The Journal of Things We Like (Lots)
Select Page

Tag Archives: Environmental Law

Two Chapters in the GIGO Mess Epic

There are two problems with cost-benefit models for environmental policymaking: the model inputs and the model outputs. This is not exactly news. Researchers and reporters have documented honest overestimates of regulatory costs, honest undercounts of regulatory benefits, and dishonest attempts to cook the cost-benefit books.1 The authors of the articles reviewed here avoid such easy targets. Instead, they strike at the heart of the welfarist policymaking preference that promotes and privileges cost-benefit analysis.

Richard Revesz challenges the orthodoxy that distributional effects should not motivate regulatory choices. Bernard Harcourt assails the myth that cost-benefit analysis offers an objective motivation for regulatory choices.

Revesz has long argued that proponents of environmental regulation must learn to love, or at least to live with, cost-benefit analysis. He readily concedes that “all other things being equal,” regulations should be designed to maximize net benefits, that is, to be economically efficient. (Revesz, P. 1490.) But in Regulation and Distribution, Revesz reminds us that inequity prevents all other things from being equal—and that profoundly unequal distributive effects demand corrective action.

Regulation and Distribution does not bother with the simplistic claim that inequity can be ignored because welfare maximization trumps all other social goals. Instead, Revesz grapples with the more nuanced argument of Louis Kaplow and Steven Shavell: that redress for distributional effects of regulatory action should occur only through the tax code rather than through regulatory decisions themselves.2

Revesz replies that using the tax code, however attractive in theory, is impossible in practice. Redistribution through provisions of tax law requires legislative action, but such legislation seems exceedingly unlikely in today’s extended period of hyper-partisan legislative gridlock.

Even if redistributive legislation were possible, a system of taxation and money transfers is ill-suited to address many types of environmental injustice. In one of the most persuasive parts of Regulation and Distribution, Revesz analyzes the deficiencies of using a tax-and-transfer approach to respond to the unequal distribution of exposure to toxic pollutants. With higher exposures come a people who confront a greater risk of becoming sick, more eventual cases of disease, and more premature deaths. Revesz shows that any attempt to respond to this inequity through the income tax system will inevitably result in undercompensation. Some aspects of the harm are too difficult to attribute to individual taxpayers. Others are too difficult to quantify and monetize. Therefore, Revesz concludes, regulatory agencies, and not the Internal Revenue Service, should figure out and adopt measures to counter the harmful distributive effects of their regulations.

How should they do this? Revesz’s proposed solutions are interesting and thought-provoking, though they also raise questions.

Before presenting his recommendations, Revesz is careful to argue that only “unusuallylarge inequities” justify intervention, lest the welfare benefits of cost-benefit analysis be overwhelmed by the routine distributional effects of regulation. (Revesz, P. 1571.) Individual regulatory agencies cannot be trusted to make the call, Revesz seems to imply, so he suggests that the Office of Information and Regulatory Affairs (OIRA) define the trigger in a guidance document. Those who are already skeptical of OIRA’s outsized power might question this idea, although Revesz would doubtless answer that OIRA’s central role in regulation is simply a fact of life. Unfortunately, Regulation and Distribution offers no advice about how large an inequity is unusually large, or even about how to begin to make that decision. This would be a nice follow-up project.

Next, Revesz proposes creation of a standing interagency working group that would craft an appropriate redistributive response to any regulation satisfying OIRA’s triggering criterion. Revesz considers two types of possible responses: directly making the rule more equitable, or indirectly mitigating the rule’s inequitable effects.

Of the two, Revesz pays much more attention to the indirect mitigation option. Having argued that the tax code cannot be used for this purpose because of legislative gridlock, Revesz advocates executive branch action under existing statutory authorities. Citing efforts by the Obama Administration to help dislocated coal industry workers, he argues that Presidents have a host of options for ameliorating the focused economic harm that sometimes results from regulations that benefit society as a whole. Many of these options involve the targeted award of federal grants or the redeployment of contingency funds such as those set aside for national emergencies. The breadth of the possibilities supports Revesz’s proposal for an interagency working group.

