TransCanada Sues U.S. Government For Rejecting Keystone Pipeline

Courtesy of the <a href="">Alberta Energy Regulator</a>

Courtesy of the Alberta Energy Regulator

On Wednesday, TransCanada filed a complaint against the United States in a federal district court in Houston alleging that the President’s rejection of the Keystone XL pipeline was invalid and unconstitutional because it was not authorized by Congress. If successful, this claim would allow construction of the pipeline.

On the same day, TransCanada filed a notice of intent to submit a claim to arbitration under the North American Free Trade Agreement (NAFTA). Even if successful, this claim would not allow construction of the pipeline, but could entitle TransCanada to money damages from the United States. The company is asking for $15 billion in damages.

Like most private lawsuits against the government, these lawsuits face long odds, but both raise important and novel legal issues that will be difficult to decide. TransCanada’s constitutional claim could change the way that the United States approves international oil pipelines. And TransCanada’s NAFTA claim could endanger the United States’ long winning-streak in NAFTA arbitrations.

TransCanada’s Constitutional Claim

The most unexpected part of TransCanada’s legal salvo was the lawsuit that it filed asking a U.S. district court to rule that President Obama’s rejection of the Keystone XL pipeline was unconstitutional. TransCanada notes that Congress has never passed a statute that gives the President authority to reject international oil pipelines and says that, without such a law, the President had no authority for his unilateral rejection of the pipeline.

Congress has never provided a legal framework for regulating oil pipelines that cross the United States’ international borders. By contrast, there are laws that establish a process for the President to decide on international natural gas pipelines and electricity transmission.

In the absence of Congressional authorization, President Lyndon Baines Johnson simply issued an executive order in 1968, Executive Order 11423, that established a process for issuing permits to proposed oil pipelines that “would serve the national interest.” Then in 2004, President George W. Bush issued a new unilateral order, Executive Order 13337 that expedited review of border crossings. Both executive orders delegate decisions on these cross-border permits to the U.S. Secretary of State.

On November 6, the current Secretary of State, John Kerry rejected the Keystone XL pipeline after seven years of review. The official U.S. Record of Decision stuck by the State Department’s controversial previous conclusion that the pipeline would improve U.S. energy security, benefit the economy, and would be unlikely to increase greenhouse gas emissions in Canada. (It also suggested that the pipeline might even decrease greenhouse gas emissions in the United States by moving oil transport from railroads to pipelines, making oil transport more efficient.) But the U.S. concluded that the pipeline was ultimately not in the national interest because it could undercut the nation’s leadership in climate talks because the pipeline was “perceived as enabling further [greenhouse gas] emissions globally.”

TransCanada’s key argument is that, in the absence of any law, the President does not have unilateral authority to reject an international oil pipeline based on this kind of consideration. Although Presidents have claimed power to decide whether a pipeline is in the national interest since President Johnson in 1968, TransCanada argues that this power has never been fully tested because the President has never rejected an international pipeline.

This creates something of a puzzle: if Congress has never passed a law governing international oil pipelines and the President does not have authority to reject an oil pipeline, then who may, in fact, regulate pipeline border crossings?

One possible answer is that international oil pipelines are primarily regulated by the states, just like domestic oil pipelines. The U.S., unlike Canada, primarily relies on state-by-state regulation for interstate oil pipelines. That is, if no law has been enacted governing international oil pipelines, then the only laws that govern them are the same ones that govern domestic oil pipelines.

President Obama’s administration will raise several counterarguments. First, it will argue that the President has inherent and unilateral constitutional authority to control the nation’s borders, so he must have some kind of ability to control international border crossings. Second, if Congress has not established any criteria for the President to use in this decision, then he is free to create his own criteria. Third, President Johnson established this process almost fifty years ago and it has been frequently used to approve pipelines so Congress has, with the passage of time, acquiesced to this process. Fourth, federal district courts have upheld the President’s unilateral decision to approve international pipelines.*

TransCanada will respond that, whatever power the President has, it does not allow him to reject a pipeline based solely on international perceptions that are inconsistent with the government’s own environmental analysis. TransCanada’s complaint also argues that, far from acquiescing in the President’s unilateral authority to reject international pipelines, recent Congresses have repeatedly sought to constrain the President’s authority, citing Congress’s frequent attempts to approve the Keystone XL pipeline. Finally, TransCanada will point to federal court decisions and executive branch opinions from nearly a century ago, which concluded that in the absence of Congressional authorization the President had, at most, limited authority to control border-crossing facilities. Though old, these opinions may remain relevant in the unusual situation where, as with oil pipelines, Congress has not established a process for permitting border crossings.

The continuing saga of the Keystone XL drama overlaid with a tangle of old and new precedents and conflicting constitutional powers will make TransCanada’s U.S. lawsuit a case to watch. If a Republican is elected President this coming November, then the issue will likely be moot because the Republican contenders say they would reverse President Obama’s decision on the pipeline. But if not, then the U.S. courts will have to resolve the thorny issues raised by TransCanada.

TransCanada’s NAFTA Claim

TransCanada’s other action, its notice of intent to submit a claim to NAFTA arbitration, alleges that the U.S. discriminated against Keystone XL’s Canadian investors, violating its obligations to afford them national and most-favored-nation treatment under Article 1102 and Article 1103 of NAFTA. TransCanada also argues that by delaying a decision on the pipeline for seven years, and then denying it, the U.S. government destroyed the value of its investment, expropriating its property in violation of NAFTA Articles 1110 and 1105.

NAFTA claims are decided by three independent arbitrators. These arbitrators are not bound by the decisions of the arbitrators that decided previous claims. Thus, it is very difficult to predict whether a NAFTA claim will be successful.

If past cases are any indication, a Canadian company like TransCanada begins at a serious disadvantage. The United States has never lost a NAFTA decision to a foreign investor. And arbitrators have sometimes gone to great lengths to avoid a finding of discrimination. In one case, California passed a law that, it admitted, used “narrowly crafted language intended to prevent approval of a specific mining project” owned by Canadian investors. But the NAFTA panel for that case held that the law was not discriminatory because, in theory, that narrowly crafted language could apply in the future if another company proposed a similar project.

On the other hand, the extraordinary facts of the Keystone XL review process could end the United States’s NAFTA winning streak. First, throughout the seven-year review, President Obama repeatedly responded to complaints from pipeline supporters by admonishing them to remember “this is Canadian oil, this isn’t U.S. oil.” And the President’s administration was, at the same time, moving to expedite domestic oil pipelines. Second, after repeatedly delaying the decision on Keystone XL and repeated environmental impact studies, the U.S. denied the permit on the basis of a perception that was not supported by the seven years of analysis it had done. It will be difficult to explain why it took seven years to analyze the pipeline if, in the end, the government chose to ignore that analysis.

Finally, TransCanada’s lawsuits may operate in tandem because one relevant set of laws that Congress has passed concerning international energy trade is the set of laws approving and implementing NAFTA. In U.S. court TransCanada will argue that even if Congress has not prescribed a specific process for international oil pipelines, it has, at least ruled out any discriminatory or arbitrary treatment of Canadian investors in those pipelines. One of the chief challenges for U.S. lawyers will be to explain why the federal government should impose a uniquely lengthy and unpredictable process on Canadian oil pipelines while expediting domestic oil pipelines.

Regardless of the outcome, TransCanada’s Keystone XL challenges set the stage for potential blockbuster decisions that will have a lasting impact on energy, constitutional, and trade law.


You can see more legal documents & analysis related to the Keystone XL pipeline and other North American oil pipelines at Oil Transport Tracker (Shortcut link:



*Full disclosure: Before my academic career, I worked in private practice and represented TransCanada in two of these earlier cases. 

Supreme Court: EPA Should Have Considered Cost When Deciding Whether Mercury Limits For Power Plants Were Appropriate

Screen Shot 2015-06-29 at 9.19.29 PMToday the United States Supreme Court held that the Environmental Protection Agency (EPA) improperly refused to consider costs when determining whether it was “appropriate and necessary” to regulate mercury emissions from power plants under the Clean Air Act. Ultimately, EPA may be able to keep the same rules after going back and explaining why the cost of the regulations is justified in the circumstances. But the decision is an important victory for advocates of cost-benefit analysis and those who think environmental agencies should pay more attention to the costs of regulation.

Section 112 of the Clean Air Act directs EPA to regulate hazardous air pollutants from power plants if it finds “regulation is appropriate and necessary.” 42 U.S.C. §7412. EPA said that regulation was “appropriate and necessary” even without considering costs because 1) power plant emissions posed risks to human health and the environment that were not eliminated by other provisions of the Clean Air Act and 2) there were controls available to reduce those dangerous emissions. So there was no need for EPA to consider costs to make its initial decision to regulate, but it promised to consider costs when adopting the actual final regulations for power plants.

