The Yale Law Journal


Preventing Policy Default: Fallbacks and Fail-safes in the Modern Administrative State

21 Feb 2012


In Prods and Pleas,1 Benjamin Ewing and Douglas Kysar make a valuable point about an often overlooked benefit of our system of federalism and separation of powers. As they explain, the system’s redundancy allows it to function even if one or more key component is dysfunctional and may even help stimulate dysfunctional components back into action. The title of their article is a reference to this possible stimulus effect.

Today, the most dysfunctional branch of government seems to be Congress.2 Although we may not approve of all the actions of other branches, these branches do seem to be able to make policy decisions and conduct normal business, both of which seem increasingly challenging for Congress. This congressional dysfunctionality is a major problem, because the normal route for national policymaking runs from Congress in setting policy to the President in agency implementation to judicial oversight and enforcement. When that route is blocked, however, federalism and separation of powers provide a plethora of byways and detours that can still be used to make progress. Ewing and Kysar provide a “field guide” to these paths for circumventing congressional inaction.3 In so doing, Ewing and Kysar add a valuable layer of nuance to the usual pictures of government based on hierarchy or separate governmental spheres.

Where they go wrong, in my view, is in overemphasizing the role of the common law as a backup system. We live in an age of statutes and administrative regulation.4 When Congress fails to respond to a national problem, courts play their most important role as interpreters of statutes, overseers of agencies, and on occasion as constitutional arbiters. Ewing and Kysar use common law litigation over climate change as the basis of their case study. In terms of climate change, however, by far the most important institutional backups have involved the Supreme Court acting in a non-common law capacity, combined with federal agency action and extensive legislation by states.

This does not mean that the common law is completely irrelevant. In a previous article, I argued that tort law can contribute to the legal system’s response to catastrophic harms such as climate change.5 In particular, it can improve incentives for disaster mitigation and risk spreading.6 I am in sympathy with Ewing and Kysar’s criticism of American Electric Power Co. v. Connecticut (AEP).7 I share their view that the Supreme Court would have done better to determine whether major carbon emitters can be liable under the federal common law of nuisance, rather than holding that the federal common law was displaced by the Clean Air Act.8 Their argument in favor of this position is richly textured and compelling. But the emphasis they place on this example of a “prod and plea” seems disproportionate to the modest potential of the common law in addressing climate change. The common law is simply not where the action is in today’s world. As Ewing and Kysar themselves note, public nuisance claims have failed to get traction, even in less contentious areas.9 Indeed, Ewing and Kysar do not provide convincing examples of tort law prompting major legislative policy shifts. The only examples that I can think of involve the creation of alternative compensation schemes to replace or complement tort law, such as the September 11th Victim Compensation Fund, rather than changes in regulatory policy or the creation of new regulatory schemes in response to common law rulings.10 While the common law can play a supporting role in regulating major social risks, it is unlikely to play the leading role.

Given this context, it seems unlikely that a contrary result in AEP would have led Congress to tackle the issue of climate change, except perhaps by overturning the decision. So far, nothing has managed to break through congressional inertia, and the political situation seems unfavorable for federal climate legislation. But if court decisions, agency initiatives, and state legislation lack the ability to unfreeze a deadlocked Congress, they can at least help fill the gap left by congressional inaction.

My response to Prods and Pleas proceeds in two parts. Part I shows that the federal courts have indeed provided important “prods” in terms of climate policy, although not in common law cases. In particular, the Court’s decision in Massachusetts v. EPA11 illustrates how public law litigation can act very effectively as a prod to executive action and a fail-safe against congressional default. Using California as an example, Part II demonstrates that state legislatures have also been prominent in filling the policy gap left by Congress.12 States like California have begun to step into the breach in the face of congressional default by creating innovative regulatory techniques to address climate change. So far, however, these state activities have not prompted congressional action.

It seems more useful to consider these actions to be stopgaps and fail-safes that operate when the main mechanism for policymaking has defaulted, rather than prods for congressional action. As my terminology indicates, agency, judicial, and state activity are second-best approaches that will ultimately have limited benefits unless Congress acts. If Congress does act, these initial actions could help pave the way by providing important information about regulatory tools and their effectiveness, as well as by nudging Congress toward action, as Ewing and Kysar suggest. But actions by courts, the Environmental Protection Agency (EPA), and states have very real policy impact. These efforts should lead to some carbon reductions and communicate to the international community that the United States is not entirely out of the picture in terms of climate policy. Thus, these actions act as fail-safes to provide at least limited remedies to pressing problems in the absence of action by the governmental units that would normally be expected to take the lead.

