A Scrivener's Error or Greater Protection of the Public: Does the EPA Have the Authority to Delist "Low-Risk" Sources of Carcinogens from Section 112's Maximum Achievable Control Technology Requirements
By Bradford C. Mank
INTRODUCTION
Since 1970, section 112 of the Clean Air Act (the Act) has required the EPA to establish national emission standards for hazardous air pollutants (NESHAPs),1 or air toxics. Section 112 regulates all stationary sources such as factories, power plants, and refineries that emit listed air toxics. The Act currently lists 187 different types of regulated hazardous air pollutants, some of which are carcinogens and some of which are non-carcinogens that pose other serious health risks.
In section 112 of the 1970 Amendments to the Act, Congress mandated that the Administrator of the Environmental Protection Agency (EPA) establish health-based standards for air toxics that provide “an ample margin of safety to protect the public health from such hazardous air pollutant.” However, the statute failed to provide adequate guidance on how the EPA should interpret or apply the “ample margin” standard. Because the EPA regulated only seven air toxics from 1970 until 1990, Congress in the 1990 Amendments to the Act substantially revised section 112 to authorize the EPA to use a primarily technology-based11regulatory scheme to reduce air toxics. The 1990 Amendments require the EPA to issue emissions standards based on the “maximum achievable control technology” (MACT) standard for each category or subcategory of major sources of air toxics. Subsection 112(c)(9)(B), however, authorizes the EPA under certain conditions to delete, or “delist,” a category or subcategory of sources from MACT standards if all sources in the category or subcategory are “low-risk.”
This article will focus on the scope of the EPA's authority to delist categories and subcategories of sources, especially those emitting carcinogens. The plain language of subsection 112(c)(9)(B) provides different requirements for carcinogenic and non-carcinogenic air toxics. Under subsection 112(c)(9)(B)(ii), the EPA has authority to exempt either categories or subcategories of sources emitting non-carcinogenic air toxics from MACT standards as long as the EPA substitutes standards that are “adequate to protect public health within an ample margin of safety” and that will cause “no adverse environmental effect.” By contrast, subsection 112(c)(9)(B)(i) explicitly provides the EPA with the authority to exempt from MACT standards only whole categories of carcinogenic sources, provided that no source in the category emits air toxics causing “a lifetime risk of cancer greater than one-in-one-million to the individual in the population who is most exposed to emissions of such pollutants from the source.” The subsection does not explicitly authorize the EPA to exempt a subcategory of sources releasing carcinogenic chemicals. The omission of the term “subcategory” in subsection 112(c)(9)(B)(i) is potentially very significant and may limit the EPA's authority to delist a subcategory of carcinogenic sources from a larger category of such sources.
Even though subsection 112(c)(9)(B)(i)'s statutory language only authorizes the EPA to exempt whole categories of carcinogenic sources, the EPA recently published a rule that purports to use that very subsection to delist a subcategory of “low-risk” sources. On July 30, 2004, the EPA published a final rule for the plywood and composite wood industries (PCWP). The rule established MACT standards for approximately 223 sources that release potentially carcinogenic air toxics, including formaldehyde. However, partly in response to an industry petition and partly in response to its own initiative, the EPA also established an exemption that could apply to over half of PCWP sources. Specifically, the EPA used its subsection 112(c)(9)(B)(i) authority to “delist” and treat as a separate subcategory “low-risk” PCWP sources that release small amounts of air toxics. The agency justified this exemption on the grounds that these PCWP facilities posed such low risks that the PCWP MACT standards were unnecessary and too costly. During 2003 and 2004, the EPA also adopted or considered similar exceptions under subsection 112(c)(9)(B) for other source categories releasing air toxics. The EPA's claim that it has the authority to exempt subcategories of so-called “low-risk” carcinogenic sources is significant because it could be used to exempt thousands of sources governed by MACT standards in dozens of industries. It would also give the Agency authority to exempt individual sources from MACT, an authority that Congress in 1990 explicitly refused to grant to the Agency.
In its PCWP rule, the EPA contended that it has the authority to exempt a subcategory of “low-risk” sources releasing carcinogenic chemicals because it assumed that Congress had made a drafting error in subsection 112(c)(9)(B)(i) by using only the term “category,” but not the term “subcategory.” In rare cases where statutory language makes little sense or contravenes Congress's likely intent, courts have recognized the doctrine of scrivener's error to correct obvious errors in a statute. Because of a strong presumption that Congress normally is careful in using statutory language, there is a heavy burden on an agency to demonstrate that a statute contains a scrivener's error, and courts limit an agency's discretion in rectifying such an error to the smallest change necessary to fix the error.
