Clean Air Rules: A New Roadmap for the Power Sector


How new market-based regulations fit with today’s programs.

Fortnightly Magazine - June 2007

The nationwide cap-and-trade program to reduce sulfur dioxide (SO2) emissions from the power sector under the Acid Rain Program has earned widespread acclaim through dramatic SO2 emission reductions, far-ranging environmental and human health benefits, and far lower-than-expected costs. The similarly-modeled program to reduce ozone season nitrogen oxide (NOX) emissions in the eastern United States, the NOX Budget Trading Program, has achieved comparable success on a regional scale (see sidebar “Emissions Trading: In the Beginning,”).

These programs serve as the model for the next generation of regulations controlling emissions from the power sector—the Clean Air Interstate Rule (CAIR) and the Clean Air Mercury Rule (CAMR). Under CAIR and CAMR, four new trading programs will begin in 2009 and 2010. More than 300 units will be facing federal SO2 and NOX requirements for the first time under CAIR, and all of the approximately 1,300 units under CAMR will face their first federal mercury requirements.

Emissions monitoring and reporting requirements are the leading edge of both of these programs: NOX monitoring and reporting begins in 2008 for all units, and SO2 and mercury monitoring and reporting begins in 2009.

A great deal of implementation-related activity is already underway. Many states have submitted CAIR and CAMR state plans. The power sector has begun actively installing pollution controls and making other compliance preparations, and the SO2 allowance market already has begun responding to anticipated changes under CAIR.

The Clean Air Visibility Rule (CAVR) also plays into these plans, but on a longer time horizon, with impact on a broader range of sources.

What do these three programs require of the power sector, and where are we on the road to implementation? Figure 1 provides an implementation timeline.

The Clean Air Interstate Rule

Building on the Acid Rain and NOX Budget Trading programs, EPA finalized CAIR in 2005, requiring further reductions in 28 Eastern states and the District of Columbia. CAIR will reduce region-wide SO2 by approximately 70 percent and NOX by approximately 60 percent from 2003 levels to prevent significant contribution to nonattainment in downwind ozone and PM2.5 nonattainment areas. CAIR includes emissions budgets for each jurisdiction based on application of highly cost-effective controls to fossil fuel-fired electric generating units in model cap-and-trade programs with two phases of reductions. However, states have discretion in deciding which sources to control to meet the budget, and whether to participate in the federally run cap-and-trade programs delineated in the model rules.

To address their contribution to unhealthy levels of fine particles in downwind states, CAIR requires 25 Eastern states and D.C. to reduce and cap annual SO2 and NOX emissions. In addition, CAIR requires 25 Eastern states and D.C. to reduce and cap ozone season (May through September) NOX emissions to address their contribution to unhealthy levels of 8-hour ozone. As shown in Figure 2, most states covered under CAIR are required to address contributions to both PM2.5 and ozone nonattainment, and therefore reduce annual SO2 and NOX emissions as well as seasonal NOX emissions.

Generally, the CAIR model rules include fossil fuel-fired boilers and combustion turbines serving a generator with a nameplate capacity greater than 25 MW and producing electricity for sale. The CAIR programs generally are aimed at the same types of sources as the Acid Rain Program and NOX SIP Call—namely electricity generating units that sell power. However, the universe of CAIR sources is somewhat more inclusive in two ways. First, CAIR affects some sources that either permanently (e.g., simple-cycle turbines, certain cogeneration units) or temporarily (e.g., independent power producers, or IPPs, with power-purchase agreements in effect) were exempt from the Acid Rain Program. These units were designed and operated to be in the business of producing electricity for sale and were part of the universe of sources that EPA demonstrated could reduce emissions in a highly cost-effective manner for purposes of CAIR. Second, CAIR affects some power-generating sources that were not regulated under the NOX SIP Call because CAIR uses a definition whereby sources are considered to be “fossil-fuel-fired” and that is consistent with the one used in the Acid Rain Program (i.e., if you combust any fossil fuel, you are considered “fossil-fuel-fired”), while the NOX SIP Call definition considers a source to be “fossil-fuel fired” if more than 50 percent of annual heat input results from combusting fossil fuels.

The majority of the approximately 320 new sources expected to be affected under CAIR are simple-cycle combustion turbines outside the NOX SIP Call region that came online prior to 1991. Most of the others are IPP units or cogeneration units that were exempt from the Acid Rain Program.