Revesz’s mitigation goal is laudable. As he acknowledges, however, achieving it would require concerted action centrally controlled from the White House. Revesz argues that Presidential administration is, like OIRA, a fact of life; it may as well be deployed in support of beneficial environmental regulation. On the other hand, if a President may use emergency funds to help coal miners weather the effect of greenhouse gas emission limitations, what is to stop a President from using emergency funds to build a border wall? At this moment in history, the scope of Presidential authority inherent in Revesz’s proposal is arresting.

Mitigation approaches, notwithstanding the Presidential power they invoke, still amount to money transfers. Effectively, they use tax revenues (or federal borrowing) to achieve distributive goals, without requiring amendment of the tax code. Therefore, Revesz notes, these approaches are not the best way to address non-monetary harms, such as the environmental injustice of disproportionate exposures to toxic pollutants. To address inequitable non-monetary consequences of proposed rules, Revesz recommends that the interagency working group should consider directly changing the rule.

Regulation and Distribution discusses this option only briefly, leaving some interesting questions unaddressed. For example, could an agency ever justifiably reject a rule that maximizes net benefits in favor of a rule that is less efficient but more equitable? Might we be willing to accept a slightly smaller pie in exchange for keeping more people alive and healthy enough to enjoy partaking? Revesz very nearly implies an affirmative answer, but never quite says so. The follow-up question, “under what circumstances should an agency do this,” would surely lead to a very interesting conversation.

Bernard Harcourt, it seems, would eagerly join such a conversation. For if Regulation and Distribution challenges one pillar of cost-benefit orthodoxy without quite trying to bring down the whole edifice, Harcourt’s The Systems Fallacy has no such compunctions.

The Systems Fallacy argues persuasively that the claim that cost-benefit analysis provides policymakers with neutral, scientific, or objective guidance, is false. Harcourt contends that any cost-benefit analysis necessarily embodies normative political values and then, by guiding policymaking, in turn reshapes normative political values.

Harcourt traces the origins of cost-benefit analysis not to welfare economics but to military operations research and systems analysis. Systems analysis approaches worked well enough, he says, for military or engineering problems addressing the performance of tangible objects. But social policy problems do not define themselves. Before even confronting the problem of determining the values and functions to use in a policy analysis, Harcourt explains, an analyst must make a series of decisions about the scope of the analysis.

Harcourt identifies and illustrates five critical choice-of-scope decisions: conceptualizing the metaphorical social system to be analyzed, defining the system’s boundaries, determining the system’s objectives, selecting policy alternatives to be analyzed within the system, and choosing criteria to evaluate system performance under the various policy alternatives. With hypotheticals and real-world examples, Harcourt shows that each of these decisions “entail[s] normative choices about political values.” (Harcourt, P. 421.)

Although Harcourt concedes that systems analysis and cost-benefit analysis are not identical, it is easy to see – and Harcourt demonstrates—that cost-benefit analysis requires the same set of value-laden choice-of-scope decisions. Moreover, he contends, once those decisions produce a policy outcome, a feedback loop engages: cost-benefit analysis determines policies; the policies dictate allocations of social resources; the allocations of social resources affect people’s lived reality, altering the society’s balance of political values. This is what Harcourt finds most offensive about allowing cost-benefit analysis to set social policy: a supposedly objective analytical tool, often entrusted to technocrats, “silently impose[s] political values on society.” (Harcourt, P. 422.)

Harcourt acknowledges that smart welfare economists, again exemplified by Kaplow and Shavell, among others, have a response. If welfare is defined broadly enough to include people’s desires to implement political values such as fairness, then cost-benefit analysis can maximize welfare while including society’s political preferences, rather than privileging only some political values through choice-of-scope decisions. But Harcourt responds, devastatingly, that this catholic vision of welfare and welfare maximization exists only in theory. A real cost-benefit analysis inevitably addresses a particular, selected social problem. Therefore, a real cost-benefit analysis inevitably makes the normative choice-of-scope calls Harcourt describes. And, Harcourt notes, maximizing net benefits within the arbitrarily-defined metaphorical system being analyzed may not actually maximize overall social welfare, broadly defined to include political values.

