ICLE Comments to the FCC RE: Amazon LEO Gen 1 Milestone Extension/Waiver Request
I. Introduction and Summary
Amazon LEO’s extension request raises important questions about how the Federal Communications Commission (FCC) should apply build-out rules in high-capital-expenditure, launch-constrained industries. These comments evaluate the non-geostationary orbit (NGSO) milestone regime from a law & economics perspective, focusing on the real-options structure of spectrum licensing and satellite-deployment investment.
A satellite license functions economically as a real option[1]—the right, but not the obligation, to deploy and bring into service a constellation at a future date. Milestone rules force exercise of that option on a regulatory timeline. In doing so, they reduce the option value that can make speculative license holding attractive.
The option value of a license differs across firms. For a speculator, option value is high because sunk costs are low and flexibility remains high. For a firm that has already made large, irreversible, deployment-specific investments, option value is much lower. As demonstrated sunk investment grows, the firm has already exercised much of the option embedded in the license. The screening rationale for forced exercise correspondingly weakens.
This framework clarifies the economic logic of milestone policy. Milestones can serve as efficient screening tools when they compel exercise of options held by speculative licensees. Rigid forfeiture rules become dynamically inefficient when applied to firms that have already committed substantial irreversible capital, because those investments demonstrate that the option has largely been exercised.
Milestone rules therefore play an important economic role in deterring warehousing. In capital-intensive, launch-constrained low-Earth orbit (LEO) satellite markets, however, rigid interim milestones may not optimally balance anti-warehousing deterrence, dynamic competition, and efficient capital deployment.[2]
II. The Screening Function of Spectrum Milestones
Spectrum milestones aim to separate genuine operators from speculators.[3] The economic logic parallels classic screening and signaling models. Milestones impose costs that fall differently across firms. For firms that intend to deploy, compliance costs are small relative to their planned investment. For speculators, milestone deadlines—combined with ex ante bond requirements—make warehousing costly. The bond, triggered when a licensee misses a milestone, creates the primary financial penalty that deters speculative holding.[4]
This framework draws on foundational insights from information economics. Spectrum licensing creates information asymmetries between regulators and licensees about deployment intentions. Milestones function as a separating mechanism. Like educational credentials or insurance deductibles, they impose costs that induce self-selection between firms that plan to deploy and firms that do not.[5]
The screening rationale assumes that compliance costs remain systematically lower for genuine deployers than for speculators. That assumption holds when milestones require physical deployment or verified capital expenditures.[6] Under some conditions, rigid forfeiture provisions can invert this relationship. External shocks—such as supply-chain disruptions, launch-vehicle shortages, or force majeure events—can raise compliance costs for all licensees regardless of intent. When that occurs, the screening mechanism can lose its separating function.
The Commission designed its milestone framework primarily for terrestrial wireless systems and early geostationary satellite markets. In those contexts, deployment timelines and capital requirements were more predictable. NGSO broadband constellations present different calibration challenges.
Satellite constellation deployment requires massive capital commitments with multi-year lead times and tightly coupled production steps.[7] Unlike terrestrial build-out—where delays in one market do not prevent deployment elsewhere—satellite manufacturing and launch operate as a single integrated production function. A delay in one launch campaign can cascade through the entire deployment schedule in ways that calendar-based milestones cannot easily accommodate.
Economic theory also recognizes that credible commitments can reveal information that reduces the need for regulatory screening.[8] When a licensee makes substantial transaction-specific investments—such as manufacturing satellites, securing launch contracts, and building ground infrastructure—those expenditures function as self-enforcing bonds that align incentives with deployment. Satellite investments are highly asset-specific. Satellites designed for particular orbital configurations have limited alternative uses, face no meaningful secondary market, and degrade on fixed operational timelines. These features make constellation investments strongly sunk in the economic sense, strengthening the self-enforcing commitment to deployment.[9]
This insight draws on Oliver Williamson’s framework of credible commitments and asset specificity.[10] Spectrum-specific sunk costs create path dependence that discourages abandonment.[11] That behavior is precisely what milestone rules aim to prevent. Once a firm commits billions of dollars to satellite manufacturing and launch, the marginal cost of continuing deployment often falls below the sunk cost already incurred. Economic incentives therefore favor completing the system even without regulatory compulsion.
