The Fifth Circuit recently vacated and remanded EPA's final rule updating its effluent limitation guidelines (ELGs) for steam-electric power plants. I'm dedicating an entire post to this case because the issues are important and Judge Duncan's 53-page opinion (joined by Judges Haynes and Ho) addressing them is fascinating and full of nerdy adlaw puzzles. Buckle up.
Judge Duncan is a talented writer. Consider, for example, his opening description of the questions presented in this case. If you're familiar with this area of law, you know what's coming: a deep dive into the world of flue gas desulfurization, leachate, bottom ash transport water, gassification wastewater, and electrostatic precipitators. But check out how Judge Duncan introduces the issue:
Steam-electric power plants generate most of the electricity used in our nation and, sadly, an unhealthy share of the pollution discharged into our nation's waters. To control this pollution, the Clean Water Act, 33 U.S.C. § 1251 et seq., empowers the Environmental Protection Agency to promulgate and enforce rules known as “effluent limitation guidelines” or “ELGs.” Id. §§ 1311, 1314, 1362(11). For quite some time, ELGs for steam-electric power plants have been, in EPA's words, “out of date.” 80 Fed. Reg. 67,838. That is a charitable understatement. The last time these guidelines were updated was during the second year of President Reagan's first term, the same year that saw the release of the first CD player, the Sony Watchman pocket television, and the Commodore 64 home computer. In other words, 1982. See id. (noting ELGs were “promulgated and revised in 1974, 1977, and 1982”). The guidelines from that bygone era were based on “surface impoundments,” which are essentially pits where wastewater sits, solids (sometimes) settle out, and toxins leach into groundwater. Id. at 67,840, 67,851. Impoundments, EPA tells us, have been “largely ineffective at controlling discharges of toxic pollutants and nutrients.” Id. at 67,840. Consequently, in 2005 the agency began a multi-year study to bring the steam-electric ELGs into the 21st century. Id. at 67,841.
In November 2015, EPA unveiled the final rule: the “Effluent Limitations Guidelines and Standards for the Steam Electric Power Generating Point Source Category,” 80 Fed. Reg. 67,838 (Nov. 3, 2015). The rule updates guidelines for six of the wastestreams that issue from plants and foul our waters. Importantly, the Clean Water Act requires setting new ELGs based on the “Best Available Technology Economically Available” or “BAT.” 33 U.S.C. § 1314(b)(2)(B). BAT is the gold standard for controlling water pollution from existing sources. By requiring BAT, the Act forces implementation of increasingly stringent pollution control methods. See NRDC v. EPA, 822 F.2d 104, 123 (D.C. Cir. 1987)(describing the Act as “technology-forcing”).
We consider a challenge to the final rule brought by various environmental petitioners. They target two discrete parts of the rule: the new ELGs for “legacy wastewater” (wastewater from five of the six streams generated before a specific date) and for “combustion residual leachate” (liquid that percolates through landfills and impoundments). These two categories account for massive amounts of water pollution. For instance, leachate alone would qualify as the 18th-largest source of water pollution in the nation, producing more toxic-weighted pound equivalents than the entire coal mining industry. The environmental petitioners' basic complaint is that EPA set an unlawful BAT for these two categories. Whereas the BAT for the other streams adopts modern technologies, they claim the agency arbitrarily set BAT for legacy wastewater and leachate using the same archaic technology in place since 1982—namely, impoundments. It was as if Apple unveiled the new iMac, and it was a Commodore 64.
The environmental petitioners challenge those portions of the rule under the Administrative Procedure Act and the well-worn Chevron test governing review of agency action. See Chevron, USA, Inc. v. Nat. Res. Def. Council, Inc., 467 U.S. 837, 104 S.Ct. 2778, 81 L.Ed.2d 694 (1984). For the reasons discussed below, we agree that the portions of the rule regulating legacy wastewater and combustion residual leachate are unlawful. Accordingly, we VACATE those portions of the rule and REMAND to the agency for reconsideration.
I defy you to find another discussion of "toxic-weighted pound equivalents" half as clear and compelling.