The Systems Fallacy is not an article about environmental law. Harcourt’s arguments are general; his illustrations concern policies aimed at crime reduction. But any environmental lawyer will recognize at once that Harcourt’s argument applies strongly to environmental policy. Pollution control regulations, which so often promise benefits that are broadly dispersed and hard to quantify in exchange for costs that are concentrated and monetary, seem to face particularly stringent cost-benefit scrutiny from all three branches of government. And the problem Harcourt identifies goes beyond pollution regulation to pervade all environmental policy. The choice-of-scope decisions Harcourt describes, for example, are awfully familiar to anyone who has ever been involved with an environmental impact statement under the National Environmental Policy Act.

If there is a weak spot in The Systems Fallacy, it is the discussion of what to do about the problem the article identifies. Harcourt quite properly insists that he is not opposed to analytical rigor or to quantifying what can be quantified. But, he asserts, policy analysis should be limited to a single dimension, thereby evading the systems fallacy by avoiding the normative choices embedded in the construction of metaphorical systems to analyze. That solution seems unconvincing and unrealistic. Choosing the dimension for analysis would also be fraught with political value judgments, and the functional relationships between variables usually would turn a unidimensional metric into a multidimensional system.

Alternatively, Harcourt argues, the solution is to politicize cost-benefit analysis and policymaking, wresting back normative power from the technocrats. It is hard to argue with Harcourt’s objective; assuring political accountability for inherently political judgments is a good idea. But at this moment in history, when alternative facts are spun to serve political agendas, overbearing technocrats may not be society’s biggest problem.

Late in 2018, law professor and former OIRA administrator Cass Sunstein, in a keynote address at a conference of Revesz’s Institute for Policy Integrity, said: “We often think that the issues that divide us are issues of values. But the fundamental divisions involve issues of fact, not values.”3 No doubt that is true, sometimes. Not always, though. Sometimes, different people really do hold different values. Sometimes, even agreed-upon facts produce different policy positions in different people. Sometimes, different values even drive different perceptions of facts.

In distinct ways, both Regulation and Distribution and The Systems Fallacy teach us to be vigilant for those possibilities. Richard Revesz and Bernard Harcourt offer new reasons to be skeptical of cost-benefit policy prescriptions. They show us that cost-benefit analysis has limits that cannot be overcome by attacking the “garbage in” problem, by collecting more data, by refining functional models. Their work should inspire us to think outside the cost-benefit box. If we pay heed, we may be able to use cost-benefit analysis more wisely and to avoid the problem of policy “garbage out.”

Cite as: Steve Gold, Two Chapters in the GIGO Mess Epic, JOTWELL (July 9, 2019) (reviewing Bernard E. Harcourt, The Systems Fallacy: A Genealogy and Critique of Public Policy and Cost-Benefit Analysis, 47 J. Legal Stud. 419 (2018); Richard L. Revesz, Regulation and Distribution, 93 N.Y.U. L. Rev. 1489 (2018)),

Should We Use the Market to Address Climate Change?

Alice Kaswan, Energy, Governance, and Market Mechanisms, 72 U. Miami L. Rev. 476 (2018).

The recent report from the Intergovernmental Panel on Climate Change this fall has made clear the urgent need to address climate change. What should be the primary policy tool that we use to address the problem? Economists have vociferously advocated for the use of carbon taxes or cap-and-trade permit systems, on the grounds that they provide the most efficient way to decarbonize global economies. Yet carbon taxes have had little success in the political arena. Many of the existing policies that countries and states have used to address carbon emissions have been regulations or subsidies, not market-based approaches. Is this a fundamental misstep on the part of policymakers?

In her recent article, Energy, Governance, and Market Mechanisms, Alice Kaswan argues that this is not a misstep, and that in fact there are good reasons—political, democratic, even economic—to prefer non-market-based instruments to advance decarbonization. Her article is ambitious in its scope but effective in raising important questions about what approach is best.

Kaswan raises a couple of key points about why non-market-based mechanisms may be superior to address the transition to a decarbonized economy. First, she argues that government coordination of climate policies can allow the achievement of multiple goals in addition to reducing carbon emissions at the least cost (which is what market-based tools excel at). For instance, we might be concerned about the distributional impacts of a transition to a decarbonized economy and adding on social equity measures to market-based tools may not be as effective as a fully integrated approach. Similarly, there are a lot of additional issues we are concerned about in energy production than simply carbon emissions (e.g., bird mortality from wind turbines, or long-term waste disposal from nuclear power), and a price on carbon alone cannot help us resolve those tradeoffs.