In short, Ewing and Kysar are right that courts should decide the merits of climate nuisance claims rather than reject the claims as nonjusticiable or precluded by statute.13 They are also importantly right about the significant judicial role when Congress has failed to set national policy on pressing issues such as climate change. But they are less persuasive in their effort to cast the common law as an important potential player in preventing policy defaults. In an era in which public policy is predominantly embodied in statutes and agency regulation, the common law necessarily has a secondary role. This does not mean that the common law is irrelevant, but it does mean that the ability of courts to issue prods and pleas is often more effectively deployed elsewhere. In the age of statutes, the crucial role that courts play in the policy arena is through public law, not the private law of torts.

I. Courts as Policy Instigators

The United States Supreme Court decided its first case about climate change in 2007, in Massachusetts v. EPA.14 The Court’s opinion was not just a “plea” but a high-voltage “prod” to the executive branch. It is a splendid example of Ewing and Kysar’s general theory, but not in the common law context on which they focus almost exclusively.

Some background is necessary to understand the case.15 The federal air pollution law, the Clean Air Act, requires the government to set limits on any air pollutant from cars that may endanger human health or welfare.16 The EPA is the federal agency responsible for enforcing federal environmental laws, including the Clean Air Act. The Bush Administration, however, argued that the federal government had no authority to regulate greenhouse gases under the statute, because greenhouse gases were not “pollutants” within the meaning of this statute. Even if it did have statutory authority, the Administration said, exercising such authority might undermine efforts to negotiate greenhouse gas reduction internationally with countries such as China. The Administration also argued that the regulations might conflict with federal rules about fuel efficiency for cars. In response to the Administration’s position, the Commonwealth of Massachusetts and several other plaintiffs sued to force the government to regulate car emissions.

The Court first had to determine whether the plaintiffs had standing to bring the case. In its discussion of standing, the Court said that “[t]he harms associated with climate change are serious and well recognized.”17 As to the second requirement for standing (causation), the EPA did “not dispute the existence of a causal connection between man made greenhouse gas emissions and global warming.”18 The EPA argued that the specific EPA regulation that the plaintiffs sought would not by itself have a significant impact on global warming. But the Court rejected the “erroneous assumption that a small incremental step, because it is incremental, can never be attacked in a federal judicial forum.”19 Finally, the Court concluded that a judicial remedy would be meaningful even though the amount of emissions involved was small compared to total global emissions: “While it may be true that regulating motor vehicle emissions will not by itself reverse global warming, it by no means follows that we lack jurisdiction to decide whether EPA has a duty to take steps to slow or reduce it.”20

By finding standing, the Court played the role that Ewing and Kysar envision: giving “voice to individuals and actors whose grievances have been neglected by the other branches of government” and “structur[ing] that voice within the pedigreed, rationalized discourse of law and its principles.”21 States, communities, and environmental groups were thus able to find a forum for expressing their concerns. Indeed, the goal of the litigation was to open such a forum before the EPA and the courts by requiring the EPA to initiate a rulemaking procedure, which, under the Administrative Procedure Act, would provide a broad opportunity for public input22 and would also be subject to judicial review by any person with standing.23

The Court’s standing holding was significant, not just for its immediate effect of opening the door to federal climate litigation, but also because it focused attention on the climate issue. Summarizing the Court’s holding on standing, Justice Stevens argued that Massachusetts had standing because the sea-level rise “has already harmed and will continue to harm Massachusetts,” the “risk of catastrophic harm” was remote but real, and the risk “would be reduced to some extent if petitioners received the relief they seek.”24 In its standing analysis, the Court thus performed the function that Ewing and Kysar describe by “holding up particular problems to the light” and thereby “implicitly or explicitly enter[ing] into the conversation about what actions are called for and why.”25

On the merits, the Court held that the EPA had misapplied the Clean Air Act in several critical respects. The EPA had argued that carbon dioxide is not an “air pollutant” within the meaning of the Clean Air Act.26 The Court found this view incompatible with the plain language of the statute, which defines pollutant very broadly.27 The Court also found that the EPA had considered impermissible extraneous factors in making its determination.28 The Court remanded the case for proceedings consistent with its holding that the EPA reconsider its decision under the correct statutory standards.29 These standards required the EPA to base its decision solely on the state of the scientific evidence bearing on whether greenhouse gases from cars were a threat to human health or welfare, rather than on its views about whether regulation under the Clean Air Act was desirable as a matter of policy.