Now for a quick summary of the Court's reasons for vacating EPA's final rule. The Court concludes that EPA acted arbitrarily and capriciously by setting a BAT limit for legacy wastewater equal to the BPT standard for surface impoundments for five reasons:
- EPA admits in the Final Rule that impoundments are ineffective at removing toxins from wastewater--a concession that, according to the Court, is "in critical tension" with EPA's choosing them as a BAT given that BAT is supposed to be the Clean Water Act's most stringent standard;
- EPA refused to set impoundments as the BATs for five of the six wastewater streams addressed in the Final Rule because doing so wouldn't result in further progress toward the Act's pollution-eliminating goals only to turn around and choose impoundments as BAT for legacy wastewater, which is nothing but those same streams generated before the compliance date. Such paradoxical behavior "signals arbitrary and capricious agency action";
- By choosing a BAT that freezes impoundments in place rather than one that would promote reasonable further progress, EPA blurs the line between the statutorily distinct concepts of BAT (the more stringent standard) and BPT (the less stringent standard);
- The evidence before the agency "strongly indicates that other available technologies are far better than impoundments at removing pollutants from . . . legacy wastewater," a reality that is "difficult, if not impossible, to square with EPA's decision nonetheless to set 1980s-era impoundments as the BAT . . . .";
- Congress intended BAT limits to be based on the performance of the single best-performing plant in an industrial field, but the Final Rule contains no hint that EPA's decision to peg BAT to impoundments was based on anything like the performance of the single best-performing plant in the field.
For remarkably similar reasons, the Court agreed with Petitioners that EPA's decision to set impoundments as BAT for leachate fails Chevron step one and, in the alternative, Chevron step two. It fails at step one, the Court explains, because it conflates BAT and BPT (as described above). It fails at step two for other reasons already discussed (setting impoundments as BAT is internally inconsistent with agency's findings that impoundments are ineffective and that there are other more effective options that are available and achievable etc.).
That's the 10,000 foot view. Now let's dig into some (though not nearly all) of the interesting details:
- The Court's description of how the case ended up in the CA5 in the first place is interesting for a couple of reasons. First, it demonstrates that some randomness is baked into the process that determines which adlaw cases wind up in which federal courts. As Judge Duncan explains:
Four separate lawsuits challenging the final rule were originally brought in the Second, Fifth, Eighth, and Ninth Circuits . . . . The four cases were consolidated by the United States Judicial Panel on Multidistrict Litigation and randomly assigned to our court.
Second, the Court's synopsis of the case's procedural history demonstrates that as complex as this 53-page treatise already is, it could have been much more complex:
In August 2017, we granted EPA's motion to sever and hold in abeyance the Industry Petitioners' and Water Company Petitioners' challenges to the final rule. In September 2017, EPA announced it would reconsider the rule's regulations concerning non-legacy FGD and bottom ash transport water. See 82 Fed. Reg. 43,494. As a result of these procedural developments, the challenges to the final rule raised by the Industry Petitioners and the Water Company Petitioners are not before us. We address only the challenges brought by the Environmental Petitioners.
In a footnote, the Court adds that
The agency's reconsideration of those aspects of the rule has been challenged in separate lawsuits brought by a coalition of environmental groups, including some of the petitioners in this case. That challenge was brought contemporaneously in the District of Columbia federal district court and the D.C. Circuit. The district court ruled it lacked jurisdiction, Clean Water Action v. Pruitt, 315 F.Supp.3d 72, 85 (D.D.C. 2018), and the D.C. Circuit transferred its case to this court, Clean Water Action v. Pruitt, No. 17-1216, Order (D.C. Cir. Feb. 1, 2018), where it is now pending as No. 18-60079.
Your tireless author did some poking around and discovered that Case No. 18-60079 is scheduled for oral argument in New Orleans in the en banc courtroom today. The panel is be Judges Higginbotham, Jones, and Costa. I can't wait to listen to that argument. Naturally, the petitioners in that case have already filed a 28j notice arguing that Judge Duncan's opinion in Southwestern Electric Power Co. v. EPA is helpful to their cause. Assessing that argument is beyond the scope of this post, but rest assured, you'll hear from me again once we get an opinion in No. 18-60079.
2. The next highlight-worthy point appears in the Court's discussion of the standard of review. It begins with the usual: a line or two about arbitrary and capricious review under the APA; the obligatory invocation of State Farm, Chenery, and the rest; and the even-more obligatory (if that's possible) recitation of the Chevron two step (or three step, or one step, or whatever). Deep within that tangle of footnotes, Bluebook cites, and nested quotations, however, the Court tosses in this interesting line: “Chevron review and arbitrary and capricious review overlap at the margins, specifically at Chevron step two." As I explain below, the Court's invocation of this gloss on the canonical standard of review turns out to be critical to its subsequent analysis.