Second, Kaswan argues that long-term planning is an essential component of a transition to a carbon-free energy system, given the interconnectedness of a wide range of elements of our energy systems and the long timeframes for many investments in those systems. According to Kaswan, market-based tools may not be the most effective in managing these kinds of planned transitions—particularly if carbon prices are low, and so far we have only observed relatively low carbon prices in practice.

Third, Kaswan argues that public participation would be more robust for non-market-based regulatory measures, and that this public participation will result in a more equitable and more accountable approach to carbon reductions. And finally, Kaswan argues that non-market-based mechanisms appear to be more politically realistic than stringent market-based tools—something that has been quite apparent this fall as the French protest against a new gas tax and Washington state voters turned down a carbon tax proposal.

One law review article will not be able to conclusively answer any of these difficult questions about the role of market-based mechanisms in climate policy—the challenge spans the entirety of the modern economy, across countries with very different political and cultural settings, and an incredible range of technical problems. But Kaswan’s piece is a vital starting point for the now-vibrant debate about which policy approaches will be more successful, and an important counterpoint to a policy discourse that has mostly been dominated by advocates of carbon pricing. Even if you don’t agree with her arguments, Kaswan’s analysis should give you important points to consider.

Cite as: Eric Biber, Should We Use the Market to Address Climate Change?, JOTWELL (February 14, 2019) (reviewing Alice Kaswan, Energy, Governance, and Market Mechanisms, 72 U. Miami L. Rev. 476 (2018)),

Better Environmental Law from an Unlikely Source

For ‘tis the sport to have the engineer / Hoist with his own petard.
William Shakespeare, Hamlet, Act III Scene iv.

Happy is the litigator who successfully turns an argument against the adversary who propounded it. The joy is no less delicious for an academic. Professor Sanne Knudsen tries to turn the trick against the conservative majority of the current Supreme Court in her tidy article, The Flip Side of Michigan v. EPA: Are Cumulative Impacts Centrally Relevant?

Knudsen has gone to war against the narrow, atomistic thinking that, in times of both regulatory advance and retrenchment, has characterized much of environmental policy. Flip Side seeks to infiltrate comprehensive analysis across a broad front of agency decision-making, strengthening environmental regulation under cover of a court decision that struck down a major pollution-control rule.

Flip Side begins by analyzing the Michigan v. EPA opinion, in which the Supreme Court, per Justice Scalia, ruled 5-4 that the Clean Air Act requires EPA to consider industry’s costs of compliance when deciding whether it is “appropriate and necessary” to regulate mercury emissions from power plants. Knudsen avoids, or rather just dips a toe into, several scholarly debates the opinion generated. Did EPA lose at Chevron step one, because the statutory term “appropriate” unambiguously includes cost considerations, or at Chevron step two, because it is unreasonable to exclude cost considerations when construing ambiguous statutory language? How strongly did the Court endorse cost-benefit analysis as a mandatory component of environmental regulation? How broadly will the mandate to consider costs be applied in other statutory contexts?

For Knudsen’s project, the resolutions of those debates are not terribly important. What matters, she argues, are three salient features of Michigan v. EPA: First, that the Court held that a statute required EPA to consider a factor not explicitly mentioned in the statute’s grant of regulatory authority. Second, that the Court conflated review of EPA’s statutory construction with review of the rationality of EPA’s decision. And third, that the Court expressed this conflation broadly rather than through a tight focus on the specific statutory words at issue. That Knudsen draws support for her reading from both Cass Sunstein and Lisa Heinzerling suggests she is on to something.

From these features, Knudsen crucially infers that the method of Michigan v. EPA need not be limited to cost considerations, need not hinge on details of statutory language, and need not be limited to the Clean Air Act. As she puts it, Michigan v. EPA stands for the proposition “that factors which are presumptively indispensable to nonarbitrary decisionmaking must be considered.” (P. 20.) And that frees Knudsen to ask her central question: what other factors, in addition to cost, fit that description?

To answer, Knudsen turns to science to argue that cumulative impacts, no less than compliance costs, are so centrally relevant to environmental policy decisions that they must be considered in any rational regulatory process. For someone with Knudsen’s background in environmental engineering—or with any background at all in ecology or environmental science—this part of Flip Side’s argument takes little effort. Citing examples of climate change, chemical exposure, water quality and supply, and introduced species, Knudsen shows convincingly that interconnectedness and interaction pervade the subjects that environmental statutes address.

