22 F.4th 1018
D.C. Cir.2022Background
- Statutory framework: Under the Energy Policy and Conservation Act (42 U.S.C. § 6313), DOE may adopt more-stringent efficiency standards than ASHRAE Standard 90.1 only if, by "clear and convincing evidence," the new standard is technologically feasible, economically justified, and would produce significant additional energy conservation; a 2007 "lookback" requires DOE to evaluate standards if ASHRAE has not amended 90.1 for six years.
- Rule at issue: In January 2020 DOE issued a Final Rule tightening efficiency standards for eight of twelve categories of commercial packaged boilers (large boilers used in commercial and multifamily heating).
- DOE’s analysis: DOE compared a no-new-standards case to a new-standards case using life-cycle-cost (LCC) modeling that required assumptions about shipments (using AHRI model listings as a proxy), random assignment of efficiencies to buildings, burner operating hours (using CBECS/RECS and a 30 Btu/h per sq. ft. heat‑load rule), and future fuel prices (based on EIA historical and forecast data, with adjustments).
- Challenges raised: Petitioners (APGA, AHRI, Spire) argued DOE: (1) did not properly apply the statutory clear-and-convincing evidentiary standard; and (2) relied on deficient data/assumptions (random assignment ignoring purchaser optimization, inappropriate fuel-price assumptions, dubious burner-hour estimates, and use of model-listings as shipment proxies).
- Procedural posture and remedy: The D.C. Circuit held DOE had articulated the clear-and-convincing standard as an alternative ground but that DOE’s substantive justifications for the amended standards were not reasonably supported in the record. Court remanded the Final Rule to DOE for corrective action within 90 days and ordered automatic vacatur if DOE fails to justify the rule within that period (unless DOE shows need for more time).
Issues
| Issue | Plaintiff's Argument | Defendant's Argument | Held |
|---|---|---|---|
| Whether DOE applied the statutory "clear and convincing" evidentiary standard | DOE failed to apply the heightened standard and even disclaimed it in the preamble | DOE asserted it satisfied the standard and alternatively treated it as an applicable ground | Court rejected plaintiffs' claim that DOE never applied the standard — DOE had stated it as an alternative ground — but proceeded to assess whether DOE reasonably met it and found the record insufficient |
| Random assignment of boiler efficiencies to sample buildings | Random assignment ignores rational purchaser behavior and overstates gains attributable to regulation | DOE said a consumer-choice model lacked sufficient data and random assignment was a practical approach with available data | Court found DOE’s cursory response inadequate under arbitrary-and-capricious review and especially deficient under the clear-and-convincing standard; remand required |
| Energy price assumptions (average vs. marginal prices; large purchasers’ discounts) | DOE used average forecast prices that overstate marginal costs actually paid by large commercial/fleet purchasers, inflating projected fuel‑cost savings | DOE maintained it used best-available data and adjusted averages with seasonal/marginal factors | Court held DOE’s conclusory response failed to show it reasonably addressed comment concerns; record does not support that clear-and-convincing threshold was met |
| Burner operating hours estimates | DOE’s estimated burner hours (based on heat-load assumptions and CBECS/RECS) produced anomalous distributions suggesting flawed assumptions or data | DOE asserted high confidence in its building-load estimation and noted lack of comprehensive burner-hour data | Court concluded DOE ignored significant criticisms and failed to explain or substantiate its estimates; remand required for reasoned response |
| Use of AHRI model‑listings as a proxy for shipment data | Model-listings do not reliably reflect actual sales distribution; use may skew LCC results | DOE argued model listings approximate shipments and validated the proxy for two categories where AHRI supplied shipment data | Court held DOE’s proxy reliance was reasonable here because DOE empirically validated it for categories with shipment data and petitioners gave no quantitative rebuttal; this issue alone did not defeat the rule |
Key Cases Cited
- Colorado v. New Mexico, 467 U.S. 310 (definition of the clear-and-convincing evidentiary standard)
- Motor Vehicle Mfrs. Ass'n v. State Farm, 463 U.S. 29 (agency action arbitrary and capricious when it fails to consider important aspects of problem)
- Sea Island Broad. Corp. of S.C. v. F.C.C., 627 F.2d 240 (judicial review asks whether it was reasonable for agency to find heightened standard met)
- Van Hollen, Jr. v. Fed. Election Comm’n, 811 F.3d 486 (agency must engage arguments and show reasoned decisionmaking)
- Allina Health Servs. v. Sebelius, 746 F.3d 1102 (vacatur is the normal remedy for unlawful agency action)
- Allied-Signal, Inc. v. Nuclear Regulatory Comm'n, 988 F.2d 146 (remand vs. vacatur balancing test)
- In re Core Commc’ns, Inc., 531 F.3d 849 (encouraging time-limited remand alternatives to open-ended remand)
- A.L. Pharma, Inc. v. Shalala, 62 F.3d 1484 (automatic vacatur if agency fails to justify rule within court-ordered timeframe)