An earlier post raised questions about the utility of analyzing greenhouse gas emission differences in the typical NEPA document/routine. These are hardly new questions. But we should be clear about two things. First is the difficulty of generalizing about NEPA documents. NEPA routines and the documents they generate are diverse. Even the same category or class of NEPA documents—all EISs or all EA/FONSIs, for example—often barely merit being considered side-by-side. Agencies take such variegated actions which are governed by NEPA and have structured their compliance routines therewith so variously that it is hard to generalize about virtually anything and whether it does or doesn’t belong in the “average” NEPA document. Secondly, we should be clear about what we mean when we say ‘climate change’ as it might appear in NEPA. Some will tell you that the relationship between GHG emissions and potentially bad/catastrophic outcomes are what “climate change” means. On this theory, there’s some sense to forcing climate change analyses on unwilling agencies as if NEPA were some kind of “tax” (or penalty!) and the more emissions attributable to that agency, the more NEPA/tax it should pay. But I will argue that as a matter of consequentialist reasons—which NEPA itself is all about—that approach to climate change in NEPA is wrong.
CEQ’s Confusing Path
CEQ first proposed to issue guidance on climate change in 1997. Back then, the memo from then-chair Kathleen McGinty stated (my emphasis) that
[w]hile uncertainties remain, particularly in the areas of the exact timing, magnitude, and regional impacts of such changes, the vast majority of scientific evidence supports the view that continued increases in [GHG] emissions will lead to climate change. . . . The NEPA process provides an excellent mechanism for consideration of ideas related to global climate change.
And that prompts many climate hawks still today to argue that NEPA routines need to include climate change. Their will to leave no stone unturned, no argument unmade, and no possible manifestation of climate change unnoticed to the powerful—those whose decisions, collectively, comprise the US’s lethargic response to the globally-scaled crisis of climate change—drives them. The ranks of the denialists may be thinning they argue, but inaction is everywhere still!
NEPA has often occupied unsafe ground between warring parties like this. But this time it’s different. The dilemma is thus: can the typically de minimis fractions of the global GHG portfolio and their being cut through the selection of some alternative be a reason for choosing one alternative over others? By the time CEQ unveiled another attempt in February 2010 to guide action agencies on climate change, it was at least aware of the dilemma. Unfortunately, it still hadn’t come up with much of a plan.
The Feb. 2010 proposal aimed to get action agencies to consider “the effects of a proposed agency action on GHG emissions.” Notice: that’s importantly different from considering a proposal’s effects on the “human environment.” NEPA demands the latter. So CEQ argued that
[b]y statutes, Executive Orders, and agency policies, the Federal government is committed to the goals of energy conservation, reducing energy use, eliminating or reducing GHG emissions, and promoting the deployment of renewable energy technologies that are cleaner and more efficient. Where a proposal . . . implicates these goals, information on GHG emissions (qualitative or quantitative) that is useful and relevant to the decision should be used when deciding among alternatives.
Of course, energy efficiency and conservation are, to a first approximation at least, laudable goals. There are some troubling questions that arise surrounding “rebound” effects (a/k/a Jevon’s paradox) and whether any given actor’s savings simply enable some other actor’s consumption in a free economy.
Let’s suppose that the rebound effect is usually oversold by conservation’s skeptics and oil companies alike and that that is why these other “statutes, Executive Orders, and agency policies” commit the government to fostering conservation and renewables. If any action agency is obliged by some other norm to weigh those variables in its thinking, interpreting NEPA to do so makes little practical difference. And if they aren’t, then CEQ’s use of those other norms to interpret NEPA as did the 2010 proposal turns a general tool (NEPA) into a specific tool (pre-existing norms on fostering energy conservation), arguably substituting the latter’s purposes for the former’s. NEPA, after all, encompasses all aspects of the human environment, not just energy efficiency/conservation. And if, as is so often the case, NEPA’s documents are already bulging at the seams testing the cognitive limits of even the most conscientious decision-maker, it had better be some kind of metric worth using if it’s stuffed into an already crowded cognitive task.
CEQ at least seemed aware of this dilemma. GHG emissions may or may not be a very good measurement of a proposal’s overall environmental impact and, in order for their consideration to make sense, the proposal should at least involve some more-than-trivial GHG quantities. CEQ proposed a 25,000 metric ton annual equivalent as a “useful, presumptive, threshold” [sic] for when and where the suggested “qualitative or quantitative” GHG analyses should appear.
Yet in the same breath CEQ retreated, cautioning that it wasn’t proposing “this as an indicator of a threshold of significant effects [NEPA § 102(2)(C)’s threshold for EISs], but rather as an indicator of a minimum level of GHG emissions that may warrant some description in the appropriate NEPA analysis. . . .” How is such a threshold “useful” to NEPA if it isn’t about effects on the human environment?
Roads to Nowhere: Quantifying Emissions Why?
If agencies should only employ NEPA to consider the consequences of actions over which they have control or authority, see Dept. of Transp. v. Public Citizen, 541 U.S. 752, 768 (2004), the question is why go to the trouble of quantifying emissions and, if over 25,000 metric tons annual equivalent, to then “consider” reducing those emissions when the emissions in question are so causally insignificant? At the risk of over generalizing, virtually every NEPA-governed decision by US agencies last year could’ve been set to some “no action” alternative so that, as for NEPA’s relationship to “climate change,” emissions would’ve looked like a big focus, and I doubt it would’ve measurably improved any good estimate of our emissions trajectory globally. What does that say about NEPA? It says that, at least for the moment, NEPA’s act-consequentialism is poorly fit to a globally-scaled problem like climate change.