Failing to regard these relations when making environmental policy, Knudsen argues, systematically understates the risks presented by environmental contamination, undervalues the public health benefits of regulation, and underestimates the environmental impact of government activity. To view each regulatory or resource management decision in isolation, ignoring the cumulative impact of that decision combined with other relevant governmental or non-governmental actions, Knudsen argues, is to ignore a critically important aspect of the problem. And that, she concludes, is arbitrary and capricious under Michigan v. EPA.

Flip Side concludes with a brief consideration of how environmental policy could improve if agencies knew that their decisions must take account of cumulative impacts in order to survive judicial review. Such a requirement, for instance, could help defend EPA’s practice of including co-benefits (those caused by, but incidental to, meeting the intended regulatory target) in regulatory impact analyses. Knudsen further gives examples of a Minnesota statute that requires consideration of cumulative impacts when issuing air pollution permits, of the potential of evaluating cumulative toxicity risks in chemicals regulation, and of an EPA Region 9 effort to consider cumulative impacts on surface water quality when issuing pollutant discharge permits. In each of these media-specific regulatory programs, Knudsen shows, requiring cumulative impact assessment would bring agency decisions into better alignment with the public health and environmental protection goals of the statutes being implemented.

For litigants that may wish to challenge future agencies that deregulate or decline to regulate without considering cumulative impacts, Flip Side provides a generalized strategy. The strategy is not without its risks. As Knudsen acknowledges only in passing, rigorous evaluation of cumulative impacts is in many contexts difficult to nearly impossible. Even where information exists that would permit some cumulative impact analysis, the exercise presents difficult line-drawing problems: if it is arbitrary and capricious to act without considering cumulative impacts, how far must the analysis extend to survive judicial review? Of course the Supreme Court, in interpreting environmental statutes including the National Environmental Policy Act, the Endangered Species Act, and the Comprehensive Environmental Response, Compensation and Liability Act, has asserted that it is easy to answer such questions by importing concepts such as proximate cause from the common law of torts. Nevertheless, the possibility that courts might misuse Knudsen’s clever argument gives one pause.

On the other hand, Knudsen’s idea may have even more potential than she presents in Flip Side. Her fundamental point, that interactions in complex systems must be considered in formulating rational environmental policy, could be applied also to the cost side of the cost-benefit computation. Experience shows that initial predictions of compliance costs are usually exaggerated. This is not surprising, given the incentives that face a regulated industry before and after a new requirement is promulgated. If Michigan v. EPA means agencies must always consider costs before deciding to regulate, Flip Side implies that courts should welcome, even demand, agency approaches that treat costs in a more dynamic, systems-oriented way.

A petard was a primitive type of grenade, an explosive charge that was intended to breach defensive walls but that often blew up the person planting it. In The Flip Side of Michigan v. EPA, Sanne Knudsen has lobbed a grenade that should explode the walls that constrain environmental regulation into a series of individual decisions analyzed as if ceteris paribus were a description of reality instead of an analytical convenience. It should open a frame for policy-making that will more fully account for regulatory benefits and for the environmental impact of proposed actions while correcting the exaggeration of regulatory and opportunity costs.

With a deeply anti-regulatory ideology currently gripping all three branches of the federal government and many of the states, Flip Side’s thesis is unlikely to be enshrined in the administrative law canon anytime soon. In the fullness of time, however, Knudsen’s argument may—to end with a different metaphor—make some lemonade out of the environmental lemon that is Michigan v. EPA.

Cite as: Steve Gold, Better Environmental Law from an Unlikely Source, JOTWELL (December 7, 2018) (reviewing Sanne H. Knudsen, The Flip Side of Michigan v. EPA: Are Cumulative Impacts Centrally Relevant?, 1 Utah L. Rev. 1 (2018)),

Encouraging Technological Innovation in Environmental and Energy Law

Zachary Liscow and Quentin Karpilow, Innovation Snowballing and Climate Law, 95 Wash. U. L. Rev. 385 (2017), available at SSRN.

