Johannes M. Bauer, June 27, 2025
Key points
- The rules established in the BEAD Restructuring Policy Notice for reasonable costs create contradictory incentives that may result in inefficient outcomes.
- To avoid undesirable outcomes, states should, as far as possible within the new rules, adopt a long-term perspective in selecting sub-awardees.
- States can also create incentives, such as appropriate penalties and damages for non-performance, that increase the likelihood of truthful offers during the benefit-for-the-bargain round.
- States should consider complementing these incentives with independent evaluations of the cost modeling assumptions informing the offers.
Background
The BEAD Restructuring Policy Notice of June 6, 2025 (PN) greatly narrows the vision embedded in the original BEAD program.[1] The previous approach recognized that digital equity and complementary policy measures, such as workforce development programs, boost the social and economic benefits of high-speed Internet access.[2] In contrast, the revised PN aims at minimizing BEAD program outlays for broadband access.
Although efficiency and expediency are desirable, the new approach risks jeopardizing even the more modest remaining goals of BEAD. For one, the PN creates tensions between short- and long-term planning that eligible entities may struggle to reconcile. This ambiguity is further aggravated by uncertainty about how the National Telecommunications and Information Administration (NTIA) will conduct final reviews.
To avoid these pitfalls, State Broadband Offices (SBOs) will have to establish appropriate incentives for the benefit-for-the-bargain round, supplemented with robust methods for the evaluation of bids. Most importantly, they must select an appropriate time horizon for determining BEAD outlays, and they must put mechanisms in place that help improve the trustworthiness of bids.
Tensions between short and long-term cost minimization
Because the PN broadens the eligibility of technologies to include satellites and unlicensed wireless, there is now a potential tension between the emphasis on minimizing BEAD program outlays and the endorsement in the PN of the longer-term outlook embedded in the Infrastructure Act. Specifically, the PN defines reliable broadband as
“The term “Priority Broadband Project” means a project that provides broadband service at speeds of no less than 100 megabits per second for downloads and 20 megabits per second for uploads, has a latency less than or equal to 100 milliseconds, and can easily scale speeds over time to meet the evolving connectivity needs of households and businesses and support the deployment of 5G, successor wireless technologies, and other advanced services.” (PN, p. 9)
The preference for fiber under the old BEAD rules resolved this tension because fiber is scalable with relatively modest incremental costs: Fiber projects that minimize the short-term subsidy needs therefore also meet the longer-term scalability goals. The same is not automatically true for fixed wireless access (FWA), unlicensed fixed wireless (ULFW), and low earth orbiting satellites (LEOs). Even though it is desirable to include these technologies, the sub-awardee selection rules must be modified to avoid inefficiencies.
Short-term minimization of subsidy requirements and longer-term scalability are only congruent if the BEAD subsidy is sufficient to generate enough cash flows for sub-awardees so that they can scale their services in response to changing demands. These include variations in the subscriber base, differentiated needs of users demanding higher than the 100/20 Mbps capacity, and, most importantly, upgrading the network to higher thresholds for broadband capacity
Experience suggests that this ideal scenario may not hold for each of the access technologies. FWA, ULFW, and LEOs have much higher incremental costs, at least at the current state of technology. ISPs will likely seek to recover these costs of investing in scalability through higher broadband prices. This may reduce adoption so that some locations may not generate sufficient cash flow, even after the initial BEAD subsidy, to meet scalability goals. Demand-side subsidies could help overcome this insufficiency, but it is unlikely that such programs will be adopted in the near future.
The record of earlier programs such as ACAM and E-ACAM raises another flag of caution. It shows that subsidizing networks with specific capabilities may create a lock-in effect that can only be overcome with additional subsidies when an upgrade to higher capacity is needed.[3]
What are reasonable costs and how should they be evaluated?
Like the old BEAD NOFO the BEAD Restructuring PN envisions a performance period of four years. After that period, all providers must be able to provide standard access to requesting customers within 10 business days. The two-tier scoring rubric primarily focuses on the subsidies needed to build 100/20 Mbps connectivity during that four-year period. Any project that is 15% below the second-best offer will automatically win. Only if offers are within 15% of each other, secondary criteria will become relevant, such as the time to deployment.
