The impact of turning to satellite internet.

In March 2025, U.S. Secretary of Commerce Howard Lutnick launched an effort to change key elements of the $42 billion Broadband Equity, Access, and Deployment (BEAD) Program, which Congress established in the Infrastructure Investment and Jobs Act of 2021. Secretary Lutnick’s stated goal for the reforms is to get the “benefit of the bargain.” 

—Howard Lutnick, Secretary of Commerce, during his confirmation hearing.

To accomplish this, Secretary Lutnick reportedly plans to shift BEAD away from the deployment of fiber networks, which can be expensive, and toward the deployment of low-earth orbit (LEO) satellite networks, which can be relatively cheap.¹ 

But who, exactly, stands to benefit from this bargain? And is LEO really the bargain that Secretary Lutnick seems to think it is?

Who will be Impacted by Changes to BEAD?

BEAD is designed to bring high-speed, reliable, and affordable broadband to areas with little to no existing internet service, i.e. rural communities. Such communities tend to be predominantly white, have lower-than-average incomes, and are represented by Republican members of Congress (by targeting certain demographics, BEAD lives up to the "Equity" part of its name). For good or ill, any changes to BEAD will have a disproportionate impact on this population.

The following table shows the ratio of BEAD-eligible locations broken down by state and congressional district.² 

The “Bargain”: Comparing LEO and Fiber

Secretary Lutnick likely has many goals for the BEAD program, but this analysis focuses on just one: increasing the amount of BEAD awards that go to LEO providers (namely Elon Musk’s Starlink and Amazon’s Project Kuiper).³ Since BEAD is currently designed to deploy fiber networks, the best way to evaluate this potential change is to compare LEO and fiber technologies. 

However, in many ways, fiber service and LEO service are not natural competitors. Fiber networks are designed to provide high-speed service to large numbers of users who live near network infrastructure. LEO networks, by contrast, are designed to provide service at lower speeds and higher latency to a limited number of users in areas so remote that they have virtually no other options. Elon Musk himself has described Starlink as for “people who simply have no connectivity right now. Or the connectivity is really bad.” In short, the technologies fill different roles. 

The reason LEO and fiber are in competition is because BEAD is currently designed to deploy fiber networks in remote areas (i.e. areas currently served by LEO), effectively bringing high-speed, reliable, affordable broadband to those who currently have little to no service. To the extent Secretary Lutnick shifts BEAD away from fiber and towards LEO, he is essentially saying that this endeavor is not worth the price.

Cost of Infrastructure

Generally speaking, fiber networks are more expensive to deploy than LEO networks. This is due to the upfront cost of labor and materials for fiber network construction. LEO networks, by contrast, are more expensive to maintain than fiber networks. This is due to the need to replace LEO satellites every five years or so. When viewed over a 30-year period (i.e., an infrastructure time horizon), the total cost of LEO infrastructure can be much higher than the total cost of fiber infrastructure.⁴ This is largely due to the fact that a fiber network is built once whereas a LEO network must be rebuilt continuously.

Cost to Subscribers

The high ongoing cost of LEO networks likely explains, in part, why LEO service is much more expensive to the subscriber than fiber service.⁵ Starlink service is twice as expensive as the most equivalent (though still much faster) fiber service, and the difference is more extreme when compared on a price-per-megabit basis, with Starlink being 20 times more expensive than fiber.⁶

In every state, affordability is the main reason people can’t get online. So if Secretary Lutnick’s goal is to “get broadband into the hands of low-income people,” then LEO will not be the best tool.

Speed and Capacity

LEO service is generally much slower than fiber service. Fiber service has virtually unlimited capacity, while Starlink service currently maxes out at 220 Mbps (for a real-world example, in recent years, Starlink’s median speed was 65 Mbps while AT&T Fiber’s median speed was 350 Mbps). Starlink’s speeds have also slowed over time as more people subscribe, leading Starlink to institute waitlists and congestion charges. Additionally, fiber offers high upload speeds, while LEO’s upload speeds are closer to those of DSL, limiting its usefulness for entrepreneurship, healthcare, employment, and AI. 

Image from CNET

Reliability

LEO service is affected by weather (e.g. clouds) and obstructions (e.g. trees, mountains, buildings); it is subject to speed-constricting data caps; and it is vulnerable to national security threats. Fiber service, by contrast, is less susceptible to interference, does not have data caps, and, being located on U.S. territory, is easier to protect from foreign threats. However, when it comes to natural disasters, fiber’s terrestrial infrastructure can be vulnerable, whereas LEO has proven to be resilient, quick to deploy, and able to support various emergency services. 