Table 1 delineates the expanding coverage of electricity generators from the Acid Rain Program to CAIR and CAMR as EPA has learned more about less traditional electricity generators and their capabilities for highly cost-effective controls. This table is intended as an overview of programs. Individuals with specific questions about applicability should contact their state air office. The National Association of Clean Air Agencies (formerly STAPPA and ALAPCO) maintains a state-by-state list of contacts at

How the New Trading Programs Work

If states choose to participate in the federal cap-and-trade programs to reduce SO2 and NOX emissions, the result will be a larger seasonal NOX program beginning in 2009, a new annual NOX program beginning in 2009, and a new SO2 program in the CAIR region with a tighter, regional cap in 2010. All three new CAIR programs require additional reductions in 2015.

The Acid Rain Program will continue to operate in addition to the new regional CAIR SO2 trading program beginning in 2010. (Title IV NOX requirements also remain unchanged under CAIR.) Sources will use Title IV SO2 allowances to demonstrate compliance with annual CAIR requirements as well as with annual Title IV requirements. As a result, banked Title IV allowances can be used for CAIR compliance, and sources in all states subject to CAIR for SO2 will be subject to two SO2 trading programs that share the same currency. Under CAIR, however, one allowance does not always cover one ton of emissions. Instead, for purposes of CAIR, SO2 allowances of vintage 2009 and earlier will each cover one ton of emissions; vintage 2010 to 2014 allowances will authorize 0.50 tons of emissions; vintage 2015 or later allowances will authorize 0.35 tons of emissions. These ratios are necessary to achieve the more stringent reductions required under CAIR, and render acid-rain compliance a foregone conclusion with CAIR compliance while maintaining the value of acid-rain allowances.

The NOX Budget Trading Program will cease to operate with the start of the seasonal NOX trading program under CAIR in 2009. As a result, sources in most CAIR states will be subject to two separate NOX trading programs, both for purposes of compliance with CAIR. However, these two programs will not share currency, as NOX allowances are not interchangeable between the annual NOX program, which comprises part of the PM2.5 control strategy, and the seasonal NOX program for ozone control.

EPA will provide NOX emission allowances to each state according to the state budget for each program. States covered by both programs will allocate both annual and seasonal allowances to sources (or other entities).

The CAIR seasonal NOX program allows the use of banked allowances from the NOX SIP Call, just as the CAIR SO2 program allows the use of banked allowances from the Acid Rain Program. The annual NOX trading program includes a limited compliance supplement pool for early reductions in 2007 and 2008, or to address issues of reliability of electricity supply in 2009.

The structure of the CAIR programs and, in particular, the provisions allowing use of banked allowances from the Acid Rain and NOX Budget Trading Programs, exemplify EPA’s effort to ensure an orderly transition to CAIR’s trading programs and strongly encourage early reductions. There is a substantial incentive for sources to begin complying with CAIR immediately, and emissions already have dropped as a result. Preliminary data for 2006, the first year following CAIR promulgation, shows an 800,000-ton reduction in SO2 emissions from 2005.

CAIR Market Projections and State Activity

SO2 prices have ranged between $450 and $550 over the last year, an increase from four to five years ago largely due to CAIR increasing the value of today’s allowances. EPA’s modeling during rule development projected that pre-2010 vintage SO2 allowances would be worth approximately $736 per allowance in 2010 (in 2007 dollars), and that 2010 to 2014 vintage allowances would be worth approximately $368 per allowance due to the 2:1 retirement ratio. With 2007 vintage allowances trading at approximately $425 per ton as of early April 2007, observers generally agree that the SO2 market is undervalued. But with allowance demand low as sources finalize compliance plans, there is little current support for higher prices.

EPA’s modeling of the NOX market projects prices in the annual market to be $1,440/ton in 2010. Although there will be two distinct markets, EPA expects that the prices in both the annual and seasonal markets will be established by the cost of controls for annual compliance. There has been trading activity in the 2009 seasonal market, but allowances for the new annual NOX CAIR market have yet to trade. Observers expect that trading will not occur until CAIR State Implementation Plans (SIPs) have been approved and NOX allowance accounts are populated later this year.

For both the SO2 and NOX markets, it will take time for buyers and sellers to continue to assess the fundamentals of the changes introduced by CAIR, but this is secondary to the achievement of the environmental accountability and results of the program.