Although the Infrastructure Act and the PN embrace scalability as an important goal, this approach undermines it, as short-term cost minimization is not automatically the best long-term solution. For durable infrastructure projects, it is economically rational to minimize costs over a longer planning horizon that considers the future cost of upgrades to the available network capacity. Thus, it is important to establish a relevant planning horizon. Given the longevity of broadband infrastructures, an examination of costs over a 20–30-year period would seem appropriate but may not be possible in the short remaining time.[4]
A pragmatic approach could build on the performance timeline established for satellites. It would solicit information on the likely total cost of a project over a 14-year period. Over this time horizon, access networks will have to support higher capacity to meet increasing demands and support continued innovation in applications and services. Although broadband policy has not established a forward-looking path for capacity needs, a modest goal would be to envision access networks that support universal gigabit connectivity by 2040 and higher speeds if customers demand them.[5]
Reasonable costs for purposes of sub-awards would then include the short-term BEAD outlays to deploy 100/20 Mbps connectivity within four years plus the discounted expected subsidies required to upgrade to gigabit connectivity by 2040. If not, such future subsidies are anticipated, this will create confidence that the short-term minimization emphasized in the scoring rubric is congruent with economic principles. To make a rational selection of sub-awardees, SBOs would ideally also solicit information on the total cost of building such networks, as this will likely influence the level and structure of prices experienced by subscribers.
This approach would help to avoid the problems created by earlier programs such as ACAM. We know that fiber can be scaled to higher capacity with incremental investment in upgraded lasers and electronics. Other technologies also require such upgrade investments. In the case of LEOs, eligible entities would have to reserve higher capacity during later phases of the performance period. Whether such upgrades require additional subsidies will depend on the specific investment scenario and may vary by location.
How can states ascertain the credibility of offers?
SBOs have accumulated considerable expertise in a short time, but they must overcome new problems of incomplete information when assessing the revised and new bids in a short period. Under the old rules, SBOs and other stakeholders could develop trusted relationships that increased confidence in the offers. Evaluating additional bidders during a pre-qualification round will mitigate the problem, but it may not prevent strategic offers. States have two options to incentivize players to submit truthful bids: sufficiently powerful penalties for non-performance and independent verification of bids.
The envisioned benefit-for-the-bargain round invites additional players and bids. States must at least conduct one more bidding round. This design has both desirable and undesirable consequences. Similar to a reverse first price sealed-bid auction, the proposed approach promises a streamlined and transparent process. Increased competition from new suppliers is generally desirable and should result in lower subsidy needs. However, a one-round bidding process in which the lowest offer is selected does not necessarily introduce efficient incentives for the bidders to make truthful offers. It may result in lower quality offers and contribute to subsequently higher monitoring and compliance costs. Furthermore, a one-round bidding process may not result in offers for all locations so that states may have to conduct multiple rounds, in a very short time window.
Assessing the credibility and accuracy of offers is particularly important under the two-tier scoring rubric of the revised rules. The rubric prioritizes BEAD program expenses and allows limited other criteria only in cases where bids are within 15% of the lowest bid. This implies that in locations where the difference exceeds 15%, the lowest bid must be chosen. Due diligence requires that SBOs carefully assess these bids regardless. Access technologies have different vulnerabilities to foliage, line of sight requirements, and weather conditions. Moreover, the costs for existing LEO services are already expended (sunk in economic terms), allowing a broader range of strategically low bids that might crowd out better offers.
One way to overcome this challenge is to embed incentives into the bidding process that reward truthful bids. Like the original NOFO, the PN establishes clawback provisions that allow eligible entities to recoup disbursed funds from sub-awardees and, under certain conditions, the federal government to clawback funds from the eligible entities.[6] These provisions work in the right direction, but they provide relatively weak incentives for sub-awardees to be truthful in their bids, as the implied penalty will often be limited. Introducing higher penalties would strengthen the incentive for bidders to make accurate and reliable offers.
This approach principally aligns with the market-based model envisioned in the PN and the stipulated benefit-for-the-bargain model. In this model, eligible entities can claim damages for non-performance. From an economic perspective, default-related damages comprise the disbursed subsidies and the monetary value of disadvantages imposed on prospective broadband users of not being connected. These costs of not being connected may range between $1,800-$2,400 per year for every household that is not connected.[7] Adding (all or part of) this social loss to the already expended subsidy would yield a more effective penalty.