These dynamics create a potential screening paradox. Firms most likely to satisfy milestone requirements are often those whose investment behavior already demonstrates commitment. At the same time, the anti-warehousing rationale weakens precisely where compliance becomes most likely.
Milestones nevertheless serve an additional coordination function. Operators in later processing rounds rely on predictable timelines from earlier-round licensees to design their systems and negotiate interference protections. Milestones calibrated to encourage expeditious deployment can therefore produce ecosystem-wide benefits beyond screening out speculators.
The key policy question is not whether milestones are necessary for committed deployers. The relevant question is whether their timing and enforcement appropriately balance anti-warehousing deterrence against the risk of penalizing firms whose deployment progress is genuine but constrained by factors beyond their control.
III. Real Options and Satellite Deployment Investment
Spectrum licenses embed real-options characteristics for which milestone policy must account.[12] Deploying a satellite constellation requires large-scale, irreversible investment under uncertainty—precisely the conditions in which option value becomes most important.[13]
Milestone rules effectively force licensees to exercise that option on a regulatory timeline.[14] For speculative holders, forced exercise can improve efficiency by preventing strategic delay.[15] For firms already deep into irreversible deployment, rigid milestone timing can force suboptimal capital allocation without providing meaningful screening benefits.
The real-options framework therefore suggests that milestone policy should consider how much option value a licensee has already extinguished through irreversible investment. A firm that has committed billions of dollars to satellite manufacturing, launch contracts, and ground infrastructure has already exercised much of its deployment option. Remaining uncertainty concerns execution timing, rather than deployment intent.
Milestone enforcement also carries broader competitive implications.[16] In nascent NGSO markets, aggressive enforcement that threatens credible deployers with forfeiture can reduce the number of viable competitors.
The emerging LEO broadband market exhibits characteristics of dynamic competition driven by rapid technological innovation. At the same time, it features network effects and substantial barriers to entry.[17] Regulatory policies that inadvertently eliminate credible competitors through rigid milestone enforcement may impose greater consumer harm than the speculative warehousing such rules seek to prevent.
The regulatory-design literature emphasizes the importance of adaptive mechanisms that incorporate new information.[18] Traditional command-and-control milestone regimes assume stable deployment environments. The NGSO sector instead operates in a context of rapid technological change, evolving launch economics, and supply-chain constraints that regulators did not fully anticipate when adopting the current framework.
Progress-contingent approaches can better align with the real-options structure of LEO investment.[19] Rather than focusing solely on point-in-time milestone compliance, regulators can evaluate a licensee’s deployment trajectory. This approach preserves anti-warehousing deterrence while reducing the risk of forfeiting licenses held by firms that demonstrate genuine deployment progress.
IV. Recommendations for NGSO Milestone Policy
The preceding analysis yields four policy recommendations.
- Preserve milestone screening as an anti-warehousing tool.
The institutional and economic case for deployment obligations remains strong. Without enforceable commitments, applicants may seek authorizations for speculative purposes—such as blocking rivals, increasing bargaining leverage, or warehousing spectrum rights. Those outcomes would leave orbital and spectrum resources idle and delay service to consumers.
- Reevaluate milestone calibration.
The Commission should assess whether interim milestones remain calibrated to current industry conditions, including satellite-manufacturing cycles, launch-vehicle availability, and upstream capacity constraints. As the economic literature emphasizes, well-designed mechanisms separate high-value from low-value entrants while minimizing wasteful expenditures that do not increase output or innovation.
A progress-contingent bonding regime could better align the screening mechanism with the underlying information problem. Under such a framework, operators would post financial bonds that are released as they achieve verified deployment milestones. The Commission could also consider a graduated milestone structure with more frequent checkpoints and proportional, rather than binary, consequences.