Innovation is a critical component of environmental progress. The dramatic reductions in emissions per-mile-travelled from automobiles over the past forty years stem from major breakthroughs like the catalytic converter. Our efforts to switch from fossil-fuel-based energy and reduce greenhouse gas emissions will depend on many different kinds of technological innovation. The dramatic price drops in both wind and solar energy, for instance, are in significant part the result of the development of new technologies.

How can environmental law facilitate the development of new technology to address the challenges of climate change and other environmental problems? The predominant position of economists has been that legal tools that force economic actors to address the full costs of their actions, including the externalities that are the basis of many environmental problems, is the appropriate approach to spur innovation. A carbon tax (or a tradable permit system which requires polluters to purchase their permits) will create incentives for firms and individuals to come up with new technologies that will reduce environmental problems. Liscow and Karpilow’s article challenges this dominant paradigm, drawing on recent significant economics research.

Yet policymakers have stubbornly ignored this advice from economists. For instance, in the 2009 stimulus bill instead of levying a carbon tax the Obama Administration put billions of dollars into subsidies and tax credits to support research, development, and deployment of new renewable energy technologies. Is this just a case of elected officials and policymakers ignoring the wisdom of economists, or is there something more going on here?

Recent research in economics has indicated that there may be something more going on. Led by Daron Acemoglu at MIT, a number of economists have concluded that in order to advance real technological progress to address environmental problems, market-based mechanisms like carbon taxes or tradable permit systems have to be paired with other policy tools, such as subsidies for research and development. The reason is that innovation is path dependent – what we research now, and what technologies we develop now, depends in large part on what research has occurred in the past.

Zachary Liscow and Quentin Karpilow spin out the possible implications of this research (what they call “innovation snowballing”) for legal efforts to address climate change. As they make clear, the implications extend far beyond the most basic question of whether subsidies in the context of research and development are a good policy choice. As it turns out, we might reconsider a range of policy and legal questions based on this research – for instance, even if we don’t use market-based mechanisms, we might nonetheless adjust the kinds of regulatory tools we use to react to climate change. In addition, there are a number of difficult questions about what kinds of research and development we might subsidize, as well as when, and how. For instance, Liscow and Karpilow point out that we might want to focus our subsidy efforts on renewable energy technologies that are unlikely to have positive spillovers for the development of fossil-fuel technology as well. Biomass energy builds on (and can support further research in) related fossil-fuel combustion technologies, so we might not wish to provide significant support for it, as opposed to support for solar energy research, which has little or no overlap with fossil fuel technology.

Liscow and Karpilow are not the only ones who have engaged with these questions. Other scholars (both inside and outside environmental law) have explored whether market-based mechanisms are the best tool to advance technological innovation in the environmental context. Examples are David Driesen’s work and Margaret Taylor’s article in the Proceedings of the National Academy of Sciences noting that cap-and-trade programs appear not to boost innovation significantly. If there is a weakness in the Liscow and Karpilow paper, it is that the authors could have engaged more with this prior research. And some of the extensions that Liscow and Karpilow address – for instance, whether innovation snowballing should lead us to think differently about government procurement programs or investment in infrastructure – could have fruitfully engaged with some of the relevant cutting edge work in environmental law, such as Sarah Light’s work on military contracting and environmental policy, or Alex Klass’s work on energy infrastructure.

But the strength of Liscow and Karpilow’s article is the depth with which they explore the follow-on questions that the original innovation snowballing research prompts. That strength makes this article well worth reading for anyone thinking about legal and policy design in the context of climate change.

Cite as: Eric Biber, Encouraging Technological Innovation in Environmental and Energy Law, JOTWELL (March 14, 2018) (reviewing Zachary Liscow and Quentin Karpilow, Innovation Snowballing and Climate Law, 95 Wash. U. L. Rev. 385 (2017), available at SSRN),

The Real World

My very first law professor, Bob Ellickson, once said to my Torts class: “You know how law professors do empirical research? They sit in a room and think very hard.”

That was in 1984. A lot has changed since then, partly because of pioneering work by Ellickson himself.1 Since 2012, more than 500 law review articles have included the word “empirical” in their titles, and probably hundreds more – including every item in the most recent issue of the Journal of Empirical Legal Studies – report or analyze empirical data without titular advertisement. Many of these papers feature linear regressions or other complex statistical analyses aiming to tease out relationships between variables. Yet there remains much value in research that simply but rigorously informs us of what actually happens in the real world. Understanding environmental law, like understanding the environment, begins with observing. This Jot acknowledges the contributions of two recent articles that help us see.

