EPA’s PM2.5 Co-Benefits PR Trick Exposed

EPA’s 2012 Mercury and Air Toxics Standards (MATS) rule, which established maximum achievable control technology (MACT) standards for mercury and other hazardous air pollutant (HAP) emissions from power plants, is again in the news. The Supreme Court on Monday rejected a petition by Michigan and 26 other states to freeze the rule.

Petitioners complained that EPA continued to implement MATS even though the Court last year deemed the rule to be unlawful. The Court held that EPA “strayed well beyond the bounds of reasonable interpretation in concluding that cost is not a factor relevant to the appropriateness of regulating [HAP emissions from] power plants.”

Although EPA did not compare costs and benefits when deciding whether to regulate power plant HAP emissions, it did compare costs and benefits when promoting the rule to Congress and the public. EPA boasted that although MATS would cost utilities $9.6 billion to implement in 2016, it would generate $37 billion to $90 billion in health benefits in the same year (77 FR 9306).

However, EPA attributed more than 99% of the quantified benefits to coincidental reductions in fine particulate matter (PM2.5)–a pollutant not directly targeted by the rule and not classified as a HAP in the Clean Air Act. Specifically, EPA’s Regulatory Impact Analysis (p. 5-93) claimed that reductions in PM2.5-related emissions would avert 4,200 to 11,000 premature deaths in 2016–annual “co-benefits” valued by the agency at $36 billion to $89 billion.

A study by economist Anne Smith of NERA Economic Consulting finds that even if we accept the epidemiological literature supporting an association between PM2.5 at current levels and mortality (skepticism is justified), the MATS rule’s co-benefit estimates are flimflam. About 99 percent of the PM2.5 reductions attributable to MATS are projected to occur in areas already in attainment with EPA’s National Ambient Air Quality Standard (NAAQS) for PM2.5. Under the Clean Air Act, NAAQS are to be set at a level requisite to protect public health, allowing an adequate margin of safety.

In monetizing the co-benefits of collateral PM2.5 reductions in MATS and numerous other rules, EPA assumes that reductions in PM2.5 levels save as many lives in areas below the NAAQS as in areas above the NAAQS. That is “inconsistent” with the basic idea of NAAQS–the lowest allowable concentration  supported by the health effects literature and erring on the side of caution via incorporation of a “margin of safety.”

According to Smith, ”nearly 100 percent” of the PM2.5 co-benefits ascribed by EPA to the MATS rule and EPA’s so-called Clean Power Plan exceed what “can be reasonably inferred” from the epidemiological studies on which the PM2.5 NAAQS is based.

In a companion article, Smith’s colleague Scott Bloomberg points out a further problem for the agency. As the air gets cleaner and ambient PM2.5 concentrations continue to decline, EPA’s claims about the PM2.5 co-benefits of its regulations will become increasingly implausible.

EPA’s PR Problem 

EPA estimated that meeting the MATS rule’s mercury emission standards would cost utilities $9.6 billion in 2016 yet would achieve only $4 million to $6 million in health benefits in the same year (77 FR 9306). EPA did not even try to estimate the benefits of the Rule’s MACT standards for other HAPs such as chromium, nickel, and acid gases. For the HAP reductions that are the Rule’s statutory purpose, estimated costs are between 1,600 and 2,400 times greater than estimated benefits.

The actual benefit-cost ratio is likely even more abysmal. EPA not only had to skew the epidemiological evidence to infer health benefits worth $4 million to $6 million from the rule’s mercury reductions. It also had to make the unprovable claim that the required mercury reductions would boost the brainpower of a typical child born into a subsistence fishing household by an unmeasurable 0.00209 IQ points.

So if we consider just the rule’s HAP reductions, MATS is the poster child of all-economic-pain-for-no-environmental-gain regulation. A big PR problem for the agency.

EPA’s solution, as noted above, was to claim gigantic co-benefits from coincidental reductions in PM2.5-related emissions.

Exposing EPA’s Co-Benefits PR Trick

Smith finds a pervasive ”quantitative inconsistency” between the rationale EPA uses to set NAAQS and the estimates of public health benefits the agency publishes in its Regulatory Impact Analyses (RIAs), non-NAAQS rulemakings, fact sheets, press releases, and speeches to inform (lobby) Congress and the public.

Smith begins that for PM2.5 and ozone, epidemiological studies do not find a sharp threshold below which all adverse health effects can be ruled out. How then does EPA decide where to set the NAAQS so that allowable pollution concentrations are low enough to protect public health with an adequate margin of safety?

Basically, it’s a matter of expert judgment. As explained in the 2012 NAAQS for PM2.5, EPA examines the literature reporting associations of PM2.5 and health effects, assesses the uncertainties, and decides “where the evidence of associations is strongest” and, conversely, where the agency “has less confidence in the reported associations.” EPA sets the NAAQS at the point where its “confidence in the magnitude and significance of the associations is reduced to such a degree that a standard set at a lower level would not be warranted to provide requisite protection that is neither more nor less than needed to provide an adequate margin of safety.”

