Air Toxics: What Progress Has EPA Made in Regulating Hazardous Air Pollutants?

Report for Congress
Air Toxics:
What Progress Has EPA Made in
Regulating Hazardous Air Pollutants?
July 22, 2002
Anne L. Hardenbergh
Resources, Science, and Industry

Congressional Research Service ˜ The Library of Congress

Air Toxics: What Progress Has EPA Made in
Regulating Hazardous Air Pollutants?
The Clean Air Act Amendments of 1990 contained four programs to ensure that
the Environmental Protection Agency (EPA) would regulate hazardous air pollutants
(HAPs) more quickly than the Agency had under the original act. The first program
directed EPA to establish emission standards for stationary sources of HAPs based
on maximum achievable control technology (MACT). The second program, known
as the residual risk program, required EPA to examine the risk remaining from these
sources after implementation of MACT standards. If warranted, EPA must
promulgate additional standards to protect public health or the environment. Third,
Congress instructed EPA to regulate HAPs that are a particular problem for urban
areas. Fourth, Congress directed EPA to evaluate mobile sources, such as cars, to
determine their HAP emissions and to regulate these sources.
EPA has made progress in regulating HAPs under some of these programs,
particularly in comparison to its pre-1990 record (when only 7 HAPs were regulated
over a 20-year period). But the pace of regulation has fallen behind that envisioned
in the 1990 Amendments. Under the MACT program, for example, the Agency
originally named 174 source categories subject to regulation for their HAP emissions.
In the statute, Congress specified that MACT standards for all of these categories
were to be promulgated by November 15, 2000. By May 15, 2002, EPA still had not
promulgated final standards for 56 source categories (a little less than one-third of
those to be regulated), covering a combined total of 84,000 major emitting sources.
Under the 1990 amendments, if EPA has not promulgated standards by 18
months after the deadline, states and industries must develop MACT standards
themselves. Rather than allow this “hammer” to strike, however, EPA issued a rule
in April 2002 that will give these sources until May 15, 2004 to develop MACT
standards. The Sierra Club has filed suit challenging the deadline extension.
The residual risk program, intended to examine whether risk remains after
MACT standards are implemented, is also behind schedule. The first residual risk
standards, for coke ovens, were due in 2001; 12 other categories are to be reviewed
before the end of 2002. While EPA has developed a methodology for determining
residual risk, the methodology and the availability of the data for conducting the
analyses have been questioned by the Agency’s Science Advisory Board. It is
unlikely that EPA will meet deadlines for residual risk standards.
Both the urban air toxics and mobile source air toxics programs have also been
delayed. Questions have been raised regarding the adequacy of the mobile source air
toxics standards; most standards for urban air toxics have yet to be promulgated.
Issues for Congress include the adequacy of Agency resources for the HAPs
programs; the feasibility of the residual risk program, given inadequacies of the
available data and concerns over the risk assessment methodology; and whether the
Agency is complying with congressional intent in key areas of the HAPs programs.
This report will be updated as events warrant.

Maximum Achievable Control Technology Standards.................1
Current Status............................................3
Remaining Issues..........................................6
The Residual Risk Program......................................6
Current Status............................................7
Remaining Issues..........................................8
The Urban Air Toxics Program...................................9
Current Status...........................................10
Remaining Issues.........................................11
The Mobile Source Air Toxics Program...........................11
Current Status...........................................12
Remaining Issues.........................................12
Conclusions .................................................13
List of Tables
Table 1. MACT Source Category Final Rules: Schedule Requirements and Actual
Promulgations as of July 22, 2002.................................4
Table 2. Urban Air Toxics and Their Major Sources.....................14
This report was prepared under the research supervision of James E. McCarthy,
Specialist in Environmental Policy.

Air Toxics: What Progress Has EPA Made in
Regulating Hazardous Air Pollutants?
Hazardous air pollutants (HAPs), or air toxics, are pollutants that are either
known or suspected to cause cancer or other adverse health effects, such as damage
to reproductive functions or birth defects. Polychlorinated biphenyls (PCBs), arsenic
compounds, formaldehyde, asbestos, methanol, and chlorine are just some of the
chemicals and chemical compounds that are considered air toxics. HAPs come from
both mobile and stationary sources.
Under the original Clean Air Act (CAA), EPA was directed to take a risk-based
approach to regulating HAPs. Congress, dissatisfied with the fact that only 7 HAPs
were regulated from 1970 to 1990 using the risk-based approach, amended the CAA
in 1990 to direct EPA to use a technology-based approach to reduce HAP emissions.
A separate risk based program was also contained in these amendments. Rather than
have EPA determine what air toxics should be regulated, as was done under the
original CAA, Congress included a specific list of 188 air toxics in the
Amendments.1 Some of the air toxics on this list are included in Table 2 at the end
of this report, along with the main sources that produce them.
The 1990 Amendments completely rewrote Section 112 of the Clean Air Act
and established four different programs for the regulation of HAPs: maximum
achievable control technology, residual risk, urban air toxics, and mobile source air
toxics. This report describes these programs, including the current status of each
program and what significant issues still remain in their implementation. While EPA
has made progress in regulating HAPs, many of the deadlines set by the 1990
Amendments have not been met. Each of the four air toxics programs are behind the
schedule envisioned by Congress, and appear likely to remain so for reasons outlined
Maximum Achievable Control Technology Standards
Section 112 of the CAA (42 U.S.C. 7412), is the main section regulating HAPs.
One method of regulation is known as maximum achievable control technology
(MACT). The purpose of MACT standards is to achieve “the maximum degree of
reduction in emissions of [HAPs]” (42 U.S.C. 7412(d)(2)). There are a number of
provisions relating to MACT. First, EPA was directed to determine major sources of
the 188 HAPs listed in §112(b) by November 15, 1991. EPA published the source list
in the Federal Register on July 16, 1992; it included 174 source categories (57 FR

