[[pp. 52977-53026]] NESHAPS: Final Standards for Hazardous Air Pollutants for

Note: EPA no longer updates this information, but it may be useful as a reference or resource.

[Federal Register: September 30, 1999 (Volume 64, Number 189)]
[Rules and Regulations]
[Page 52977-53026]
From the Federal Register Online via GPO Access [wais.access.gpo.gov]
[DOCID:fr30se99-27]

[[pp. 52977-53026]] NESHAPS: Final Standards for Hazardous Air Pollutants for
Hazardous Waste Combustors

[[Continued from page 52976]]

[[Page 52977]]

sources under the CAA to title V permitting is discussed in a separate
section.
1. How Are the Title V Permitting Requirements Applicable?
We intend, by placing the new standards only in 40 CFR part 63 and
not cross-referencing them in RCRA regulations, to rely on existing air
programs to implement the new requirements, including operating permits
programs developed under title V. All hazardous waste combustors
subject to the MACT standards promulgated in this rule will thus be
subject to title V permitting requirements for air emissions and
related operating requirements (this includes hazardous waste
combustors that are considered area sources under the CAA, as discussed
in more detail below). In this rule, we are not amending any of the
existing air permitting procedures. The procedures of 40 CFR part 71
for federal operating permits, or a State title V program approved
under part 70, will remain applicable. Thus, all current CAA
requirements governing permit applications, permit content, permit
issuance, renewal, reopenings and revisions will apply to air emissions
from hazardous waste combustors pursuant to promulgation of the
hazardous waste combustor MACT standards.²⁹³
---------------------------------------------------------------------------

\293\ Requirements of other CAA permitting programs, such as
construction permits, will continue to apply, as appropriate, to the
HWC's sources subject to today's rule.
---------------------------------------------------------------------------

The public participation requirements for title V permits in parts
70 and 71, such as allowing an opportunity for public hearing and
public comments on draft permits, also apply (see 40 CFR 70.7(h) and
71.11). We are committed to enhancing public participation in all of
our programs. In 1996, we published a guidance manual on public
involvement in the RCRA program intended to improve cooperation and
communication among all participants in the RCRA permitting process
(RCRA Public Participation Manual, EPA530-R-96-007, September 1996).
Although the Manual is written in the context of the RCRA program, the
principles are common to all program areas. For example, the Manual
encourages early and meaningful involvement for communities and open
access to information. It also acknowledges the important role of
public participation in addressing environmental justice concerns.
Since these principles are applicable in all situations, we encourage
air programs and sources subject to the hazardous waste combustor MACT
standards to refer to the RCRA manual for additional guidance on
implementing effective public participation activities.
2. What Is the Relationship Between the Notification of Compliance and
the Title V Permit?
The hazardous waste combustor MACT standards promulgated in this
final rule include emissions limitations for several hazardous air
pollutants, as well as detailed compliance, testing, monitoring, and
notification requirements. Under these provisions, you not only
demonstrate compliance with the emissions limitations, but also
demonstrate that you have established operating requirements and
monitoring methods that ensure continuous compliance with those limits.
These demonstrations are made during a comprehensive performance test
and subsequently documented in an NOC.
We are requiring, in Sec. 63.1210(f), that you comply with the
general provisions governing the NOC codified in Sec. 63.9(h). Those
provisions specify that in addition to describing the air pollution
control equipment (or method) for each emission point for each
hazardous air pollutant, the NOC also must include information such as:
methods that were used to demonstrate compliance; performance test
results; and methods for determining continuous compliance (including
descriptions of monitoring and reporting requirements and test
methods). We also are requiring in Sec. 63.1207(j) that you comply with
the all of the operating requirements specified in the NOC upon
submittal to the Administrator.
Although these requirements are self-implementing, in that you must
comply in accordance with the time frames set forth in today's rule,
the requirements are ultimately implemented through title V operating
permits (see 40 CFR parts 70 and 71). Section 63.1206(c)(1) specifies
that: (1) You can only operate under the operating requirements
specified in the DOC or NOC (with some exceptions as laid out in the
regulations); (2) the DOC and NOC must contain operating requirements
including, but not limited to, those in Sec. 63.1206 (compliance with
the standards and general requirements) and Sec. 63.1209 (monitoring
requirements); (3) operating requirements in the NOC are applicable
requirements for the purposes of 40 CFR parts 70 and 71; and, (4)
operating requirements in the NOC must be incorporated into the title V
permit. In addition, because title V permits can only be issued if,
among other conditions, ``the conditions of the permit provide for
compliance with all applicable requirements'' (see Secs. 70.7(a)(1)(iv)
and 71.7(a)(1(iv)), parts 70 and 71 are clear that title V permits must
contain the operating requirements documented in the NOC.
As mentioned above, you must comply with all operating requirements
specified in the NOC as of the postmark date when the NOC is submitted
to the Administrator. Operating requirements documented in the NOC must
be included in your title V permit--either through initial issuance if
you do not yet have a title V permit, or through a permit revision if
you already have a permit. Including information from the initial NOC
in title V permits should not create the potential for any compliance
conflicts. Because it is the first time the NOC operating requirements
are incorporated into the permit, there would be no requirements
already on permit with which the NOC would conflict.
However, the potential for compliance conflicts could be created
when a subsequent NOC is submitted. For example, you are required to
conduct periodic comprehensive performance testing (see
Sec. 63.1207(d)(1)). Subsequent to each test, you must submit another
NOC to the Administrator. Because of the dynamics of the testing and
permitting cycles, it is possible that once you have information from
the initial NOC in the permit, you could find yourself, after
subsequent testing, in a situation where there might be potentially
conflicting requirements with which you must comply (i.e., requirements
in the title V permit and requirements in the most recently submitted
NOC). This might occur, for example, if any of the operating
requirements changed from the previous test.²⁹⁴ The
potential for compliance conflicts that might arise from this situation
can be avoided, however, by following the guidance presented below.
---------------------------------------------------------------------------