Although EPA said it ignored costs when it made its initial decision to regulate, it still estimated the costs and benefits of the final rules that it adopted. EPA estimated that its rules would cost power plants $9.6 billion dollars a year. EPA couldn’t estimate all the possible benefits of limiting mercury emissions, but the little it could quantify came to about $5 million a year—less than 0.1% of the cost of the rule. On the other hand, EPA said that cleaning up mercury would have massive side benefits: it would lower sulfur dioxide emissions and these reductions would be worth between $37 and $90 billion per year. So these ancillary benefits far outweighed the costs of EPA’s rule, but if you didn’t count them, EPA’s rule imposed costs far in excess of its benefits.

Justice Scalia, writing for a 5-4 majority, held that EPA must consider the costs of regulation before making its initial decision to regulate, reasoning that “No regulation is ‘appropriate’ if it does significantly more harm than good.” The four dissenters conceded that, generally speaking, “an agency must take costs into account in some manner before imposing significant regulatory burdens” but agreed with EPA’s argument that the agency could consider those costs later when adopting regulations for specific source categories.

The Supreme Court’s decision may not have much impact on mercury regulation. Power utilities are already complying with the mercury rules that the court struck down in this case. And the case will now go back to the appellate court, which could decide to leave the rules in place while the agency rethinks whether these rules are “appropriate and necessary” factoring in the costs that they impose. EPA already determined that the benefits of the rules far outweighed their costs if you consider ancillary benefits, so it will probably reach the same decision. On the other hand, the Court’s decision raises very important questions for the future.

First: Can agencies consider ancillary benefits? At oral argument, some justices seemed to suspect it was inappropriate to consider the benefits associated with pollutants other than mercury. After all, if the other pollutants are the problem, why not adopt regulations aimed at the other pollutants? On the other hand, it has long been standard practice for agencies to consider ancillary or “co-benefits” of reducing pollutants other than the main target of regulation. If an agency is going to consider all the important costs of a regulation, why shouldn’t it consider all the important benefits? In some ways, the mercury rule may just be an outlier case because EPA estimated that the co-benefits of reducing sulfur dioxide were 10,000 times greater than the direct benefits of reducing mercury itself. But over half of the benefits of EPA’s Clean Power Plan come from co-benefits in reducing pollution other than greenhouse gases, so the question does have wider importance.

Second: How much cost-benefit analysis will the Court require for other regulations? Today’s decision may be seen as part of a trend that is making cost-benefit analysis a kind of default background principle for agency decision-making. Just fourteen years ago, Justice Scalia wrote an opinion for eight justices, holding that EPA could not consider the cost of regulation when the Clean Air Act demanded a standard at the level “requisite to protect the public health.” In that case, Justice Scalia explained that EPA could consider costs later when it implemented the standard. Last year, the Court held that EPA could consider the cost of emissions controls when it decided whether a State “contributed significantly” to air pollution in another state; Justice Scalia dissented. Now, the Court holds that EPA must consider the cost of regulation when it determines whether regulation is “appropriate and necessary.” Justice Scalia writes the opinion, and all justices agree that EPA must consider costs at some stage. Observing this trend, litigants will feel increasingly bold to demand that EPA consider the costs at each stage of adopting new environmental regulations.

Supreme Court Leaves Room for State Regulation of Natural Gas Sales

Yesterday, the U.S. Supreme Court held that the federal Natural Gas Act does not preempt the field of state antitrust regulation of natural gas prices, which means states can apply their own policies to natural gas sales as long as those policies do not conflict with federal law.

Vanderbilt’s Jim Rossi has just posted an analysis of the case at SCOTUSblog. As he notes, U.S. courts have been struggling with how to draw a line between state and federal authority in both electricity and natural gas markets.

Under both the Natural Gas Act and the Federal Power Act, the federal government has authority over wholesale energy sales, while the states retain authority over retail sales of natural gas and electricity. As a result, the Supreme Court’s decision on the Natural Gas Act may have important implications for electricity markets as well. And it is another important precedent in the courts’ struggles to balance state’s traditional authority over their own energy markets with increasingly integrated interstate energy markets.

Professor Rossi writes:

According to the majority opinion, written by Justice Stephen Breyer and joined by five other Justices, the [Natural Gas Act] “was drawn with meticulous regard for the continued exercise of state power, not to handicap it or dilute it in any way.” Under Section 1(b) of the [Natural Gas Act], wholesale transactions fall squarely – and even exclusively — within the jurisdiction of federal regulators. For nearly seventy years, the Court has acknowledged the sharp clarity of this federal-state division of authority over wholesale and retail sales, sometimes even calling it a jurisdictional “bright line.”

… The difficult question this case presented was what to do when a practice affects both types of sales.

… Notwithstanding (somewhat confusing) language in the opinion that purports to place this matter on the state “side” of any dividing line, the majority questioned whether the [Natural Gas Act] contains any sharp dividing line at all: “Petitioners and the dissent argue that there is, or should be, a clear division between areas of state and federal authority in natural gas regulation. But that Platonic ideal does not describe the natural gas regulatory world.”

Here is Professor Rossi’s full post and here is the Supreme Court’s opinion.

U.S. Supreme Court Narrows Greenhouse Gas Rules: What It Means for the U.S. Climate Agenda

Today, in Utility Air Regulatory Group v. Environmental Protection Agency (EPA), the U.S. Supreme Court struck down a portion of the United States’ first regulations for greenhouse gas emissions from industrial sources. The Court held that the Environmental Protection Agency (EPA) may not apply its “Prevention of Significant Deterioration” (PSD) program to new industrial sources on the basis of their greenhouse gas emissions. Instead, EPA can only regulate greenhouse gas emissions from new sources that are already subject to the PSD program because they emit other pollutants.

This is the first Supreme Court decision on EPA’s authority to regulate greenhouse gases from industrial sources, so it has important implications for EPA’s future climate agenda—including its recently proposed rule for the electricity sector. And the varied opinions offered by the Supreme Court justices offer hints about how courts will approach the inevitable legal challenges to those regulations.

EPA’s PSD program has two basic requirements:

1) You need a permit before you build a new major industrial source of air pollution.

2) And to get a PSD permit, you must show that you are using the “best available control technology” for the air pollutants that you emit.

In this case, the Supreme Court held:

1) EPA may not require new sources to get a PSD permit simply because they will emit large amounts of greenhouse gases. The Court held that it would be unreasonable for greenhouse gases to trigger the permit requirement, because the PSD permit program is only meant to apply to the thousands of industrial sources that emit conventional pollutants, not the millions of sources that emit significant amounts of greenhouse gases.

2) But if a source needs a PSD permit anyway, because it emits other pollutants, then EPA may require it to adopt the “best available control technology” for greenhouse gases, along with other air pollutants.

I will not say anything more about the complexities of the decision, because I described them extensively in a previous post, which read the tea-leaves of oral argument in the case, and suggested the Supreme Court would reach exactly this compromise. So you can read that post both for a description of the statutory interpretation question and an explanation of the reasoning that the court eventually followed.

The most pressing question raised by the case today may be its implications for the United States’ future climate agenda, including EPA’s recently proposed rule for existing power plants. There are three important implications, and each could spell trouble for EPA’s climate agenda.

First, the Court suggested that one reason for rejecting EPA’s rule is that “it would bring about an enormous and transformative expansion in EPA’s regulatory authority without clear congressional authorization” because millions of sources would be subject to a greenhouse gas permit requirement. EPA, it is true, had suggested it would only regulate a reasonable number of them, but the court was not willing to leave that decision in the agency’s hands.  The court noted: “When an agency claims to discover in a long-extant statute an unheralded power to regulate a significant portion of the American economy, we typically greet its announcement with a measure of skepticism.”

This passage will trouble EPA. In the agency’s recent proposal to cap greenhouse gas emissions from state power sectors, which the agency calls the “Clean Power Plan,” the agency is using a long-ignored statutory provision, Clean Air Act §111(d), to overhaul the nation’s electricity sector. As noted in a previous post, §111(d) has rarely been used, and it is so obscure that when Congress passed the Clean Air Act amendments in 1990, no one even noticed that the House and Senate had passed two different versions. Talk about unheralded.

Second, the Court expressed some skepticism about controlling greenhouse gas emissions through energy efficiency, which is an important part of EPA’s climate agenda. Carbon dioxide, the most common greenhouse gas, is the inevitable result of burning fossil fuels. Clean combustion of clean fossil fuels emits carbon dioxide and water. And once carbon dioxide is emitted, it is hard to pull out of the air. So most attempts to limit carbon dioxide emissions are really attempts to limit fossil fuel combustion. The only other option is carbon capture and storage, which is usually too costly to be feasible. EPA’s Clean Power Plan and its guidance on what is the “best available control technology” under the PSD program both rely on encouraging energy efficiency.