1The amendments originally specified 190 pollutants. One, hydrogen sulfide, appeared on
the list due to a clerical error and was removed in 1991 by Pub. Law 102-187. A second,
caprolactum, was removed by EPA in 1996 using its authority under Section 112(b)(3).

31576). Second, the CAA Amendments directed EPA to develop a schedule of
promulgation dates for source category standards by November 15, 1992. EPA
published the schedule December 3, 1993 (58 FR 63941). Section 112(e) of the
Clean Air Act partially determined EPA’s schedule by requiring promulgation of
emissions standards for:
!40 categories by November 15, 1992
!25% of categories by November 15, 1994
!An additional 25% of categories by November 15, 1997
!All categories by November 15, 2000.
Based on EPA’s original listing of 174 source categories, the mandated
percentage targets listed above require standards for:
!40 source categories within 2 years of the enactment of the 1990 Amendments
!A total of 44 source categories within 4 years (an additional 4)
!An additional 43 source categories within 7 years
!All 174 within 10 years (an additional 77).
Aside from these specific deadlines, EPA had discretion in prioritizing
promulgation of standards based on: adverse effects on public health and the
environment, the quantity and location of HAPs emissions, and the efficiency of
grouping source categories based on pollutants rather than based on processes.
The standard to which sources will be held, the MACT standard, is codified in
Section 112(d)(2) as
the maximum degree of reduction in emissions ... (including a prohibition on
such emissions, where achievable) that the Administrator, taking into
consideration the cost of achieving such emission reduction, and any non-air[-]
quality health and environmental impacts and energy requirements, determines
is achievable for new or existing sources.
The law requires that standards for new sources be no less stringent than the
emissions control achieved in practice by the best-controlled similar source. Existing
sources may have a less stringent standard than new sources, but not less stringent
than either the average control achieved by the best-performing 12% of existing
sources, or the average control achieved by the best-performing 5 sources, if there are
fewer than 30 sources of the HAP. These minimum standards for new and existing
sources are known as “MACT floors.”
Section 112 distinguishes between major sources and area sources. A major
source is a stationary source that emits, or has the potential to emit, 10 tons or more
per year of any single HAP or 25 tons or more per year of any combination of HAPs.
An area source is a stationary source that emits less than these threshold amounts.
Although area sources do not emit as much pollution as major sources, area sources
may create local pollution problems, particularly in densely populated urban areas.
In regulating area sources, the Administrator of the EPA may decide to base emission
standards on “generally achievable control technology,”or GACT, rather than

Congress also provided an assurance that air pollutants would be regulated even
if EPA failed to promulgate standards for source categories on time. This is known
as the “MACT hammer.” Section 112(j) states that if EPA has not promulgated
standards by 18 months after the deadline, then owners and operators will be forced
to apply to states for permits, and states will have to create regulations for the
remaining unregulated source categories. The last hammer date was May 15, 2002,
for MACT standards that were due November 15, 2000.
Current Status.EPA missed the May 15 deadline for 56 source categories.
Final rules for 10 source categories were promulgated as of July 22, 2002, meaning
46 source categories remain with no final MACT standards. However, some of these
source categories are the subject of proposed MACT standards, which might be of
assistance to states and industries should they be required to develop MACT
standards themselves.
Table 1 shows the evolution of EPA’s MACT schedule. The first row gives the
number of additional source categories due at each statutory deadline, based on the
percentage targets given in the 1990 Amendments. For instance, the standards for 4
source categories due by November 15, 1994 are in addition to the standards for 40
source categories EPA had to complete by November 15, 1992. EPA’s original
schedule was published December 3, 1993. The numbers in the second row are based
on this schedule. EPA has updated its schedule several times since then; the numbers
in the third row reflect EPA’s February 2002 schedule.
EPA refers to source categories as being in “bins;” source categories with
standards due November 14, 1992 are the “2-year bin” and source categories with
standards due November 15, 1994 are the “4-year bin.” The numbers in the last row
reflect how many source categories in each of the bins have final standards. For
instance, of the 6 source categories in the 2-year bin, EPA has promulgated MACT
standards for all of them; thus, EPA is finished with the 2-year bin. EPA is also
finished promulgating standards for source categories in the 4-year and 7-year bins.
It should be noted that the numbers in the last row do not reflect how many source
categories had standards promulgated by the deadlines. Rather, the numbers reflect
how many source categories scheduled for that deadline have been finalized. A
number of the standards in the 2-year, 4-year, and 7-year bins were promulgated after
their respective deadlines had passed.
The “Total” column of Table 1 reflects the total number of source categories
EPA has listed. Of the 196 source categories EPA had identified by July 2002, EPA
is finished with 150 of them – 87 source categories have final standards and 63
source categories have been delisted (indicating EPA has decided not to promulgate
a standard for that category) or subsumed (incorporated into a broader category for
which a MACT standard has been promulgated). Thus, 46 source categories remain
with no final MACT standards. All of these source categories are in the 10-year bin.

