Chapter Comments on the Clean Power Plan (corrected)

The following chapter letter (drafter: Jim Wilson) was posted at the request of the President and Secretary on the comments section of the EPA website in mid-January:

James H. Wilson
Trout Unlimited
Northern Virginia Chapter
EPA-HQ-OAR-2017-0355
Comments to the Docket

Our Trout Unlimited chapter (1900 members) has a number of comments on the proposed repeal of the Clean Power Plan. Comments on the repeal proposal are followed by comments on the Regulatory Impact Analysis (RIA). In short, we are against repealing the Clean Power Plan (CPP), for the following reasons:
All of the recent climate research, including the most recent (2017) National Climate Assessment, links climate change with increased fossil fuel burning. The CPP is expected to reduce carbon emissions by more than 400 million tons per year within 15 years. These are important emission reductions that are needed to meet national and global emission reduction targets.
The electric utility sector that this regulation targets is one of the simplest to regulate with a finite set of large sources whose carbon emissions are already measured and tracked via Continuous emission monitors. This sector also has a long-term planning horizon, so it seeks regulatory certainty about what technologies it needs to invest in for units that have 50-year lifetimes. The Clean Power Plan provides such certainty.
In the repeal proposal, EPA proposes a change in the legal interpretation of section 111(d) of the Clean Air Act. The proposal goes on to say that “Notwithstanding the CAA, all of the EPA’s other section 111 regulations are based on BSER consisting of technological or operational measures that can be applied to or at a single source.” The problem with this argument is that carbon dioxide is not like all of the other (criteria air) pollutants that are regulated under section 111(d). EPA acknowledged this in the analysis that supported the endangerment finding by, for example, examining how New Source Review might be applied for greenhouse gases in a way that was consistent with how criteria pollutants were handled, but also set appropriate (i.e., higher) emission thresholds. Setting the same emission thresholds for New Source Review as criteria pollutants would result in greater regulatory burden with very little associated emission reduction. Setting BSER for fuel and technology combinations will either result in negligible emission benefits (if they are based on heat rate improvements), or will be prohibitively expensive (if they are based on carbon capture and storage). Therefore, the CPP system of setting statewide limits and letting the state governments work with their EGUs to determine a compliance strategy is an appropriate and efficient way to achieve needed carbon dioxide emission reductions.
EPA solicits comment from the public regarding the approach to estimating and reporting PM-related foregone benefits. The 2017 RIA says that EPA previously overestimated the benefits of reducing fine particle precursor air pollutants. EPA’s estimated value of those benefits is now lower (2017 analysis) because many areas no longer have PM concentrations near the levels of the ambient air quality standard. Estimates were calculated assuming that the number of PM-2.5 attributable deaths falls to zero at PM-2.5 levels at or below that lowest measured level of each of two epidemiological studies used to quantify PM-2.5 related risk of death (Krewski et al, 2009: LML=5.8 ug/m3, Lepeuli et al 2012; LML=8 ug/m3) or below the NAAQS.
The CPP repeal analysis assumes the existence of a threshold at the standard below which no health benefits occur. The health literature does not support such an assumption, a point made clear in a 2010 summary of expert opinions.
Potentially more defensible is the assumption of no effect below the lowest observed concentration in the data from the key studies. In this case, the data cannot identify the shape of the concentration-response function at lower PM-2.5 levels. At a minimum, the analysis of the foregone health benefits should not include the below the NAAQS case because the data clearly show mortality risk at fine PM levels below 12 ug/m3.
One of the key differences between the 2015 and 2017 RIAs reflects different accounting conventions. The 2015 RIA cost estimates include significant cost savings associated with demand-side energy efficiency programs. For the 2017 RIA, these savings are counted as a foregone benefit. This change has a significant effect on the cost-benefit accounting. This change in treatment appears to come from the Office of Management and Budget Guidance for Implementing Executive Order 13771. This change in analysis and the OMB guidance runs counter to 40 plus year of cost-benefit accounting being taught in US universities and implemented in EPA Regulatory Impact Analyses and other federal government analyses. Plus, this change does not make any sense. Positive costs are costs and negative costs are savings. We have known for years how to conduct cost analyses. The effort in the past 20-30 years has been on how to monetize benefits. Treating cost savings as benefits is more confusing than enlightening.
