Blog Post

State Implementation of the Clean Air Act: Frequently Asked Questions

Apr 1, 2014 | Dallas Burtraw

In 2007, the US Supreme Court affirmed the authority of the Environmental Protection Agency to regulate greenhouse gases under the Clean Air Act. That set in motion a series of events leading the Obama administration to explore how to regulate greenhouse gases using the Clean Air Act and, in particular, how to regulate existing stationary sources such as power plants. According to RFF Darius Gaskins Senior Fellow Dallas Burtraw, this is the most important climate-related decision that any administration has faced in the United States, given the potential emissions reductions and the precedent of regulating existing stationary sources for the first time.


  1. Was the Clean Air Act designed to regulate greenhouse gases?
  2. How is the Clean Air Act being used to regulate greenhouse gases today?
  3. What is the timeline for regulating existing power plants under the Clean Air Act?​

Implementation Approaches

  1. What are some ways that states might regulate greenhouse gas emissions under the Clean Air Act? 
  2. Have any of these approaches that states might use to implement these regulations been tested?
  3. How is a tradable performance standard different from cap and trade? 
  4. How could a tradable performance standard or cap and trade impact electricity prices? 
  5. Could each of the states choose a different approach for implementing the Clean Air Act regulations?

Stringency and Costs

  1. How stringent will the regulations on existing power plants be, and how much will it cost to implement the regulations? Are these dimensions related?
  2. Is using the Clean Air Act a cost-effective approach for regulating greenhouse gas emissions from power plants? 

Challenges and Advantages

  1. What challenges might the states face as they begin developing their implementation plans for Clean Air Act regulation? 
  2. What are the advantages or disadvantages with a state-centered approach for implementing the Clean Air Act regulation? 

1. Was the Clean Air Act designed to regulate greenhouse gases?

The Clean Air Act doesn’t say much about greenhouse gases in the statute directly. It’s an institution that is designed to consider scientific evidence of new environmental challenges and it’s been used that way many times.

The Clean Air Act was originally going to be used to regulate emissions from power plants back in 2001. The Bush administration had suggested that emissions from power plants—greenhouse gas emissions—should be regulated. Christine Todd-Whitman, the administrator for the US Environmental Protection Agency (EPA) at that time, was putting together plans to initiate such regulation. She was going on the Sunday morning talk shows discussing how regulation was going to take shape under the Clean Air Act.

But then President Bush famously decided that greenhouse gases did not qualify as air pollution as defined under the Clean Air Act. That’s what began a chain of dominoes.

One of the special provisions of the Clean Air Act is the citizen lawsuit provision. It’s not unique, but it’s rare, and it’s very powerful. Citizens and states have the right to sue EPA for enforcement and have other various rights under the Clean Air Act. This provision and the special solicitude toward states—as well as the status of states as holders of lands vulnerable to climate change (sea level rise)—led to a lawsuit and decision by the Supreme Court in 2007 (in Massachusetts v. Environmental Protection Agency) to affirm EPA’s authority to regulate greenhouse gases under the Clean Air Act.

With that authority affirmed, and under the threat of additional lawsuits, EPA began to look at whether or not greenhouse gases constituted a threat to human health and welfare. Although many other scientific bodies already found that to be the case, EPA made its formal finding in 2009 that greenhouse gases were a threat. It was compelled to begin to mitigate that threat and, in fact, was faced with another round of private lawsuits to try to force it to do so.

Return to Top

2. How is the Clean Air Act being used to regulate greenhouse gases today?

EPA took a three-pronged strategy to regulating greenhouse gases under the Clean Air Act. In January 2011, EPA released the Mobile Source Standards, which are quite important. They lead to a 5 percent per year improvement in the energy efficiency of the vehicle fleet in the United States. Those now are going to unfold all the way through 2025, leading to a fuel efficiency standard of around 54 miles per gallon, on average, for new vehicles in that year.

The second prong was requiring pre-construction permits for major stationary sources. That was also put in place in January 2011.

The third had to do with standards of performance for operation of the facility. It’s been “start again, stop again, start again” as the EPA has tried to figure out how to approach this issue—and it’s the most important one. They've chosen to develop this approach under the portion of the Clean Air Act that issues performance standards, Section 111. These operating performance standards will be applied to all the major sectors, starting with electricity, but also including refining, pulp and paper, iron and steel, and so on.

