The downward trend of ozone levels has slowed in recent years despite regulatory efforts at the federal and state levels. New research suggests that regulation that targets emitters when high-ozone events are most likely could be a cost-effective way of further reducing ozone levels.
Ground-level ozone causes smog, exacerbates asthma and other respiratory conditions, and damages ecosystems. Under the Clean Air Act, the US Environmental Protection Agency (EPA) sets standards for ozone concentrations, and EPA may strengthen those standards this year. Cap-and-trade programs, which put a limit on the emissions that are allowed in a given jurisdiction, also have been an important tool in reducing ozone levels by reducing emissions of nitrogen oxides (NOx), a key chemical ingredient in ozone formation. Yet, although EPA standards and cap-and-trade programs have reduced NOx emissions and ozone levels, many parts of the country continue to violate the ozone standards. To reduce ozone violations, EPA could continue tightening emissions caps and expanding the group of emitters to whom the caps apply.
In a new working paper, we explore an alternative strategy: targeting emissions that are most likely to cause violations of the ozone standards. We show that targeted regulations can substantially reduce the costs of achieving the ozone standards, as long as the relationship between NOx and ozone—which can vary widely over time and space—is predicted with sufficient accuracy.
Ozone’s Fickle Formation and Ozone Trends
Ground-level ozone forms through a complicated process. Under hot and sunny conditions, volatile organic compounds react with NOx to form ozone. The amount of ozone that forms depends on concentrations of these volatile organic compounds and NOx, as well as weather conditions. Major sources of NOx emissions include power plants, industrial facilities, and vehicles.
EPA set the current ozone standard in 2015. An area’s air quality violates the standard if the ozone concentration exceeds a certain limit too frequently. Over the past two decades, NOx emissions from power plants have decreased substantially, thanks partly to the EPA cap-and-trade programs that now cover most of the eastern United States. The precipitous decline in NOx emissions has, as intended, reduced ozone levels and high-ozone events (hours during which the concentration exceeds the limit) (Figure 1).
Figure 1. Reductions in High-Ozone Events and Nitrogen Oxide Emissions Over Time in the Northeast Region of the United States
The dashed red line indicates average NOx emissions from power plants
NOx emissions, ozone levels, and high-ozone events steadily declined until around 2013. Since then, however, improvement has been much more gradual. Part of the problem is that the electric power sector has shifted from coal to natural gas and renewable energy sources such as wind and solar. A further reduction of NOx emissions requires even greater improvements in rates of emissions by the energy system.
Toward a More Targeted Policy
Because sunlight and heat affect ozone formation, EPA designates May through September as “ozone season” and regulates NOx more strictly during this time frame. Moreover, NOx emitted from a smokestack can travel many miles and linger in the air for hours or even days before it facilitates the formation of ozone far from its source.
EPA enforces a “good neighbor” rule, which ensures that a state is responsible for the ozone in other states that its NOx emissions may cause. The importance of targeting ozone-season emissions with regional cap-and-trade programs for NOx emissions directly relates to the importance of sunlight, heat, and the long-distance transport of emissions in facilitating ozone formation. EPA also administers annual caps, though these caps haven’t been as stringent as those in the cap-and-trade programs.
EPA could try to reduce high-ozone events and help states meet air-quality standards by continuing to tighten NOx emissions caps. However, an alternative may be more cost-effective: targeting ozone-season emissions that are most likely to contribute to violations of the ozone standards. This targeting could be accomplished by introducing trading ratios, such that emissions rules are stricter for firms at locations or times of day that are more likely to see ozone violations.
Economic theory indicates that two conditions need to hold for such differentiated policy to be cost-effective: the effects of emissions on ozone levels should vary across space or time, and this variation should be predicted accurately. In our new paper, we use data from 2001 to 2019 for the Northeast and Mid-Atlantic regions of the United States to see if both conditions hold.
Figure 2 illustrates our estimates of the effects of NOx emissions on high-ozone events by hour of day for the region surrounding Baltimore and Washington, DC. At 3:00 p.m., one ton of NOx emissions increases the probability of a high-ozone event by almost three percent; at midnight, the increase is closer to one percent. These sensitivities vary regionally and over longer time horizons, too, depending on factors such as the presence of volatile organic compounds. We also find that the hourly relationship between emissions and high-ozone events is stable over time, which means that the correlation can be predicted accurately.
Figure 2. Hourly Effect of NOx on Ozone Formation in the Region Surrounding Baltimore, Maryland, and Washington, DC
How much more cost-effective would a differentiated policy be than a simple tightening of the emissions caps? We compare the costs of two hypothetical policies: one that varies an emissions price (or, equivalently, introduces trading ratios) in proportion to the marginal hourly effect of NOx on ozone formation (Figure 2), and a second that has a uniform emissions price (or, equivalently, no trading ratios).
A differentiated NOx price reduces costs by 15 to 21 percent compared to a uniform price, but the advantage depends on the precision with which the NOx-ozone relationship can be estimated. If weather conditions or other factors that contribute to ozone formation are mostly uncertain, a policymaker would have difficulty in properly setting a price.
Sophisticated tools for modeling air-parcel transport and weather could provide enough information to construct a differentiated emissions-pricing policy that is precise enough to cost-effectively reduce high-ozone events. Our research indicates that, at least in some cases, a differentiated policy may substantially reduce costs, but EPA would need to work out the details for implementation.