But the Supreme Court was not willing to endorse this approach. First, it stated that it didn’t need to decide whether energy efficiency could be the “best available control technology” because EPA also said states could consider carbon capture and storage. Second, it said that even if EPA could mandate energy efficiency at new sources, it could not redesign the source, require it to consume less electricity, or otherwise micromanage industrial source proposals. In doing so, the Supreme Court handed industry arguments to use against regulators in permit proceedings.

A third important takeaway from the case is that Justice Scalia, the conservative justice that authored the Supreme Court’s opinion, was able to convince Justice Kennedy to join his opinion limiting EPA’s authority to regulate greenhouse gases. Justice Kennedy is generally considered the Court’s swing vote and he was a deciding vote on the Court’s 5-4 decision in Massachusetts v. EPA, which required EPA to consider the climate consequences of greenhouse gases from cars and trucks.

Justice Kennedy has seemed very supportive of EPA’s efforts to regulate greenhouse gases. At oral argument, he admonished industry’s lawyers that he would continue to follow “both the result and the reasoning” of Massachusetts v. EPA—and the reasoning of Massachusetts v. EPA stressed the possible benefits of greenhouse gas regulation. Until now, EPA may have been justified in believing that the Court’s swing justice would sympathize with the challenges they face in adapting the Clean Air Act to address global warming and give them the benefit of the doubt. But today’s decision shows that Justice Kennedy’s sympathy only goes so far: he is quite willing to strike down overly broad climate regulations. That may have much longer-term implications for EPA’s climate agenda—only the coming years will tell.



Full disclosure: Before entering my academic career in 2011, I represented some of the petitioners in their challenge to EPA’s regulations.

Greenpeace v. Canada: Symbolic Blow to the Nuclear Industry, Game-changer for Everyone Else?

  • Please welcome guest blogger, Martin Olszynski,  who is an Assistant Professor at the University of Calgary’s Faculty of Law. Martin has written extensively on environmental assessment, so I am delighted to publish his thoughts on the Federal Court of Canada’s recent decision in Greenpeace v. Canada, which may have important implications for several high-profile energy projects that are currently under consideration in Canada, as well as environmental assessment law in general.

By Martin Olszynski

In a rather lengthy (431 paragraphs) decision, the Federal Court of Canada agreed with Greenpeace and other environmental groups that portions of the Joint Review Panel report for the Darlington New Nuclear project proposed by Ontario Power Generation were inadequate. See Greenpeace Canada v. Canada (Attorney General), 2014 FC 463 (CanLII). Justice Russell held that the environmental assessment conducted by the Panel failed to comply with the Canadian Environmental Assessment Act, SC 1992 c 37 (essentially Canada’s version of the United States National Environmental Policy Act). Specifically and as further discussed below, there were gaps in the treatment of hazardous substances emissions and spent nuclear fuel, and a failure to consider the effects of a severe “common cause” accident such as the combined earthquake and tsunami that caused the Fukushima nuclear disaster.

As noted by the media, while the decision is of limited effect on a project already indefinitely postponed by the provincial government, “it is a symbolic blow to an industry coping with the public and political fallout from Japan’s 2011 Fukushima meltdown.” As further discussed below, the decision is also likely to have implications for environmental assessment in Canada generally and several other projects currently making their way through either the regulatory process or the courts, including three projects in Western Canada that have received international attention: Taseko’s New Prosperity mine, Enbridge’s Northern Gateway pipeline and Kinder Morgan’s Trans Mountain pipeline.


In the fall of 2006 and under direction from the Ontario Minister of Energy, Ontario Power Generation applied to the Canadian Nuclear Safety Commission for a site preparation license for several new reactors at its existing Darlington nuclear plant in Bowmanville, Ontario. Application for this license, as well as for authorizations under the federal Fisheries Act RSC 1985 c F-14 and the Navigable Waters Protection Act RSC 1985 c N-22 (now the Navigation Protection Act), triggered the application of the-then CEAA (since replaced with the Canadian Environmental Assessment Act, 2012 SC 2012 c 19). The project was referred to a “joint review panel” in 2008 and a three-member panel was appointed in 2009. Following 284 information requests and seventeen days of hearings in the spring of 2011, the panel submitted its final report to the federal Minister of the Environment in August of that same year, concluding that the project was not likely to result in significant adverse environmental effects. Greenpeace and the other applicants challenged the adequacy of the environmental assessment and panel report shortly thereafter.

Justice Russell summarized the applicants’ argument as follows:

 [127] As identified in the Report, the [joint review panel] itself found that key information about the proposed Project was absent from the [environmental assessment] documentation. For example, the Panel found that no specific nuclear reactor technology, site design layout, cooling water option, used nuclear fuel storage option, or radioactive waste management option has been selected. Thus, at the present time, federal decision-makers still do not know: (a) the particulars of the specific project to be implemented at the Darlington site; (b) the full range of site-specific or cumulative environmental effects; or (c) whether there are feasible mitigation measures over the project’s full lifecycle. These and other fundamental gaps are attributable to the fact that what the [joint review panel] had before it was not a “project”, but merely a plan for future planning, assessment, and decision-making.

 (See paras 218 – 220 for the full list of alleged gaps and deficiencies)

The reason that so many project components remained unspecified was that Ontario Power Generation, with the panel’s blessing, had prepared its environmental impact statement based on a “plant parameter envelope” or “bounding scenario” approach. As described by Ontario, “this approach involves identifying the salient design elements of the Project and, for each of those elements, applying the ‘limiting value’ (the value with the greatest potential to result in an adverse environmental effect) based on the design options being considered” (at para 5). The respondents argued that such an approach was consistent with the requirement, pursuant to section 11 of the CEAA, to conduct the assessment as early as practicable in the planning process and before irrevocable decisions are made (at para 66), and further that it was supported by the case law (at para 72).


After a thorough review of the statutory regime and associated jurisprudence, Justice Russell concluded that there was nothing that precluded the adoption of the plant parameter envelope approach per se (at para 181). However, he did find inadequacies with the panel’s treatment of three specific issues (at para 228):

  • The failure of the Panel to insist on a bounding scenario analysis for hazardous substance emissions, in particular liquid effluent and stormwater runoff to the surface water environment, and for the sources, types and quantities of non-radioactive wastes to be generated by the project;
  • The Panel’s treatment of the issue of radioactive waste management; and
  • The Panel’s conclusion that an analysis of the effects of a severe common cause accident at the facility was not required at this stage, but should be carried out prior to construction.

In assessing these matters, Justice Russell accepted the applicants’ argument – unchallenged by the respondents and somewhat unique to the Canadian legislation and the broader Parliamentary context within which it operates – that the environmental assessment process under CEAA is fundamentally different from future licensing or regulatory processes (at para 230), and that there is therefore a limit as to the extent to which the consideration of environmental effects and their mitigation can be left to those later processes (Justice Russell described this as a matter of improper delegation):

[232] Under the CEAA, the ultimate decision-maker for projects referred to review panels is the Governor in Council (in practical terms, the federal Cabinet), which decides whether the responsible authorities will be permitted to take steps to enable the project to move forward. Parliament chose to allocate this decision to elected officials who are accountable to Parliament itself and, ultimately, to the electorate…

[235] The most important role for a review panel is to provide an evidentiary basis for decisions that must be taken by Cabinet and responsible authorities. The jurisprudence establishes that gathering, disclosing, and holding hearings to assemble and assess this evidentiary foundation is an independent duty of a review panel, and failure to discharge it undermines the ability of the Cabinet and responsible authorities to discharge their own duties under the Act [citing Pembina Institute for Appropriate Development v Canada (Attorney General), 2008 FC 302 (CanLII) at paras 72 – 74]…

[237] In short, Parliament has designed a decision-making process under the CEAA that is, when it functions properly, both evidence-based and democratically accountable. The CNSC [Canadian Nuclear Safety Commission], in considering future licensing decisions, will be in a fundamentally different position from the Panel that has conducted the EA. The CNSC will be the final authority making the decision, not merely an expert panel. Although the CNSC approaches this role with considerable expertise, it does not have the same democratic legitimacy and responsibility as the federal Cabinet.

(Emphasis added)

With respect to the respondent’s plant perimeter envelope approach, which the panel acknowledged was a departure from typical EA practices, this meant that it “was incumbent on the Panel to ensure the methodology was fully carried out” (at para 247), bearing in mind also the challenges that such an approach poses for public participation: “The less specific the information provided… the more difficult it may be for interested parties to challenge assumptions, test the scientific evidence, identify gaps in the analysis, and ensure their interests are fully considered” (at paras 247, 249).