On September 28, 2004, the Natural Resources Defense Council (NRDC) and the Environmental Integrity Project filed a petition for reconsideration (the Petition) with the EPA's then-Administrator, Michael Leavitt, asking him to indefinitely stay the PCWP rule. The Petition argues that the plain language of subsection 112(c)(9)(B)(i) explicitly allows the EPA to delist only an entire source category of carcinogenic air toxics and only if every single source in the category is low-risk. Accordingly, the Petition contends that the Agency may not delete a limited subcategory of sources emitting carcinogens, such as some but not all PCWP sources.
This article concludes that the EPA's creation of a low-risk subcategory of PCWP sources is improper because the plain language of subsection 112(c)(9)(B)(i) limits the Agency's delisting authority to whole categories of carcinogenic sources. The EPA has failed to meet its heavy burden in attempting to demonstrate that Congress *82 made a drafting error when it omitted the term “subcategory” in subsection 112(c)(9)(B)(i) for carcinogenic chemicals. The doctrine of scrivener's error is inapplicable to the plain language of subsection 112(c)(9)(B)(i). There are plausible reasons why Congress would have used the term “subcategory” in subsection 112(c)(9)(B)(ii) for non-carcinogenic chemicals that were assumed to have a safe threshold but deliberately omitted the term “subcategory” in subsection 112(c)(9)(B)(i) for carcinogenic chemicals that were assumed in 1990 to have no safe threshold other than zero emissions.
Part II examines the differences between “non-threshold” carcinogens and “threshold” non-carcinogens and the difficulties that the EPA had in regulating air toxics under the 1970 Amendments to section 112. Part III explains the 1990 Amendments to section 112, which shifted the statute to a largely technology-based focus. Part IV examines two provisions in Section 112-- subsections 112(d)(4) and 112(c)(9)(B)--that treat non-threshold carcinogens differently than threshold non-carcinogens. Part V introduces the PCWP Rule's process for delisting a “low-risk” subcategory of carcinogenic sources under subsection 112(c)(9)(B)(i). Finally, Part VI examines the EPA's argument that the omission of the term “subcategories” in subsection 112(c)(9)(B)(i) is a drafting error. It concludes that the EPA has failed to meet its burden of proving a scrivener's error because there are plausible reasons why Congress might have allowed the delisting of subcategories of non-carcinogenic sources, but limited the delisting of carcinogens to entire source categories.
Since 1970, section 112 of the Clean Air Act (the Act) has required the EPA to establish national emission standards for hazardous air pollutants (NESHAPs),1 or air toxics. Section 112 regulates all stationary sources such as factories, power plants, and refineries that emit listed air toxics. The Act currently lists 187 different types of regulated hazardous air pollutants, some of which are carcinogens and some of which are non-carcinogens that pose other serious health risks.
In section 112 of the 1970 Amendments to the Act, Congress mandated that the Administrator of the Environmental Protection Agency (EPA) establish health-based standards for air toxics that provide “an ample margin of safety to protect the public health from such hazardous air pollutant.” However, the statute failed to provide adequate guidance on how the EPA should interpret or apply the “ample margin” standard. Because the EPA regulated only seven air toxics from 1970 until 1990, Congress in the 1990 Amendments to the Act substantially revised section 112 to authorize the EPA to use a primarily technology-based11regulatory scheme to reduce air toxics. The 1990 Amendments require the EPA to issue emissions standards based on the “maximum achievable control technology” (MACT) standard for each category or subcategory of major sources of air toxics. Subsection 112(c)(9)(B), however, authorizes the EPA under certain conditions to delete, or “delist,” a category or subcategory of sources from MACT standards if all sources in the category or subcategory are “low-risk.”
This article will focus on the scope of the EPA's authority to delist categories and subcategories of sources, especially those emitting carcinogens. The plain language of subsection 112(c)(9)(B) provides different requirements for carcinogenic and non-carcinogenic air toxics. Under subsection 112(c)(9)(B)(ii), the EPA has authority to exempt either categories or subcategories of sources emitting non-carcinogenic air toxics from MACT standards as long as the EPA substitutes standards that are “adequate to protect public health within an ample margin of safety” and that will cause “no adverse environmental effect.” By contrast, subsection 112(c)(9)(B)(i) explicitly provides the EPA with the authority to exempt from MACT standards only whole categories of carcinogenic sources, provided that no source in the category emits air toxics causing “a lifetime risk of cancer greater than one-in-one-million to the individual in the population who is most exposed to emissions of such pollutants from the source.” The subsection does not explicitly authorize the EPA to exempt a subcategory of sources releasing carcinogenic chemicals. The omission of the term “subcategory” in subsection 112(c)(9)(B)(i) is potentially very significant and may limit the EPA's authority to delist a subcategory of carcinogenic sources from a larger category of such sources.