Long-Term Investment

Fiber networks are a long-term asset to the communities they serve. Even if an individual ISP fails, the fiber network remains, and another ISP can take over and continue to provide service. In this way, a community is virtually guaranteed to get the full value out of a fiber network’s 35-year lifespan (50 years, by some estimates, and longer from the associated conduit and utility poles). Moreover, because a fiber network is physically located in the community it serves, the jobs and revenue it generates can remain local. By contrast, LEO infrastructure is not an asset to the community it serves, and the jobs and revenue LEO networks generate typically materialize elsewhere. 

Additional Factors

Other factors to consider when comparing LEO and fiber networks include:

• A fiber network’s ability to provide backhaul for other services, such as cell phone service. 
• The economic benefits when a community gains access to reliable, affordable,  high-quality connectivity (e.g. increased property value and business revenue, better health and education outcomes).
• A LEO network’s reliance on spectrum, a finite and in-demand resource.
• A LEO network’s creation of space debris and environmental damage (an issue recognized by the first Trump administration that will be harder to manage with a depleted Office of Space Commerce).
• The leadership characteristics of the various ISPs. For example: Starlink, the dominant LEO provider, is primarily owned by one man, SpaceX founder, CEO, and major shareholder Elon Musk, who has repeatedly threatened to turn Starlink service “off” to achieve his personal goals. By contrast, the dominant fiber providers are publicly traded and governed by boards of directors. Their decisions, which can still be harmful, are generally constrained by market incentives.

Don’t Bargain Away Our Broadband

To be clear, LEO service is better than no service. But fiber will provide affordable, reliable, high-speed service for decades, whereas LEO service doesn’t even classify as broadband today (which is why LEO areas are eligible for BEAD support). If Secretary Lutnick’s goal is "to get broadband into the hands of low-income people"⁷ in the most cost-efficient way, than, in most cases, LEO is not the best tool. Fiber is. 

Fortunately, if Secretary Lutnick simply makes no changes to BEAD, then he will accomplish his goal and his legacy will be one of bringing affordable, reliable, high-quality fiber connectivity to millions of Americans. However, if he instead changes BEAD to prioritize LEO, his legacy will be one of swapping in a poor substitute at the last minute, leaving millions of Americans stranded in the digital divide for years, if not decades. 

The real bargain is using BEAD to close the digital divide for generations. Let’s not squander this opportunity.

Drew Garner is the Director of Policy Engagement at the Benton Institute for Broadband & Society.

Nick Pappin, Assistant Director at Washington State University Extensions Program for Digital Initiatives, supplied data for the congressional district analysis.

Footnotes

1. Elon Musk’s Starlink also stands to potentially gain tens of billions of dollars from such a change.

2. Note, this analysis is based on the FCC map, which has a slightly different definition of “location” than BEAD maps. BEAD maps measure service to buildings, FCC maps measure service to individual residences. Thus, on the BEAD maps, an apartment building would count as a single location, but on the FCC maps, each apartment unit within building would count as a single location. To the extent apartment buildings are more common in Democratic areas, this analysis will overrepresent Democratic eligibility for BEAD.

3. There are multiple ways to do this, including a price cap per location and a loosening of BEAD’s fiber preference. See Evan Feinman’s exit email for more.

4. Infrastructure refers to direct network construction and maintenance costs; it does not include staffing, R&D, backhaul, satellite ground stations, control software, electronics, trucks, leases, and facilities. Assumptions: fiber infrastructure in low density areas costs $4,050 for construction and $230/yr for maintenance per passing (source). This yields a total cost of around $11,000 over 30 years. Starlink infrastructure currently costs around $800,000 per satellite (source, source) and each satellite can serve roughly 570 homes (source, source, source), the marginal cost per launch is $15,000,000 (source) and each launch can carry roughly 22 satellites (source). If satellites are replaced every five years, the total cost is roughly $16,000 per home over 30 years. Notes: actual costs vary greatly per project; numbers are current and may change over time.

5. Not including Starlink’s variable equipment fees, which BEAD will presumably cover.

6. When comparing plans of similar speeds. Assuming a fiber offering of $50-$60 for a ≥100mbps plan (BroadbandNow, USAToday, author survey of offerings) and $120 for ~25-220mbps plan (Starlink).

7. Note: Congress established the BEAD program to bring high-quality broadband to all un- and underserved households, not just low-income households.

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 2025. 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