Applying this standard to hazardous substance emissions and on-sitechemical inventories, Justice Russell concluded that the environmental assessment came up short. He noted Environment Canada’s submissions to the panel that, notwithstanding several information requests to Ontario Power Generation, the remaining gaps prevented the department from assessing effects with respect to effluent and storm water management (at paras 257 – 259). The panel itself noted that “[Ontario Power Generation] did not undertake a detailed assessment of the effects of liquid effluent and storm water runoff to the surface water environment” but that it “committed to managing liquid effluent releases in compliance with applicable regulatory requirements and to applying best management practices for storm water” and on this basis concluded that the project was not likely to result in significant adverse environmental effects (at paras 264, 265).

In a passage that is sure to interest administrative law scholars and practitioners (and discussed further below), Justice Russell held that while such a conclusion may be reasonable, it did not comply with CEAA:

[272] To repeat what is stated above, because of its unique role in the statutory scheme, a review panel is required to do more than consider the evidence and reach a reasonable conclusion. It must provide sufficient analysis and justification to allow the s. 37 decision-makers to do the same, based on a broader range of scientific and public policy considerations. One could say that the element of “justification, transparency and intelligibility within the decision-making process” (Dunsmuir, above, at para 47; Khosa, above, at para 59) takes on a heightened importance in this context.

[273] In this case, there are references to commitments by [Ontario Power Generation] to comply with unspecified legal and regulatory requirements or applicable quality standards, and to apply good management practices. There are references to instruments that may or may not contain relevant standards or thresholds based on the information before the Court (e.g. the Ontario Stormwater Management Planning and Design Manual (March 2003)). And there are references to thresholds or standards in statutory instruments (e.g. Fisheries Act, Canadian Environmental Protection Act) without specific information about how these are relevant to or will bound or control the Project’s effects…

[275] In essence, the Panel takes a short-cut by skipping over the assessment of effects, and proceeding directly to consider mitigation, which relates to their significance or their likelihood. This is contrary to the approach the Panel says it has adopted (see EA Report at p. 39), and makes it questionable whether the Panel has considered the Project’s effects at all in this regard.

(Emphasis added)

This is not to suggest that future regulatory processes “have no role to play in managing and mitigating a project’s environmental effects” (at para 241). For Justice Russell, a conceptual distinction can be made between two kinds of situations where a panel, despite some uncertainty, might conclude that significant adverse environmental effects are unlikely (at para 280):

(a)  Reliance upon an established standard or practice and the likelihood that the relevant regulatory structures will ensure compliance with it; or

(b)  Confidence in the ability of regulatory structures to manage the effects of the Project over time.

The latter approach is problematic in that it “may short-circuit the two-stage process whereby an expert body evaluates the evidence regarding a project’s likely effects, and political decision-makers evaluate whether that level of impact is acceptable in light of policy considerations, including “society’s chosen level of protection against risk” (at para 281, referring to the Government of Canada’s policy on the application of the precautionary principle and adopted by the Darlington panel).

Turning next to the issue of spent nuclear fuel, there does not appear to have been any real dispute between the parties that the panel’s treatment of this issue was cursory. Rather, the respondents’ position was that this was something that Canada had mandated the Nuclear Waste Management Organization (NWMO) to study. Justice Russell disagreed:

[297] In my view, the record confirms that the issue of the long-term management and disposal of the spent nuclear fuel to be generated by the Project has not received adequate consideration. The separate federal approvals process for any potential NWMO facility, which has not yet begun…will presumably ensure that such a facility is not constructed if it does not ensure safety and environmental protection. But a decision about the creation of that waste is an aspect of the Project that should be placed before the s. 37 decision-makers with the benefit of a proper record regarding how it will be managed over the long-term, and what is known and not known in that regard. (emphasis added)

According to Justice Russell, the management and storage of spent nuclear fuel was not a “separate issue” (at para 312). Rather, the environmental assessment “is the only occasion…on which political decision-makers at the federal level will be asked to decide whether that waste should be generated in the first place” (ibid). Nor was there anything in the Terms of Reference that suggested that this issue was not to be addressed (at para 313).

Finally, with respect to a severe “common cause” accident (e.g. as a result of both an earthquake and flooding), the problem was not that Ontario Power Generation failed to assess the risks of accidents associated with its new build (at para 327) but rather that it failed to assess these risks in conjunction with the existing Darlington plant. The panel itself recognized this gap and recommended an evaluation of “the cumulative effect of a common-cause severe accident involving all of the nuclear reactors in the site study area” prior to construction (Recommendation # 63). For Justice Russell, however, that was insufficient:

[334] In my view, the one conclusion that is not supported by the language of the statute is the Panel’s conclusion that the analysis had to be conducted, but could be deferred until later. Rather, in my view, it had to be conducted as part of the EA so that it could be considered by those with political decision-making power in relation to the Project.

In light of these three deficiencies, and as was the case in Pembina Institute cited above, the Court remitted the Report to the panel for further consideration, pending which the relevant government agencies have no jurisdiction to approve the project.


Justice Russell’s thorough treatment of the federal environmental assessment regime means that the decision is likely to have implications going forward. This is so because, notwithstanding the fact that Greenpeace dealt with the prior CEAA regime and CEAA 2012 is in many ways different, the provisions dealing with a panel’s duties and political decision-making are effectively unchanged. These implications, as well as Justice Russell’s somewhat unprecedented but in my view correct approach to judicial review in this context, are further discussed below.

Failure to Assess Environmental Effects “Short-Circuits” the CEAA Regime

Perhaps the most important take-away message from Greenpeace is that, generally speaking, Panels must do the work of actually assessing potential environmental effects and their mitigation. This is a necessary consequence of CEAA’s two-step decision-making process, which Justice Russell describes as “evidence-based and democratically accountable.” Democratic accountability is hindered where the evidence with respect to potential adverse environmental effects is missing, inadequate or postponed to some future regulatory proceeding. This finding, supported by prior jurisprudence and the 2008 Pembina Institute decision in particular, is likely to cause problems for both Taseko’s proposed New Prosperity mine and Enbridge’s proposed Northern Gateway pipeline.

I have previously written about Taseko’s New Prosperity project here and here. Briefly, the second federal panel that reviewed Taseko’s revised project concluded – like the first one – that the project is likely to result in significant adverse environmental effects. In the first of my posts, I suggested that this result was at least partially its own undoing, and its refusal to provide sufficient information to the panel in particular. Like Ontario Power Generation, Taseko was of the view that such “details” could be dealt with at the regulatory phase (see e.g. its final written submissions to the panel at p 8 – 11), an approach that the New Prosperity panel ultimately rejected (see New Prosperity Report at p 22). In its December 2013 application for judicial review, Taseko argues, inter alia, that the panel erred in law when it did so. Greenpeace suggests that this aspect of Taseko’s challenge is unlikely to succeed.

Greenpeace also lends support to the recent letter to the Prime Minister, signed by 300 scientists, which urges him to reject the Northern Gateway Joint Review Panel report. Amongst five major flaws, the signatories to the letter allege inappropriate reliance on yet-to-be-developed mitigation measures:

…Northern Gateway omitted specified mitigation plans for numerous environmental damages or accidents. This omission produced fundamental uncertainties about the environmental impacts of Northern Gateway’s proposal (associated with the behaviour of bitumen in saltwater, adequate dispersion modeling, etc…). The panel recognized these fundamental uncertainties, but sought to remedy them by demanding the future submission of plans… Since these uncertainties are primarily a product of omitted mitigation plans, such plans should have been required and evaluated before the [panel] report was issued.

Whether or not the foregoing is an accurate characterization of the JRP’s conclusions and recommendations (a quick glance of the National Energy Board’s 209 conditions does suggest that these scientists are likely onto something), the letter’s characterization of the environmental assessment process as one intended to offer “guidance, both to concerned Canadians in forming their opinions on the project and to the federal government in its official decision” (at page 3) could have been written by Justice Russell himself.

What is Separate?

A related aspect of Greenpeace worth discussing is the Court’s approach to the management of spent nuclear fuel. As noted above, Justice Russell concluded that this was not a separate issue, and that the “creation of nuclear waste” was “an aspect of the Project that should be placed before [Cabinet]” (at para 297).

Such dicta could prove useful to those, such as the City of Vancouver in the context of the National Energy Board’s Trans Mountain pipeline application, arguing that the environmental assessments for major pipelines (including Northern Gateway) should assess the climate change implications of the increased oil production enabled by the construction of such pipelines (in its application, Trans Mountain states that the pipeline is in response to requests for increased capacity “in support of growing oil production”). Although the matter is not free from doubt, the statutory language on this front certainly is broad (see CEAA, 2012 para 5(1)(a): “…a change that may be caused…”). It also seems plain that panels cannot arbitrarily decide to exclude certain environmental effects, nor is deference to government policy or other initiatives appropriate (e.g. the NWMO in Greenpeace, or the Province of Alberta’s intensity-based regulatory approach to greenhouse gas emissions in Pembina Institute).

