The Volokh Conspiracy

Mostly law professors | Sometimes contrarian | Often libertarian | Always independent

Climate Change

Federal Court Bars Federal Agencies from Considering the Costs of Climate Change in Rulemakings [Updated]

A federal district court has taken the unusual step of enjoining an Executive Order setting forth an Administration's regulatory priorities.


On his first full day in office. President Biden issued Executive Order 13990, requiring that federal agencies begin reversing Trump Administration environmental policies, particularly those related to climate change. One provision of EO 13990 declared that it is "essential that agencies capture the full costs of greenhouse gas emissions as accurately as possible," and created an Interagency Working Group (IWG) tasked with issuing an interim "social cost of carbon" (SCC), estimating the costs imposed by greenhouse gas emissions. The EO further provided that "agencies shall use" this cost estimate "when monetizing the value of changes in greenhouse gas emissions resulting from regulations and other relevant agency actions" pending the development of a more robust social cost of carbon estimate.

As is common these days, a number of states filed suit against this new Biden policy, challenging the new Administration's authority to order consideration of the social cost of carbon. One suit, filed in Missouri, was dismissed on standing grounds, as has happened with past efforts to challenge regulatory policy Executive Orders in court. As occurred when progressive groups and blue states challenged Trump Administration EOs, the district court in Missouri v Biden concluded that the plaintiff states needed to challenge a specific final regulatory action that harms then, and could not challenge the presidential directive itself. This case is now on appeal.

A second suit was filed by a second group of states in Louisiana, and has produced quite different results. On February 11, Judge James Cain of the U.S. District Court for Western Louisiana granted a preliminary injunction against the Biden Administration, barring federal agencies from considering the IWG's social cost of carbon estimates or "independently relying upon the IWG's methodology considering global effects, discount rates, and time horizons," and requiring federal agencies to follow the guidance of a George W. Bush Administration memo (Circular A-4) in conducting regulatory analyses. Whatever one thinks of the IWG's social cost of carbon estimates or their role in regulatory policy, this is a bizarre and highly problematic opinion.

The threshold problem with Judge Cain's opinion in Louisiana v. Biden is its conclusion that the case is justiciable. The plaintiff states' claims, and the relief sought, concern instructions for how agencies are to consider the climate impacts of their decisions, and does not target anything that could remotely be considered a "final agency action." (Nor, for that matter, is the President an "agency" under the APA for purposes of judicial review.) If the plaintiff states were challenging how a specific agency relied upon the IWG's SCC estimates to reach a particular regulatory decision, such claims could be raised in a legal challenge to that specific agency decision. Such a challenge could even include arguments that the IWG's estimates are arbitrary and capricious. What is not generally allowed, however, is for plaintiffs to challenge a presidential directive prescribing the manner in which agencies are to go about making regulatory decisions before such regulatory decisions are made. It is not clear how such claims satisfy the requirements of Article III standing, nor are such claims ripe.

Judge Cain notes that the plaintiff states are concerned that the IWG's social cost of carbon estimates, when considered in the context of particular programs, may result in more stringent regulatory requirements. They might. And when a given agency imposes a more stringent regulatory burden on a state or private entity due to the social cost of carbon, affected parties (including the plaintiff states) will have ample opportunity to challenge that regulatory action and the analysis upon which that action was premised. At that point, it would be perfectly appropriate for a court to consider whether that agency was allowed to consider the potential costs of climate change in this way, and whether its doing so was consistent with the APA's requirement of reasoned decisionmaking. Yet that is quite different from declaring, as the court does here, that no agency may rely upon these estimates in any rulemaking or other agency action going forward, without regard for the particular statutory authorities or constraints applicable to that specific agency action.

The opinion's analysis then turns to the plaintiff states' arguments that EO 13990 worked "an enormous and transformative expansion in [its] regulatory authority without clear congressional authorization," in violation of the "major questions doctrine." As much as I like the major questions doctrine (and have even argued for its aggressive use), it has no place here.

Judge Cain writes:

The Court finds that EO 13990 contradicts Congress' intent regarding legislative rulemaking by mandating consideration of the global effects. The Court further finds that the President lacks power to promulgate fundamentally transformative legislative rules in areas of vast political, social, and economic importance, thus, the issuance of EO 13990 violates the major questions doctrine.

