21 F.4th 841
D.C. Cir.2021Background
- In April 2020 the FCC issued an Order opening the 6 GHz band (5.925–7.125 GHz) to unlicensed devices (Wi‑Fi routers and client devices) while imposing safeguards to prevent harmful interference with incumbent licensed users (fixed microwave links, mobile broadcasters, utilities, public safety).
- The Order distinguishes standard‑power access points (must use an Automated Frequency Coordination (AFC) system and are barred from mobile subbands) from low‑power indoor access points (limited to 5 dBm/MHz, must use contention‑based protocols, be indoor only, have integrated antennas, no batteries, and no weatherproofing); client devices face stricter limits.
- The FCC relied heavily on technical analyses, notably a CableLabs Monte Carlo simulation showing negligible interference in scenarios modeled, and it reran and adjusted an AT&T‑submitted worst‑case study to conclude interference risk is insignificant.
- Petitioners (AT&T Services, Lumen, electric utilities, APCO, and the National Association of Broadcasters) challenged the Order as arbitrary and capricious under the APA and inconsistent with the Communications Act, arguing the FCC understated interference risks and relied on flawed studies and inadequate protections/enforcement.
- The D.C. Circuit largely upheld the FCC’s technical judgments and deference to agency expertise, denying most petitions, but remanded for the FCC to respond to the National Association of Broadcasters’ specific request that the agency reserve a slice of the band for licensed mobile operations and to address concerns about precedent from the 2.4 GHz band.
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Whether the FCC unlawfully underestimated the risk of harmful interference | Petitioners: FCC intended to eliminate all interference; any non‑zero risk makes Order arbitrary and capricious | FCC: Order aims to make risk "insignificant," not zero, and includes enforcement mechanisms | Court: Rejects petitioners; deference to FCC's predictive technical judgments; FCC did not promise zero risk |
| Reliance on technical studies (CableLabs Monte Carlo; AT&T worst‑case study) | Petitioners: CableLabs is a "black box," used unrealistic assumptions (building loss, activity factor, sample size); AT&T was improperly adjusted | FCC: disclosed methodology sufficient, Monte Carlo appropriate, adjustments to AT&T study justified to produce probabilistic assessment | Court: Upholds FCC’s reliance and explanations; affords deference on technical modeling absent highly persuasive contrary evidence |
| Adequacy of low‑power rules (5 dBm/MHz, contention‑based protocol, indoor restrictions) | Petitioners: Power limit arbitrary; contention‑based protocol won’t protect against focused microwave beams; indoor safeguards cannot prevent outdoor use; enforcement inadequate | FCC: limit chosen via engineering judgment and testing; CBP reduces continuous transmissions; design limits (no battery/weatherproofing/integrated antenna) make outdoor use impractical; Enforcement Bureau can investigate and remedy interference | Court: Sustains FCC; measures reasonable and supported by agency expertise and record |
| Whether FCC failed to address mobile licensee concerns / reserve spectrum for mobile operations (NAB) | NAB: FCC did not respond to concerns about mobile protection and precedent from 2.4 GHz; requested reservation for mobile operations | FCC: Relied on studies and experience but did not adequately address NAB’s 2.4 GHz concerns in the record | Held: Grant petition in part; remand to FCC to respond to NAB’s concerns about mobile protection and explain why no reservation is needed; denial of vacatur pending explanation |
Key Cases Cited
- Motor Vehicle Mfrs. Ass’n v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29 (U.S. 1983) (standard for arbitrary and capricious review)
- Mobile Relay Assocs. v. FCC, 457 F.3d 1 (D.C. Cir. 2006) (great deference to FCC spectrum‑management judgments)
- EarthLink, Inc. v. FCC, 462 F.3d 1 (D.C. Cir. 2006) (deference to agency predictive judgments)
- Mozilla Corp. v. FCC, 940 F.3d 1 (D.C. Cir. 2019) (agency must consider public‑safety implications where appropriate)
- American Radio Relay League, Inc. v. FCC, 524 F.3d 227 (D.C. Cir. 2008) (limitations on agency reliance when record shows data cherry‑picked)
- American Trucking Ass’ns, Inc. v. EPA, 283 F.3d 355 (D.C. Cir. 2002) (agencies need not publicize raw underlying data for every technical study)
- Marsh v. Ore. Nat. Res. Council, 490 U.S. 360 (U.S. 1989) (deference to agency expertise on technical matters)
- Allied‑Signal, Inc. v. Nuclear Reg. Comm’n, 988 F.2d 146 (D.C. Cir. 1993) (vacatur/remand balancing test)