\294\ On the other hand, if the limits did not change, there
would be no conflict between the NOC and the permit.
---------------------------------------------------------------------------

The requirements in parts 70 and 71 govern the timing and
procedures for permit issuance, revisions, and renewals, and you should
refer to those requirements when obtaining or maintaining your permit.
For today's rule, we provide guidance on what we recommend as to how
operating requirements in the NOC should be incorporated into title V
permits.²⁹⁵
---------------------------------------------------------------------------

\295\ We are recommending this approach as guidance in the
preamble, but not including any associated regulatory provisions.
This guidance is essentially an interpretation of the current part
70 and 71 rules.

---------------------------------------------------------------------------

[[Page 52978]]

For incorporating information from an initial NOC into a title V
permit, when you have an existing title V permit, we recommend that you
and your permitting agency follow the procedures for significant
modifications. The primary rationale for using these procedures is to
afford the public an opportunity to review all of the information
pertinent to your compliance obligations. We want to ensure a level of
public involvement when including operating requirements in title V
permits that is commensurate with that under RCRA. In RCRA, operating
parameters are initially developed pursuant to trial burns and
incorporated into permits either through initial issuance (in the case
of facilities operating under RCRA interim status) or through a RCRA
class 2 or 3 permit modification (in the case of new facilities). In
either situation, significant opportunities exist for public review and
input parallel to those under initial title V permit issuance or
significant permit modification procedures.
With regard to a subsequent NOC developed pursuant to periodic
performance tests, we prefer an implementation scheme for this rule
that avoids unnecessary permit revisions. Thus, we recommend that you
coordinate your five-year comprehensive performance testing schedule
with your five-year permit term to the extent possible. This would
allow changes in the NOC to be incorporated into the permit at renewal
rather than through separate permit revisions. This also helps to
minimize the number of permit revisions, as well as, the likelihood of
having two sets of requirements with which to comply.
We recognize, however, that such coordination may not always be
possible or feasible. At times, it may be necessary to include
information from the most recent NOC through a permit revision. We
expect that this will be accomplished using, at most, the minor permit
modification procedures in Sec. 70.7(e)(2) or Sec. 71.7(e)(1). Keeping
in mind that the information from the initial NOC was included either
as part of the initial permit issuance or as a significant revision,
the information was already subject to review by both the regulatory
agency and the public. Thus, the public should have a clear
understanding of your compliance obligations. The obligation to comply
with the emissions limitations in Secs. 63.1203, 63.1204, or
Sec. 63.1205 does not change even if any of the associated compliance
information, such as operating limits, is revised pursuant to
subsequent performance tests. Given our experience in regulating (under
RCRA) the types of sources subject to today's MACT standards, we do not
expect the information in a NOC to change significantly over time. We
have been regulating these sources for almost twenty years; the testing
and monitoring requirements we are promulgating in this rule reflect
the ``lessons learned'' over time. Thus, the initial set of compliance
parameters are likely to need primarily minor changes over time. You
and your regulatory agency also are experienced in setting operating