Under this approach, missing a single milestone would not automatically trigger license suspension or revocation. Instead, partial bond forfeiture tied to each checkpoint would maintain continuous incentive pressure while preserving the operator’s ability to demonstrate compliance at later intervals. This structure would address the bluntness of the current two-milestone regime—50% and 90% deployment thresholds—by providing more granular visibility into deployment progress. Any recalibration should also align with the proposed extension of NGSO license terms to 20 years under the Space Modernization for the 21st Century NPRM.
- Better incorporate evidence of sunk investment.
The Commission’s waiver process allows operators to present sunk-investment evidence when seeking milestone extensions. The absence of clear criteria, however, has produced inconsistent outcomes. In some cases, operators cite aggregate expenditure figures without verifiable benchmarks demonstrating meaningful deployment progress.
The Commission should formalize objective, verifiable metrics—such as completed satellite manufacturing, binding launch contracts, or finished ground infrastructure—as structured criteria for evaluating whether a licensee has effectively exercised its deployment option. These metrics would shift milestones from automatic forfeiture triggers toward rebuttable screening tools.
When operators can credibly demonstrate substantial sunk investment, the marginal benefits of rigid calendar-based deadlines fall, while the marginal costs—including false-positive forfeiture risk and entry barriers—rise. NASA’s commercial-resupply program offers a useful institutional model: staged readiness assessments and technical reviews, rather than binary date-certain forfeiture.
- Account for dynamic and regulatory competition.
Milestone enforcement sends signals that extend beyond individual licensees. In LEO broadband markets with high fixed costs and strong first-mover dynamics, revoking a near-deployment authorization can reshape market structure. Such actions may weaken potential competition, accelerate market tipping, and redirect investment toward foreign regulatory regimes.[20]
Milestone rules serve an important economic function in allocating scarce orbital and spectrum resources. Their efficiency, however, depends on calibration to industry structure and input constraints. The Commission’s task is not to decide whether milestones should exist. The relevant question is whether enforcement should distinguish between speculative reservation of rights (where option value remains high and screening is most valuable) and demonstrated, irreversible commitment to deployment (where the option has largely been exercised and the screening rationale weakens).
The Commission should evaluate the pending request through a forward-looking, economically informed framework that considers screening accuracy, dynamic investment incentives, global regulatory competition, and the long-run structure of LEO markets. The FCC’s Space Modernization for the 21st Century NPRM recognizes that legacy rules may not reflect contemporary LEO deployment economics.[21] The same recognition should guide the Commission’s approach here.
[1] See Avinash K. Dixit & Robert S. Pindyck, Investment Under Uncertainty (Princeton Univ. Press 1994); Avinash K. Dixit, Entry and Exit Decisions Under Uncertainty, 97 J. Pol. Econ. 620 (1989).
[2] See Peter Cramton, Innovation and Market Design, 9 Innovation Pol’y & Econ. 113 (2009) (analyzing market-design tools that select high-value entrants while minimizing waste).
[3] See George A. Akerlof, The Market for “Lemons”: Quality Uncertainty and the Market Mechanism, 84 Q.J. Econ. 488 (1970); Paul Milgrom & John Roberts, Price and Advertising Signals of Product Quality, 94 J. Pol. Econ. 796 (1986).
[4] See Akerlof, supra note 3; see also Peter Q. Blair & Bobby W. Chung, A Model of Occupational Licensing and Statistical Discrimination, 111 AEA Papers & Proc. 91 (2021).
[5] See Joseph D. Piotroski & Suraj Srinivasan, Regulation and Bonding: The Sarbanes-Oxley Act and the Flow of International Listings, 46 J. Acct. Res. 383 (2008) (finding that high compliance costs screen out smaller firms and drive jurisdictional sorting).
[6] See Blair & Chung, supra note 4; Xing Xia, Barrier to Entry or Signal of Quality? The Effects of Occupational Licensing on Minority Dental Assistants, 71 Labour Econ. (2021).