A second option to verify bids is to seek independent reviews of the planning costs used for the bids. The PN relies on self-certification, but this does not provide strong safeguards to assure reliable bids. Access technologies differ widely in their cost structure. For example, fiber has high initial investment costs; low operations and maintenance (O&M) costs, and its capacity can be scaled at modest incremental costs. Fixed wireless access can be rolled out at lower initial investment costs, it has higher O&M costs, and the incremental costs of capacity upgrades are not fully known. The costs of expanding the capacity of satellite networks to meet broadband demand are even more difficult to anticipate.
The Infrastructure Act and the PN establish scalability as an important goal. Therefore, as was explained above, the rational approach is to adopt a forward-looking approach that considers costs over an extended time horizon. This is also an economically rational approach. Variations in the investment and O&M costs of alternative broadband access technologies imply that the least-cost solution will not only depend on local conditions but also on the time-horizon over which costs are minimized and the choice of an appropriate discount rate to determine present values for future expenses.
Summary and recommendations
- The rules established in the BEAD Restructuring Policy Notice for reasonable costs create contradictory incentives that may result in inefficient outcomes.
- To avoid such outcomes, states should, as far as possible within the new rules, adopt a long-term perspective in selecting sub-awardees.
- States can also create incentives, such as appropriate damages for non-performance, that increase the likelihood of truthful offers during the benefit-for-the-bargain round.
- States should consider complementing these incentives with independent evaluations of the cost modeling assumptions informing the offers.
[1] See National Telecommunications and Information Administration (NTIA), Broadband Equity, Access, and Deployment (BEAD) Program: BEAD Restructuring Policy Notice, p. 12. retrieved on June 16, 2025, from https://www.ntia.gov/sites/default/files/2025-06/bead-restructuring-policy-notice.pdf.
[2] See J.M. Bauer, An Anticipatory Assessment of Proposals to Reform the Broadband Equity Access and Deployment Program (BEAD) (June 05, 2025). Available at SSRN: https://ssrn.com/abstract=5296353.
[3] See FCC Announces E-ACAM Support to Expand Broadband to Rural Communities, retrieved on June 24, 2025, from https://www.fcc.gov/document/fcc-announces-e-acam-support-expand-broadband-rural-communities. See also J.M. Bauer, An
[4] See the discussion in A. Afflerbach, Fixed Wireless Technologies and Their Suitability for Broadband Delivery, Benton Institute for Broadband and Society, June 2022. Retrieved on June 25, 2025, from https://www.benton.org/sites/default/files/FixedWireless.pdf. An appropriate long-term cost comparison will have to discount https://www.benton.org/sites/default/files/FixedWireless.pdf. Because the cost distribution over time matters, an appropriate long-term cost comparison will have to discount future costs to determine a present value.
[5] Much of the United States already is already served by networks that can support gigabit capacity. According to data by Ookla, the median U.S. download speed in April 2025 was 291 Mbps (see Ookla, Speedtest Global Index, retrieved June 15, 2025, from https://www.speedtest.net/global-index). Although the exact timeline to higher capacity is contingent on the needs of applications, user preferences, and other factors, it seems plausible that universal gigabit connectivity should be reached no later than 2040.
[6] See NTIA, Notice of Funding Opportunity, Broadband Equity, Access, and Deployment Program, NTIA-BEAD-2022, p. 51, retrieved June 16, 2025, from https://broadbandusa.ntia.doc.gov/sites/default/files/2022-05/BEAD%20NOFO.pdf.
[7] These are estimates of a lower bound of the effects. The Michigan Economic Development Corporation found annual economic benefits of $1,850 per household; George W. Zuo, Wired and Hired, American Economic Journal, 2021, found income effects of $2,000 per household and year; John Horrigan, in various studies, estimated annual effects per household of $2,000-$3,150. I consider the range of $1,800-$2,400 as a safe lower bound. States and communities are encouraged to consider additional possible effects, such as benefits of improved access to healthcare, education, and government services.
Updated on July 18, 2025. The views expressed do not necessarily reflect the position of the Quello Center or of Michigan State University.