parameter limits and monitoring systems to ensure compliance with
performance standards. Again, this expertise and experience suggests
that primarily minor adjustments will need to be made. In light these
factors, we are confident that changes in the NOC may be appropriately
incorporated into title V permits using the minor permit revisions
procedures. Furthermore, regulatory agencies are obligated under
Sec. 63.1206(b)(3) to make a finding of compliance based on performance
test results. This requirement provides an additional administrative
safeguard to ensure that you are setting the proper operating limits.
The minor permit modification process will allow you to meet your
compliance obligations under Sec. 63.1207(j) and begin to comply with
the conditions in the NOC upon submittal (i.e., post-mark). Under
Secs. 70.7(e)(2)(v) and 71.7(e)(1)(v), you may make the change proposed
in the minor permit modification application immediately after filing
such application. Following this, you must comply with both the
applicable requirements governing the change and the proposed permit
terms and conditions (i.e., the information from the NOC that you are
incorporating into your permit). The provisions in this section also
ensure that you will not be in the position of having to choose between
compliance with the NOC or compliance with your permit because this
section also specifies that during this time period, you need not
comply with the existing permit terms and conditions you seek to
modify.²⁹⁶ Since the NOC is submitted to the Administrator
once you have a title V permit (see Sec. 63.9(h)(3)), we expect that
you will submit the NOC together with a minor permit modification
application. Any modifications added to the permit through this process
can be reviewed by the public at the time of permit renewal.
---------------------------------------------------------------------------

\296\ If, however, the source fails to comply with its proposed
permit terms and conditions during this time period, the existing
terms and conditions it seeks to modify may be enforced against it
(Secs. 70.7(e)(2)(v) and 71.7(e)(1)(v)).
---------------------------------------------------------------------------

We encourage permitting authorities to develop permits in a way
that minimizes the need for future permit revisions and is consistent
with the requirements in parts 70 and 71. For example, you may request
that your permitting authority develop a permit that contains
alternative operating scenarios. This would allow you to alternate
among various approved operating scenarios while concurrently noting
the change in your operating record.
3. Which RCRA Permitting Requirements Are Applicable?
The RCRA permitting requirements particular to incinerators and
boilers and industrial furnaces are found in 40 CFR 270.19, 270.22,
270.62, and 270.66. These permitting requirements apply to new
facilities, to those operating under interim status while they pursue a
permit, and to sources seeking to renew their permits. In today's final
rule, we amend the introductory text in each of these sections to
reflect that RCRA permitting requirements for hazardous waste combustor
air emissions and related operating parameters will not apply once you
demonstrate compliance with the requirements of the new MACT standards
by completing a comprehensive performance test and submitting a NOC to
the Administrator.²⁹⁷ The timing for the deferral of the
RCRA permitting requirements is consistent with the timing in today's
rule for the deferral of applicable standards in 40 CFR parts 264 and
265.
---------------------------------------------------------------------------

\297\ The final rule language in these sections differs from
that in the NPRM to reflect placement of the standards only in part
63 and deferral of RCRA controls to the air program.
---------------------------------------------------------------------------