ACT Source Category Final Rules: Schedule Requirements and Actual Promulgations as of July 22, 2002
Deadlines 11/15/1992 11/15/1994 11/15/1997 11/15/2000 TOTAL
Mandateda 40 4 43 87 174
inal Scheduleb639c42d87174
A’s Most Recent Schedulee63948103f196g
iki/CRS-RL31515ationsh6 final38 final25 final18 final87 final
g/w1 delisted23 delisted or subsumed39 delisted or subsumed63 delisted or subsumed
s.or27 proposed27 proposed
leak19 no action19 no action
://wikiese numbers were determined using the CAA Amendments promulgation schedule and the EPAs original 174 source categories. 57 Fed. Reg. 31576.
httpese numbers are based on EPAs original schedule, published December 3, 1993. 58 Fed. Reg. 63941.
is number includes the source categoryCoke Ovens: Charging, Topside, and Door Leaks that had a statutorily mandated promulgation of December 31, 1992.
is number includes the source categoryPublicly Owned Treatment Works that had a statutorily mandated promulgation date of November 15, 1995.
ese numbers are based on EPAs most recent schedule, published February 12, 2002. 67 Fed. Reg. 6521.
is number includes two source categories (wet-formed fiberglass mat production and coal- and oil-fired electric utility steam generating units) that were listed late enough to be
subject to Section 112(c)(5). This section gives EPA two years to develop MACT standards for source categories listed after November 15, 1998. These source categories were
not technically due November 15, 2000.
A has added, delisted, and subsumed a number of source categories since the publication of the original source category list in 1992. Of the original 174 source categories listed,
17 have been delisted, 46 have been subsumed into other categories, and 22 have been added, giving a total of 133 source categories requiring standards as of July 22, 2002.
e numbers in this row reflect how many source categories scheduled to be promulgated by the corresponding deadline have been finalized. The numbers do not reflect how many
standards EPA had promulgated by the deadline.
Congressional Research Service, using Federal Register notices and rules and information available on EPAs Air Toxics Website:
y 22, 2002).

Under the Clean Air Act, owners and operators of facilities for which MACT
standards were not finalized were to submit complete permit applications by May 15,
2002. But, on April 5, 2002, EPA promulgated a final rule to ease the requirements
for these sources, allowing them an additional two years to submit most of the
information required in a permit application.
State permit applications (or Title V permits) are submitted in two parts. Part
1 contains basic information about the facility, while Part 2 is much more detailed
and requires identification of MACT, the MACT floor, and other information. At a
minimum, EPA noted that Part 1 would have to be submitted by the hammer date to
comply with the statute. Therefore, every facility that could reasonably be expected
to be subject to MACT standards had to submit a Part 1 application by May 15, 2002.
However, EPA extended the deadline to submit Part 2 of the application to May 15,
2004. EPA anticipates promulgating all MACT rules by this later date. Increasing the
time to submit the Part 2 application ensures that the estimated 84,000 facilities that
were subject to the MACT hammer provisions will not needlessly expend resources
determining case-by-case MACT standards.
The Sierra Club filed suit against the EPA challenging the April 5 rule
extending the deadline. In the Sierra Club’s view, EPA’s action was clearly illegal
under the CAA. As of July 2002, this case was in settlement negotiations.2 According
to EPA’s Office of Air Quality Planning and Standards, a settlement with the Sierra
Club may include decreasing the amount of time for permit submission from two
years to 18 months or even one year.3
Another recent controversial MACT matter involves a new source category. In
December 2000, EPA added coal- and oil-fired electric utility steam generating units
to its list of source categories. Under Section 112(n)(1)(A), Congress directed EPA
to study HAP emissions from these sources and regulate them if appropriate and
necessary.4 Congress also directed EPA to study mercury emissions from these
sources.5 In the 2000 notice, EPA remarked that it did not find regulation of HAP
emissions for natural-gas-fired electric utility steam generating units to be necessary.
However, coal- and oil-fired units would be subject to MACT standards for mercury
emissions. Two utility groups filed suit shortly after the ruling to challenge EPA’s
listing of the source category.6 Although the case was dismissed, this rulemaking, a
proposal for which is expected in 2003, may be one of the most controversial MACT
standards that EPA will issue because some utilities do not believe they should be
subject to MACT for mercury emissions and are likely to file lawsuits once EPA
issues a MACT standard.