These performance standards are distinguished between standards that apply to newly constructed sources (because they should be state-of-the-art), and standards that apply to existing sources, which could be upgraded such that they approach state-of-the-art when it’s feasible for them to do so. EPA issued its draft final standards for new fossil-fired electricity generating units in September 2013. They aren't finalized yet, but nearly so.

The main event is how EPA chooses to regulate the existing electric generating units under the performance standard provision of the act. These units are responsible for nearly 40 percent of the country’s carbon dioxide emissions.

Return to Top

3. What is the timeline for implementing regulations on existing power plants under the Clean Air Act?

EPA now is operating under instructions from President Obama. In his June 2013 climate speech he described a schedule. He charged EPA, under Section 111 of the Clean Air Act, with regulating and finalizing the standards as they would apply to new sources and then to meet a schedule with respect to the existing sources.

In June 2014, draft guidelines are supposed to be in place for the existing sources. In June 2015, final guidelines are supposed to be in place. The states have 13 months—until June 30, 2016—to develop implementation plans that describe how they will meet the guidelines that EPA has put in place. It’s a state-driven process in which the states develop implementation plans and then EPA approves those plans, depending upon whether or not they meet the guidelines that EPA has put in place.

Return to Top

4. What are some ways that states might regulate greenhouse gas emissions under the Clean Air Act?

I think that there are three general approaches that introduce flexibility in some form and might be used to regulate under the Clean Air Act. The first is a tradable emissions rate performance standard. I know that’s a mouthful. The way to think about it is that its emissions rate averaging.

An emissions rate is the tons of the pollutant, of carbon dioxide, per megawatt hour of electricity generation from a facility. As emissions rate averaging suggests, this approach would enable two or more facilities to average their emissions rate in order to achieve a relatively clean emissions rate performance standard.

Averaging is important because some facilities can reduce their emissions rate at a lower cost than other facilities. But even more importantly, averaging allows a company to move its generation away from an electric generating unit that has a high emissions rate and do more generation at units that have lower emissions rates, and lower its overall average for emissions rate in the process.

You can imagine flexibility to allow that kind of averaging within commonly owned units by a company or all the units in a state, or possibly intrastate averaging on a national basis because greenhouse gases, carbon dioxide in particular, don’t have any local affects, specifically. So averaging on a broad geographic basis is quite compelling.

A second approach would be for EPA to provide a methodology that converts the emissions rate into a tonnage count of emissions, or describes how the states could do that. It is likely EPA starts with an emissions rate approach because performance standards have typically been defined in terms of an emissions rate, and the portion of the Clean Air Act that EPA has chosen to regulate addresses standards of performance. But one can convert to emissions tonnage by taking the rate—the tons per megawatt hour—and multiplying it by the activity level, or the megawatt hours, to end up with just tons. Then the states have a budget to work with. It’s essentially like a cap. Flexibility in this context would be cap and trade—maybe within the state, maybe across states. And, states might choose to use an emissions fee rather than a cap-and-trade program to reduce emissions and achieve the cap. It would not be a federally driven program; it would rather be a state-driven program in which EPA authorizes the flexibility, including the possibility of trading emissions allowances across state lines.

The third approach is a resource planning approach. Some states might want to be given an emissions rate standard or a mass-based target, but they may not want to launch a trading program. They might want to just try to demonstrate that all the other measures that they are already doing—such as renewable portfolio standards to promote renewable energy, and energy efficiency programs, and other policies—might lead to enough emissions reductions already or might be ramped up so they can achieve the emissions target that’s handed to them by EPA. I would call this a portfolio approach, because it accounts for a package of activities implemented at the state level.

Return to Top

5. Have any of these approaches that states might use to implement these regulations been tested?

A performance standard was actually the first major experiment that we have in incentive-based regulation. It was used in environmental policy in the United States in the 1980s with the phase-out of lead from gasoline. It’s widely viewed as having been successful and saved hundreds of millions of dollars for refineries compared to the alternatives that were being considered at the time.

The conversion from an emissions rate to a mass-based approach that gives states a budget to work with has also been done before. It was used under the NOx Budget Program, which involved 19 states in the northeast in a large trading program to reduce emissions of nitrogen oxides (NOx) from power plants.

That was also developed by beginning with an emissions rate standard that was tied to technology that had recently become available (selective catalytic reduction) and multiplied the emissions rate by activity levels at each of the states to arrive an emissions budget. EPA calculated a budget for each state and then gave the states the latitude to opt into a trading program if they chose to do so—and every state chose to do so because they could save so much money doing it that way.