Consider the example of a person (or an organization) developing a cost-benefit analysis as input for car purchase decisions, where a range of technology options were being weighed. Besides the initial purchase cost, one of the key variables in assessing annualized costs would be the difference in expected annual fuel costs/savings. A car meeting the current fuel economy standards could be used as the baseline, and then the fuel savings for a higher mile per gallon gasoline-powered vehicle, or an electric vehicle would yield fuel savings, which should be counted as lower expected operating and maintenance costs over the expected vehicle lifetime. Fuel costs or savings need to be in the cost analysis with appropriate discounting to present value or equivalent annual cost.
The other key difference between the 2015 RIA for the CPP and the 2017 RIA for the repeal is in estimating the social cost of carbon. In the 2017 RIA, climate benefit estimates have been estimated using a measure of the domestic social cost of CO2. The 2015 RIA used a measure of the global social cost of carbon. The social cost of carbon is a metric that estimates the monetary value of impacts associated with marginal changes in CO2 emissions in a year. It includes a wide range of anticipated climate impacts, such as net changes in agricultural productivity and human health, property damage from increased flood risk, and changes in energy systems costs, such as reduced costs for heating and increased costs for air conditioning. It is typically used to assess the avoided damages resulting from regulatory actions. As noted at the beginning of this paragraph, the social cost of carbon estimates used in the 2017 RIA focus on the direct impacts of climate change that are anticipated to occur within U.S. borders.
So, what is the difference between a domestic focus versus a global focus on this social cost? While both the 2015 and 2017 RIAs include a wide range of social cost estimates, I used the central values from each analysis at a standard 3 percent discount rate for a simple comparison. For emissions occurring in the year 2030, the 2017 RIA uses a domestic social cost of carbon estimate of $7 per metric ton of CO2 emissions. The corresponding global social cost estimate used in the 2015 RIA was about $40 per metric ton CO2. This makes the estimated benefits of reducing CO2 emissions in the 2017 RIA to be less than 20 percent of what they were in the RIA performed two years ago.
A global measure is appropriate, since greenhouse gases contribute to damages around the world no matter where they occur. Furthermore, using a low value for avoided climate change related damages flies in the face of recent estimates of extreme weather events in the United States alone. A recent NOAA report shows that 2017 was an historic year for billion-dollar weather disasters in the United States. NOAA estimates that extreme weather events during 2017 cost the U.S. $306 billion—with $265 billion hurricane-related, and $18 billion from wildfire damage. While there has been limited statistical attribution work performed thus far for these weather-related disasters, the analyses performed for Hurricane Harvey show that as much as 15% of the damage from this storm is attributable to climate change. The estimated damage from Hurricane Harvey is $125 billion. Fifteen percent of this amount is $19 billion. If climate change increased the damage from all of the extreme weather events during 2017, then the climate change related damages would be $45 billion. These social cost figures of US-only damages are far more than the expected annual cost of the Clean Power Plan.
Another disappointing part of the 2017 RIA is that it does not include any new assessments of the CPP (or its repeal) using the electric utility sector model (the Integrated Planning Model) that has been the key analytic tool that EPA as relied for every major rule affecting this industry since the 1980s. EPA says that they will make available IPM-based analyses before any action that relates to the CPP is finalized. The RIA that was provided should have included IPM results. If IPM inputs have not changed since 2015, then EPA should just say that the model results are the same as in the 2015 RIA. EPA should provide any new IPM-based analyses to the public and allow an adequate period for review and comment before the final rule is issued.