2Sierra Club v. EPA, No. 02-1135 (D.C. Cir. filed April 25, 2002).
3Steve Cook, “Settlement Talks Could Make 80,000 Sources Subject to ‘MACT Hammer,’
EPA Official Says,” Daily Environment Report (28 Jun. 2002): A-1.
4EPA, Study of Hazardous Air Pollutant Emissions from Electric Utility Steam Generating
Units – Final Report to Congress. February 1998. EPA-453/R-98-004.
5EPA, Mercury Study Report to Congress. December 1997. EPA-452/R-97-003. (June 28, 2002).
6Utility Air Regulatory Group v. EPA, No. 01-1074 (D.C. Cir. filed February 16, 2001).

Remaining Issues.Lawyers in the Sierra Club case have expressed
concern that EPA will not be able to promulgate all MACT standards by 2004. If that
is the case, the amount of money the states would have to spend to determine site-
specific standards could be significant. EPA estimated in its April 5 rule that the cost
to affected facilities to file Part 1 of the permit would be $9,000,000 (67 FR 16593).
Businesses, on the other hand, have estimated that the costs to determine case-by-
case MACT (Part 2 of the permit application) could be between $10,000 and $50,000
per affected facility. With EPA’s estimate of 84,000 facilities that are subject to the
MACT hammer, the costs could run from $840 million to $4.2 billion to determine
case-by-case standards.
By contrast, EPA spent only $17 million annually in FY2001 and 2002 on the
MACT program. The Agency’s request for 2003 calls for a cut in program spending,
to $12 million. EPA’s position is that spending can be cut because the regulation7
development program is winding down. It is the case that standards remain to be
issued for less than half of EPA’s identified source categories. However, delays
continue in the program and the outcome of the MACT hammer extension is still
unclear. EPA may not need as large a budget because proposed standards for many
of the remaining source categories have already been developed. However, the
decreased budget request may also reflect a sense that some of the burden for
developing MACT standards may now fall to states and industries.
The Residual Risk Program
A second program of the 1990 CAA Amendments is known as the residual risk
program. This program, found in Section 112(f) consists of two parts: a report to
Congress and the development of additional HAP emissions standards. The report
to Congress, due November 15, 1996 was required to include a number of items:
!available methods of reducing residual risk and their costs
!public health significance of residual risk
!actual health effects to people living near emissions sources
!risks from background concentrations of HAPs
!negative health or environmental effects created by reducing residual risk
!available epidemiological and other health studies
!methods of calculating residual risk
!uncertainties in the risk assessment methodology
!recommendations regarding legislation on residual risk.
EPA issued its Residual Risk Report to Congress in March 1999.8 Because EPA
had not issued any residual risk standards at the time of the report, many of the
mandated items were answered generally rather than specifically. EPA did not
provide concrete information for any of the first six items listed above. For instance,

7Personal Communication, Tom Coda, U.S. EPA Office of Air Quality Planning and
Standards, July 2, 2002.
8EPA, Residual Risk Report to Congress. March 1999. EPA-453/R-99-001. (June 11, 2002).

EPA addressed broad methods of reducing residual risk, but did not address any
specific technological or commercially available methods to reduce risks.
In the report, EPA did broadly outline its residual risk framework. This
framework is based on methodology developed by the National Research Council of
the National Academy of Sciences and the Commission on Risk Assessment and
Risk Management.9 EPA noted that uncertainties associated with this framework fall
into four categories: uncertainties surrounding exposure, uncertainty in models,
uncertainties in input values, and inherent data variance. The Agency identified some
questions that can be asked to reduce the uncertainty associated with residual risk
calculation. EPA elected not to make any recommendations regarding legislation on
residual risk, as the Agency felt the current legislation gave it sufficient authority and
flexibility. Finally, EPA noted in the report that it plans to set priorities for assessing
residual risk based on “achieving the largest, most cost-effective risk reductions
first,” while still respecting statutory deadlines based on MACT promulgation.10
The second part of the residual risk program is the promulgation of additional
standards. Congress, in Section 112(f)(2), directed EPA to develop more stringent
standards for the source categories identified under the MACT program if they would
be necessary to “provide an ample margin of safety to protect public health” or to
prevent adverse environmental impacts. When assessing whether adverse
environmental impacts should be prevented, EPA is to take into account costs,
energy, safety and other relevant factors. If MACT standards have not reduced the
lifetime cancer risk of the most exposed person to one in one million, then EPA
should promulgate additional standards.
Residual risk standards are to be promulgated within 8 years (9 years for MACT
standards due November 15, 1992) after the promulgation of MACT standards. The
standards become effective immediately upon promulgation. However, existing
sources have up to 90 days to meet the standards and can apply for a waiver of up to
2 years. Section 112(f)(5) exempts EPA from assessing residual risks of area sources
subject to generally achievable control technology (GACT) standards. However, EPA
has interpreted the legislation to mean that residual risk must still be assessed for area
sources subject to MACT standards.11
Current Status.The first MACT standards were promulgated September 22,
1993. Because these standards were due November 15, 1992, EPA has 9 years to
issue the standards. Thus, the latest date to promulgate residual risk standards for the
first group of sources is September 22, 2002. EPA promulgated standards for coke
ovens on November 27, 1993. Because these standards had a statutorily mandated
promulgation date of December 31, 1992, only 8 years are allowed for residual risk
standard setting. Thus, under this schedule, EPA was slated to issue a residual risk

