Environmental protection can be an important government function, in particular because private incentives, as reflected in market prices, often do not capture the full social value of environmental quality, or perhaps more precisely, changes in that quality. In the standard analytic framework, private actors cannot capture the value of environmental improvements, or do not bear the full costs of environmental degradation, unless “transaction” (negotiating) costs are zero, so that resulting environmental quality is lower than the optimal level.
At the same time, the political incentives shaping environmental policies and their implementation are hardly immune from various kinds of distortions, the upshot of which is promulgation of environmental policies that might be too stringent, insufficient, or counterproductive. Given that such policies have the effect of transferring wealth among interest groups, economic sectors, and geographic regions, it is not difficult to predict that the complex congressional bargaining process yielding actual policies might fail systematically to result in “optimal” outcomes, particularly given the shifting nature of majority coalitions.
Note also that environmental protection in some ways is a classic “collective good,” which is why private incentives might lead to inefficient outcomes. This is hardly a new concept; but far less familiar are the implications of the fact that the consumption of resources in the production of such collective goods as environmental protection accrues to the benefit of all (although individuals certainly might value it differently). In particular, any majority coalition that forms around an environmental policy has powerful incentives to transfer some of those resources to itself, away from society writ large.
Protecting Our Natural Resources: Cui Bono?
In other words, the majority can reduce the use of resources for environmental protection accruing to the benefit of all, shifting those resources toward the provision of goods benefitting members of the majority. This means that democratic decisionmaking under a variety of majoritarian decisionmaking rules will likely result in the under-provision of collective goods, just as is the case for the private provision of such goods, although the modern literature offers a rebuttal to the standard view of private incentives under a range of circumstances.
Accordingly, it is clear that the formulation and implementation of environmental policies could be improved in ways achieving equivalent outcomes at a lower resource cost, or superior outcomes without increased costs, or perhaps even a reduction in costs combined with improved environmental results. That is: Laws and the implementation of them could be changed so as to yield incentives for decision-makers leading to improved outcomes.
One problem with this framework is straightforward: If “improved outcomes” in the context of environmental policy are defined as fewer resources consumed for given outcomes, then why do democratic processes not pursue the latter, with the savings to be redistributed to some set of political winners? After all, politics is the art of wealth redistribution, and resource-savings can be expected to further that goal.
One answer is that “waste” accrues to the benefit of some subset of interests; the environmental policy bureaucracy is an obvious candidate. That the environmental bureaucracy’s budget, which is already large, continues to grow in size and complexity means that the budget is a function of itself—that is, that Congress finds it increasingly difficult to monitor the budget and to determine the minimum cost of providing the agency’s lump-sum basket of output. It is unsurprising, therefore, that we find vigorous enforcement of regulatory policies that are self-defeating—the Endangered Species Act is a good example—even as they increase the budget and authoritative powers of the environmental bureaucracy.
Accordingly, there are good reasons to believe that environmental policy can be improved in the fundamental sense that incentives for policymakers can be changed so as to produce resource-savings for given environmental outcomes. Note that the political willingness to bear the costs of environmental improvements is heavily a function of aggregate wealth, so that environmental policies that impose economic costs greater than benefits are likely to sap political support for environmental protection.
I have been asked to write a Liberty Forum essay on how U.S. environmental policy can be reformed, but of course this is a multidimensional subject too vast for a short essay. Let me at least address in some detail two basic topics: reforming the National Environmental Policy Act of 1969 and reforming the Chevron doctrine of deference to the executive branch bureaucracy’s interpretation of law.
National Environmental Policy Act of 1969
The National Environmental Policy Act of 1969 is the basic law under which federal reviews of the environmental impact of proposed construction projects and the like are conducted. NEPA is in need of substantial reform by Congress: It has created a heavy bias in favor of the infrastructure status quo and against new projects, even if the latter would result in important environmental improvements compared with the existing state of affairs—in particular, in terms of the likelihood or levels of damage, accidents, or the emission of various effluents This status quo bias is exacerbated by the “completeness” requirement and by a crucial cost-shifting problem, the combination of which prevents sound benefit/cost analysis of proposed projects and other environmental concerns under NEPA.
All environmental policy both in principle and as applied is (or should be) an exercise in benefit/cost analysis: Are the benefits of a given policy or project prospectively larger or smaller than the potential adverse effects, when the environmental effects of relevant alternatives are included in the analysis?
There could scarcely be any project or, indeed, other human endeavor that does not create some adverse environmental effect, however broadly defined. Clearly, we are not willing to reject all new projects—an extreme outcome even among extreme outcomes—in substantial part because 1) a growing population demands more physical capital, 2) shifts in demand and cost conditions across sectors imply resource flows among those sectors, including capital investment, and 3) the inexorable physical depreciation of the existing capital stock means that disallowing any and all new investment would return humanity to a state of nature.