Even though we rely on the title V permitting program to address
air emissions from hazardous waste combustors, we still need RCRA
permits at these sources to address: (1) Other RCRA regulations
applicable to all types of RCRA units, including hazardous waste
combustors, that are not duplicated under the CAA; (2) any risk-based
emissions limits and operating parameters, as appropriate; and (3)
other RCRA units at the facility. Also, new facilities (including new
hazardous waste combustor units) must obtain RCRA permits prior to
starting construction. Thus, the remaining RCRA permitting requirements
in 40 CFR part 270 governing permit applications and permit content
continue to apply. These

[[Page 52979]]

include the provisions in Secs. 270.10(k) and 270.32(b)(2), which
together provide authority to require a facility owner or operator to
submit information necessary to establish permit conditions and to
impose site-specific conditions, including risk-based conditions,
through the RCRA permit.
Even though you will still have two permits, the scope and subject
matter of each are distinguishable. The title V permit will focus on
the operation of the combustion unit (e.g., air emissions and related
parameters) while the RCRA permit will continue to focus on the other
basic aspects of hazardous waste management. The RCRA permit would thus
include conditions to ensure compliance with relevant requirements in
40 CFR part 264, including: General facility standards; preparedness
and prevention; contingency planning and emergency procedures;
manifesting; recordkeeping and reporting; releases from solid waste
management units; closure; post-closure; financial responsibility;
corrective action; storage; materials handling; and air emissions
standards for process vents and equipment leaks from tanks and
containers.
The only time we foresee that conditions in both RCRA and title V
permits may govern the same hazardous waste combustor operating
parameters and limits is when there is a need to impose more stringent
or more extensive risk-based conditions, e.g., under RCRA omnibus
authority, to ensure protection of public health and the environment.
This situation is discussed in greater detail in Part Three, Section IV
(RCRA Site Specific Risk Assessment Decision Process).
4. What Is the Relationship of Permit Revisions to RCRA Combustion
Permitting Procedures?
In June, 1994, we published a proposed rule for RCRA Expanded
Public Participation and Revisions to Combustion Permitting Procedures
(59 FR 28680, June 2, 1994). The proposal contained amended procedures
for interim status combustion facilities during the trial burn period
that were intended to make the procedures for interim status facilities
more like those governing permitted facilities. We finalized the
expanded public participation requirements (see section immediately
below), but did not finalize the proposed permitting revisions. At the
time we began to finalize the proposal, we were already committed to
issuing comprehensive air emissions standards under MACT. It was
anticipated that there would be overlap between the emissions standards
in the proposed MACT rule and the combustion permitting procedures in
the June 1994 proposed rule. It did not make sense to finalize
provisions in one rulemaking effort only to propose changing them yet
again in another rulemaking effort. Now, given the approach being
adopted in today's final rule to permit hazardous waste combustor air
emissions under title V of the CAA, there is no longer as strong a need
to pursue the amended procedures for RCRA permitting in the June 1994
proposal. We do not, therefore, intend at this time to finalize these
proposed permitting amendments.
5. What Is the Relationship to the RCRA Preapplication Meeting
Requirements?
In 1995, we finalized the expanded RCRA public participation
requirements (60 FR 63417, December 11, 1995). These included
requirements for a facility to advertise and conduct an informal