9The Commission on Risk Assessment and Risk Management was a temporary body
established by Section 303 of the 1990 CAA Amendments. This commission consisted of
10 appointees and was to evaluate several aspects of risk assessment, as well as the results
of the National Academy of Sciences report.
10EPA, Residual Risk Report to Congress, p. 113.
11EPA, Residual Risk Report to Congress, p. 106.

assessment for coke ovens by November 27, 2001. However, EPA has not yet issued
residual risk standards for any source categories.
EPA did release a draft residual risk case study of secondary lead smelters in

2000 for review by its Science Advisory Board. The Science Advisory Board (SAB)

is an advisory body of independent experts who provide scientific advice and review
to the Agency and relevant Congressional committees. Although the SAB stated that
EPA’s document used assumptions consistent with current practice and consistent
with the Report to Congress methodology, the SAB also expressed concerns about
missing elements and data.
Remaining Issues.Not long after its review of the draft residual risk case
study on secondary lead smelters, the SAB commented unfavorably on EPA’s ability
to carry out residual risk assessment in a “credible manner.”12 The SAB’s concern
stems from the dearth of scientific and monitoring information available regarding
HAPs emissions. The SAB is not certain that “scientific analysis will be able to13
generate the type of information envisioned in the [Clean Air Act Amendments].”
This comment on the residual risk program and the SAB’s request in the comment
that Congress review the mandates of the CAA Amendments led to a Senate
Environment and Public Works Committee hearing on October 3, 2000 that was
partially intended to examine issues with residual risk.
In their testimony, a witness from Sciences International, Inc., a witness from
the SAB, and a witness representing the American Chemistry Council (ACC)
addressed the lack of resources available to EPA to complete residual risk
assessments.14 The SAB witness and the ACC witness questioned the viability of the
models used by EPA on account of a lack of peer review. The ACC witness and the
Sciences International witness noted that EPA’s risk information database contains
outdated information. The witness from the ACC indicated concern about the
inability of EPA to make residual risk determinations by the statutory deadlines. The
witness from the SAB and the witness from Sciences International criticized EPA for
using overly conservative estimates in its assessments. The Sciences International
witness noted that such estimates could result in a significant number of
overestimated risks. The witness representing the ACC also criticized the residual
risk program because of the mandate that industry comply with residual risk
standards within 90 days of their promulgation. This witness noted that meeting this
timetable may be impossible for many industries.
However, a witness from the National Wildlife Federation noted the importance
of the residual risk program. Positive comments from this witness about the residual
risk program included its ability to be comprehensive in its approach, to address harm
from persistent bioaccumulative toxics, and to incorporate cumulative risk. EPA also

12SAB, Executive Committee Commentary on Residual Risk Program, EPA-SAB-EC-COM-

00-005, p. 2. (June 11, 2002).

13SAB, Executive Committee Commentary, p. 2.
14Since this hearing, appropriations for the residual risk program have more than tripled, but
they are still only $5.2 million in FY2002.

defended its residual risk approach, stating that its residual risk estimates are
plausible because they reflect the suggestions of the scientific community. Further,
EPA stated that it does not use conservative assumptions for all the parameters that
go into a residual risk assessment. In fact, EPA noted that “where we lack adequate
information to support a quantitative assessment for a pollutant, we implicitly assume
that the risks from that pollutant are zero.”15
In July 2001, EPA requested information from the public on improvements to
the Integrated Risk Information System (IRIS). This database is the one that was
criticized by certain witnesses at the Senate hearing for being outdated and
inaccurate. EPA indicated it would use the information provided to set the IRIS
2002-2005 agenda, as well as determine the need for updating IRIS more frequently
and for adding new chemicals to the database. A summary of this needs assessment
is to be posted on EPA’s IRIS webpage when completed. As of July 2002, this
summary had not yet been posted.
It is unclear whether EPA’s methodology will be successful in meeting the goals
Congress envisioned in setting up the residual risk program. While this methodology
has been reviewed by the scientific community and the SAB asserts that the Agency
has made a “good faith start”16 in using the methodology, it may be that the science
and data available at this point in time are not adequate for such rigorous assessment.
EPA has missed the deadline for coke oven residual risk standards; decisions on 12
other standards are due by the end of 2002. It is unlikely that any of these standards
will be promulgated by that deadline.
The Urban Air Toxics Program
Under Section 112(k)(3) of the Clean Air Act, EPA is required to develop a
strategy to address air toxics in urban areas. These air toxics are generally emitted
from area sources. Area sources can be distinguished from major sources, as
discussed in the MACT section. Area sources are defined as those sources emitting,
or having the potential to emit, less than 10 tons of any single HAP or less than 25
tons of any combination of HAPs per year.
Congress ordered EPA to identify at least 30 air toxics emitted from area
sources that represent the greatest threat to urban public health and to identify the
sources of these air toxics. Area sources that represent 90% of emissions of each of
the 30 air toxics must be subject to standards such as MACT or GACT (Section
112(k)(3)(B)(ii)). Congress also directed that this strategy “shall achieve a reduction
in the incidence of cancer attributable to exposure to [HAPs] emitted by stationary
sources of not less than 75 [percent]” (Section 112(k)(3)(C)). The strategy was to be
submitted to Congress by November 15, 1995.