Assessing the climate change effects of increased oil production would not amount to “a trial of modern society’s reliance on hydrocarbons,” as the National Energy Board’s outgoing chief recently stated in an interview with the Financial Post, and which he described as a policy question belonging “to the world of policy-making and politics, in which we are not involved at all.” With respect to its obligations under CEAA, 2012 at least (recognizing that the Board is dealing with a dual mandate here, the other coming from s 52 of the National Energy Board Act RSC 1985 c N-7 and which does actually require it to reach a conclusion with respect to the public interest), it would be the exact opposite. Although complex, it would entail an evidence-based analysis of whether Trans Mountain or Northern Gateway may contribute to an increase in oil production and, if so, the greenhouse gas emissions associated with that. Importantly, the final Environmental Impact Statement for Keystone XL determined that this was not likely to be the case for that particular pipeline.

A Green Shade of Reasonableness Review?

In the final part of this post, I want to briefly discuss Justice Russell’s approach to reasonableness review. For convenience, the relevant passage is as follows:

[272] To repeat what is stated above, because of its unique role in the statutory scheme, a review panel is required to do more than consider the evidence and reach a reasonable conclusion. It must provide sufficient analysis and justification to allow the s. 37 decision-makers to do the same, based on a broader range of scientific and public policy considerations. One could say that the element of “justification, transparency and intelligibility within the decision-making process” (Dunsmuir, above, at para 47 [this case sets out a doctrine of deference to administrative agencies somewhat similar to Chevron]; Canada (Citizenship and Immigration) v. Khosa, 2009 SCC 12 (CanLII) at para 59) takes on a heightened importance in this context. (emphasis added)

In my view, this is precisely the kind of analysis that Justice Binnie had in mind when he stated, at para. 59 of Kosa, that “[r]easonableness is a single standard that takes its colour from the context.” In the environmental assessment context, judicial review is not available on the merits of government-decision making – as Justice Russell observed that is a matter of democratic accountability. In this context, judicial review should function in the service of democratic accountability by ensuring the integrity of the decision-making process, a process that government predictably and – where it has been adequate – justifiably relies on to gain support for its political decisions. In the context of Northern Gateway, for example, the Prime Minister and then Minister of Natural Resources Joe Oliver were reported as saying that they will “make a decision only after considering the recommendations of the ‘fact-based’ and ‘scientific’ review panel.” Mr. Oliver also released a statement where he described the panel report as “a rigorous, open and comprehensive science-based assessment.” In this context, the role of a reviewing court should be to ensure that the environmental assessments do in fact meet these standards, failing which there can be no democratic accountability.

An earlier version of this post appeared at ABlawg, the University of Calgary Faculty of Law blog.

FERC’s Demand Response Strategy Hits a Snag: D.C. Circuit Vacates Order 745 in Electric Power Supply Association v. FERC

  • I am delighted to welcome guest blogger Sharon Jacobs. Sharon was my colleague at Harvard Law School and will be an Associate Professor at Colorado Law beginning this summer.  Sharon’s scholarship focuses on administrative, energy and environmental law and she has a forthcoming article on federalism and demand response programs, so she is the perfect person to discuss the D.C. Circuit’s recent decision in Electric Power Supply Association v. FERC, which invalidated a federal order designed to encourage demand response. -James Coleman

By Sharon B. Jacobs

It is a poorly kept secret that D.C. Circuit judges do not exactly clamor to be assigned Federal Energy Regulatory Commission (FERC) cases. The notable exception is now-Senior Judge Stephen Williams, who loves them. His grasp of the intricacies of energy regulation is unparalleled on any court in the country. It is unfortunate, therefore, that Judge Williams was not assigned to the D.C. Circuit panel that recently handed down Electric Power Supply Association v. FERC. In a 2-1 opinion authored by Judge Janice Rogers Brown and joined by Judge Laurence Silberman, the panel vacated FERC’s final rule on compensation for demand response resources in wholesale energy markets. Judge Harry Edwards offered a well-reasoned and ultimately more persuasive dissent.

Demand response is the reduction of electricity use in response to a price signal. In other words, customers are paid not to consume energy. Demand response has been called the sale of “negawatts,” although the phrase is an imperfect description of the actual transaction. Where demand response bids are accepted, market administrators need not purchase as much generation (supply) to meet aggregate demand. Because the cost of electricity goes up as demand increases, especially at times of peak consumption, demand response can lead to significant savings.

Electricity markets are divided into two spheres: retail (sales to end-use customers) and wholesale (sales for resale). For the most part, states regulate the former, while FERC controls the latter. FERC’s demand response strategy affects both markets. In an earlier order, FERC allowed aggregating companies to bid retail customers’ demand response commitments directly into wholesale markets. In the rule challenged in this case, Order 745, FERC sought to further eliminate barriers to demand response participation in wholesale markets by requiring market administrators to pay demand resources the “locational marginal price” or “LMP” for each megawatt not consumed. The locational marginal price is the same price that generators receive when they bid their megawatts of power into wholesale markets. It reflects the value of energy at a specific location at the time of delivery. PJM, the market administrator for the mid-Atlantic region, explains that the LMP fluctuates like taxi fares—lighter electricity traffic yields a lower, steadier fare, whereas congestion on the wires causes the fare to rise. FERC included a caveat in its rule: demand response resources would only receive the LMP when their participation in wholesale markets would be cost effective, as determined by a specified “net benefits” test.

The bulk of the opinion concerned a threshold question: whether FERC acted within the scope of its jurisdiction under the Federal Power Act when it established compensation and other rules for retail demand response resources participating in wholesale markets. Under the Act, FERC has clear jurisdiction over rates for wholesale sales of electric energy in interstate commerce as well as rules, regulations and practices affecting those rates. FERC argued that it could set wholesale rates and other rules for demand response in wholesale markets because they were practices “directly affecting” wholesale sales. The panel majority disagreed, instead characterizing what FERC did as indirect regulation of the retail market for electricity.

There were three major problems with the opinion.  First, the majority found the Federal Power Act’s jurisdictional provisions much clearer than they are in fact.  It applied the normally deferential Chevron test, under which the court will defer to the agency’s reasonable interpretation of an ambiguous statutory provision it is authorized to administer, to FERC’s jurisdictional claims. Though some have argued that allowing the agency to determine the scope of its own jurisdiction when statutory language is ambiguous is analogous to permitting the fox to guard the henhouse, the Supreme Court recently affirmed the propriety of this practice in City of Arlington v. FCC. The Federal Power Act’s grants of jurisdiction did not anticipate demand response and the statute’s application to the phenomenon, as the dissent recognized, is unclear. In other words, the statutory provisions at issue, as applied to demand response, are ambiguous. Thus, the court should have deferred to FERC’s reasonable interpretation of those provisions at Chevron step two.

Second, Judge Brown found that the Federal Power Act foreclosed FERC’s reading because the Commission’s interpretation “has no limiting principle.” In an argument reminiscent of Justice Scalia’s warning in his Massachusetts v. EPA dissent that Frisbees and flatulence could be regulated under EPA’s capacious definition of “air pollutant,” Judge Brown warned that FERC’s interpretation of its “affecting” jurisdiction would authorize it to regulate “steel, fuel, and labor markets.” As the dissent pointed out, however, the limiting principle could not be clearer. Under the D.C. Circuit’s own holding in CAISO v. FERC, FERC may only regulate practices that “directly affect” wholesale rates or are “closely related” to those rates, “not all those remote things beyond the rate structure that might in some sense indirectly or ultimately do so.” As Judge Edwards pointed out in his dissent, this language clearly precludes regulation of “steel, fuel, and labor markets.”

Third, to the extent that the true motivation for the decision was general unease about federal encroachment on traditional areas of state regulatory power, the decision overlooked a key aspect of FERC’s demand response rules that mitigate any unwanted impact on state authority. An earlier FERC order, Order 719, offered state and local regulatory authorities an “opt-out”: those who did not want their retail customers participating in wholesale markets for demand response could prohibit them from doing so via legislation or regulation. Order 745’s pricing scheme was layered on top of this jurisdictional compromise. In Judge Edwards’s words, “[t]his is hardly the stuff of grand agency overreach.”