Quoting EPA’s 2008 NAAQS for PM2.5, Smith brings out the inescapable implication. The likelihood that emission reductions will produce public health benefits must be less in NAAQS attainment areas than in non-attainment areas. In other words, EPA must have less confidence that emission reductions will produce health benefits in NAAQS attainment areas than in non-attainment areas.

In fact, Smith argues, EPA’s confidence in the health benefits of emission reductions in attainment areas must be close to zero:

In order for a selected NAAQS level to be deemed as requisite to protect the public health, EPA’s subjective probability that the relationship exists at and below the selected NAAQS level must, logically, be very nearly zero. (Indeed, the subjective probability of continued effects must fall to nearly zero at an ambient concentration somewhere above the selected NAAQS level. This is because the NAAQS needs to include at least some margin of safety, and thus must be set at least somewhat lower than the level where expected risk is deemed to become too small to be considered a public health concern.)

Yet in the MATS and Clean Power Plan rules, EPA monetizes PM2.5 co-benefits as if the probability of health effects at levels below the NAAQ were equal to the probability above the NAAQS. The inconsistency is glaring:

At the same time that EPA is setting NAAQS at levels where it has minimal confidence that the public health is affected at lower concentrations, the Agency’s RIAs are giving the same weight to risks calculated for population exposures below the NAAQS level as they do to risks calculated for population exposures above the NAAQS level. That is, RIAs assume elevated hazards exist with 100 percent certainty for all ambient pollutant exposure levels down to a zero concentration, inconsistent with EPA’s judgments (formed when assessing those pollutants’ hazards), which imply nearly 0% certainty. EPA does not explain or try to justify why data that are too uncertain to use in the NAAQS preamble context are certain enough to use in the RIA context.

PM2.5 Co-Benefits from MATS and Clean Power Plan Nearly Zero 

The MATS rule’s collateral PM2.5 reductions will supposedly save 4,200 to 11,000 lives a year. But, notes Smith, “A figure in the MATS RIA reveals that over 99 percent of those projected benefits are projected to occur in areas where the PM2.5 levels will already be below the PM2.5 NAAQS of 12 μg/m3 (Figure 5–15 on p. 5-102 of Ref. 21).” The point? “If the MATS rule’s co-benefits are calculated probabilistically, accounting for the very low subjective probability that EPA assigned to the existence of the PM2.5-health effects relationships at levels below the NAAQS, the resulting estimate of expected benefits from the MATS rule becomes nearly zero.”

EPA’s RIA for the Clean Power Plan projects PM2.5 co-benefits of up to 4,100 lives saved in 2020 and 6,200 in 2030. In this case, too, Smith comments, “more than 99 percent of the co-benefits would be discounted if health risks below the NAAQS are assigned a much lower probability (or confidence weight) of risks above the NAAQS.”

PM2.5 Co-Benefits Increasingly Implausible

In a commentary article, NERA economist Scott Bloomberg notes EPA’s overstatement of PM2.5 co-benefits increases over time:

As the ambient air in the U.S. gets cleaner, a greater share of the population will be living in areas where confidence in the continued association between PM2.5 and mortality is near zero. Thus, the degree of overstatement in co-benefits estimates from one regulatory analysis to the next has been increasing over time and will continue to do so.

The RIA for the Clean Power Plan projects tens of billions of dollars in annual health benefits from PM2.5 reductions. Recall that EPA’s NAAQS for PM2.5 is 12 μg/m3. According to Bloomberg, recent NERA analysis ”determined that 99 percent of the 2025 PM2.5 precursor emission reductions in the proposed climate rule were projected to occur in counties with an expected PM2.5 concentration in 2020 less than 12 μg/m3, of which 97 percent are below 10 μg/m3 and 55 percent are below 7.5 µg/m3.”

Bloomberg finds that “the proposed climate rule has double the co-benefits in areas with expected PM2.5 concentrations very far below the ambient air quality standards (55 percent versus 27 percent in areas less than 7.5 µg/m3), while the fraction of co-benefits in areas at least 15 percent below the air standard (i.e., less than 10 µg/m3) has also increased substantially (i.e., to 97 percent from 89 percent).”

Why so?  ”The simple reason for this is that the mercury co-benefits were based on air quality projected in 2015, while the Clean Power Plan co-benefits were based on air quality from 2020 and later—a much cleaner environment due to a very large number of emissions regulations poised to take effect after 2015.”

In short, the climate rule’s PM2.5 co-benefits ”are even more unreliable and overstated because a far greater share of those co-benefits are associated with ever lower PM2.5 concentrations for which the EPA itself has significantly reduced confidence in PM-2.5 mortality associations.”

Bloomberg concludes that, unless EPA mends its ways and makes its impact assessments consistent with its NAAQS determinations, the agency’s benefit estimates will become increasingly overstated and less credible “in each incremental air regulatory impact analysis going forward.”