural consumers, but the NPRM does not engage with affordability as a criterion for service adequacy. This is a meaningful gap. Stakeholders might raise questions including:

• At $80–$120/month, is LEO satellite service within reach for households that currently receive Lifeline-discounted service through a rate-of-return carrier?
• The FCC's own broadband benchmark is now 100/20 Mbps. Does meeting that speed threshold, without a comparable affordability standard, actually advance the Section 254 goal of "reasonably comparable rates"?

Carefully chosen language on satellite.

The NPRM asks whether federally supported terrestrial networks should be classified as "primary infrastructure" while satellite serves as a "secondary and redundant layer"—and then asks whether such treatment is consistent with the FCC's technology-neutral approach. This framing appears to set up a tension the FCC will need to resolve: the BEAD Program, which the National Telecommunications and Information Administration has restructured to be technology-neutral, has already approved plans that include LEO satellite as a primary delivery technology in some states. The NPRM does not reconcile these two federal postures.

The BEAD coordination question.

The NPRM raises the question of how USF and BEAD should be aligned going forward, but does not provide a framework for that coordination. State broadband offices should note that the NPRM's approach to defining "enforceable commitments"—which could trigger reductions in USF support—could directly affect how BEAD-funded projects interact with existing high-cost carrier territories. Which states are most exposed to the interaction between BEAD implementation and possible CAF BLS reductions? If a state's BEAD final proposal has already been approved and includes a location that is also served by a rate-of-return carrier receiving CAF BLS, how does the FCC propose to handle the overlap?

Adoption as a condition of support.

The NPRM floats—but does not commit to—conditioning full support on a 70 percent broadband adoption rate within a carrier's service area. This would be a significant departure from current practice, which ties support to deployment (making service available) rather than adoption (subscribers actually taking service). Adoption is heavily influenced by a number of factors—like income, digital literacy, device access—largely outside a carrier's control. Which communities are most likely to fall below a 70 percent adoption threshold, and would the adoption condition effectively penalize carriers serving the most disadvantaged populations?

The Lifeline Program

The NPRM is silent on how changes to high-cost support would interact with Lifeline—the USF program that subsidizes service for low-income consumers. If a rural carrier loses high-cost support and exits the market, Lifeline subscribers in that area lose their service provider. The NPRM does not address this potential impact. 

Tribal Lands

The NPRM does not appear to distinguish areas on tribal lands, which historically have had significantly worse broadband access and different regulatory considerations (tribal sovereignty, federal trust responsibilities, the FCC's own tribal priority rules). 

The USF Contribution Mechanism

The NPRM notes that the High-Cost Program's total budget is $4.5 billion annually, but it does not address the ongoing debate over whether the contribution system—how the fund is financed—is sustainable. This is a structurally related issue that many stakeholders consider equally urgent.

Any specific numbers on who would lose service.

As noted above, the NPRM presents the "do nothing/let it sunset" option (#3 above) without offering any estimate of how many carriers might exit the market or how many locations would be affected. Absent data from the FCC to help estimate potential losses, stakeholders may want to offer their own and raise concerns about widening the digital divide. What would the transition look like for the communities currently served?

Is LEO a true substitute?

Stakeholders may ask a number of questions about broadband service delivered via LEO. 

• Coverage on the National Broadband Map does not equal subscription. What is the actual adoption rate of LEO satellite service in rural rate-of-return territories today, and among which demographic groups, especially those with lower incomes?
• The NPRM cites Starlink availability data but not performance data in practice. How does Starlink's service reliability compare to terrestrial fixed broadband in areas with extreme weather or difficult terrain?
• What about competition? What happens to Starlink pricing if it becomes the only viable option in a territory and terrestrial competition disappears?

Steps From Here

This NPRM was circulated for tentative consideration at the May 20, 2026 FCC open meeting. There are many steps to go before final Commission action. To begin, the issues raised remain subject to change until the full FCC adopts the item. The NPRM must be published in the Federal Register, which could be a month or more after the May 20 vote. Comments will then be due 60 days after Federal Register publication; reply comments will be due 90 days after. The FCC is likely to then take a few months to review the comments and arrive at changes to its rules. Those changes will then be voted on at another open FCC meeting.

The proceeding is assigned to WC Docket Nos. 26-96 and 10-90. There will be much debate to come.

Introducing this item, FCC Chairman Brendan Carr said, "The communications sector never stands still, and neither should our rules." He added that the NPRM "explores how the agency’s legacy high-cost programs can be refreshed to reflect today’s marketplace."

This article is meant to serve as a quick summary of the proposed proceeding. For details on the NPRM and an FCC fact sheet, see Reforming the High Cost Program for an All-IP Future.

Notes

1. The Broadband Serviceable Location (BSL) Fabric is a government-maintained database identifying every location—such as a home or small business—in the country where fixed broadband service could be installed. A BSL is distinct from locations that are actually served, and is not equivalent to households or census units.

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

Kevin Taglang
Executive Editor, Communications-related Headlines
Benton Institute
for Broadband & Society
1041 Ridge Rd, Unit 214
Wilmette, IL 60091
847-220-4531
headlines AT benton DOT org

Share this edition:

Broadband Delivers Opportunities and Strengthens Communities