[7] See Kristian Stout, Satellite-Spectrum Policy Changes Are Needed, Int’l Ctr. for L. & Econ. (2025), https://laweconcenter.org/resources/satellite-spectrum-policy-changes-are-needed; Michael Calabrese, Jessica Dine & Kristian Stout, Ex Parte Letter of ICLE and OTI Re: SB Docket Nos. 25-180, 25-157 & 25-306, Int’l Ctr. for L. & Econ. & New Am. Found. Open Tech. Inst. (2026), https://laweconcenter.org/resources/ex-parte-letter-of-icle-and-oti-re-sb-docket-nos-25-180-25-157-25-306.
[8] See Edward B. Rock, Securities Regulation as Lobster Trap: A Credible Commitment Theory of Mandatory Disclosure, 23 Cardozo L. Rev. 675 (2002).
[9] See Oliver E. Williamson, Credible Commitments: Using Hostages to Support Exchange, 73 Am. Econ. Rev. 519 (1983); Richard Baldwin, Sunk-Cost Hysteresis, Nat’l Bureau of Econ. Rsch., Working Paper No. 2911 (1989).
[10] See Williamson, supra note 9; see also Baldwin, supra note 9.
[11] See Baldwin, supra note 9 (showing that sunk costs create path dependence that discourages exit after deployment).
[12] See Dixit & Pindyck, supra note 1; Lenos Trigeorgis, Real Options and Investment Under Uncertainty: What Do We Know?, Nat’l Bureau of Econ. Rsch., Working Paper No. 22 (2002).
[13] See Dixit, supra note 1; Scott R. Baker, Nicholas Bloom & Steven J. Davis, Measuring Economic Policy Uncertainty, 131 Q.J. Econ. 1593 (2016).
[14] See Nancy L. Stokey, Wait-and-See: Investment Options Under Policy Uncertainty, 21 Rev. Econ. Dynamics 246 (2018).
[15] See Cramton, supra note 2; Stokey, supra note 14.
[16] See Richard J. Gilbert, The Role of Potential Competition in Industrial Organization, 3 J. Econ. Persp. 107 (1989); David S. Evans & Richard Schmalensee, Some Economic Aspects of Antitrust Analysis in Dynamically Competitive Industries (MIT Press 2002).
[17] See Michael Calabrese et al., Low-Earth Orbit Satellites: Policies to Promote Spectrum Sharing, Foster Competition, and Close Digital Divides—A Report of the LEO Policy Working Group, Int’l Ctr. for L. & Econ. & New Am. Found. Open Tech. Inst. (2025), https://laweconcenter.org/resources/low-earth-orbit-satellites-policies-to-promote-spectrum-sharing-foster-competition-and-close-digital-divides-a-report-of-the-leo-policy-working-group.
[18] See Warren E. Walker, Vincent A.W.J. Marchau & Darren Swanson, Addressing Deep Uncertainty Using Adaptive Policies, 77 Tech. Forecasting & Soc. Change 917 (2010).
[19] See generally Ian Ayres & John Braithwaite, Responsive Regulation: Transcending the Deregulation Debate (Oxford Univ. Press 1992); see also Diane S. Koons et al., Risk Mitigation Approach to Commercial Resupply to the International Space Station, Nat’l Aeronautics & Space Admin. (2010), https://ntrs.nasa.gov/api/citations/20100014822/downloads/20100014822.pdf (describing NASA’s staged readiness reviews and milestone-based risk management).
[20] See generally Victor Fleischer, Regulatory Arbitrage, Univ. of Colo. L. Sch. Legal Stud. Res. Paper No. 10-11 (2010), https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1567212; Anthony Van Fossen, Globalization, Stateless Capitalism, and the International Political Economy of Tonga’s Satellite Venture, 22 Pac. Stud. 1 (1999); Steven Vogel, International Games With National Rules: How Regulation Shapes Competition in “Global” Markets, 16 J. Pub. Pol’y 167 (1996).
[21] See Space Modernization for the 21st Century, 90 Fed. Reg. 56,338 (2025) (FCC).