15Robert Brenner, “Testimony of Robert Brenner, Deputy Assistant Administrator, Office
of Air and Radiation, U.S. Environmental Protection Agency, Before the Committee on
Environment and Public Works, United States Senate, October 3, 2000.” Available at: (July 1, 2002).
16SAB, Executive Committee Commentary, p. 1.

EPA released the final Integrated Urban Air Toxics Strategy on July 19, 1999
(64 Fed. Reg. 38706-38740). This strategy identified 33 urban HAPs, with 30 of
these having significant contributions from area sources. The HAPs and their main
sources can be found in Table 2 at the end of this report. EPA identified 29 area
source categories of the 33 urban HAPs. In releasing the strategy, the Agency also
delineated its time line to implement the rest of the requirements under Section
112(k). By 2004, EPA expects to promulgate standards for the new area source
categories it identified in the strategy (some area source categories were already
regulated or were proposed for regulation under MACT standards or other authorities
of the CAA). By 2006, EPA expects to promulgate additional standards to meet the
statutory requirement that sources representing 90% of the emissions of each urban
HAP be controlled. EPA expects to finish promulgating these standards in 2009.
Because compliance is required by no later than 3 years from promulgation of the
standard, EPA projects that all sources will be in compliance by 2012. However,
under the original deadlines outlined in the CAA, all sources were scheduled to have
been in compliance with the strategy by November 15, 1999.
Congress also directed EPA to establish a research program to study the health
effects of HAPs (Section 112(k)(2)). EPA’s nationwide program is known as the
National-Scale Air Toxics Assessment (NATA). NATA gives an inventory of air
toxics, their ambient concentrations, estimates of population exposures, and a
characterization of the potential public health risk of these urban air toxics. EPA will
use this information to develop baseline information, determine the progress made
in meeting reduction goals, and prioritize the list of urban HAPs. In the future,
NATA is expected to evaluate all 188 HAPs and diesel particulate matter.
Current Status.In May 2002, EPA finished releasing the first NATA data.
The first set of NATA data is based on 1996 emissions data for the 33 urban HAPs.
These data show that when risks from all urban air toxics are combined, 200 million
people are estimated to have a potential cancer risk greater than ten in one million.
For 20 million people, the potential cancer risk is estimated to be greater than 100 in
one million. NATA found benzene, chromium, and formaldehyde to pose the greatest
potential risk of cancer across the entire United States while arsenic, coke oven
emissions, 1,3-butadiene and polycyclic organic matter pose the greatest potential
risk of cancer regionally. Table 2, at the end of this report, identifies the main sources
of these urban HAPs.
EPA also used NATA to assess the cancer risk from background sources of
urban HAPs. EPA estimates that the entire population of the United States exceeds
a cancer risk of ten in one million due to background concentrations of air toxics
compounds. This assessment partially answers one question EPA left open in its

1999 Residual Risk Report to Congress.

EPA’s Integrated Urban Air Toxics Strategy noted that the Agency would
conduct a research program at the urban level. This urban research program takes a
case study approach – studying a few cities that will serve as models for
understanding urban HAPs risk. The first city selected as a pilot for this project is
Cleveland, Ohio. The pilot project began in March 2001. The three goals for the
project are to reduce air toxics in Cleveland, sustain efforts to reduce air toxics over
time, and replicate the approach in other urban areas. The pilot project’s initial