The most controversial part of Order 745, and the real reason the rule was the subject of such concerted opposition, got the least airtime in the opinion. In what was billed as an alternate holding in Part IV (but felt more like dicta), the panel found that Order 745’s locational marginal pricing scheme was arbitrary and capricious. In under two pages of text, the opinion declined to “delve now into the dispute among experts” yet asserted that the Commission had not “adequately explained how their system results in just compensation.” “If FERC thinks its jurisdictional struggles are its only concern with Order 745,” the opinion cautioned, “it is mistaken.” In a much more nuanced discussion of the Commission’s choice and the deference due to FERC “in light of the highly technical regulatory landscape that is its purview,” Judge Edwards concluded that the Commission provided a “thorough explanation” for selecting the locational marginal price as the appropriate level of compensation. In a nutshell, FERC’s argument was that the compensation level was necessary to overcome barriers to participation by demand response resources in wholesale markets and that it accurately reflected the value demand response provided to those markets.

Prior to this ruling, FERC had been successfully pursuing a policy of what I call, in a forthcoming article, “bypassing federalism”:  working a de facto rather than a de jure reallocation of regulatory power by extending its influence through the expansion of wholesale markets. In the context of demand response, that strategy was undermined by the Commission’s aggressive posture on pricing in Order 745. It was the idea that demand response resources would be paid the LMP for their “negawatts,” thereby competing directly with generation in wholesale markets, that triggered the groundswell of opposition from generation resources. The decision will not go into effect until seven days after the disposition of any motion for rehearing, and FERC is still considering its options as well as the decision’s impact on its rules and related programs. The panel’s decision may yet be reversed by the D.C. Circuit en banc or by the Supreme Court. But, as a policy matter, the Commission might have avoided a direct confrontation over its demand response rules by moving more deliberately on the pricing question.  As I have written elsewhere, for agencies whose regulatory schemes face concerted opposition, discretion is sometimes the better part of valor.

State Energy Policy and the Commerce Clause: Spotlight on Colorado and Minnesota

By Alexandra B. Klass
University of Minnesota Law School

Within the past month, two federal district courts—one in Colorado and one in Minnesota—have issued important decisions on the constitutionality of state clean energy policies. Both cases raised the same legal issue, namely, whether the state laws in question regulate extraterritorially in violation of the dormant Commerce Clause of the U.S. Constitution. But the courts reached different results in each case and, more importantly, the Minnesota and Colorado policies reviewed by each court were quite different from each other even though both involved efforts to promote clean energy within the state. Some of the recent commentary on the two cases has downplayed the significant differences between the two state policies in question, leading to confusion about the implications of the courts’ rulings.

First, a bit about the dormant Commerce Clause. The Commerce Clause of the U.S. Constitution grants Congress the authority to regulate interstate commerce. But the Supreme Court has also interpreted that provision to contains a “dormant” aspect that limits states from interfering with the free flow of commerce among the several states. A law can violate the dormant Commerce Clause if: (1) it facially discriminates, has a discriminatory purpose, or is discriminatory in effect; (2) the law is facially neutral and there is no evidence of discriminatory intent or effect but the burdens of the law on interstate commerce outweigh the in-state benefits; or (3) the law attempts to control conduct completely outside its borders and thus regulates “extraterritorially.” The dormant Commerce Clause has been applied to state laws for over 100 years, including laws banning or limiting out-of-state imports of goods or services, out-of-state exports of goods or services, minimum pricing laws tied to prices in other states, laws attempting to regulate trucks and trains in interstate transportation, and a variety of other state laws intended to promote in-state businesses as well as environmental, health and safety interests over similar out-of-state interests.

Now, onto the recent state energy policy cases. Both cases involve efforts by states to encourage the use of renewable electricity resources in the state and limit the generation of electricity that emits significant amounts of CO2 in an effort to address climate change. But the two state policies under constitutional challenge involve very different ways of reaching that goal. The Colorado lawsuit involves a challenge to a state renewable energy mandate. Such laws, known as renewable portfolio standards, renewable energy standards, clean energy mandates, or renewable energy mandates, have been adopted in over half the states. Such laws require utilities and other electricity providers in the state to generate or purchase a certain percentage of their electricity for retail sale from renewable energy sources by a particular date, often 15%, 20%, or 30% by 2020 or 2025, with lower amounts mandated between now and the targeted date. Such laws encourage the use of wind, solar, geothermal, or hydropower energy with significant variation among the states with regard to what resources “count” and the percentages required.

By contrast, the Minnesota lawsuit does not involve a challenge to the state’s renewable energy mandate, even though Minnesota has one of the most aggressive renewable energy mandates in the nation. Instead, the lawsuit involves a challenge to another Minnesota energy policy that limits the construction, use, or import of new coal-fired power in the state by prohibiting the construction of facilities that emit a certain amount of CO2 each year or imports from such facilities. Only a few states (New York, Oregon, California) in addition to Minnesota have such limits on coal-fired power. All of those states impose those limits on coal-fired electricity in addition to imposing a renewable energy mandate on electricity providers in the state.

The Colorado case

In Energy and Environmental Legal Institute v. Epel, __ F. Supp. 2d __, 2014 WL 1874977 (D. Colo., May 9, 2014), a non-profit organization representing and promoting coal energy interests along with one of its members challenged the state’s renewable energy standard, which requires Colorado electric utilities and other retail electricity providers in the state to provide up to 30% of their retail electricity sales from renewable energy sources by a certain date. Electricity providers can meet their renewable energy mandate by either generating or buying renewable power directly or by purchasing renewable energy credits. The plaintiffs argued on summary judgment that the renewables mandate places a restriction on how out-of-state goods are manufactured, and requires out-of-state electricity to be generated according to Colorado’s terms. Thus, according to the plaintiffs, by projecting Colorado’s policy decisions onto other states, the law regulates extraterritorially in violation of the dormant Commerce Clause.

The court rejected this argument and held that the law did not impact wholly out-of-state transactions. If a Wyoming coal company generates electricity and sells it to a South Dakota business, the Colorado law does not apply. Instead, the court found that the law applies only to energy generators that choose to do business with a Colorado utility and, even then, the law only applies in determining whether the energy the Colorado utility purchases counts towards its renewables mandate. The court agreed that the Colorado law would influence the profits of out-of-state companies whose electricity could not be used to fulfill the mandate, but held that the dormant Commerce Clause “neither protects the profits of any particular business, nor the right to do business in any particular manner.” The court also found that the law did not discriminate against interstate commerce or unduly burden interstate commerce.

The Minnesota case

In North Dakota v. Heydinger, __ F. Supp. 2d __, 2014 WL 1612331 (D. Minn., Apr. 18, 2014), the State of North Dakota, North Dakota lignite coal interests, and multi-state electric cooperatives in the upper Midwest sued the State of Minnesota over limits on coal-fired power in its Next Generation Energy Act. The provisions of the law at issue state that after a certain date, no person shall, without CO2 offsets: (1) construct a new “large energy facility” (defined to encompass coal-fired power plants but not most natural gas-fired plants) in the state; (2) import power from a new large energy facility from outside the state; or (3) enter into a long-term power purchase agreement that would contribute to statewide power sector CO2 emissions. The plaintiffs argued on summary judgment that the limits on imports of coal-fired power from outside the state regulated extraterritorially in violation of the dormant Commerce Clause and discriminated against interstate commerce. Notably, even though Minnesota has a renewable energy mandate that is also part of the state’s Next Generation Energy Act, the plaintiffs in the Minnesota case did not challenge Minnesota’s renewable energy mandate at all. As a result, the state energy policy at issue in the Minnesota case is quite different from the state energy policy at issue in the Colorado case, even though both state policies are intended address climate change by imposing requirements on state electricity providers.

In an April 2014 decision, the U.S. District Court for the District of Minnesota agreed with the plaintiffs that the limits on coal-fired electricity imports regulated extraterritorially. Because the court struck down the import limits on those grounds, it did not reach the claims that the law also discriminated against interstate commerce. In reaching its decision, the court adopted an extremely broad interpretation of the law, finding that it applied to any electric power provider selling electricity on the multi-state, regional electric grid (encompassing more than 10 states), rather than applying only to persons located in or operating in Minnesota. The court pointed to statements made by the Minnesota Department of Commerce in earlier regulatory proceedings that indicated the agency might apply the law to multi-state electric cooperatives based outside the state but with members in Minnesota if the cooperative generated coal-fired power outside the state and sold it into the multi-state grid. Because electrons cannot be tracked once they have entered the electric grid, the court found such a transaction could apply where the buyers and seller were all outside of Minnesota because some of the electricity might enter the state of Minnesota. Because such an application of the law would apply even when no party to the transaction was based in Minnesota, the court found that the law regulated extraterritorially in violation of the dormant Commerce Clause. The court rejected the argument that it should not interpret the law so broadly to encompass all sales of electricity into the multi-state grid even though the state had never actually applied the law to these types of out-of-state transactions that did not directly involve a Minnesota-based actor intending to import coal-fired power to the state.


