Photo from Tony Mendoza
It’s no secret that many parts of the Eastern US suffer from poor air quality from smog, more technically known as ground-level ozone. What’s less well known is how much these local air quality problems literally flow from power plant emissions in upwind states. EPA has a chance to address this problem in a new cross-state air pollution rule, and I’d like to explain how EPA can include improvements in such a rule to build on its prior efforts to more effectively protect the public from dangerous ozone pollution. That’s all laid out below, but also in a more concise set of slides right here.
The Ozone Transport Opportunity
But first: when I was a kid, I learned about ozone as a beneficial molecule that helps protect the Earth’s surface from the sun’s most intense ultraviolet radiation. At the time, countries were coming together to try and shrink the growing hole in the ozone layer by agreeing to phase out chlorofluorocarbons—ozone-destroying refrigerants then in wide use in our air condition systems. Like so many things in life, context matters. While ozone in our stratosphere helps shield the Earth from harmful UV rays, ozone that forms at ground level can be devastating to our lungs and to our bodies.
Breathing ozone can be like getting a sunburn on your lungs. Ozone is a potent asthma trigger and can cause or contribute to a host of other adverse respiratory and cardiovascular effects and even premature death. Based on its harmful impacts to human health, EPA in 2015 tightened the maximum level allowable in the ambient air—the National Ambient Air Quality Standard—from 75 parts per billion over 8 hours to 70 parts per billion. Many places in the country, including a number of urban areas in the East, continue to fail to meet this health-based standard.
Ground-level ozone forms when certain “precursor” pollutants—nitrogen oxides and volatile organic compounds—interact in the presence of heat and sunlight, in the air after they are emitted. Nitrogen oxides or “NOx,” which are the primary driver of ozone formation in the Eastern United States, are produced by a wide range of combustion processes. An outsize share of the NOx pollution in the United States comes from fossil fuel-fired power plants. Although the amount of coal-fired power generation has been in decline in recent years, coal plants remain some of the largest sources of NOx pollution in the country, and gas-fired power plants are major sources of NOx pollution as well.
Ozone and ozone precursor pollutants travel on the wind and can be carried long distances. Consequently, the areas of high ground-level ozone concentrations can be far removed from the original sources of the pollution. The problem of “ozone transport” is particularly acute for states on the East Coast, which are located downwind from fossil fuel plant heavy Mid-Atlantic and Midwestern states.
In fact, a number of the downwind states that suffer from the worst air quality in the East generate only a small fraction of their ozone-forming pollution in-state. Connecticut, for example, which routinely registers the highest ozone levels east of the Mississippi River is responsible for only 5-6 percent of the ozone problem at its worst air quality monitors. By contrast, other upwind states, including New York, New Jersey, and Pennsylvania, each contribute between 8 and 20 percent of Connecticut’s ozone problem. This means that for Connecticut to meet EPA’s ambient ozone standard, it is dependent on pollution reductions occurring in upwind states.
Fortunately, Congress understood this problem when it drafted and updated the Clean Air Act, including in the Act a “Good Neighbor” provision requiring that each time EPA strengthens an ambient air quality standard, all states must adopt measures prohibiting emissions that contribute significantly to difficulties faced by any other state in achieving and maintaining that new air quality standard. If upwind states fail to adopt adequately protective measures—as has routinely occurred—the obligation shifts to EPA to fill that gap.
EPA has been working for some time to address cross-boundary transport of ground-level ozone under the Act’s “Good Neighbor” provision. It has developed a series of transport rules intended to tackle upwind states’ transport obligations for the 1997 and 2008 ozone standards. These rules have generally focused on states that contribute at least 1 percent of the level of the air quality standard (currently set at 70 parts per billion) to a nonattaining downwind state, and on reducing NOx emissions from power plants in those upwind states. In developing transport rules for ground level ozone, EPA determines where the state-to-state pollution linkages are, and then evaluates the emission reductions and air quality benefits that could be achieved by controlling emissions from upwind power plants at a range of cost thresholds, historically selecting a single cost threshold applicable to all upwind states. Using that cost threshold, the agency then develops state-specific NOx emission caps or “budgets.” Power plants subject to EPA’s transport rules are required to hold emission allowances covering their NOx emissions. The allowances are freely tradable, so if a power plant’s actual emissions exceed its allowance allocation, it can purchase allowances from other plants, although total emissions in an upwind state cannot exceed the state cap by more than a fixed percent without incurring penalties. These rules have resulted in important reductions in NOx emissions from power plants, but they have yet to meet the Act’s promise and direction to bring downwind states into attainment of ozone ambient air quality standards.