3. Despite the similarity between the petitioners' objections to EPA's legacy wastewater and leachate regulations, petitioners leveled leveled them under different legal standards--one under the APA arbitrary and capricious standard and the other under Chevron. As I've already explained, the Court agreed with petitioners on both scores, citing essentially the same core problem with the agency's approach to both aspects of the final rule.
Anyone who has studied administrative law will be familiar with the ongoing--and apparently intractable--debate regarding whether and to what extent review under APA arbitrary and capricious review and step two of the Chevron test overlap. The Court's opinion gives us some indication about how Judges Duncan, Ho, and Haynes would answer that question. Not only do they expressly say that the standards overlap, they also proceed to explain precisely how and why a single infirmity in EPA's approach required vacatur of two parts of the final rule under each standard.
So what, you ask? Well, the debate over whether and to what extent Chevron step two overlaps with arbitrary and capricious review is not merely academic. Far from it. Whether a case is decided under Chevron or State Farm or both can make a big difference. When a court finds that an agency action is arbitrary and capricious, it will usually remand to the agency for further consideration. See, e.g., U.S. Sugar Corp. v. EPA, 830 F.3d 579, 630 (D.C. Cir. 2016). When a court concludes that an agency interpretation fails Chevron, though, it has found that the agency acted outside of the compass of its delegated authority or that the agency's action rests on an impermissible construction of its authorizing statute. In such circumstances, the agency's action is effectively vacated. See, e.g., Michigan v. EPA, 135 S. Ct. 2699, 2711 (2015); Am. Library Ass'n v. FCC, 406 F.3d 689, 708 (D.C. Cir. 2005).
4.The Court returns to the overlap point later in its opinion and ups the ante:
Because Chevron step two and the APA share the “arbitrary and capricious” standard, “[t]he APA reflects the principles of Chevron,” and analysis under the two standards proceeds similarly. Nutraceutical Corp. v. Von Eschenbach, 459 F.3d 1033, 1038 (10th Cir. 2006); see also, e.g., PIERCE § 3.6 (suggesting that Chevron step two has “complete overlap” *1029 with APA test of whether a rule adopts an “unreasonable” statutory interpretation) (quoting Animal Legal Def. Fund v. Glickman, 204 F.3d 229, 234 (D.C. Cir. 2000))
Am I misreading that passage, or has the Court really concluded that the existence, in practice, of an overlap between the State Farm and Chevron standards proves that "the APA reflects the principles of Chevron" ? If that is indeed the Court's position, don't tell Aditya Bamzai. He has argued extensively (and forcefully if you ask me) that Chevron marked a dramati departure from the APA standard rightly understood. Here's his summary of that argument:
Congress enacted the APA in 1946 in part to stop this deviation from the traditional interpretive rules and to recapture the interpretive methodology that prevailed before the Court’s experimentation with the law-fact distinction during the 1940s. The APA’s text, drafting history, and early scholarly interpretations all point in this direction: they suggest that Congress sought to cabin the discretion that the Court had recently granted administrative agencies. But the APA’s text and drafting history were quickly forgotten. In the time between the APA’s adoption and Chevron, courts relied interchangeably on cases applying the mandamus standard, cases applying the traditional contemporary and customary canons, and cases applying the 1940s approach breaking down the distinction between judicial review of questions of law and questions of fact. The result was, as Judge Friendly observed in Pittston Stevedoring Corp. v. Dellaventura, a bewildering and often contradictory set of rules to govern judicial review of agency statutory interpretation. Chevron cleared up this confusion by departing from, rather than seeking out, the meaning of the APA’s text and the traditional interpretive methodology.
As an aside, Aditya Bamzai is my hero. I highly recommend that you go forth and read his brilliant article on the origins of judicial deference to executive interpretation as soon as possible.
5. At several points, the Court's State Farm analysis drifts into Chevron territory or vice versa. Here are some examples:
That paradoxical action signals arbitrary and capricious agency action. See, e.g., Chamber of Commerce, 885 F.3d at 382 (“Illogic and internal inconsistency are characteristic of arbitrary and unreasonable agency action.”); see also, e.g., GameFly, Inc. v. Postal Regulatory Comm'n, 704 F.3d 145, 148 (D.C. Cir. 2013) (explaining that agency action “ ‘illogical on its face’ ” may be arbitrary and capricious) (quoting Am. Fed'n of Gov't Emps., Local 2924 v. Fed. Labor Relations Auth., 470 F.3d 375, 380 (D.C. Cir. 2006)). It also strongly suggests that EPA has contravened the plain language of the CWA, which defines BAT as the technology that “will result in reasonable further progress” toward pollutant discharge elimination. 33 U.S.C. § 1311(b)(2)(A) (emphasis added); see also, e.g., Nat'l Crushed Stone, 449 U.S. at 74-75, 101 S.Ct. 295*1017 (discussing “reasonable further progress” component of BAT).