In response to the Supreme Court’s ruling, the EPA ultimately issued a finding that greenhouse gases endanger human health and safety,30 and the EPA has begun developing regulations to reduce greenhouse gases based on this finding.31 Designing these regulations presents difficulties because the provisions of the Clean Air Act concerning emissions from factories and electrical generators do not fit very well with the kinds of controls needed for greenhouse gases.32 The statute is primarily designed to deal with local pollution problems, rather than global ones. However, the EPA has been trying to design rules that are reasonably suitable and do not violate the statute. Its initiatives to date include permitting requirements, enhanced fuel efficiency rules, and a schedule for imposing greenhouse gas restrictions on power plants and refineries.33 The implementation challenges faced by the EPA reflect the difficulties of applying a highly detailed statutory scheme to a problem that falls within its scope but has not received direct attention from Congress.

Despite these implementation difficulties, the EPA’s action is especially significant because of the current gridlock in Congress. If Congress fails to take effective action, the EPA may well end up creating the primary mechanism for controlling American greenhouse emissions through this administrative process. Potentially, these regulations could provide the basis for a comprehensive federal regulation of greenhouse gas emissions. Obviously, it would be preferable for Congress to create a new legislative framework for greenhouse gases. Such a major policy development ideally should have the direct imprimatur of Congress, and in any event new legislation could target climate change better than the Clean Air Act, a general-purpose statute that primarily addresses other forms of pollution. Given political realities, however, EPA action may be the best available option, at least in the near term.

The Supreme Court’s role in initiating this process may not have been completely indispensable, but it was clearly substantial. The Court foreclosed the Bush Administration’s statutory interpretation argument and required that the EPA exclude all factors except the science of climate change from the endangerment decision. It also left no doubt that a decision of some kind by the EPA was mandatory—climate change was not an issue that the EPA could simply refuse to consider. The result was a set of national regulations that have already gone into effect pending judicial review. Although the Obama Administration might well have decided to pursue such regulations anyway, because of the Court’s ruling the Administration was assured of a solid legal foundation, and it was also able to take advantage of considerable preparatory work that took place under President Bush. In its standing holding, the Court also made it clear that climate science is sufficiently strong to provide the basis for a judicial finding of harm.

Notably, none of the Court’s decisionmaking was done under the aegis of the common law. Rather, it involved a bit of constitutional law in applying standing doctrine under Article III, some interpretation of the Administrative Procedure Act in determining the grounds for reviewing the agency’s action, and interpretation of the Clean Air Act in reversing the Bush Administration’s decision.34

The lower federal courts have also considered climate issues and subsequently prompted administrative action. For instance, one court held that impacts on climate change are a sufficient basis to require the creation of an environmental impact statement before a major federal project or regulation proceed.35 This ruling led the government to develop an economic analysis of the harm caused by climate change for the first time, resulting in an official determination of the social cost of carbon.36 Again, the courts were able to prod the executive branch in a useful way, but not via the common law.

II. State Regulation as Fallback And Fail-Safe

Congress’s default in addressing climate change—and the executive branch’s default prior to Massachusetts v. EPA—created a regulatory vacuum that state governments attempted to fill. Indeed, Massachusetts v. EPA itself can be seen as a successful “prod and plea” directed by state governments to the Supreme Court and thence to the EPA. Perhaps surprisingly,37 state governments have moved much more aggressively than the federal government to address climate change.38 By 2006, every state had taken steps of some kind to address climate change.39 Much more than the public nuisance lawsuits against emitters that they emphasize, these state initiatives exemplify Ewing and Kysar’s vision of climate governance:

  Rapidly evolving, globally interconnected, and wickedly complex, such systems do not yield to straightforward command-and-control regulation or other familiar lawmaking forms. Instead, “governance” only emerges from the decentralized, overlapping, and continually evolving interventions of public and private actors—each operating at different levels and from different spheres of authority, utilizing a range of policy tools both hard and soft, and representing diverse interests and stakeholder groups.40  

In terms of climate change, this regulatory network has clearly emerged at the state level. Quite apart from its role in encouraging federal action, the state efforts are significant in their own right. They illustrate how, in an era of limited government and unlimited harm, one level or branch can have an important impact regardless of whether it successfully incentivizes other levels or branches to act.