meeting with the neighboring community to discuss anticipated
operations prior to submitting a RCRA Part B permit application. Since
hazardous waste combustors subject to the new MACT standards (and title
V permitting) still need RCRA permits for other hazardous waste
management activities, you are still subject to the RCRA preapplication
meeting requirements in 40 CFR 124.31. Even though operations and
emissions associated with the combustor unit are now to be addressed
primarily under CAA requirements, we anticipate that the public will
continue to exhibit a great deal of interest in combustor activities at
RCRA meetings. They may not always be familiar with our administrative
``boundaries'' dictated by the various environmental statutes. Given
this potential lack of familiarity, and because combustor units and
emissions are already discussed at these meetings, we strongly
encourage you to continue including combustor unit operations in
discussions during RCRA preapplication meetings. Furthermore,
conditions for hazardous waste combustor activities may sometimes be
imposed under RCRA, for example, in cases where the results of a site-
specific risk assessment indicate a need for conditions more stringent
or more extensive than those imposed under MACT. You should be prepared
to discuss the site-specific risk assessment process and how it may
result in additional conditions being included to their RCRA permits.
All other public participation requirements in 40 CFR part 124
associated with the RCRA permitting process continue to apply. These
include requirements for public notice at application submittal, public
notice of the draft permit, opportunity for public comments on the
draft permit, and opportunity for public hearings. These requirements
also are explained in the RCRA Public Participation Manual (EPA530-R-
96-007, September 1996), which provides guidance on how to implement
RCRA public participation requirements, as well as, recommendations on
how to tailor public involvement activities to the situation at hand.
For example, if the community around a facility does not speak English
as a primary language, the manual encourages use of multilingual fact
sheets. As mentioned previously, we encourage you and States to apply
the principles contained in the RCRA manual to hazardous waste
combustor MACT compliance and title V activities as well.
C. Is Title V Permitting Applicable to Area Sources?
Under today's rule, hazardous waste combustors meeting the
definition of an area source will be subject to today's MACT standards
(see discussion in Part One, Section III.B). As discussed in the May
1997 NODA, under Sec. 63.1(c)(2), area sources subject to MACT are
subject to title V permitting as well, unless the standards for that
source category (e.g., subpart EEE for hazardous waste combustors)
specify that: (1) States will have the option to exclude area sources
from title V permit requirements; or (2) States will have the option to
defer permitting of area sources. We received several comments on our
NODA discussion (see 62 FR 24215) on the issue of subjecting area
sources to title V permitting. The comments were fairly evenly split--
several supported requiring area sources to obtain title V permits,
while several were against it. After considering the comments, we have
chosen not to provide the option to the States to exclude hazardous
waste combustor area sources from title V permitting requirements or to
defer permitting of these sources.
Commenters that support the Agency's position affirm that title V
permits serve an important role to incorporate all requirements
applicable to a source in one enforceable permitting document. They
maintain that the compliance certifications and opportunities for
public involvement inherent in the title V program will serve a useful
and valuable public service. Other supporters note that requiring all
hazardous waste combustors to obtain title V permits will help to
ensure that the permits are both consistent and adequate. The idea of