CAIR requires that covered states submit SIPs to EPA by September 2006. The agency also promulgated a federal implementation plan (FIP) that implements the model-trading rules for every CAIR state and offered to leave it in place for states not wishing to submit a SIP. Eleven of the states have thus far submitted the required SIPs and several other states have chosen to leave the FIP in place, at least temporarily. All of the SIPs adopt the model-trading rules, and EPA expects all 29 affected jurisdictions to participate in the EPA-run program.

States choosing to adopt the model rules have some flexibility in participating, including determining NOX allowance allocations independently; non-EGUs from the NOX Budget Trading Program; and the opt-in methodologies in the CAIR model rules. Nearly all states submitting SIPs thus far establish their own allocation methodologies, often including special set-asides both for new sources and providing incentives for various state priorities, like renewable energy or add-on controls. In some cases, states roll any unclaimed set-aside allowances back into the main allowance pool; others hold them over for possible distribution in the future.

Most states thus far have chosen to include the model rule provisions that allow sources to opt-in. Of the 19 states plus D.C. that are subject to the NOX SIP Call and also CAIR (note that Rhode Island was included in the former, but not the latter), all but five have indicated they will include the NOX Budget Trading Program is non-EGUs in the applicability for the CAIR NOX ozone-season program.

Whether sources in a state are subject to a SIP or a FIP, NOX allowances under CAIR will be allocated by the end of this year (SO2 allowances already have been allocated under Title IV). Figure 3 shows the advanced SO2 and NOX controls already in place in the CAIR region as well as those already committed or projected under CAIR.

The Clean Air Mercury Rule

CAMR requires all 50 states, D.C., and two tribes to regulate mercury emissions from coal-fired EGUs. CAMR establishes “standards of performance” limiting mercury emissions from new and existing coal-fired power plants and, like CAIR, creates a model cap-and-trade program with two phases of reductions. The first phase cap is 38 tons, taking advantage of “co-benefit” reductions—mercury reductions achieved by reducing SO2 and NOX emissions under CAIR—to fulfill EPA’s requirement to act on mercury emissions. The second phase, beginning in 2018, goes further to reduce emissions to 15 tons upon full implementation. CAMR sets an emissions-reduction requirement in the form of an annual budget for each state and two tribes in accordance with the two caps. New coal-fired power plants will have to meet new-source performance standards in addition to being subject to the caps.

EPA established annual budgets for each state, and states must ensure that current and future mercury emissions from coal-fired EGUs do not exceed the annual state budget. Like CAIR, CAMR does not exempt the units that may be exempt under the Acid Rain Program. The summary of applicability across programs in Table 1 includes general CAMR applicability for comparison.

Furthermore, under CAMR, affected coal-fired electric utility units will be required to continuously monitor mercury mass emissions for the first time, regardless of whether or not they will be participating in the trading program. Monitoring technologies will be subject to rigorous certification and quality assurance/quality control require- ments under 40 CFR Part 75. Affected sources are required to install and certify continuous emissions or sorbent trap monitoring systems by Jan. 1, 2009. This new requirement is one of the primary areas of focus for EPA’s CAMR implementation efforts.

Recent work by both EPA and industry has advanced mercury monitoring systems, reference testing methods, and calibration standards to a point that measuring capabilities that had limited feasibility a few years ago now are fully or nearly ready and even commercially available. Over the past year, the performance and reliability of mercury monitoring systems substantially have improved as a result of field demonstrations and testing by EPA and industry. EPA continues to work closely with the regulated community, monitoring equipment and software vendors, academia, and other organizations to ensure timely implementation of a technically sound, effective CAMR mercury monitoring program.

The Mercury Trading Program and State Activity

The trading program under CAMR will work very similarly to existing programs and the SO2 and NOX programs under CAIR, with two notable differences between the CAMR and CAIR trading programs. First, there are no opt-in provisions included in CAMR; second, allowances under CAMR are measured in ounces, rather than tons.

Even with some states choosing to control mercury emissions directly, EPA expects a robust trading program. In July 2006, EPA conducted limited modeling meant to be illustrative of a reduced market based on the states and tribes EPA projected would participate in the national trading program at the time. This more limited market represents states that allocated close to 69 percent of the initial budget of mercury allowances and comprises more than 700 units representing more than 200 GW of capacity—nearly equivalent to the number of coal-fired units in the successful NOX Budget Trading Program across a larger number of states. As with the NOX Budget Trading Program, EPA expects a viable market will result.

Based on this modeling, prices for mercury allowances are expected to be the same or lower as in a full national market. This is because several high-demand states such as Pennsylvania and Illinois will not be participating. Overall, however, states opting to participate in the trading program generally are characterized by larger percentages of coal-fired generation.

