Skip Ribbon Commands
Skip to main content
Home | Support RFF | Join E-mail List | Contact
RFF Logo
Skip navigation links


Join E-mail List
Please provide your e-mail address to receive periodic newsletters and invitations to public events

The Potential Impact on Energy-Intensive Trade-Exposed Industries of Clean Air Act Regulation of GHGs

Richard D. Morgenstern

Pricing carbon emissions, through either a cap-and-trade system or an emissions tax, will not only adversely affect electricity and primary energy producers, but also hurt the competitive performance of heavy fossil-fuel users in downstream industries, especially in trade-exposed sectors like steel and chemicals. Furthermore, some of the environmental benefits might be eroded (“emissions leakage”) if increases in U.S. manufacturing costs from uneven international carbon pricing caused economic activity to shift to nations with weaker greenhouse gas mitigation policies or none at all.

The Waxman-Markey bill (HR 2454) would have addressed these concerns by identifying energy-intensive trade-exposed (EITE) industries based on measurable criteria and adopting an allowance allocation method, known as output-based rebating, for these sectors. A related provision in these bills provides rebates to local distribution companies to cushion impacts on both industrial and residential electricity consumers. Both measures attempt to level the near-term playing field between domestic and foreign producers, thereby reducing both the competitiveness and the leakage impacts.

With the demise of legislative options, attention is now shifting to the use of the Clean Air Act (CAA) authorities to reduce emissions of greenhouse gas (GHG) emissions. What effect would CAA-based regulations have on EITE industries? How would the likely impacts on EITE industries differ from those of emissions pricing with output-based rebate schemes embodied in the legislative approach?

There are clearly many unknowns about how, if at all, the U.S. Environmental Protection Agency (EPA) might proceed with regulating stationary sources of GHGs. However, one commonly discussed approach is for the agency to adopt New Source Performance Standards (NSPSs)—technology-based regulations—for electric generating units and, quite possibly, for high-emitting industrial sources. Presumably, these industrial sources would be many of the same sectors Congress has categorized as EITE industries. At this point, no additional findings are required prior to issuing GHG-based NSPSs for more than 60 source categories and subcategories covering major stationary sources, including coal, oil, and gas power plants; refineries; cement plants; and various other industrial categories for which the agency made an endangerment finding many years ago based on non-GHG pollutants.

At some point, however, he believes people will make decisions that are rational and efficient. If we cannot find common ground on an emissions cap, he says, it may be time to turn to a carbon tax, which would introduce a new dynamic into the market while raising the money needed to make critical improvements along the path toward low-carbon growth. “At the same time,” he says, “we have to summon the political will and the flexibility to find solutions that are not driven by ideology but reflect instead a common desire to achieve the best possible results on a global scale.”

An NSPS, based on best demonstrated technology, would apply to both new units and existing sources that make major modifications. EPA also has the authority to establish guidelines for states to use in setting technology standards for existing sources. As my colleagues Nathan Richardson, Art Fraas and Dallas Burtraw (2010) have written, the NSPS approach is “the most predictable, likely and practical, that is, knowable pathway.”

Below I describe two cases, one in which EPA issues an NSPS for electric generating units (EGUs) only, and the other in which the agency also issues a series of NSPSs for EITE industries. Initially, imagine that EPA issued an NSPS for coal power plants requiring certain low-cost actions that would yield modest but meaningful emissions reductions. As Richardson et al. hypothesize, efficiency improvements and biomass cofiring might be good candidates.

Arguably, a cleverly designed NSPS might actually result in smaller consumer impacts than under full-blown cap and trade, largely because only the actual resource costs would be passed on to consumers, at least in the cases where electricity rates are set by the states. This contrasts with a national cap-and-trade scheme where the cost pass-through would be determined in most cases by the systemwide marginal costs, which would, on average, likely be higher than the cost of the resources needed to achieve the near-term reductions. Although it is not possible to estimate the actual price differences at this time, calculations by Burtraw et al. (2009) indicate they could be substantial.

How would these potentially modest but nonzero costs of an NSPS for electric generating units affect these industries? And how would the impacts differ from those driven by a comparable Waxman-Markey style emissions reduction strategy? If EPA could, indeed, identify the optimal investments that would be chosen by EGUs under a cap-and-trade system, then the electricity price impacts would likely also be modest, even in the absence of Waxman-Markey style rebates. While states could still offer rebates, it is almost certain that they would not be uniform across the country. In fact, one could imagine a beggar-thy-neighbor competition whereby states might use overly generous rebates to attract energy-intensive industry.