There are several errors here, starting with the conclusion that the EO (an action by the president) is a legislative rule (or even that it is a final agency action subject to judicial review). If it would be an error for an agency to consider global effects in the adoption of a particular rule, given the particular statutory authorization for that rule, a court could enjoin that agency action once it is finalized, but that is not what is happening here. Rather, the court seems to accept that directing agencies to consider global effects in future rulemakings is itself the sort of legislative rule that requires express congressional authorization under the major questions doctrine. This is wrong on multiple levels and poses a severe threat to White House oversight of agency rulemaking.

Nothing in EO 13990 expands federal regulatory authority. This EO, like prior regulatory EOs issued by prior presidents, sets out the current administration's regulatory priorities. In other words, it gives agencies direction as to how they should use the regulatory authority they already have. Also as with prior EOs issued by prior administrations, it directs agencies to consider things they may not have considered in the past or to consider things in a new way. There is nothing unlawful about this. Asking agencies to consider, say, the global effects of their actions no more expands federal regulatory authority than did prior EOs that asked agencies to begin considering cost-benefit analysis, comparative effectiveness analysis, or Takings analysis. Calling upon agencies to consider the social cost of carbon, where they are not statutorily prohibited from doing so, no more implicates the major questions doctrine than did the Trump Administration EO telling agencies to adopt a form of regulatory budgeting. Indeed, even if a statute bars an agency from making such costs a factor in its decisionmaking, this does not preclude an agency from making the assessment for informational purposes. (So, for instance, the EPA does cost-benefit analyses of proposed NAAQS revisions, even though it may not consider costs when setting the NAAQS.)

The opinion also embodies a bizarre notion of executive power, under which the (allegedly) consistent approach to regulatory analysis by prior administrations somehow bars future administrations from changing course, within the bounds proscribed by relevant statutes. Specifically, Judge Cain suggests that because prior administrations adopted a particular approach to estimating costs and incorporating discount rates, the Biden Administration is somehow bound to follow suit, and that a federal court has the authority to prescribe compliance with prior administrations' policies outside the context of a specific agency action that is under review.

In seeking to justify an order that agencies adopt the regulatory analysis policies of a prior administration, the opinion also mangles the history of executive branch regulatory review, and even manages to misrepresent the authorities upon which he relies. Here I will rely on the comments of Duke's Jonathan Wiener, co-author (with Michigan's Nina Mendelson) of one of the papers on which Judge Cain purported to rely:

I noticed that on p.5 of this court's slip opinion, it cites an article that Nina Mendelson and I wrote in 2014, as a reference for what this court calls "the consensus on cost/benefits analysis required by Presidents Nixon, Ford, Carter, Reagan and Clinton. See Nina A. Mendelson & Jonathan B. Wiener, Responding to Agency Avoidance of OIRA, 37 Harv. J.L & Pub. Pol'y 447, 454–57 (2014)."  That's partly right, but actually what Nina and I wrote is that the bipartisan consensus across presidencies in favor of benefit-cost analysis (BCA) started with the EOs issued by Carter (12044, plus OMB guidance) and Reagan (12291), whereas Nixon and Ford had focused on "quality of life," "inflation," and "economic impact" (perhaps the latter embraces BCA); and then, building on Clinton's EO 12866 (which solidified bipartisan commitment to BCA), this bipartisan consensus continued in the Bush and Obama administrations (e.g. via Circular A-4 in 2003, and EO 13563 in 2011) (see our paper at pp.457-463).  More recently I have written here and here that the Trump administration departed from that consensus (by neglecting benefits), that good BCA should consider all important impacts (including target benefits, costs, co-benefits, and countervailing risks), and that the Biden memo on Modernizing Regulatory Review (Jan. 20, 2021) returns to the prior bipartisan consensus (by reaffirming EOs 12866 and 13563), plus calls for further measures yet to come.

It is not simply that the opinion gets the history wrong. It is that the opinion uses this stylized historical account as the basis for claiming that a particular approach to regulatory analysis, that embodied in the Bush Administration's Circular A-4, is somehow legally required, despite the lack of any legislation to that effect, and that a federal court has the authority to tell federal agencies, prospectively, what things they may or may not consider when developing regulations. The opinion also flubs the details of specific regulatory requirements (such as those related to NAAQS standards, noted above), but this post is long enough as it is, so I will forebear detailing these mistakes.