Karen Bradshaw’s Settling for Natural Resource Damages explores a component of environmental law practice that, she persuasively argues, has received too little scholarly attention. Regardless of whether she correctly identifies the predominance of settlement as the reason for scholars’ indifference, she is unquestionably correct to observe that claims by governmental trustee agencies for natural resource damages (NRDs) – like most civil litigation – almost always settle. But how often? For what types of claims? For how much? These are the questions Bradshaw asks.

To find answers, Bradshaw takes the simplest approach imaginable. She asks the people who know. Through Freedom of Information Act requests, Bradshaw sets out to determine the number of NRD claims each federal trustee agency has settled under its particular legal authorities, the aggregate amount of damages those settlements recovered, and the distribution of recovery amounts. She acknowledges that the quality of her data is limited by the inconsistent and uncertain quality of agency responses to her information requests. Nevertheless, the data in Settling for Natural Resource Damages appear to paint a reasonably robust picture of most types of NRD settlements by federal agencies.

To her credit, Brashaw defines NRD claims broadly. She includes liability to federal land management agencies for damaging the resources they manage as well as the better-known NRD liability for hazardous substance and oil spills provided by the Comprehensive Environmental Response, Compensation and Liability Act (CERCLA) and the Oil Pollution Act (OPA, together with predecessor provisions of the Clean Water Act). As Bradshaw points out, the limited literature on NRD liability has focused almost entirely on CERCLA and OPA.

The results of Bradshaw’s inquiry, however, seem to vindicate earlier scholars’ emphasis. Federal agencies have settled more than 300 claims for damage to resources in national park units and national marine sanctuaries. These claims account for nearly half of the settlements in Bradshaw’s data set, but the reported combined value of those settlements is less than one percent of the total for all NRD settlements. The other ninety-nine percent came in settlements under CERCLA, OPA and the Clean Water Act.

Even excluding the outsize settlements of NRD claims related to the Deepwater Horizon blowout ($8.1 billion) and the Exxon Valdez grounding ($900 million), Bradshaw reports, the cumulative nominal value of federal NRD settlements under CERCLA, OPA and the Clean Water Act is nearly $1.4 billion. All that money is supposed to compensate the public for injury to, loss of, or destruction of natural resources, and is supposed to be used only to restore, replace, or acquire the equivalent of the resources harmed. How well have agencies’ settlement practices served the goal of obtaining appropriate levels of compensation? How successful have the federal trustees’ resource restoration efforts been? Should the tort-like NRD liability provisions of these statutes be emulated in addressing other environmental problems? How does the performance of non-federal trustees – which, as Bradshaw notes, includes some settlements that were subjected to withering criticism – compare to the performance of the federal trustees?

These are urgent questions. Settling for Natural Resource Damages makes no attempt to answer them, but invites efforts to do so. Bradshaw’s call for further research is well taken, in light of the scale of NRD settlements that she documents. That in itself is a significant contribution. In the world of NRD settlements, observing what’s going on out there is the first step in understanding what’s really going on out there.

James W. Coleman’s How Cheap Is Corporate Talk? observes a different aspect of environmental law practice. Coleman addresses the “regulator’s dilemma”: agencies must identify technically and economically feasible ways to meet statutory environmental goals, but their main source of information about technology and costs is the industry to be regulated – when the industry has obvious incentives to portray any regulation as infeasible and expensive. Coleman argues that one way to test the claims an industry makes to a regulator is by comparing them to what the industry tells its investors about the same regulatory risks – when the industry has obvious incentives to portray any potential regulation as no big deal. An industry’s “two-audience problem,” Coleman contends, can ease the regulator’s dilemma.

As Coleman acknowledges, it is hardly news that companies and trade associations facing potential environmental regulations (as well as many other actors in many contexts) talk out of both sides of their mouths depending on their audience. But Coleman strives to go beyond anecdote. He attempts a rigorous analysis to demonstrate this phenomenon in operation in a way that might be helpful to regulators, to investors, and to corporate counsel who give advice about statements made to both regulators and investors.