In the interest of space, I will focus on California’s climate change legislation. California, which by itself constitutes one of the ten largest economies in the world,41 leads all states with legislation aimed at reducing greenhouse emissions from automobiles and electrical generators, as well as with an ambitious mandate to reduce emissions to 1990 levels by the end of the decade.42 A brief description of California’s regulatory effort is useful to understand what can be achieved at the state level and how far this surpasses the kind of remedy that a court could supply.

California’s efforts focusing specifically on climate change can be traced back to 1988, when a law required the first inventory of in-state greenhouse gas emissions.43 Since then, California has continued to pursue a wide range of policies to reduce greenhouse gas emissions. In 2006, Governor Schwarzenegger signed into law the capstone of the state’s climate policy, the California Global Warming Solutions Act of 2006, also known as A.B. 32.44

A.B. 32 sets a binding greenhouse gas emissions target, requiring California to reduce emissions to the 1990 level by 2020.45 This law generated worldwide attention, including a statement by the British Prime Minister that its signing represented a “historic day for the rest of the world as well.”46 The Prime Minister and the Governor of California also entered an agreement to “share best practices on market-based systems” and to “cooperate to investigate new technologies”;47 similar agreements now exist between California and states and provinces in Australia and Canada.48 In the November 2010 elections, a ballot initiative to suspend indefinitely the operation of A.B. 32 was soundly defeated with 61% of Californians voting to keep A.B. 32 in effect.49 The vote showed that there is significant grassroots support for climate change legislation, at least in California.

In implementing A.B. 32, the California Air Resources Board developed nine “discrete early action greenhouse gas emission reduction measures”50 designed to go into effect before the cap on carbon emissions is implemented. Four of these actions focus on reducing emissions of high global warming potential (GWP) gases, which are gases whose impact on the climate is hundreds or thousands of times greater than that of carbon dioxide.51 In order to highlight how much more room for initiative legislatures and agencies have compared with courts, it is useful to list these items. The nine discrete early actions are52:

  1. Establishing a low-carbon fuel standard, per Executive Order S-01-07,53 to reduce the greenhouse gas intensity of transportation fuels by 10% by 2020;54
  2. Reducing emissions from small containers of automotive refrigerants with high global warming potential;55
  3. Increasing capture of methane from landfills;56
  4. Establishing aerodynamic efficiency standards (based on the U.S. EPA’s voluntary SmartWay program) for heavy-duty tractors and trailers to improve fuel efficiency;57
  5. Creating a tire pressure program that requires automobile service providers to check and inflate the tires of each vehicle they service;58
  6. Reducing diesel emissions from ports by providing electricity to berthed ships;59
  7. Setting a limit on emissions from pressurized gas dusters with high global warming potential;60
  8. Reducing emissions of perfluorocarbons (PFCs) from the semiconductor industry;61 and
  9. Reducing sulfur hexafluoride (SF6) emissions in non-electric and non-semiconductor applications.62

Beyond these early action items, the California Air Resources Board has decided to implement a cap-and-trade system.63 The system is designed to achieve a 15% reduction in greenhouse gases by 2020, but it is also designed to take into account the needs of different industries and to allow a smooth transition into the more rigorous requirements.64 In developing its cap-and-trade program, California worked closely with the Western Climate Initiative (WCI), a group of western U.S. states and Canadian provinces developing a regional cap-and-trade program.65

Renewable portfolio standards are another important option for state regulators seeking to regulate climate change.66 These programs require that a certain percentage of retail electricity sales be derived from renewable sources. The programs are quite diverse in their ambition and effectiveness,67 but California’s program requires that 33% of electricity be generated from renewable sources by 2011.68

California has also taken the lead in the transportation sector, which is a critical part of climate change regulation. A statute known as A.B. 1493, or the “Pavley Act,”69 requires the state to set standards for greenhouse gas emissions from new cars. The Air Resources Board has a statutory mandate to reduce carbon dioxide emissions from new car models by 30%.70 The Air Resources Board must adopt regulations that achieve “the maximum feasible and cost-effective reduction of greenhouse gas emissions from motor vehicles.”71 (The Board may not, however, impose fees or taxes, ban sports utility vehicles or light trucks, or impose speed limits.72) California is also moving toward adoption of a low-carbon fuel standard, which will encourage the use of biofuels and electrical vehicles.73

It bears emphasis that California is far from alone among the states in addressing climate change. Some of the most interesting initiatives are regional rather than state-based.74 Regional programs include the Northeast Regional Greenhouse Gas Initiative (RGGI), which is the best-known, as well as initiatives in New England, the Great Plains, the Southwest, and the West Coast.75 Indeed, even a relatively conservative Republican governor, Tim Pawlenty of Minnesota, signed the Next Generation Energy Act of 2007, which sets ambitious goals for statewide emissions reductions.76