In short: At some point the marginal costs of protecting the environment exceed the marginal benefits, which is why virtually no one chooses to live in a pristine state of nature. Consider a homo sapiens baby born in a cave some tens of thousands of years ago, in a world with environmental quality essentially untouched by man. That child at birth would have had a life expectancy of about 10 years. Had it been given the choice, it no doubt would have opted for a certain decline in environmental quality in exchange for better housing, food, water, safety conditions, medical care, ad infinitum. In other words, that child would accept, eagerly, a massive investment program in infrastructure at the expense of some environmental quality. Which is to say that environmental quality is one important dimension of the capital stock among many, and across which there are tradeoffs.
Accordingly, modern societies evaluate the tradeoffs between capital investments and other such projects, on the one hand, and environmental effects on the other, applying a broad range of anticipated effects and parameters. This explicit or implicit benefit/cost analysis properly considers the effects of a given project not only on its own terms but compared with the status quo. Proper analysis should balance the adverse effects of both insufficient review (too little attention to the potential adverse effects of the proposed project) and those of excessive delay (too little attention to the potential benefits of the proposed project). This latter tradeoff is similar to the standard “type 1/type 2” error problem in statistics, in which the type 1 error is rejection of the null hypothesis when it is true, while the type 2 error is acceptance of the null hypothesis when it is false.
The Status Quo Bias
NEPA reviews concentrate only on the potential adverse effects of the proposed project under consideration, even if that project, whatever its attendant asserted problems, would significantly reduce the likelihood of environmental damage, or reductions in the costs of achieving lower levels of risks. Consider, for example, a proposed pipeline that would transport petroleum products currently moved by railroad or by trucks. The following table summarizes this comparison of adverse incidents for the United States from 2005 through 2009.
Petroleum Transport Adverse Incidents, 2005-2009
Mode Average ton-miles/year Average incidents/year Incidents/billion ton-miles
Trucking 34.8 695.2 19.95
Railroad 23.9 49.6 2.08
Liquid pipeline 584.1 339.6 0.58
Natural gas pipeline 338.5 299.2 0.89
Source: Diana Furchtgott-Roth and Kenneth P. Green, “Intermodal Safety in the Transport of Oil,” Table 8, monograph published by the Fraser Institute, October 2013.
The vastly greater safety of pipelines for the transport of petroleum products as compared to trucking and rail transport is manifest; but NEPA reviews of proposed pipeline projects shunt this larger context aside, focusing only on the environmental effects of the proposed pipeline itself. This myopia is inconsistent with the larger goals of improved safety and reduced environmental risks, but is a direct consequence of the implementation of NEPA as written. This is particularly the case as technological improvements and other such advances enhance the environmental performance of new infrastructure projects relative to existing ones. A reform of this law by the Congress would bring environmental improvement and reduced costs for capital investment.
The “Completeness” Requirement
In Scenic Hudson Preservation Conference v. Federal Power Commission (1965), the U.S. Court of Appeals for the Second Circuit held that the project under consideration could be approved only if “the record on which it bases its determination is complete.” The need for a “complete” record is an obvious route to endless litigation and delay, in that there is no limiting principle that would exclude consideration of any given potential environmental impact, regardless of how trivial or speculative. Such delay is inconsistent with the need for any modern economy to improve and replace infrastructure as it depreciates or becomes obsolete, whether economically or physically. And, again, it is inconsistent with the increased aggregate wealth needed for a growing population to maintain and improve environmental quality.
For any project there is a hierarchy of potential effects, from the large and significant, to the small and insignificant, to the trivial. The number of potential effects almost literally is infinite. A major proposed capital investment would create environmental risks while providing a stream of beneficial services over time. Any reasonable review of such a proposed project cannot do much better in terms of environmental protection than to focus on major impacts while insisting on lower-risk designs, ongoing inspections, and other procedures intended to avoid or mitigate risks and adverse events as they emerge.
Ex ante examination of any and all risks—“completeness”—is preposterous in an economy in which capital investments must be made so as to avoid impoverishment and, indeed, environmental degradation. A reform of NEPA in this context would require that the Congress define the nature and magnitude of significant risks and environmental impacts, with less rather than more interpretive flexibility for the administrative agencies. This change would be made under the reasonable assumption that the list of less significant, small, and trivial risks is too lengthy to examine in detail, and that the very large number of such factors will tend to cancel them out as a whole, in particular when such less important impacts are viewed across the vast array of proposed projects.
The Problem of Cost-Shifting
Not all environmental impacts are worth avoiding. That is, the benefits of a given project may outweigh any adverse environmental impacts, however defined, a truism that is the beginning of sensible benefit/cost analysis in this context. In order for decision-makers systematically to achieve that end, they must be confronted with both the political benefits and costs of their decisions. Because the NEPA regulatory approach does not require compensation for asset owners—unlike the case under a takings approach—the law in effect allows Congress to demand a maximalist protection of environmental values without bearing any of the costs of doing so, in this case in terms of some sort of required budget outlay.