For his project, Coleman mines statements about a cleverly-selected instance of environmental regulation: EPA’s annual efforts to promulgate a Renewable Fuel Standard pursuant to the Energy Independence and Security Act of 2017. This rulemaking and statute, which underlie the notices at gas pumps that ethanol has been blended into the fuel, suit Coleman’s purpose for several reasons. The statute’s requirement for annual revisions to the Standard generates recurrent opportunities for affected industries to state clear views about the impact of proposed and final regulations in documents that are easily paired – comments submitted to EPA during the rulemaking process and Form 10-K disclosures released to investors. The number of affected companies is small enough for thorough review of the documents yet large enough to provide results amenable to statistical analysis. Perhaps most usefully, implementation of the Standard affects different industries differently. Increasing the market share of ethanol and other biofuels threatens the profits of petroleum refiners and allies but offers growth opportunities for agricultural producers and allies. Thus Coleman could observe, by reviewing more than 10,000 pages of documents, the way industries with opposing interests depicted the effects of the same rule in statements to both regulators and investors.

How, though, can the comparison be made? Simply counting contradictions will not do, for as Coleman understands, “[m]ost actors facing a two-audience problem are smart enough to avoid direct factual contradictions.” (P. 66). Instead, Coleman codes the substance of individual statements about the Standard’s impact. Essentially, he counts the number of arguments about the standard that each company made to each of its two audiences (in regulatory comments and Forms 10-K) about how the Standard would affect the company.

Coleman found that companies with dominant interests in petroleum predicted more negative impacts of the Standard, on average, in their regulatory comments than in their Forms 10-K (2.78 versus 0.78). By contrast, companies with dominant interests in biofuels predicted more positive impacts of the Standard, on average, in their Forms 10-K than in their regulatory comments (0.58 versus 0.21). Applying two straightforward statistical techniques, Coleman implicitly tested the null hypothesis that each of the two sets of comparable discussions were drawn from a population with the same mean number of negative or positive impacts. The observed differences were statistically significant (P < .004 by either test) for the companies with petroleum interests and barely not significant (.05 < P < .055 by either test) for the companies with biofuels interests. Thus Coleman demonstrates that with respect to the Renewable Fuel Standard, “oil companies warn regulators and reassure investors.” (P. 70.)

There are methodological issues. Despite Coleman’s efforts to ensure data quality, the taxonomy of distinct predictions of negative and positive impacts of the rule, and the assignment of individual statements to codes within that taxonomy, must to some degree be subjective. And it is not clear that counting arguments is the best way to detect the “exaggeration, ambiguity and omission” that Coleman correctly describes as the hallmark of assertions by intelligent actors facing a two-audience problem. Moreover, affected interests in environmental regulatory disputes surely make many statements about potential rules, targeted to both regulators and investors, outside of officially submitted comments and formal 10-K disclosures.

These quibbles do not denigrate the work’s contribution. By choosing a consistent and limited set of documents to compare, Coleman avoids any possible allegation of cherry-picking. Counting arguments may miss differences in tone and emphasis, but is unlikely to understate the contrast between a company’s pitches to regulators and investors. Coleman’s analysis confirms, in a systematic rather than anecdotal way, our intuition that regulated industry tells different stories to different audiences to suit different ends. His findings suggest that regulators and investors should keep saltshakers handy when hearing from an industry to be affected by a pending regulatory choice – including an industry that stands to benefit. The lesson bears demonstration, even during an Administration more likely to deregulate than to regulate.

Karen Bradshaw’s Settling for Natural Resource Damages and James W. Coleman’s How Cheap Is Corporate Talk? have relatively modest goals. These papers are not burdened with grand theory or elaborate statistics. Instead, they collect and present data to describe clearly phenomena that we know exist, but that we have comprehended only vaguely. They help us see. 

Cite as: Steve Gold, The Real World, JOTWELL (August 18, 2017) (reviewing Karen Bradshaw, Settling for Natural Resource Damages, 40 Harv. Envtl. L. Rev. 211 (2016) and James W. Coleman, How Cheap Is Corporate Talk? Comparing Companies’ Comments on Regulations with Their Securities Disclosures, 40 Harv. Envtl. L. Rev. 48 (2016)),