On May 16, 2025, the U.S. Court of Appeals for the Fifth Circuit issued a significant ruling in a longstanding dispute between the Texas Commission on Environmental Quality (“TCEQ”) and the U.S. Environmental Protection Agency (“EPA”) over sulfur dioxide emissions compliance.[1] The court reversed its previous position, siding with the TCEQ and vacating the EPA’s rejection of Texas’s State Implementation Plan (“SIP”) for sulfur dioxide under the Clean Air Act.
Crucially, the court’s updated opinion establishes a new standard for classifying areas as “unclassifiable” regarding air quality. This designation means that if the EPA’s data doesn’t “reliably support” a finding of either meeting or failing to meet air quality standards, the area must be labeled “unclassifiable.” Consequently, these areas would avoid the stricter pollution controls imposed on regions failing to meet standards. This revised definition will now be the benchmark across the Fifth Circuit’s jurisdiction: Texas, Mississippi, and Louisiana.
Court Action
This shift followed a petition for a rehearing by industry groups. Instead of a full court rehearing, the panel opted to replace its original opinion after further legal arguments. Judge Southwick, who initially sided with the EPA, authored the new opinion, now joined by Chief Judge Elrod, who had previously dissented.
In her earlier dissent, Judge Elrod expressed concerns that the court had given too much weight to the EPA’s air quality modeling choices, which led to the classification of two Texas counties as not meeting the 2010 sulfur dioxide standard of 75 parts per billion.[2] The Trump administration had initially deemed Rusk and Panola Counties as “unclassifiable.” However, the Biden EPA, using computer modeling provided by the Sierra Club based on emissions from the Martin Lake coal plant, concluded these counties were in violation of the standards. Areas failing to meet standards must develop state implementation plans (“SIPs”) to reduce pollution, a requirement not applicable to “unclassifiable” areas.
Fifth Circuit Reasoning
In the new opinion, Judge Southwick directly addressed the EPA’s reliance on the Sierra Club’s modeling, stating, “We disagree that Sierra Club’s modeling, under these circumstances, provided a sufficient basis for EPA’s nonattainment designation.” He further clarified the court’s interpretation of the Clean Air Act: “[W]e interpret [the Clean Air Act] as requiring EPA to designate an area as ‘unclassifiable’ if the available evidence does not allow for a meaningfully reliable determination of attainment or nonattainment. We also explained that EPA can know an area should be designated ‘unclassifiable’ when there is not much evidence, the competing evidence is too closely balanced, or the evidence is dubious.”
Applying this new test, the court found that “the evidence before EPA implicated all three categories—EPA relied solely on Sierra Club’s modeling that had conceded limitations and that was further called into question by conflicting monitoring data. Given this, EPA should have designated the areas as unclassifiable or rationally explained why an alternative designation was clear and not debatable.” The opinion concluded that “EPA seems to have forced a result on sparse and suspect evidence,” which violates the Administrative Procedure Act (“APA”) and cannot withstand “searching review.”
The Loper Bright Factor
This revised opinion also reflects the impact of the Supreme Court’s 2024 ruling in Loper Bright Enterprises v. Raimondo,[3] which limited the judicial deference given to federal agencies’ interpretations of ambiguous laws, effectively overturning the Chevron doctrine. Industry groups had argued that the Fifth Circuit’s initial deference to the EPA on technical matters was inconsistent with the Loper Bright ruling, which emphasizes courts’ independent interpretation of statutes.
While acknowledging the Loper Bright decision, Judge Southwick clarified that it does not eliminate all deference to agency fact-finding, noting the Supreme Court’s reliance on the APA’s provisions for reviewing factual findings. However, under its own independent review of the Clean Air Act, the Fifth Circuit concluded that the law requires an “unclassifiable” designation when the available data does not reliably support either attainment or nonattainment.
The court deemed two other arguments from the industry groups as no longer relevant. The first argument claimed the EPA failed to treat similar cases consistently by relying on the Sierra Club’s modeling in this instance but rejecting it in others. The court reasoned that since the EPA must reevaluate the Texas plan, this argument is now moot. The second argument concerned the EPA’s assertion that it lacked the discretion to wait for more monitoring data. Given the time that has passed and the requirement for the EPA to reconsider the available data, the court also found this issue to be moot in the current context.
In her concurring opinion, Chief Judge Elrod indicated ongoing concerns about the EPA’s refusal to consider the plant operator’s alternative computer model. However, she stated that under the circumstances, she would not fully disagree with the majority’s analysis on this point, noting that the EPA will be required to apply the court’s new interpretation of “unclassifiable” upon remand.
Key Takeaways
- Reversal of prior holding: The Fifth Circuit had previously upheld the EPA’s disapproval of Texas’s SIP, citing noncompliance with modeling data standards. The new decision reverses that holding, finding that the EPA acted arbitrarily and capriciously in rejecting Texas’s approach.
- State deference restored: The court emphasized the statutory deference owed to states under the Clean Air Act in designing SIPs. Texas’s alternative modeling approach, although nontraditional, was found to be reasonable within the framework of the law.
- Impacts beyond Texas: This decision could ripple beyond Texas, potentially emboldening other states to challenge EPA SIP rejections. It also sets a precedent limiting federal influence in areas traditionally governed by state environmental agencies.
- Industry implications: Power plants and refineries in sulfur dioxide nonattainment zones may now have a clearer regulatory pathway, potentially reducing compliance costs if states are granted wider latitude.
Next Steps
- Regulated entities in Texas may want to actively monitor for TCEQ guidance updates reflecting the court’s decision and any revised SIP submissions.
- Environmental counsel should assess how this decision affects ongoing or pending SIP disputes in other jurisdictions.
- State regulators may reevaluate their strategies for balancing EPA expectations with localized air quality planning.
The EPA may seek either panel rehearing or en banc review, or appeal to the Supreme Court, though such steps are discretionary. Affected stakeholders should prepare for potential regulatory whiplash depending on further judicial developments.
Texas v. U.S. Env’t Prot. Agency, No. 17-60088 (5th Cir. May 16, 2025). ↑
Texas v. U.S. Env’t Prot. Agency, 91 F.4th 280 (5th Cir. 2024) (Elrod, J., dissenting). ↑
603 U.S. 369 (2024). ↑