These shortcomings in the agency's explanations strongly indicate that its BAT decision simply defaults to the outdated BPT standard that has been demonstrated to be a poor performer by the agency's own analysis. That is antithetical to the statutorily-mandated BAT standard.
These examples are particularly interesting because by hinting that arbitrary and capricious actions strongly suggest that EPA has contravened the plain language of the Act, the Court implies an overlap between Chevron step one and State Farm (more on this below). Blurry lines indeed!
6. At one point in its analysis under Chevron step one, the Court explains that EPA's final rule "conflates the BAT and BPT standards in a way not permitted by the statutory scheme." Here is the relevant passage:
We agree with petitioners, however, that the leachate rule conflates the BAT and BPT standards in a way not permitted by the statutory scheme. The rule pegs BAT for leachate to the decades-old BPT standard, without offering any explanation for why that prior standard is now BAT. That is flatly inconsistent with the Act's careful distinction between the two standards.
. . . .
To be sure, we do not say that EPA is precluded by the Act from ever setting BAT equivalent to a prior BPT standard. But given the plain distinction between the two standards marked out in the Act, the agency would at least have to offer some explanation for its decision that speaks to the statutory differences between BAT and BPT. Here we are given nothing along those lines. Consequently, the only conclusion we can draw from this record is that, in setting BAT for leachate, the agency simply defaulted to the prior BPT. As explained, however, the statutory scheme does not confer authority on the agency to collapse the carefully-wrought distinction between BAT and BPT in this manner. See, e.g., Texas v. United States, 497 F.3d 491, 501 (5th Cir. 2007) (“Chevron deference comes into play, of course, only as a consequence of statutory ambiguity, and then only if the reviewing court finds an implicit delegation of authority to the agency.”).
I was surprised to see this objection to the sufficiency of EPA's explanation raised part of a Chevron step one discussion. After all, aren't illogical agency explanations the stuff of State Farm analysis? Even assuming Chevron and State Farm do overlap, the overlap is due to the focus on reasonableness common to the State Farm test and step two of Chevron. As already mentioned, that is how the Court itself describes the supposed overlap.
Here, though, the Court seems to read into the statue a requirement that EPA provide a particular sort of explanation for a particular sort of rule, and it does so smack dab in the middle of its Chevron step one analysis. In effect, the Court holds that the plain language of the CWA unambiguously required EPA to provide a fuller explanation than it gave in the final rule. I haven't read the CWA in a while, but I don't recall seeing such an express requirement anywhere in the actual text of the Act.
Of course, this is not the first time that a court's rejection of an EPA rationale under Chevron arguably had the effect of reading an explanation requirement into a statute. The Supreme Court arguably did just that in Michigan v. EPA. In Michigan, though, the Court was applying Chevron step two. Here, the Court was applying step one. I can't claim to have read every Chevron case, but I'm not aware of another example of this phenomenon. If you are, please let me know!
There's a lot more to say here. How does this sort of merging of Chevron step one and State Farm fit with Chenery? Mead? Brand X? I could go on and on, but I would lose my day job if I did. So for now I'll leave further analysis to those with more time and sharper legal minds.
A few thoughts in conclusion. While I'm not sure I agree with every single move the Court makes in this opinion, I agree with the vast majority of its moves and wholeheartedly agree with the results it reached. I also really admire Judge Duncan's writing style, his attention to detail, and his bold approach to these issues. I can't wait to see more from him.
So what have we learned? In my view, the key takeaway here is that standards of review really matter in these cases. A clever judge like Judge Duncan can make a subtle gloss on the applicable standards do a whole lot of work. The petitioners' decision to lodge their twin challenges to EPA's final rule under different legal standards presented the panel with a puzzle. By focusing on where the tests arguably overlap, though, Judge Duncan found a solution that gave the Court the flexibility necessary to slide seamlessly from one standard to the other. In this post, I've attempted to highlight some of the interesting implications such artful drifting might have for adlaw theory more generally. But I know I've barely scratched the surface. I'd love to hear what you think about this opinion. Did you find any aspects of the Court's analysis not discussed here particularly interesting? Let me know!
Oh yeah, bonus points to readers who can guess the rationale for the picture I chose as a theme for this post . . .