Monetizing Carbon to Make it More Considerable?
One way to change things is to give GHG emissions a price tag. With that, agencies could conceivably factor in the “cost” of their choices’ GHG emissions versus other monetizable costs. In the case most credit with having prompted OMB’s “social cost of carbon” working group and now controversial cost estimates, Center for Bio. Diversity v. NHTSA, 538 F.3d 1172 (9th Cir. 2008), the agency was sorting out controls for a significant fraction of global emissions—emissions from new US motor vehicles—and could’ve been rather stringent. Best estimates were that ~40 million metric tons of CO2 equivalent separated NHTSA’s preferred action from the more stringent, fuel saving standards it could have imposed. (A metric ton of CO2 equivalent is emitted from the combustion of ~103 gallons of gasoline.)
NHTSA had estimated the costs of imposing the stringent (but feasible) fuel economy requirements but had not estimated the costs of foregoing them. Not surprisingly, it chose to forego stringent fuel economy standards. The Ninth Circuit found this arbitrary and capricious under 5 U.S.C. § 706(2)(A): monetizing costs on only one side of a ledger looks like stilted accounting at least. See 538 F.3d at 1200-02. And it also found NHTSA’s EA/FONSI “inadequate” for failing to evaluate the “incremental impact” the excess emissions “will have on climate change or on the environment more generally” if aggregated with other “past, present, and reasonable foreseeable actions such as other [fuel economy] standards.” Id. at 1216.
Would things be different with a “social cost of carbon” price estimate in hand? Yes—if monetization is the norm and all of the environmental costs and benefits in a NEPA analysis are reduced to net present (market) values. I’m not sure we want that for the average NEPA document, though. (If monetization isn’t the norm, the Ninth Circuit’s distaste for stilted accounting works in the opposite direction.) The social cost of carbon estimates have been rightly criticized as omitting a great deal. (See, for example, this report on the uncounted costs of increased wildfire risks.) Whatever their merits, they estimate global costs, not costs to US citizens and that presents a particularly tricky challenge for US agency officials charged with safeguarding the welfare of Americans first. Finally, as many economists have now begun acknowledging, using mathematical tools to average out seemingly concrete numbers like $38 per ton can mask huge risks of error, misinterpretation, and irrationality given how little reason we have to believe the Earth’s climate will just gradually, gently warm instead of falling off some geophysical cliff.
None of this is to suggest that NEPA documents involving the building of roads, bridges, beaches, runways, etc., shouldn’t consider the risks expected for the project’s existence in a shifting climate. As CEQ proposed in 2010,
a proposal for long-term development of transportation infrastructure on a coastal barrier island will likely need to consider whether . . . design parameters may be changed by the projected increase in the rate of sea level rise.
Beyond the easy cases, though, even this kind of analysis is still incredibly hard, consists mostly of mathematical modeling, and may exceed the average action agency’s competence.
A Bigger Plan?
The linking of marginal emissions to climate change’s expected effects is still in its infancy. Thus, without some bigger plan for why this kind of “analysis” should be forced into NEPA documents, it is hard to conclude that CEQ’s February 2010 proposal was on the right track.
The chasm between climate change’s effects and the emissions trade-offs that surface in the typical NEPA routine has become a major impediment to NEPA plaintiffs, as this ably-argued student work shows.
But CEQ aims to reduce paperwork, to foster a NEPA process that identifies and assesses the “reasonable alternatives to proposed actions that will avoid or minimize adverse effects of these actions upon the quality of the human environment,” 40 C.F.R. § 1500.2(e), and seeks better decisions, not better documents. “NEPA’s purpose is not to generate paperwork—even excellent paperwork—but to foster excellent action.” Id. at § 1500.1(c).
A better approach would be to plan for uses of those NEPA documents that do attempt the kind of marginal analyses described here in subsequent work improving those very analyses. If most will be wrong, the plan should be to learn from failure.
NEPA’s grandest failure to date has been its unparalleled capacity to “fly-speck” the federal government’s discrete actions while contributing too little to its overall, comprehensive reform. NEPA Section 102(2)(A) orders all agencies to
utilize a systematic, interdisciplinary approach which will insure the integrated use of the natural and social sciences and the environmental design arts in planning and in decisionmaking which may have an impact on man’s environment.
If CEQ wants to integrate NEPA’s primary tool—the impact/alternatives analysis—with system-wide planning that could replace indifference toward future generations with real risk analysis, it should study this passage. Past NEPA deliberations and documents could help agencies study their own experiences with cause/effect, uncertainty, and proof in choosing acceptable risks. What are these agencies doing to “mine” the resource of past NEPA documents? Action agencies alone possess the countless “negative” NEPA findings that have amassed over the years. Studying these documents for what they got right and wrong might help agencies correct their own biases (or those of their leaders), eventually helping us all to understand what an “integrated use of the natural and social sciences” looks like in massive public problems like climate change. And when the unusually big (perhaps really big) choice does come along, have we recognized it as such? Have we identified its critical factors? Have we analyzed them well? Answering these questions by hindsight will surely be easier than by peering into an unknown future. Future posts here will elaborate, but check out this article for more.