EPA’s Problematic Case for Rescinding Its Endangerment Finding

This post was originally published by The Regulatory Review. Reprinted with permission.

Last week, the U.S. Environmental Protection Agency (EPA) overturned its 2009 finding that greenhouse gas emissions from vehicles endanger human health and welfare. EPA argued that it lacked the legal power to regulate these greenhouse gas emissions. As I have written elsewhere, EPA’s arguments are difficult to square with a landmark decision by the U.S. Supreme Court, Massachusetts v. EPA.

EPA advances its five most significant arguments for its latest deregulatory action while brushing aside the Massachusetts precedent. Even apart from Massachusetts, the five arguments advanced by EPA are quite problematic.

Defining Air Pollution. First, the relevant statutory provision—Section 202(a) of the Clean Air Act—empowers EPA to regulate the “emission of any air pollutant” from vehicles under certain circumstances. EPA now contends that the term “air pollutant” includes only those substances that cause harm by direct exposure to the substance in the air. EPA calls these “local and regional” pollutants. EPA now says that the Clean Air Act does not cover substances like greenhouse gases that cause harm indirectly rather than by their “mere presence” in the air.

One problem with EPA’s current argument is that the Clean Air Act broadly defines “air pollutant” to include any substance emitted into or entering the air, rather than limiting it to a certain way of causing harm. EPA’s effort to explain away this language is unconvincing. Another problem is that some greenhouse gases also cause direct harm: Methane is harmful when inhaled, and carbon dioxide causes ocean acidification. That makes them “air pollutants” even under EPA’s definition. In addition, the Act defines public welfare to include “weather, visibility, and climate,” terms which go beyond direct exposure effects.

The Linkage Requirement. Second, EPA now argues that vehicle greenhouse gas emissions lack the right kind of connection with danger to human health and welfare. Section 202 of the Clean Air Act requires EPA to regulate whenever vehicles “cause or contribute” to pollution that “endangers public health or welfare.” The 2009 endangerment finding analyzed these two elements separately. Under that analysis, EPA could regulate vehicle emissions of any substance causing significant harm, even if vehicles were only one of many sources of the substance. EPA now says that U.S. vehicle emissions, standing alone, must cause significant harm. Moreover, EPA says, this must be proven for each specific category of motor vehicles considered separately.

There is no sensible basis for limiting this interpretation to greenhouse gases, yet EPA now claims its rulemaking will have no effect on its regulation of conventional pollutants, which have not followed this new template EPA has created for greenhouse gases. That seems indefensible. Also, under EPA’s theory, the agency could manipulate endangerment findings by either lumping types of vehicles together or subdividing them into multiple categories. Most significantly, this theory would prevent EPA from regulating new sources of pollutants when different types of vehicles collectively caused significant harm, but no single source category by itself crossed the significance threshold.

Relatedly, EPA argues that the statute prevents it from bifurcating its process to consider endangerment first and the required regulatory response separately. This argument seems to run afoul of the Court’s emphasis in Vermont Yankee Nuclear Power Corporation v. Natural Resources Defense Council on the extremely wide discretion agencies have in shaping their own procedures. And the Clean Air Act itself often takes this bifurcated procedural approach in other contexts, requiring that an agency make a risk determination first and then set a regulatory standard.

The Major Question Doctrine. Third, relying on West Virginia v. EPA, the agency argues that regulation of greenhouse gases necessarily involves a major question, which requires extra-clear statutory authority. West Virginia did hold that a particular climate regulation violated the major questions doctrine, which states that, for an agency to regulate a matter of exceptional political or economic significance, it needs clear, explicit authorization by the U.S. Congress. But that was because of the scope of that particular regulation. In Utility Air Regulatory Group v. EPA, which the current EPA also relies on, the Court struck down one part of a climate regulation as posing a major question, but it upheld another part of the same regulation. In its latest deregulatory action, EPA seems to be overreading these precedents.

Insignificant or De Minimis Harm. Fourth, the United States is the world’s second-largest emitter of greenhouse gases, and transportation is the U.S. sector with the highest emissions. Yet EPA currently claims these emissions are too insignificant to matter, causing only 1 percent of warming by its reckoning. But a 1 percent increase can be significant when it is 1 percent of an extremely serious harm. By EPA’s reasoning, a pollutant causing 1 percent of U.S. mortality would not be a material danger worthy of regulation. There are roughly three million U.S. deaths per year, so 1 percent would be 30,000 deaths, which clearly would be significant. Given current estimates of the social cost of carbon, a rough calculation suggests that a 1 percent increase in U.S. climate impacts over the multi-decade period used by EPA to estimate costs would about equal the costs EPA attributes to greenhouse gas regulation that the agency terms “immense.”

EPA’s Fallback Argument. Finally, EPA argues that, even if the endangerment finding were valid, it still could not legally regulate vehicle greenhouse gases. EPA says that “under any legal standard, it is unreasonable for the EPA to impose trillions of dollars in costs on manufacturers and American consumers in exchange for results that do not materially further congressional objectives—at least absent an extraordinarily clear indication in the statutory text.” This is a questionable argument on several counts. To begin with, as noted above, EPA’s effort to trivialize the benefits of regulating vehicle emissions is shaky.

Moreover, the claim of trillions of dollars in costs is quite a stretch. EPA’s fact sheet refers to $1.3 trillion in increased vehicle costs over 28 years, which is not “trillions.” Remarkably, EPA also found that eliminating the endangerment finding would cost consumers nearly $1.5 trillion due to the much higher maintenance and repair costs for gas vehicles compared to electric vehicles. So basically, the endangerment finding saved Americans money even by EPA’s current count.

On their own merits, the first two of EPA’s five arguments—defining “air pollution” and demanding a tighter linkage between greenhouse gases and human health—are the most plausible of the lot, although they are still problematic. But these arguments are the hardest to square with Massachusetts v. EPA. This all amounts to a questionable basis for an agency charged with environmental protection to turn its back on a problem of this magnitude.