I am sympathetic to the argument that the Bush Administration's approach to discount rates is preferable to that of the Biden IWG, but the idea that it is unlawful for a President to order use of the IWG's approach where agencies have the discretion to do so is quite radical, and without meaningful legal precedent. If Congress wants agencies to adopt a particular methodology for conducting regulatory analysis, it is free to do so. But unless and until it does, the White House may direct agencies to prioritize or emphasize particular concerns within the bounds of existing statutory constraints. In this regard it is notable that Judge Cain cites no statutory authority for precluding the Biden Administration's approach across-the-board. At best all he can cite are program-specific requirements that agencies consider particular things.

There may well be strong policy arguments against the Biden Administration's approach to climate change and regulatory policy (see, e.g., here). But such policy disagreements do not provide license for federal courts to dispatch with traditional administrative law doctrines or invent new ones.

I expect this decision to be appealed and would like to think the appellate court will not repeat the district court's errors.

P.S. In the meantime, the New York Times reports, this ruling is slowing a wide range of agency actions, including the approval of permits of oil and gas development. In other words, in the name of challenging regulations, these state plaintiffs may have actualy increased some regulatory burdens.

UPDATE: Devin Watkins of the Competitive Enterprise Institute takes issue with my post, and argues that Judge Cain's opinion is "reasonable," and "far from the crazy opinion" my post suggests. To be frank, I thought I had gone easy on Judge Cain, as I did not identify every error nor point out every oddity (such as the opinion's reference to the "separation of powers clause"). Judge Cain's opinion was far from reasonable. It was wrong many times over and made a mess of relevant law.

Watkins' attempted defense of the Louisiana v. Biden decision makes multiple errors of its own, both misstating and misapplying black-letter administrative law concerning legislative rules and what constitutes a final agency action, while also engaging in a bit of bait-and-switch as to what is being challenged and which agencies were enjoined.

The IGW estimate cannot be a legislative rule because it, by itself, does not obligate anyone to do anything. Any obligation agencies have to apply the social cost of carbon estimate comes from the Executive Order. Even if the EO could be challenged as a final agency action, it expressly says it does not augment or constrain preexisting agency authority or create any legally binding obligations, nor does it imbue the IWG with the power to take action with the force of law.

If an agency uses the IWG's social cost of carbon estimates to do something that is unlawful or arbitrary, that action can be challenged, but there is no basis to challenge a number that was issued pursuant to an Executive Order unless and until it is incorporated into an agency action that has legal effect. That a federal agency might, in the future, reject a state submission doesn't cut it.

That various statutes direct agencies to consider particular factors does not bar the President from asking agencies to consider other matters (though it may preclude agencies from relying upon such matters in making their decisions). If a given statute precludes the consideration of specific factors by an implementing agency, this could serve as the basis for a challenge to a specific action taken pursuant to that statute, but not the basis of a global challenge to preclude any and all consideration of those factors by any agency acting pursuant to other statutes.

I am particularly surprised that Watkins seems to endorse Judge Cain's view that OMB Circular A-4—a non-binding guidance –cannot be modified or departed from without notice-and-comment because OMB solicited comment when it promulgated A-4. This is wrong as a matter of law and, if taken seriously, would hamstring Presidential efforts to set agency priorities. If a given degree of process were required for an agency to take an initial action, no less process is required for the agency to change course. But (and this is key), an agency's voluntary choice to provide more process than is necessary (as the OMB did) does not operate as a ratchet forever obligating that agency to provide such process in the future. To claim otherwise in the case of a non-binding guidance, such as A-4, is neither reasonable nor remotely correct.

I understand the objections to the IWG's SCC estimates (see Roger Pielke Jr.'s critique of the Biden Administration's approach here), but Judge Cain's opinion is indefensible, and it gets worse with every repeated reading. My original post and this update barely scratch the surface in this regard.

It also turns out (as noted in the P.S. above) that Judge Cain's ruling is having some unintended effects, and creating conflicts for agencies that are under legal obligations to take broader account of climate-related costs (including those overseas). For more on the consequences of the ruling (and more on how OIRA review of agency rulemakings actually operates), see this declaration in support of the government's motion for a stay in Louisiana v. Biden by Dominic Mancini, Deputy Director of OIRA.