Moreover, twenty-one states submitted state plans to EPA by the November 2006 CAMR deadline and additional state plans have been received since. The remaining states actively are working on plans. Of all affected jurisdictions, 33 states and two tribes are planning on participating in the CAMR trading program. Thirteen states have indicated they will not participate in the trading program, and at least one state is still undecided. Three states (Idaho, Vermont, and Rhode Island) and the District of Columbia do not have any coal-fired EGUs and thus have zero budgets.

States not participating in the trading program must ensure they meet their state budget with other methods, often through control requirements implemented through percent-reduction provisions based on an analysis of control options states have evaluated as feasible. Some states have chosen to do this in phases like EPA has, though the start of the second phase is in some cases accelerated. Unlike a capped program, percent-reduction programs do not necessarily guarantee emissions will remain below a state’s budget, often because of the uncertainty of new source growth over time. Therefore, states are often coupling these programs with caps to ensure the state’s budget will be maintained.

Where mercury trading programs are enacted, EPA expects overcontrol in 2010. This is because sources are likely to optimize the controls installed for CAIR to reduce as much mercury as possible in anticipation of increasing prices for mercury allowances under the lower second-phase cap.

In December 2006, EPA proposed a CAMR federal plan to be finalized in states that either fail to submit a CAMR state plan or whose state plan is somehow deficient. EPA is evaluating comments but has not finalized a CAMR federal plan. The proposal would put into place a cap-and-trade program in any State in which the federal plan is finalized.

The Clean Air Visibility Rule

CAVR supplements the emission reductions of CAIR by requiring emission controls known as best available retrofit technology (BART) for industrial facilities emitting air pollutants that reduce visibility by contributing to regional haze. The pollutants include PM2.5 and their precursors, such as SO2, NOX, volatile organic compounds and ammonia. The BART requirements apply to facilities built between 1962 and 1977 that have the potential to emit more than 250 tons a year of visibility-impairing pollution and cover 26 categories, including utility and industrial boilers and large industrial plants such as pulp mills, refineries, and smelters.

Many of these facilities have not been subject to federal pollution-control requirements for these pollutants. Under the 1999 regional-haze rule, states are required to set periodic goals for improving visibility in the 156 “Class I” natural areas such as national parks. CAVR includes guidelines for states to use in determining which facilities must install controls and the type of controls the facilities must use. States must develop their implementation plans by December 2007, identify the facilities that will have to reduce emissions under BART, then set emissions limits for those facilities and require installation of BART in 2014.

For CAIR-affected EGUs, participation in the CAIR programs is considered to meet their federal source-specific BART requirements, because CAIR was determined to be better than BART controls of CAVR in the CAIR region. Specifically, controls for EGUs subject to CAIR will result in more visibility improvement in natural areas than BART would have provided. States could, however, require additional reductions.

Projected Controls in Response to Clean Air Suite

Although aspects of CAIR and CAMR are being litigated, implementation moves ahead. Briefs have been filed by industry and stakeholders. EPA has filed a brief on CAMR and will file on CAIR this month. Having worked very hard to develop the details of these critically important environmental programs, EPA has gone on to work with states, which are now working aggressively to put implementing rules in place. The regulated community is going forward with installing equipment for CAIR, entering into contracts for construction of mercury controls and putting monitoring systems in place.

Sources have begun responding to the new requirements with investments and application of retrofit technology. EPA estimates that in 2010, 60 percent of total capacity will be scrubbed, increasing to 73 percent by 2020. Modeling shows that the percentage of advanced controls will go up (with the amount of capacity with advanced controls projected to increase even faster) and the number of units without advanced controls will go down, especially for larger units.

As observed with previous programs, the regulated community responds with a sense of purpose and alacrity to cap-and-trade programs. EPA, using existing Clean Air Act authority, is moving to address interstate transport of SO2 and NOX emissions and lower mercury emissions with CAIR, CAMR, and CAVR. We expect these programs to deliver significant human health and environmental improvements in a cost-effective manner by harnessing market forces to achieve substantial required emissions reductions.

EPA has worked diligently to promulgate rules to meet regional needs, all the while preserving to the greatest extent possible the essential accountability, simplicity, flexibility, high level of compliance, and efficiency of the cap-and-trade mechanism. If our experience to date is any indication, the next generation of programs promises to continue this legacy of results.