The third approach, a portfolio of policies implemented at the state level, is also something that has been tried before. That is what we see in very many states today, where policies have contributed to a reduction in the growth or an actual decline in greenhouse gas emissions. So these approaches are not new, and they are the major alternatives that are on the table for EPA to consider.

Return to Top

6. How is a tradable performance standard different from cap and trade?

The tradable performance standard approach has one advantage, politically speaking, in that it cannot be branded as cap and trade because there is no emissions cap. It’s actually a rate-based approach, although it could be very stringent. Nonetheless, if you were to generate more electricity, you would have more emissions. The performance standard would ensure that you're doing so in an as efficient way as possible.

That could be a political advantage because in the contemporary policy debate, cap and trade has been painted as controversial. EPA has said that it will not use the Clean Air Act to impose cap and trade through the back door when Congress already has decided not to do so through comprehensive climate legislation.

The way that cap and trade might emerge is if states decide to opt into using cap and trade, but apparently they are not going to be mandated to do so by EPA.

Return to Top

7. How could a tradable performance standard or cap and trade impact electricity prices?

A tradable performance standard is remarkable for the fact that it will lead to virtually no change in the national average electricity price. There will be some regional differences, but on an average, nationally, the change in electricity price will not be observable to the average rate payer.

Cap and trade—if the states were to opt into cap and trade—can lead to bigger changes in the electricity price but it can also be designed such that that doesn’t happen. For example, in California’s cap-and-trade program, the allowance value is returned to consumers and it nets out the cost of the program. Most electricity consumers do not see a change in their electricity prices due to cap and trade in California. In the northeast trading program—the Regional Greenhouse Gas Initiative—the allowance value is invested in end use energy efficiency that helps reduce overall demand and keep electricity prices low. I think that one or the other of these approaches would be embodied in what many states would do, if they chose to do cap and trade. In principle, states could use a revenue raising auction or even a tax to achieve comparable outcomes and those options would raise electricity prices, but they appear less likely to happen.

The disadvantage to economists about this is just that—we don’t see a change in electricity prices. Economists want consumers to see that change because it gives an incentive for them to go out and buy more efficient air conditioners and refrigerators and other appliances and make decisions about energy reflecting the full social cost of the energy they use.

Because electricity prices would not change, one of the unanticipated, or unnoted, outcomes of the Clean Air Act strategy for greenhouse gases is that it’s locking us into a regulatory approach for achieving energy efficiency goals in the economy. Those energy efficiency goals won’t be price driven if we're not seeing an increase in our electricity prices. To the extent that we achieve further energy efficiency as part of the strategy to reduce emissions—and many people feel there’s much opportunity to do so—we're going to have to rely on regulation by states and the federal government to get more efficiency in terms of the way people use energy in their homes and businesses.

Return to Top

8. Could each of the states choose a different approach for implementing the Clean Air Act regulations?

I don’t think we will end up in a world where there are 50 different approaches in each of the states. For one reason, many states do not have the resources to really give form to an approach that has not already been thought through by their neighbors. There’s going to be leading states and maybe some states will follow.

But it is true that with the state-driven process, there are important differences—geographically, in terms of types of resources and fuels that are used, as well as varying approaches to regulating the electricity sector—so we may see some differences across the country. An important issue that is emerging now is how these differences might affect the costs of the program, and how that might be avoided.

Return to Top

9. How stringent will the regulations on existing power plants be, and how much will it cost to implement the regulations? Are these dimensions related?

Regulation under the Clean Air Act could be stringent. I emphasize that because many people approach this thinking that with the failure to pass comprehensive climate legislation in Waxman-Markey in 2009 and 2010, the country is now operating in the minor leagues of climate policy, but that’s not the case.

Regulation via the Clean Air Act could be as stringent or more than what would have been required under the Waxman-Markey legislation. The key to that outcome will be establishing a technical record by EPA about how these emissions reductions might be achieved and the fact that there are technologies and operating approaches in place, or available, that can be implemented without incurring huge costs.

Stringency, however, is not separated from cost-effectiveness. Stringent regulations that are very expensive will face a number of challenges—both in approval by the Office of Management and Budget and in court—because in the portion of the Clean Air Act that’s being used for greenhouse gas rules for stationary sources, cost is an explicit criterion to be considered by the states and EPA.