The 2015 Ozone Standard Transport Rule Offers EPA Another Opportunity to Eliminate the Upwind-Downwind Problem
Now, thanks to litigation by the Sierra Club and allies and by impacted downwind states, EPA is under a court-ordered deadline to release a draft of its newest transport rule to implement the 2015 ozone standard. EPA has an opportunity to build upon and strengthen the design of its prior transport rules, which have not proven sufficient to address downwind ozone pollution or ameliorate most upwind states’ significant pollution contributions.
There are two primary reasons that EPA’s prior ozone transport rules have underachieved. First, by relying on a single cost threshold in setting the budgets for all states and at the same time making sure no upwind state carries more than its share of the emission reductions, EPA has created a lowest common denominator approach that exacerbates under-control and forgoes more effective control measures in upwind states that contribute the most to downwind pollution problems. Second, EPA has routinely allowed the state caps to lose their effectiveness in limiting NOx and driving implementation of cost-effective pollution control strategies by carrying forward a large bank of inexpensive allowances accumulated during prior programs or earlier years of the current program, failing to adjust the caps when units retire, and allowing trading of emissions across state borders.
EPA Should Distinguish the Biggest Ozone Contributors and Require Them to Clean Up More in its Next Rule
On the first issue, in EPA v. EME Homer City Generation, L.P., the Supreme Court upheld EPA’s transport rule for the 1997 ozone NAAQS, opining on the question of balancing the potential for overcontrol of upwind emissions against the risk of under-control. Writing for the majority, Justice Ginsburg cautioned against both over- and under-control of emissions from upwind states: While no upwind state could be asked to do more than eliminate its “significant” contribution to a downwind state’s nonattainment, EPA was not to let significant contributions persist if downwind air quality problems remained unresolved. Nevertheless, in developing transport rules following EME Homer City, EPA has consistently erred in the direction of under-control to the detriment of air quality and public health in downwind states.
One primary cause of this chronic under-control is EPA’s use of a single, uniform cost threshold to determine available emission reductions and set allowance budgets for all upwind states. When coupled with the Agency’s rigid avoidance of over-control, this approach yields transport rules that, at best, eliminate significant contributions for the most peripherally connected upwind states while having little impact on the highest contributing states and on improving downwind air quality.
In its forthcoming rule, EPA should correct this problem by adding a second, higher cost threshold applicable to upwind states that more significantly contribute to downwind ozone. Under this approach, budgets for some states (those that contribute above the 1 percent threshold, but below a cutoff (e.g., 2 parts per billion) to a downwind nonattainment or maintenance area) would be based on controls achievable at one cost threshold, and other states (those that contribute above that higher contribution cutoff (the 2 parts per billion) to a downwind nonattainment or maintenance area) would be based on controls achievable at a second, higher cost threshold.
EPA adopted this two-tier approach for a different pollutant—sulfur dioxide, a precursor to fine particulate matter—in a prior transport rule. The Supreme Court acknowledged EPA’s approach, noting that, although EPA’s $500 per ton threshold resolved air quality problems for one group of upwind states, “[f]or another group of States . . . this level of controls would not suffice,” and EPA selected a $2,300 per ton cost threshold for these states. EME Homer City Generation, L.P., 572 U.S. 489, 502 n.7 (2014). Although the Supreme Court did not expressly rule on the use of multiple tiers, this approach is more faithful to the Court’s instructions in EME Homer City than EPA's current approach because it can be implemented in a manner that continues to avoid over-control, but significantly reduces the degree to which the rule results in under-control in states that are heavily linked to downwind nonattainment.