programs include a car pooling and public transportation campaign, low sulfur diesel
fuel for bus transportation, and a toxics collection day.
Section 112(k)(5) also required EPA to provide reports to Congress in 1998 and
2002. In these reports, Congress directed EPA to include information on the actions
taken to reduce public health risks from area source HAP emissions and to identify
urban areas that have high risks to public health due to continuing emissions from
area sources. EPA submitted its first urban air toxics report to Congress in 2000.17
In this report, EPA did not identify the specific urban areas with continuing health
risks. EPA did outline the steps it will take to reduce public health risks from HAPs
based on the final Integrated Urban Air Toxics Strategy released in 1999. EPA also
identified thirteen subject areas in need of further research in the 2000 report.
Remaining Issues.EPA has not yet regulated at least 13 of the area source
categories identified in its 1999 Integrated Urban Air Toxics Strategy. Since issuing
the strategy EPA has added 19 area source categories, bringing the total number of18
area source categories to 48. Of these additional area source categories, 18 have yet
to be regulated. EPA’s schedule indicates that the area sources responsible for 90%
of the emissions of each of the urban HAPs will not all be regulated until 2009. In
addition, EPA still needs to identify the metropolitan areas with increased health
risks due to continued emissions of HAPs. The recently released NATA data are
most meaningful at the state or national level and therefore do not help EPA identify
metropolitan areas with increased health risks.
The Mobile Source Air Toxics Program
Congress included in the CAA Amendments of 1990 a program to evaluate and
regulate air toxics from mobile sources. Under Section 202(l) of the CAA, codified
as 42 U.S.C. 7521, Congress directed EPA to complete a study on the need for and
feasibility of controlling mobile source air toxics (MSATs). The focus for the study
was to be on those MSATs posing the greatest risk to human health or that have
significant uncertainty associated with their risks to health; in particular, benzene,
formaldehyde and 1,3-butadiene were to be included in the study. The study was due
May 15, 1992. EPA completed the study in 1993, and updated it in 1999.
Subsequent to the study, EPA was directed to promulgate rules containing
“reasonable requirements” to control MSAT emissions. The regulations are to
achieve the greatest degree of emission reduction through the application of available
technology, taking into consideration costs, noise, energy, safety, lead time, and
standards previously set under Section 202(a) (42 U.S.C. 7521(l)(2)). At a minimum,
Congress charged EPA with promulgating rules applying to benzene and

17EPA, National Air Toxics Program: The Integrated Urban Strategy, Report to Congress.
July 2000. EPA-453/R-99-007. (June 11,


18EPA added secondary aluminum production as an area source on January 30, 2001 (66
Fed. Reg. 8820-8829) and added another 18 area sources on June 26, 2002 (67 Fed. Reg.


formaldehyde. These regulations are to be promulgated under Sections 202(a)(1) and


Section 202(a)(1) requires EPA to regulate MSAT emissions from new motor
vehicles or new motor vehicle engines when there is a danger to the public health and
welfare. Section 211(c)(1) allows EPA to control or prohibit the manufacture,
introduction or sale of fuel or fuel additives that either 1) contribute to air pollution
that endangers the public health or welfare or 2) impairs the performance of emission
control devices that are either currently in use or are expected to be in use in a
reasonable time. Section 202(a)(1) applies only to motor vehicles and motor vehicle
engines, while Section 211(c)(1) applies to fuel being used in motor vehicles, in
motor vehicle engines, in nonroad engines, or in nonroad vehicles.
Current Status.In March 2001, EPA issued its final MSAT emission control
rule (66 Fed. Reg. 17230-19273). This rule identified 21 MSATs, including 13 that
are also considered urban air toxics. Twenty of these air toxics are listed in Section
112(b) of the CAA. The last MSAT is diesel particulate matter and diesel emission
organic gases (DPM + DEOG), which are a part of diesel exhaust. Diesel exhaust is
not included in the list of 188 air toxics developed by Congress, though many of its
constituent chemicals are. In its final rule, EPA noted that no new vehicle standards
(Section 202(a)(1)) were necessary at the time to reduce impacts on the public health
and welfare. Several programs that were put in place to meet other requirements of
the Clean Air Act have reduced MSAT emissions, including the National Low
Emission Vehicle Program, limits on gasoline volatility, the reformulated gasoline
program, and the phase out of leaded gas.
EPA did finalize a fuel-based standard under Section 211(c)(1) authority in the
March 2001 rule. This fuel-based standard is known as the gasoline toxic emission
performance requirement, or TPR. TPR requires that gasoline not exceed the average
total toxic emissions emitted from 1998-2000 beginning January 1, 2002. These
emissions include benzene and formaldehyde, among others, thus complying with
Congress’ desire that EPA regulate these two MSATs. This standard essentially
codifies the fact that many refineries were overcomplying with emissions standards
during the 1998-2000 period. Thus, the rule prevents future increases in MSAT
emissions that might have occurred had refineries stopped making gasoline that
overcomplied with the previous standards. In the rule, EPA committed to a future
rule-making by 2004.
EPA’s first NATA assessment provided some information on the hazards of
mobile sources. According to the first set of NATA data, which again is based on
1996 emissions, more than 100 million people face a cancer risk of greater than ten
in one million from mobile sources. EPA’s assessment also estimates that
approximately 200 million people face a noncancer risk (respiratory irritation) from
emissions of acrolein that are above reference concentrations. However, EPA
believes that by 2007, existing standards for new vehicles will have decreased risks
of onroad exposures to 50% of 1996 levels. Risks of nonroad exposures are also
expected to decrease.
Remaining Issues.The March 2001 MSAT rule was not without
controversy. Because EPA had originally proposed to regulate only the benzene