NEPA demands in section 101(a) that regulators
use all practicable means and measures, including financial and technical assistance, in a manner calculated to foster and promote the general welfare, to create and maintain conditions under which man and nature can exist in productive harmony, and fulfill the social, economic, and other requirements of present and future generations of Americans.
Note the absence of any cost considerations or benefit/cost balancing parameters. This means that regulators are empowered to—indeed have a duty to—regulate projects in such a way that marginal costs are guaranteed to exceed marginal benefits, because government is instructed in effect to “protect the environment” without consideration of the explicit or implicit costs of doing so. As an aside, this system provides perverse incentives for private parties as well, for they bear all of the costs of environmental protection while most of the favorable effects accrue to the benefit of others. Hence the “shoot, shovel, and shut up” phenomenon.
An economy in need of constant capital investment in the face of a growing population, economic shifts, technological advances, depreciation of existing capital, and growing demands for environmental improvement should strive to balance such needs with the imperative of good stewardship of the natural environment. NEPA as currently written and enforced is inconsistent with that basic benefit/cost goal, a condition that should induce lawmakers to rewrite this statute so as to eliminate the pitfalls discussed above.
Damage Done by the Chevron Doctrine
Article III of the Constitution vests in the judiciary the power to decide “all cases, in law and equity, arising under this Constitution, the laws of the United States, and treaties made, or which shall be made, under their authority.” Notwithstanding that provision, the Supreme Court ruled in Chevron U.S.A. v. Natural Resources Defense Council (1984) that whenever the law is ambiguous, courts reviewing agency interpretations should defer to that interpretation unless it is unreasonable or clearly inconsistent with statutory language. In no part of the Constitution is the judiciary empowered to delegate that power to the executive branch, a crucial aspect of the structure of the Founding charter that allows the judiciary to serve as a check on the powers of the executive branch. Furthermore, the power of executive branch agencies to interpret the law under the Chevron doctrine infringes upon the legislative powers vested in Congress in Article I.
As a practical matter, even apart from the ambiguity of the term “ambiguous,” the central problem with the Chevron doctrine is its implicit assumption that the executive agencies are disinterested enforcers of the relevant statutes, a premise that ignores the bureaucratic benefits attendant upon rising budgets, themselves a function of agency interpretations of law and the resulting scope of regulatory activity allowable for a given agency.
The bureaucracy, in short, is an interest group. Notice also that agencies’ interpretations of their statutory authority can shift sharply with a change in administration, particularly given that the political incentives and preferences of successive agency leaders almost certainly will differ from one to the next. Consider the sharply differing interpretations of section 111(d) of the Clean Air Act as applied to the electric power sector by the Obama and Trump administrations, respectively. (The Obama administration argued that a state’s entire power system rather than individual electric-generating plants is the “source” of greenhouse gas emissions for regulatory purposes, a stance utterly at odds with decades of regulatory practice and legal interpretation.) This reality alone belies the implicit Chevron premise that the agencies’ “reasonable” interpretations of law should be deferred to by judges but not by successive Presidents or the heads of agencies whom they appoint.
Such changing agency interpretations of the law allow expansion of agency authority beyond anything that Congress might have envisioned when enacting legislation; and the changing interpretations obviously have the effect of creating increased uncertainty for the private sector, with ensuing effects upon investment and other parameters that cannot be salutary. Moreover, Chevron creates incentives for Congress to enact poorly defined laws—that is, to absolve itself of the burden of making difficult choices. That effect is a distinct erosion of the separation of powers giving Congress the authority to enact laws and the President the duty to enforce them.
Be it noted that it was then-judge Stephen Breyer who criticized judicial Chevron-type deference as an “abdication of judicial responsibility.” Similar criticisms have been made by then-judge Neil Gorsuch, as the Constitution vests the power to interpret the law in the judiciary, not the executive branch agencies, for cases before the courts.
Bringing the Bureaucrats to Heel
These two reforms are intended to impose constraints upon the ability of interest groups, whether private or public, to drive policymaking toward inefficient outcomes. For NEPA, current practice engendered by the language of the law leads the system to ask the wrong questions, a process that allows litigation pursued solely for purposes of delay and increased costs for private investment. The Chevron doctrine is inconsistent with the structure of the Constitution, and facilitates inefficient outcomes by assuming the neutrality of expertise housed in agencies driven by political agendas, overseen by a Congress interested in less accountability. Both reforms would yield systematically improved outcomes over the longer term.
Only the wealth accrued over the last few centuries, built upon well-defined property rights, has enabled humans to even worry about the environment.
In the decades since the movement began in 1970, we have discovered that we don’t have to choose between environmental protection and economic growth.
NEPA is a quite limited statute; and judicial deference to federal agencies can be good if the agencies are pursuing deregulatory measures.
Whose reforms would work best—mine or my respondents’—cannot be decided on the basis of first principles.