This approach is also more equitable than EPA's current transport framework. There is presently a large disparity between upwind and downwind states in terms of the cost per ton being used to determine whether to require certain NOx emission controls. Recent transport rules have been based on control cost thresholds of $1,400-1,600/ton NOx for upwind states while some downwind states (e.g., Connecticut) are using cost/ton thresholds above $13,000/ton NOx to endeavor to bring in-state areas into attainment with the ozone standard, as required by the Clean Air Act. Under the proposed approach, states that are larger contributors to downwind nonattainment would be required to control their sources more strictly than those that are more modest contributors, at a cost that is closer to what the downwind states themselves are imposing on their own in-state sources. There are still huge amounts of coal-fired generation without modern pollution controls installed, as shown in the table below.
To implement this suggested two-tier approach, EPA could begin as it has with other transport rules and select an initial control efficacy that would apply to all upwind states linked to downwind nonattainment and maintenance issues, at the level where emission reductions and air-quality improvements are maximized relative to the cost-per-ton of the controls considered. EPA would develop state-specific NOx budgets based on this control efficacy and verify that it had avoided over-control for all upwind states (and that it has not resolved nonattainment and maintenance issues for all downwind areas). Rather than ending its analysis there, however, if downwind nonattainment or maintenance issues persist, EPA would then focus on a subset of upwind states with large linkages to those remaining downwind nonattainment or maintenance monitors (e.g., above a 2 parts per billion contribution threshold). For this subset of states, EPA would conduct a second modeling analysis and develop a second, higher cost/ton threshold that would be used to establish revised (tighter) budgets for these states. The final NOx budgets for these larger contributor states would reflect the additional emission reductions that could be achieved at the higher cost threshold. EPA would again verify that no overcontrol was occurring by ensuring that no upwind state was required to reduce its contributions below the initial 1 percent screening threshold and that no downwind area could be brought into attainment with the ozone standard if all upwind contributors applied controls at a lower cost threshold.
In theory, so long as downwind nonattainment or maintenance issues were modeled to persist even after this second round, this approach could be extended to additional tiers and rounds of budget tightening based on higher cost/ton thresholds for the states with the largest contributions to those nonattainment and maintenance areas. However, adding even a single additional round of budget tightening, as described above, would significantly improve the robustness of a future transport rule.
EPA Should Improve Requirements Around Trading and Banking of Pollution Allowances in its Next Rule
On the second issue, EPA could further ensure the achievement of the statutory goal, within the framework upheld by the Supreme Court, by scaling back or eliminating the option to bank allowances for use in future years, by accounting for unplanned retirements of fossil-fuel-fired electric generating units, and by restricting the interstate trading of allowances.
In the past, EPA has permitted sources to carry some allowances over from a previous trading program designed to address the interstate transport of ozone under the former standard. In doing so, however, the agency converted the number of banked allowances to a lower quantity, but one that was still large enough to allow the sources in all states to emit NOx up to their state’s variability limit, an amount of additional NOx emissions that EPA allows above a state’s budget to accommodate unpredictable changes in power-sector operations. This approach provided compliance flexibility and an incentive for early emission reductions, while lessening the risk that excessive banked allowances, available at low prices, would lead sources to exceed their state’s variability limit. EPA could take a similar approach in its rule for the 2015 ozone standard, reducing the number of allowances banked to the sum of the states’ variability limits, or some lower, predetermined level, for each group of states at a certain cost threshold (see above for more on the concept of multiple tiers). Beyond addressing allowances from previous programs, EPA has also generally allowed allowances to carry over from year-to-year within a program—but that approach could also be changed, such as by using this same conceptual maximum level. Thus, in each subsequent year of the program, EPA could again apply an adjustment factor to the banks to arrive at the predetermined level, which offers an appropriate degree of compliance flexibility without imperiling the integrity of state caps.