content of gasoline, but then finalized the rule as one that regulates gasoline toxic
emission performance, many refineries were “surprised” by the final rule. Some in
the industry noted that the final rule could be very expensive for refiners. EPA,
however, believed that the rule would not impose large costs on the industry because
the rule did not ask industry to install any new technology. Environmental groups and
others disagreed with industry and sued EPA, claiming that the rule did not go far
enough in regulating mobile air toxics.19 These groups consider EPA to be in
violation of the CAA because the Agency has not promulgated any rules to control
MSAT emissions. The case has not yet been adjudicated.
Congress amended the CAA in 1990 in part to ensure that EPA would regulate
HAPs more quickly than it had in the previous 20 years. Although EPA has made
much more progress under the programs created in 1990 than it had under the 1970
Act, all four of the air toxics programs are behind schedule. While EPA has
promulgated MACT standards for 87 source categories as of July 22, 2002, 46 source
categories remain without final standards. If EPA’s extension of the MACT hammer
deadline is found to be unlawful, states and industries will have to determine their
own MACT standards on a case-by-case basis; such an endeavor could be costly.
No standards have been issued in the residual risk program. Decisions on 12
standards are statutorily mandated for the end of 2002 but are unlikely to be finished.
According to EPA’s own urban air toxics schedule released in 1999, the urban air
toxics program is about 13 years behind the schedule established by Congress.
Finally, the fuel-based standards for mobile source HAP emissions have been
challenged for not going far enough in regulating these sources.
In light of the deadlines that were missed and will continue to be missed by
EPA, Congress might decide to examine the factors that have contributed to the
Agency’s difficulty in implementing the four air toxics programs under the schedule
Congress established. These factors might include lack of adequate resources,
limitations of the available data and analytical tools, or management issues within
EPA, among other possibilities.

19Sierra Club v. EPA, No. 01-1228 (D.C. Cir. filed May 24, 2001).

Table 2. Urban Air Toxics and Their Major Sources
Urban Air ToxicMain Sources of Emissions
AcetaldehydedResidential fireplaces; wood stoves
AcroleindSmoking; automobiles; oil and coal power plants
AcrylonitrileAcrylic acid and modacrylic fiber manufacture
Arsenic compoundsc,dInorganic arsenic emissions are mainly from
metal smelters, soil, and burning plywood
treated with an arsenic wood preservative
Benzeneb,dTobacco smoke; industries that manufacture or
use benzene
Beryllium compoundsWorkplaces where beryllium is mined,
processed, or converted into alloys or chemicals

1,3-Butadienec,dOil refineries; chemical manufacturing plants;

plastic and rubber factories
Cadmium compoundsFossil fuel use; incineration of municipal wastes
Carbon TetrachlorideaManufacture or use of the chemical
ChloroformWastewater; pulp and paper mills; drinking
water; swimming pools; hazardous waste sites;
sanitary landfills; workplaces that manufacture
and use chloroform
Chromium compoundsb,dFerrochrome production
Coke oven emissionsa,cAluminum, steel, graphite, electrical, and
construction industries
1,2-DibromoethaneProduction and process facilities; past use in
(ethylene dibromide)aleaded gasoline and as a fumigant
1,2-DichloropropaneProduction and use of the chemical
(propylene dichloride)
1,3-DichloropropeneApplication, manufacture, or formulation as a
soil fumigant
Ethylene dichloride (1,2-Production, storage, use, transport, and disposal
dichloroethane)of the chemical
Ethylene oxideSmoking; occupations that manufacture, process
or use to sterilize; fumigation
Formaldehydeb,dPower plants; manufacturing facilities; smoking;
incinerators; exhaust

HexachlorobenzeneIndustrial sites or waste facilities where
chemical is used/disposed
HydrazineUse as a chemical blowing agent, photography
chemical, or pharmaceutical intermediate
Lead compoundsdLeaded gasoline combustion; tobacco smoke;
combustion of solid waste, coal, and oils; lead
smelters; lead-based paints and lead pipes
Manganese compoundsdIron and steel plants; power plants; coke ovens
Mercury compoundsdMethyl mercury: eating contaminated fish;
Elemental mercury: occupational exposures
Methylene chlorideSpray painting or other chemical uses in the
(d ichloromethane) workplace
Nickel compoundsdNickel production, processing, and use; contact
with items containing nickel
Polychlorinated biphenylsRedistribution of PCBs already present in soil
(PCBs)and air
Polycyclic organic matterFormation during combustion
(P O M ) c,d
QuinolineManufacture of dyes; metallurgical processes
2,3,7,8-tetrachlorodibenzo-Combustion of fossil fuels and wood; byproduct
p-dioxin & congeners &of some industrial processes that use chlorine
TCDF congeners (dioxin)
1,1,2,2-TetrachloroethaneChemical production activities in which it is an
intermediate product
TetrachloroethyleneDrycleaning; industries using or manufacturing
(perchloroethylene)the chemical
TrichloroethyleneFactories where it is manufactured or used;
adhesives and spot removers
Vinyl ChlorideUse, production, transport, storage, and disposal
of the chemical in the workplace
Source: EPA,Appendix: HAP Profiles.” National Air Toxics Program: The Integrated Urban
Strategy, Report to Congress. July 2000. EPA-452/R-99/007. Available at (June 11, 2002).
a These three HAPs have emissions mainly from major sources rather than area sources.
b These three HAPs are associated with the greatest potential risk of cancer across the United States.
c These four HAPs have been identified to pose the greatest risk of cancer regionally.
d These twelve urban air toxics are also considered mobile source air toxics.