The unavoidable weakness of state efforts is that, by their nature, they are limited to the specific states that choose to participate, leaving untouched the emissions from those laggard states that fail to regulate air pollutants for ideological or economic reasons. Public nuisance suits under federal common law conceivably could help fill that gap. On the other hand, state regulation has reached much of the nation’s population and, in many ways, has gone far beyond anything that one can imagine being ordered by a court in a nuisance case. Courts do not create cap-and-trade schemes, design regional alliances, impose fuel efficiency standards, or provide multi-faceted regulatory initiatives like California’s early action items. Thus, the richness and depth of state regulatory actions far exceeds those of judicial remedies.

Federal courts do have certain advantages. The life tenure of the federal judiciary and its relative political insulation may enable federal courts to render judgments that may be unpopular with crucial special interest groups opposed to climate change regulation. Indeed, this advantage was on display in Massachusetts v. EPA. But common law remedies are blunt instruments for dealing with a problem as complex as climate change. Even if the Court had ruled in favor of the plaintiffs in AEP, they would have faced serious barriers and probably years of litigation before any remedy would be in sight. Legislatures can move much more quickly. Common law decisions by the Supreme Court are also subject to congressional override. The same is true of state legislation, but states have the advantage of having their own congressional delegations to protect their interests.


According to Ewing and Kysar, “the banner of ‘prods and pleas’ stands for the important capacity of divided authorities to push each other to action when changing social conditions require it.”77 In the battle against climate change, courts engage in judicial review of administrative agencies like the EPA and the California Air Resources Board, and state legislatures are at the front of the column marching under that banner. Common law courts are, at best, limping along at the rear of the march.

There is an important insight at the heart of Ewing and Kysar’s argument. They aptly observe that, in trying to hinder governmental overreach, “we nevertheless may have splintered and hobbled our government too well.”78 The result can sometimes be an inability to deal with important national problems through the conventional process of new legislation followed by executive and judicial enforcement. But while fragmented government authority may be part of the problem, it can also be part of the solution. As Ewing and Kysar put it, “Overlapping governance mechanisms help to span jurisdictions and to marshal different fact-finding competencies, remedial powers, and value orientations. They ensure a fuller and more inclusive characterization of emerging threats to social and environmental well-being.”79 Fortunately, when Congress fails to address pressing social policy issues, our institutional structure provides fallbacks and fail-safes to prevent governmental default. We live in an age of statutes, and these backup mechanisms are generally based on statutory authority—either implementation of existing statutes by agencies and courts, or enactment of new statutory initiatives at the state level.

Ewing and Kysar invoke climate policy as an example of where prods and pleas can be effective, and rightly so. As we saw in Part I, the Supreme Court played a major role in prompting regulatory attention to climate change and led the EPA to begin an extensive effort to regulate greenhouse gases. As Massachusetts v. EPA illustrates, judicial interpretation of existing statutes can not only be only a prod to action by other branches, but also can act as a fail-safe when they default. Congress has yet to take action on climate change, but the EPA rules prompted by the Court’s interpretation of the Clean Air Act may significantly reduce greenhouse emissions even if congressional action remains delayed.80

Part II demonstrated the creativity that state legislatures and regulators have shown in addressing climate change. The result of these “overlapping governance mechanisms”81 has been to ameliorate Congress’s regulatory default in dealing with this issue. This is an important point that should challenge reductionist constitutional visions based on hierarchy and separate domains of action.

Yet despite a brief mention of the role of public law, Ewing and Kysar chose to make common law torts cases the paradigm of what they call “prods and pleas.” Although the role of the common law in addressing social problems should not be dismissed, their emphasis on private law is misplaced. Tort law can certainly help fill gaps in the legislative scheme, and perhaps, on occasion, it can help draw public and legislative attention to an issue. Particularly given the pressing nature of the climate issue, help from any quarter is welcome, even one as peripheral as the federal common law of nuisance. But in terms of policy impact, the common law is the “least dangerous branch”82 of judicial activity.

Daniel A. Farber is the Sho Sato Professor of Law and Chair of the Energy and Resources Group at the University of California, Berkeley.

Preferred citation: Daniel A. Farber, Preventing Policy Default: Fallbacks and Fail-safes in the Modern Administrative State, 121 Yale L.J. Online 499 (2012),