Whatever the ultimate size of the NSPS-induced electricity price hikes, however, the pattern of impacts on EITE industries would be quite different under an EGU-only NSPS compared to an economy-wide cap-and-trade scheme. The table on page 46 compares the impacts on a highly disaggregated set of manufacturing industries of an economywide policy versus an EGU-only approach, based on a modeling analysis using census data from the 1990s. The per-ton charge on carbon inputs is equal for the two policies. However, the former policy affects all carbon inputs, both direct and indirect, used in the manufacturing sector while the latter affects only carbon used in the production of electricity. Manufacturing industries relying heavily on electricity or nonfuel inputs that, in turn, are heavy electricity users—for example, the makers of aluminum car parts purchased by the auto industry—would be most adversely affected by an electricity-only policy.

The left side of the table displays the 10 industries hardest hit by the economywide policy and their corresponding ranking for the EGU-only policy among 361 manufacturing industries analyzed. Eight of the 10 industries hardest hit by the economywide policy rank lower (or the same) for the EGU-only policy, in most cases considerably lower. The right side of the table displays the 10 industries hardest hit by the EGU-only policy along with their corresponding ranking for the economywide approach. The hardest hit, aluminum, ranks 13 for the economywide policy. All of the top 10 under the EGU-only policy rank lower (or the same) for the economywide policy, often substantially lower. The conclusion of this policy comparison is clear: manufacturing industries are affected very differently by the economywide pricing of carbon emissions compared to regulation of the electric power sector only.

Industry NSPSs

Now we turn to the second case, wherein EPA issues an NSPS for EITE industries, in addition to the one for electric generation. Similar to the EGU NSPS, the industry ones would take the form of technology-based standards for new sources and existing sources making major modifications. The agency could, in principle, also set rules for states in their review of existing sources, although short of imposing a draconian federal implementation plan, it is difficult to imagine strong, uniform implementation across the different states.

The first thing to note about such standards is that they could be much more fine grained than the congressionally mandated procedures under Waxman-Markey, which focused on the six-digit North American Industrial Classification Scheme (NAICS). In setting regulations under the CAA, EPA could adopt a much more refined approach, focusing on the subindustrial categories that had the greatest emissions and/or the most cost-effective potential to reduce their emissions—equivalent, perhaps, to a 7- or 8-digit NAICS. In addition, EPA could potentially exclude certain subsectors of congressionally defined EITE industries and include others not covered by the Waxman-Markey definitions.

This is important because studies have shown that there is great heterogeneity in energy use and carbon dioxide emissions among industries, both within and outside the EITE sectors. Consequently, the agency’s ability to reach down in a granular fashion to the large emitters opens up the potential for highly efficient regulation. While there is certainly no assurance the agency would achieve a higher degree of efficiency in this process, the potential to do so under CAA versus a six-digit NAICS Waxman-Markey approach clearly exists.

Beyond the potential for fine-tuning of the regulations under the CAA, the rest of the story for EITE industries is not particularly promising when compared to the legislative approach. Emissions trading across industries has not been tested and would certainly face an uphill court battle. Absent that authority, the potential for within-sector trading is limited by both legal and practical factors, not the least of which is the potential market power of individual firms in particular industrial sectors. This contrasts with the electric sector where, despite early concerns by some groups in the sulfur dioxide debates, experience has shown that the large size of that market enables firms to trade allowances without fear of competitiveness concerns among themselves.

Even if an interindustry emissions trading program was established, presumably by the states, it is unlikely that an effective output-based rebating scheme could be put in place. As noted, there is a potential for beggar-thy-neighbor actions. In the absence of a uniform system, it would be virtually impossible to develop a credible rebate mechanism to address competitiveness and leakage concerns without violating the rules of the World Trade Organization.

The bottom line is that carbon dioxide regulation under the Clean Air Act, presumably via an electricity-only NSPS or by the addition of a series of NSPSs covering EITE industries, could potentially result in smaller initial energy price impacts than an economywide cap-and-trade scheme. In the case of an electricity-only policy, the pattern of affected industries would clearly be different than under a broader-based mechanism. Going forward, and given the agency’s inability to offset higher energy costs as in Waxman-Markey, the challenge for the agency will be to craft rules that impose only modest costs, thereby obviating the need for offsetting mechanisms that are likely infeasible under a CAA approach.

Further Reading

The Incidence of U.S. Climate Policy: Alternative Uses of Revenues from a Cap-and-Trade Auction.
Dallas Burtraw, Richard Sweeney, Margaret A. Walls.
RFF Discussion Paper 09-17-REV. June.

The Near Term Impacts of Carbon Mitigation Policies on Manufacturing Industries
Richard D. Morgenstern, Mun Ho, Jhih-Shyang Shih, and X. Zhang.
Energy Policy. November. Vol. 32, No. 16. 1825-1841

Greenhouse Gas Regulation under the Clean Air Act: Structure, Effects, and Implications of a Knowable Pathway
Nathan Richardson, Arthur G. Fraas, Dallas Burtraw
RFF Discussion Paper 10-23. April.

RFF Home | RFF Press: An Imprint of Routledge Terms of Use | Privacy Policy | Copyright Notice
1616 P St. NW, Washington, DC 20036 · 202.328.5000 Feedback | Contact Us