Because neither the statute nor the Court requires any carry over of allowances, EPA also has the option to decide not to carry over any allowances from the previous program, or from year to year in its new program. As noted above, EPA has a lot of experience with the ozone transport program, since its first transport rules in 1998, and that experience has shown that large allowance banks tend to develop quickly, and pose a threat to the integrity of the program. EPA could proactively address this problem by disallowing banking. There may be less of a need to prompt early emission reductions (a potential benefit of allowing the carryover of allowances from year to year) if EPA establishes budgets based on more-effective controls in the highest-contributing states, as discussed above. Moreover, there are other ways to promote compliance flexibility. For example, EPA could create a reserve of allowances available, up to the state’s variability limit, at a premium price at or above the pollution-control cost threshold that determined the state’s budget. The higher price and limited number of allowances would guard against exceedances of the variability limit, i.e., the level that EPA has determined will reasonably accommodate variation in power sector operations.
One factor that has contributed to outsized banks is that greater-than-anticipated retirements have supplied far more allowances than were needed to cover emissions from replacement generation, resulting in excessive allowances and a lost opportunity to achieve greater emissions reductions. To address this issue, EPA could adjust state budgets from year to year to account for unplanned source retirements, just as it does initially in taking planned source retirements into account when developing state budgets in the first instance. Here’s how that could work: EPA could adjust state budgets after each ozone season by removing allowances equivalent to a retired electric generating unit’s allocation, for example. The retired unit could immediately lose those allowances going forward, or it could continue to receive its allocations for a specified time—in either case, EPA would remove the appropriate number of allowances from the overall budget (perhaps from a set-aside of allowances that EPA could establish within the budget that would be held in reserve and ordinarily—i.e., when there are no unscheduled retirements—redistributed to existing units proportionally to their allocations). The latter approach could maintain a financial incentive for units to retire sooner while preserving other units’ preexisting allocations—although such an incentive might not be needed or desirable. To parallel the drawdown on allowances from unscheduled retirements, EPA could add allowances to a state’s budget reflecting anticipated emissions from unscheduled additions of electric generating units. This approach would better reflect the true makeup of the fleet in each state, ensuring that all of the operating units have a consistent incentive to deploy the controls necessary to eliminate the state’s significant contribution to downwind nonattainment. Indeed, as noted above, EPA already accounts for scheduled retirements and additions in budget-setting; doing the same for unscheduled retirements and additions would improve consistency and remove the unsupported assumption in previous rules that unscheduled retirements more or less cover emissions from new units, even though the retired capacity may be replaced by much lower-emitting capacity. Failing to account for unscheduled retirements sacrifices public health benefits, compromises the integrity of the program, and ultimately means that states could fail to eliminate their significant contributions to downwind nonattainment by deploying cost-effective pollution controls at their remaining sources.
Finally, there is no statutory requirement to permit interstate trading, and the availability of interstate trading of allowances (and thus emissions) in past programs has meant that states struggling to remain within their budgets could rely on emission reductions in other states to comply, thereby worsening downwind air quality (and public health). In a new rule, EPA could prohibit interstate trading altogether, or restrict interstate trading such that sources in a state whose budget reflects a given cost threshold can only trade allowances with sources in a state whose budget reflects the same cost threshold. Restricting trading to sources within states at the same cost threshold would prevent flows of emissions from states with less stringent budgets to states required to do more to address their large contributions to downwind air-quality problems. On the other hand, allowing trading across all sources in the program could prompt more emission reductions in states with budgets reflecting a lower cost threshold. But, whether or not EPA allows some limited interstate trading, the agency should ensure that all states will eliminate their significant contributions to downwind air-quality problems while achieving the greatest overall emission reductions.
Closing Thoughts
EPA now has an opportunity to improve and revitalize its program to address transport of ozone pollution and improve air quality in downwind states. Learning from years of experience implementing similar regulatory frameworks, EPA can adopt reforms that would better fulfill states’ and EPA’s obligations to eliminate significant contributions to nonattainment and maintenance issues in neighboring states. EPA should consider adopting multiple pollution-control cost thresholds for states contributing different levels of ozone pollution to downwind states, annually adjusting or eliminating banks of allowances, accounting for unscheduled retirements of fossil-fuel-fired electric generating units, and limiting interstate trading of emissions. Each of these reforms, individually and in combination, would advance the Clean Air Act’s goals and enable rapid compliance with the more health-protective 2015 ozone standard.