Even though subsection 112(c)(9)(B)(i)'s statutory language only authorizes the EPA to exempt whole categories of carcinogenic sources, the EPA recently published a rule that purports to use that very subsection to delist a subcategory of “low-risk” sources. On July 30, 2004, the EPA published a final rule for the plywood and composite wood industries (PCWP). The rule established MACT standards for approximately 223 sources that release potentially carcinogenic air toxics, including formaldehyde. However, partly in response to an industry petition and partly in response to its own initiative, the EPA also established an exemption that could apply to over half of PCWP sources. Specifically, the EPA used its subsection 112(c)(9)(B)(i) authority to “delist” and treat as a separate subcategory “low-risk” PCWP sources that release small amounts of air toxics. The agency justified this exemption on the grounds that these PCWP facilities posed such low risks that the PCWP MACT standards were unnecessary and too costly. During 2003 and 2004, the EPA also adopted or considered similar exceptions under subsection 112(c)(9)(B) for other source categories releasing air toxics. The EPA's claim that it has the authority to exempt subcategories of so-called “low-risk” carcinogenic sources is significant because it could be used to exempt thousands of sources governed by MACT standards in dozens of industries. It would also give the Agency authority to exempt individual sources from MACT, an authority that Congress in 1990 explicitly refused to grant to the Agency.
In its PCWP rule, the EPA contended that it has the authority to exempt a subcategory of “low-risk” sources releasing carcinogenic chemicals because it assumed that Congress had made a drafting error in subsection 112(c)(9)(B)(i) by using only the term “category,” but not the term “subcategory.” In rare cases where statutory language makes little sense or contravenes Congress's likely intent, courts have recognized the doctrine of scrivener's error to correct obvious errors in a statute. Because of a strong presumption that Congress normally is careful in using statutory language, there is a heavy burden on an agency to demonstrate that a statute contains a scrivener's error, and courts limit an agency's discretion in rectifying such an error to the smallest change necessary to fix the error.
On September 28, 2004, the Natural Resources Defense Council (NRDC) and the Environmental Integrity Project filed a petition for reconsideration (the Petition) with the EPA's then-Administrator, Michael Leavitt, asking him to indefinitely stay the PCWP rule. The Petition argues that the plain language of subsection 112(c)(9)(B)(i) explicitly allows the EPA to delist only an entire source category of carcinogenic air toxics and only if every single source in the category is low-risk. Accordingly, the Petition contends that the Agency may not delete a limited subcategory of sources emitting carcinogens, such as some but not all PCWP sources.
This article concludes that the EPA's creation of a low-risk subcategory of PCWP sources is improper because the plain language of subsection 112(c)(9)(B)(i) limits the Agency's delisting authority to whole categories of carcinogenic sources. The EPA has failed to meet its heavy burden in attempting to demonstrate that Congress *82 made a drafting error when it omitted the term “subcategory” in subsection 112(c)(9)(B)(i) for carcinogenic chemicals. The doctrine of scrivener's error is inapplicable to the plain language of subsection 112(c)(9)(B)(i). There are plausible reasons why Congress would have used the term “subcategory” in subsection 112(c)(9)(B)(ii) for non-carcinogenic chemicals that were assumed to have a safe threshold but deliberately omitted the term “subcategory” in subsection 112(c)(9)(B)(i) for carcinogenic chemicals that were assumed in 1990 to have no safe threshold other than zero emissions.
Part II examines the differences between “non-threshold” carcinogens and “threshold” non-carcinogens and the difficulties that the EPA had in regulating air toxics under the 1970 Amendments to section 112. Part III explains the 1990 Amendments to section 112, which shifted the statute to a largely technology-based focus. Part IV examines two provisions in Section 112-- subsections 112(d)(4) and 112(c)(9)(B)--that treat non-threshold carcinogens differently than threshold non-carcinogens. Part V introduces the PCWP Rule's process for delisting a “low-risk” subcategory of carcinogenic sources under subsection 112(c)(9)(B)(i). Finally, Part VI examines the EPA's argument that the omission of the term “subcategories” in subsection 112(c)(9)(B)(i) is a drafting error. It concludes that the EPA has failed to meet its burden of proving a scrivener's error because there are plausible reasons why Congress might have allowed the delisting of subcategories of non-carcinogenic sources, but limited the delisting of carcinogens to entire source categories.