Two questions related to both stringency and cost are these: How much flexibility can be adopted within the state? And will there be provisions for flexibility between states, such as an interstate trading program?

Return to Top

10. Is using the Clean Air Act a cost-effective approach for regulating greenhouse gas emissions from power plants?

Along with saying that the Clean Air Act could be stringent, I would say that the Clean Air Act could be cost effective. There are several ways this could be achieved. If EPA chooses to use approaches such as a tradable performance standard or converting the performance standard to a budget for the states and giving the states the flexibility to do cap and trade on their own (10 states already have a cap-and-trade program in place), or to pursue other measures with flexibility, this could keep the costs down and achieve a cost-effective outcome.

Basically, any of the flexible approaches are designed to achieve cost-effectiveness. What I mean by cost-effectiveness is that they're achieving emissions reductions from the set of regulated sources at the least cost possible. So it’s the biggest bang for the buck.

It’s not what economists would describe as the best possible outcome because it’s cost-effective only within the set of regulated sources. It’s not cost-effective for the entire economy because there may be other ways to achieve emissions reductions elsewhere in the economy that would be cheaper than doing it the way we will be doing it at power plants.

But it could be cost-effective for the industry—the set of sources that are regulated—and power plants are the most important emissions source that we have in the country. It is also where the lowest-cost emissions reductions are likely to be found at first, so it is promising from this standpoint.

Return to Top

11. What challenges might the states face as they begin developing their implementation plans for Clean Air Act regulation?

I think one of the key things that we can look at is whether EPA has identified a rate-based approach as a compliance pathway or if it has identified a mass-based approach and given the states an emissions budget to work with.

If the states are given a rate-based approach—an average emissions rate that needs to be achieved for the whole fleet on average, or maybe for coal units distinct from gas units, for example—then the states may decide to try to convert to a cap type of approach themselves.

What I’m going to be looking for in either outcome is what I would call the federalists’ dilemma. Cooperative federalism is a term that is used to characterize the Clean Air Act and most people hold it up as quite a successful approach. It describes a partnership between the states and the federal government.

The dilemma here is that states themselves are the jurisdiction of authority, but states—sometimes parts of states, and sometimes whole states and often many states—all coexist within common power pools in the country. Power pools are the control regions wherein dispatch or operational decisions about the electricity system are made.

You could have, in one power pool, different states using different approaches with respect to how they're going to regulate greenhouse gases from power plants. A kind of prisoner’s dilemma may emerge where some states make decisions that disadvantage their neighbors or potentially disadvantage themselves vis-à-vis their neighbors. There’s a coordination problem that requires attention.

It’s possible that EPA, in considering state plans, takes into consideration this question of how states have coordinated their efforts. Since flexibility is something that we're looking for, it could be that EPA decides whether a certain flexible approach or a different flexible approach is approved.

A consideration might be that if you're going to innovate, or deviate from EPA’s guidance or model rule, maybe you can do so as long as you have solved this problem of coordinating with your neighbors. If there is a failure to do so, then maybe these kinds of deviations would not be approved.

Return to Top

12. What are the advantages or disadvantages with a state-centered approach for implementing the Clean Air Act regulation?

Whether a state-based system of regulation under the Clean Air Act is viewed as an advantage or a disadvantage sort of depends on where you sit. Some would like to see a fully designed program coming from the federal government because they think that it can be designed at least cost. This reminds me of what we've pursued for 10 or 15 years with respect to international negotiations, thinking that we can design a program from above and then impose it on the jurisdictions that are going to be affected.

Climate change is emerging as a social movement, an environmental challenge of our lifetime. And because carbon dioxide emissions are so ubiquitous in our society and in our economy, regulations to limit emissions are going to touch us in many ways. Also, the consequences of climate change are going to touch us in many ways.

I think it compels us to have a process that has a lot of citizen participation—a lot of citizen understanding about what’s being asked of them and why they're doing it and what they can expect in return. I see an advantage in an approach that allows states to choose their own path, under strict guidelines from EPA in terms of the expectations about what should be achieved.

The collection of policies that emerges among the states is sometimes negatively characterized as a patchwork quilt. But I’d like to remember and quote Mary Nichols, the head of the Air Resources Board in California, who points out that a patchwork quilt is a uniquely American product. It’s been used before to shape our environmental regulation and we're seeing it now used again.

Return to Top