So what should we take away from these two decisions? First, it is important to keep in mind what was not at issue in either case. For some time now, there has been concern among policymakers and scholars regarding state renewable energy mandates that preference in-state renewable resources over out-of-state renewable resources through multipliers and other provisions that encourage the use of in-state wind, solar, or hydropower. Many state laws contain such a preference for in-state renewable resources because such preferences allowed legislators to argue that a renewable energy mandate would not only promote the use of clean energy but would also help promote new, in-state industries. While this is certainly good politics and may be good policy, such preferences raise dormant Commerce Clause concerns because they expressly benefit in-state industries over identical out-of-state industries. But the Colorado renewable energy mandate at issue does not contain such preferences and thus treats in-state and out-of-state renewable and non-renewable electricity resources alike. Likewise, even though the Minnesota renewable energy mandate was not even at issue in the Minnesota litigation, it is important to point out that Minnesota, like Colorado, does not preference in-state renewable resources over out-of-state renewable resources.

Second, states attempt to meet clean energy and climate change goals through a variety of policies. States have significant authority to regulate electricity sales, transportation, and industrial facilities and in recent years have used that authority to enact renewable energy mandates, place bans on coal-fired power, and impose other regulatory requirements on industrial facilities, fuel providers, electricity providers, and other businesses that contribute to CO2 emissions. Each type of policy has a different impact on in-state businesses and out-of-state parties that do business in the state. As a result, each type of policy raises different legal issues. Thus, the fact that the courts in the Colorado and Minnesota cases reached different results is significant, but it is also important not to lose sight of the fact that each court reviewed state energy policies that have similar goals, but were designed in completely different ways and have very different impacts on in-state and out-of-state actors.

Last, each court’s decision relied in large part on how broadly it found the state law to apply. In the Colorado case, the court stated that the law applied only to Colorado electricity providers and thus did not impact electricity generators in other states except when they chose to do business with electricity providers in Colorado. By contrast, in the Minnesota case the court interpreted the law limiting the use of new-coal fired power to apply to any party selling electricity into the multi-state electricity grid if there was some chance that those electrons could flow into Minnesota. Whether the language of the statute supports such a broad interpretation of the law remains to be seen and will likely be an issue on appeal. The fact remains, however, that how broadly courts interpret the reach of state energy policies will impact significantly whether those laws can withstand dormant Commerce Clause scrutiny.

For more information on the dormant Commerce Clause, its potential application to state energy policy, and recent litigation, see Alexandra B. Klass & Elizabeth Henley, Energy Policy, Extraterritoriality, and the Dormant Commerce Clause, San. Diego J. of Climate & Energy L. (forthcoming 2014), at

Federal Court Strikes Down Minnesota’s Limits on Coal Power Imports: A Critical Moment for State Regulation of Imported Fuel & Electricity

State of North Dakota, et al., v. Beverly Heydinger, et al., Case No. 11-cv-3232, (D. Minn., Apr. 18, 2014).

On April 18, the U.S. District Court for the District of Minnesota struck down the State of Minnesota’s restrictions on importing electricity from coal power plants in other states. The court held that these restrictions improperly regulated electric generators and utilities outside the state. The decision sets a precedent that could threaten state regulations of imported fuel and electricity, such as the numerous renewable power standards and California’s low carbon fuel standard. These regulations have been a flashpoint for conflicts between in-state and out-of-state interests, including Canadian energy producers who believe that the standards discriminate against them.

Minnesota adopted the restriction on electricity imports in its 2007 Next Generation Energy Act, which placed a moratorium on construction of new coal power plants within the state. The point of the moratorium was to limit greenhouse gas emissions from coal burning, which contributes to climate change. Without the import restriction, Minnesota’s moratorium might have little effect: companies looking to build a new coal plant could simply build in neighboring states, exporting electricity to Minnesota and increasing greenhouse gas emissions. So Minnesota declared that “no person shall . . . import or commit to import from outside the state power from” new coal plants or “enter into a new long-term power purchase agreement that would increase statewide power sector carbon dioxide emissions.” Minn. Stat. § 216H.03, subd. 3. New coal plants could only avoid this ban if they paid to reduce emissions elsewhere or qualified for an exception.

North Dakota and utilities with coal power plants brought a lawsuit alleging that Minnesota’s restrictions unconstitutionally regulated outside of Minnesota’s territory, and the court agreed. The U.S. Constitution’s Commerce Clause gives the federal government the authority to regulate interstate commerce and implies that states cannot “discriminate against or unduly burden interstate commerce” without congressional authorization. This rule is called the “dormant commerce clause” because it applies when congress has not authorized state regulation. One aspect of this rule is that states cannot adopt a regulation that “has the practical effect of controlling conduct beyond the boundaries of the state.”

The court held that the import restriction necessarily regulated out-of-state conduct because electricity on the grid “does not recognize state boundaries.” Electricity is not like a package that is shipped from a seller to a buyer. Instead, the interstate electric grid creates a pool of power. Electric generators contribute electricity and consumers withdraw electricity. It is as though one group was emptying buckets of water into a lake and another group was filling buckets of water from a lake. Companies may talk about purchasing electricity “from” a specific utility, but that is an accounting convention, not a description of a physical process—the electricity purchased comes from an undifferentiated pool. Thus, when a North Dakota utility sells to a North Dakota customer some of the electricity might be diverted into Minnesota, violating Minnesota’s import restriction. So Minnesota’s law regulates out-of-state conduct, and the court held that it violated the U.S. Constitution and enjoined any enforcement.

The decision raises two potential problems for state regulation of imported electricity and fuel. First, more than half of the fifty states have renewable power standards that apply to imported electricity. Under the court’s decision these standards would be invalid unless they exempted incidental imports from out-of-state utilities serving out-of-state customers. The Harvard Environmental Law Program’s Policy Initiative’s Energy Fellow Ari Peskoe has suggested some ways that states could try to insulate their regulations from a similar challenge.

Second, the court suggested that there may be strict limits on a state’s ability to regulate imported fuel and electricity through renewable portfolio standards or low carbon fuel standards. The usual rule under the dormant commerce clause is that states “may not attach restrictions to exports or imports to control commerce in other states” or otherwise “project” their regulation into other states. But the entire point of state restrictions on imported fuel and electricity is to affect out-of-state greenhouse emissions. States want to regulate imported fuel and electricity because they are concerned that out-of-state energy producers are contributing to climate change—they don’t want to import oil from places where it takes a lot of greenhouse gas emissions to produce oil and they don’t want to import electricity from states that are producing it using a lot of greenhouse gas emissions. And that concern makes sense: even if those greenhouse gas emissions take place in other states or countries, they’re just as bad for the entire world’s climate. As a result, the U.S. Court of Appeals for the Ninth Circuit recently suggested that the dormant commerce clause’s prohibition on extraterritorial regulation is only meant for extraterritorial price-regulation, so it doesn’t threaten California’s low carbon fuel standard or, presumably, state renewable power standards.

The Minnesota court, however, rejected the Ninth Circuit’s reasoning, noting that the Supreme Court and several appellate courts have held that states may not project their regulation into neighboring states, even when the regulation was not about prices. This conflicting reasoning comes at an important moment for state regulation of imported fuel and electricity. There is still no legal consensus on the validity of these regulations, which are being challenged in several lawsuits around the country., a website created by the Harvard Environmental Law Program’s Policy Initiative, is tracking these lawsuits.

Second, there is no consensus on whether these state import restrictions are a wise way to make climate policy. Although states have good reason to be concerned about the fossil-fuel industry in their trading partners, other states and countries worry that these import regulations are aimed at burdening out-of-state industry. Canada doesn’t think California should tell it how to produce oil, and is concerned that California’s regulation has been rigged to harm it. Quebec believes that state renewable portfolio standards discriminate by refusing to credit its hydropower exports as renewable. And states like North Dakota have the same concerns about Minnesota’s regulation. These conflicting interests may create conflicting regulations and state-to-state trade wars that would splinter interstate energy markets. In a forthcoming article in Fordham Law Review, titled “Importing Energy, Exporting Regulation,” I argue that the federal government should address this problem by supervising state regulation of imported energy, exempting non-discriminatory regulations from dormant commerce clause review.

No one yet knows how this legal and policy debate will be resolved. The Minnesota decision frames the legal debate through its searching dormant commerce clause review and clarifies the stakes by striking down a closely watched state electricity regulation. The one certainty is that the debate will continue.

Oral Argument Hints that Supreme Court May Trim Back U.S. Industrial Source Greenhouse Gas Regulations

Today the Supreme Court heard oral argument in Utility Air Regulatory Group v. Environmental Protection Agency (EPA), in which petitioners challenged the EPA’s “Prevention of Significant Deterioration” (“PSD”) regulations for stationary industrial sources of greenhouse gases. These regulations, finalized in 2010, require sources that emit over 100,000 tons of greenhouse gases to obtain a PSD permit and adopt the “best available control technology” for every pollutant that they emit, including greenhouse gases.

The oral argument provided few surprises: it was as complex as expected in this case of arcane statutory interpretation. As described below, the argument did hint, however, that the Supreme Court might adopt a compromise position, holding that 1) industrial sources cannot be required to obtain a PSD permit purely on the basis of their greenhouse gas emissions, but 2) can be required to adopt best available control technology for their greenhouse gas emissions if they need a PSD permit anyway due to their emissions of other pollutants. This ruling would likely have little impact on EPA’s broader agenda on greenhouse gas emissions.

The argument follows from the Supreme Court’s landmark decision in Massachusetts v. EPA, a 2007 case in which the Supreme Court held that greenhouse gases were a pollutant under the general terminology of the U.S. Clean Air Act. Relying on this decision, EPA has adopted greenhouse gas standards for new cars and trucks. It has also proposed greenhouse gas standards for new coal and natural gas power plants. And it is due to propose standards for existing coal and gas plants at some point this summer. This case, however, concerns a separate set of standards, adopted under the Clean Air Act’s catch-all for industrial sources, the Prevention of Significant Deterioration requirement that requires state and local permitting agencies to ensure that new major sources adopt the “Best Available Control Technology.” Petitioners today made clear that they are not challenging EPA’s rules for cars, or its proposed rules for individual source categories such as power plants. Instead, they challenge only EPA’s PSD catch-all.

EPA’s argument is simple: the Clean Air Act requires PSD regulation for sources of “any air pollutant” and Massachusetts v. EPA said that a greenhouse gas is a pollutant. Furthermore, the Clean Air Act language for PSD is the same as the language EPA used for the car rules and the power plant rules that petitioners are not disputing.

But there’s a catch. The Clean Air Act requires a PSD permit from any new source that emits over 250 tons of “any air pollutant.” That threshold makes sense for pollutants like lead and sulfur dioxide, but far too many sources emit that level of greenhouse gases, so EPA raised the level to 100,000 tons to avoid regulating hundreds of thousands of sources, which EPA acknowledges would be absurd.

Petitioners’ argue that, rather that re-write the statutory thresholds, EPA should not have included greenhouse gases in its PSD program. They say “any air pollutant” can mean different things in different parts of the act. It may be hard to imagine that Congress used the same word to mean different things in different places, but it’s even harder to imagine that Congress used the word “250” to mean “100,000.”

That left three arguments at play in today’s arguments:

1) The government defended its entire regulation: sources that emit over 100,000 tons of greenhouse gases need a PSD permit, and a PSD permit requires the “best available control technology” for greenhouse gases.

2) Industry argued the opposite: emitting greenhouse gases cannot trigger a need for a PSD permit, and even if an industrial source needs a PSD permit because it emits other pollutants, it should not have to adopt “best available control technology” for greenhouse gases. That is, greenhouse gases are not included in the PSD program at all.

3) The Justices spent most of their time pressing both sides why they should not adopt some version of a compromise suggested by one petitioner and a dissenting circuit court judge: emitting greenhouse gases cannot trigger a PSD permit, but if a source needs a permit, it must adopt the best available control technology for greenhouse gases.

This compromise would not rely on altering the 250-ton threshold set by the statute. Justice Kennedy, the swing-vote, noted that the government had not cited any case that would allow that type of statutory re-write. At the same time, the compromise would force the biggest industrial facilities, which need a PSD permit anyway, to adopt best available control technology for greenhouse gases. Professor Jody Freeman, President Obama’s Counselor for Energy and Climate Change, recently suggested that perhaps EPA should have adopted this approach from the beginning to avoid the risk of Supreme Court reversal.

Both petitioners and the government tried to suggest that this fallback was inadequate. This was difficult for the petitioners, given that one of the petitioners had proposed that fallback. And Justice Kennedy, the presumed swing vote, emphasized that he was looking for an argument that followed “both the result and the reasoning” of Massachusetts v. EPA, which stressed the possible benefits of greenhouse gas regulation.

But the government also had a difficult time explaining why it could not accept the proposed compromise. Nearly all sources that emit 100,000 tons of greenhouse gases emit over 250 tons of some other pollutant, so they would require a PSD permit in any case. (And, under the compromise, this would mean they must adopt best available control technology for greenhouse gases.) The only sources that would be excluded, under the compromise, would be the few sources that emit threshold levels of greenhouse gases, but not any other pollutant. EPA estimated that its regulation would cover 86% of greenhouse gases emitted by facilities over the statutory threshold, whereas the compromise would cover 83%. Justices Ginsburg, Roberts, Breyer, and Sotomayor all mentioned this distinction, suggesting that there was very little difference between the government’s position and the proposed compromise.

On the other hand, Chief Justice Roberts, another potential swing-vote, noted that this compromise might require two definitions of “pollutant” within the statutory section on PSD: one definition for the kind of pollutant that triggers the need for a permit, and another definition for the kind of pollutant that must be controlled with the best available technology. Even if it is okay to have one definition for cars and another for PSD, it is somewhat troubling to have inconsistent definitions within the PSD program itself.

The government added a final wrinkle to the compromise suggestions. Justice Sotomayor, who seemed friendly to the government, asked the government if it must lose, how it would like to lose. (See pages 67-72 of the oral argument transcript.) In answer, Solicitor General Donald Verrilli, suggested that “pollutant” should still include all greenhouse gases except carbon dioxide, which is the most common greenhouse gas, and the reason that EPA changed the threshold. This is a particularly complex suggestion, and has already earned a critique from RFF’s Nathan Richardson.

In sum, oral argument suggests that there is some appetite for a compromise among the Supreme Court’s swing votes, and even among some of the government’s supporters. But, as usual, there are too many factors at play for a firm prediction.


Two disclaimers:

1) Before entering my academic career in 2011, I represented some of the petitioners in their challenge to EPA’s regulations.

2) I have omitted some details of the regulations and the petitioners’ arguments to avoid belaboring an already complex argument.




Eighty percent of success is showing up: Or “How a pro se farmer won a default against the United States in his suit to invalidate the permit for half of Keystone XL (& why it probably won’t last)”

On April 25, Michael Bishop, a farmer acting pro se, filed a lawsuit in the U.S. District Court for the Eastern District of Texas to revoke TransCanada’s permit to construct the southern half of the Keystone XL project.  This part of the project, known as the “Gulf Coast Project” or “Phase III”, travels from Cushing, Oklahoma to the Gulf Coast.  Bishop sued the Army Corps of Engineers and its Commanding General, Thomas Bostick, because the Army Corps issued the permit to TransCanada.  The complaint that Bishop filed asked the court to order the Army Corps to revoke Keystone’s permit. Bishop then served this complaint on the Army Corps of Engineers, its officers, and the Attorney General of the United States.

Now, you might not like the chances of a pro se farmer aligned against the U.S. Attorney General, the Army Corps of Engineers, and TransCanada.  But as Sheriff Bell would say: “even in the contest between man and steer the issue is not certain.”  And, as it turns out, no one showed up to contest the lawsuit.  Even though the permit at issue belonged to TransCanada, it is not a defendant.  It was up to the government, and the government did not show up.  As a result, on Wednesday, the clerk entered a default against the Army Corps and its officers.
Mr. Bishop had won, and national news stories trumpeted his victory–e.g. Bloomberg “Texas Farmer Wins Entry of Default in Keystone Lawsuit“.  He told Bloomberg, “Tomorrow I’m going to ask the judge for everything I had in my original petition. I’m going to ask him to revoke the permit and effectively shut this pipeline down until they comply with the law.”
The victory will likely prove short-lived, however. On Thursday, the U.S. Attorney’s office for the Eastern District of Texas filed an emergency motion to vacate the clerk’s entry of default.  Although acknowledging that the AG, Army Corps, and officers had been served, the government pointed out that the U.S. Attorney’s office had not been served, a requirement under Federal Rule of Civil Procedure 4(i).  As a result, the government also suggested that the complaint itself should be dismissed “due to failure of service.”
In the end, it seems unlikely that a lawsuit of this importance will end in a default.  But it’s an important reminder of three things: 1) the variety of legal venues and strategies available to environmental plaintiffs looking to slow the flow of oil, 2) the difficulty of keeping track of the myriad resulting lawsuits, and 3) the importance of showing up.

Cross posted on ABlawg: The University of Calgary Faculty of Law Blog.