[[Page 52980]]

consistency being a desirable end result is echoed by others as well.
One commenter points out that area sources in several other source
categories are not exempt from title V permitting requirements, and
recommends that hazardous waste combustor area sources also be subject
to title V to maintain consistency with the rest of the MACT program.
Finally, some commenters state that if the Agency were not to pursue
title V permitting for hazardous waste combustor area sources, then the
Agency would have to strengthen the nontitle V permitting programs with
respect to public involvement and agency approval of modifications
relating to facility emissions.
We agree with these points. Title V permits clarify your regulatory
obligation, thereby making it easier for you to keep track of your many
compliance obligations across several air programs. Clarifying the
regulatory obligations improves compliance in many cases; we have seen
an increase in compliance among air sources with the advent of the
title V permitting program. For example, through the process of
applying for and issuing title V permits, applicable requirements of
which a source is unaware or with which it is found to be out of
compliance are identified. Once these requirements are included in a
title V permit, the source must certify compliance with these
requirements both initially and then on an annual basis.
We concur with commenters about the benefits of the public
involvement opportunities afforded by the title V permit program. Our
experience in the RCRA combustion program has shown that many of the
sources that would fall into the area source classification (e.g., some
commercial incinerators and cement kilns burning hazardous waste as
fuel) are the ones in which the public is generally most interested.
Subjecting hazardous waste combustor area sources to title V permitting
will ensure that the public will continue to be involved in permit
decisions under the CAA, as they have been under RCRA. For example, the
public will have an opportunity to comment on and request a public
hearing for a draft title V permit. They have access to State or
Federal court to challenge title V permits, depending upon whether the
permit is a part 70 or part 71 permit. Title V also provides greater
access to information about sources in many cases. Under title V,
States and EPA cannot deny basic information about sources to citizens
unless it is protected as confidential business information.
Conversely, there could be disparity in what information citizens might
be able to obtain under State non-title V operating permits.
Consistency is a key objective as well. Part 70 sets out the
minimum criteria that a State program must meet. If a State fails to
develop and implement a program that meets these minimum criteria, then
a part 71 federal operating permits program is put into place. These
minimum criteria provide for consistency across State and Federal title
V permitting programs, which might not occur under other State air
permitting programs. Consistency within CAA programs is not the only
concern. We also are, as part of our approach to integrating regulation
of these sources under RCRA and the CAA, striving to maintain
consistency with how sources have been regulated under RCRA. Under
RCRA, all of the sources that would fall into an area source
classification are currently treated the same as the sources that are
classified as major under the CAA. It is appropriate to continue
treating all hazardous waste combustor sources in the same manner
(i.e., to apply the same permitting requirements to all of these
sources) under the CAA.
Commenters that do not support applying title V requirements to
area sources generally base their position on three arguments. First,
they argue that Congress had consciously differentiated between area
and major sources when developing the CAA, so that there would be a
strong incentive for facilities to limit emissions and thus avoid the
additional requirements imposed on major sources. These commenters
maintain that subjecting area sources to title V requirements would
create a disincentive for these sources to minimize emissions.
Secondly, they suggest that other CAA permitting mechanisms, such as
federally enforceable state operating permits, might be more
appropriate for the hazardous waste combustor area sources. One
commenter notes that some sources have already invested a lot of time
and effort working with permitting authorities to develop federally
enforceable state operating permits that limit their potential to emit
below major source levels, and that the Agency's action subjecting
these sources to title V permits would render this work meaningless.
Finally, they assert that this would be the first time the Agency did
not provide the option to the States to either defer title V permitting
for area sources or exempt them entirely, and they express concern
about the precedent that would be set if the Agency were to start
requiring area sources to obtain title V permits in this rule.
After careful consideration, we are not persuaded by these counter-
arguments. Although the CAA does differentiate in some provisions
between area and major sources, it did not specify that area sources
should be exempt from the title V permitting program. On the contrary,
it provides discretionary authority in section 502(a) for the
Administrator to decide whether to exempt a source category, in whole
or in part, from title V permitting requirements. Furthermore, the
implementing regulations in 40 CFR 70.3(b)(2), 71.3(b)(2), and
63.1(c)(2) specify that the Administrator will determine whether to
exempt any or all area sources from the requirement to obtain a title V
permit at the time new MACT standards are promulgated. Clearly, the
decision to subject area sources to title V permitting is intended to
be made in the context of both the source category and the applicable
standards. The exemption from title V may only be provided if
compliance with the requirements would be ``impracticable, infeasible,
or unnecessarily burdensome.'' CAA section 502(a). Given that the
hazardous waste combustors subject to today's rule, including those
that may meet the definition of area sources, have all been subject to
common permitting regulations under RCRA, subjecting these sources to
title V permitting is not impracticable, infeasible, or unnecessarily
burdensome. Furthermore, if we exempt area sources from title V
permitting requirements, we would most likely have continued to apply
RCRA permit requirements for stack emissions to these sources. Thus,
the area sources would have been subject to dual permitting regimes
(e.g., federally enforceable state operating permits under the CAA and
RCRA permits) and the resulting burden associated with duplicative
regulation. This would be contrary to a major goal of today's rule. In
conclusion, we decided that it is appropriate to subject all hazardous
waste combustor sources subject to today's MACT standards to title V
permitting requirements. As noted earlier in this preamble, this is
also consistent with the Congressional scheme under RCRA that mandates
regulation of all hazardous waste combustors for all pollutants of
concern.
Although we provided the option to defer title V permitting for
some area sources subject to other MACT standards, this rule is not the
first time we have not allowed States to defer area sources from title
V requirements. See, e.g., 64 FR 31898, 31925 (June 14, 1999) (NESHAP
for Portland Cement Manufacturing Industry to be codified at

[[Page 52981]]

40 CFR part 63, subpart LLL). Moreover, EPA regulations governing other
categories of solid waste combustors under CAA section 129 do not
differentiate between major and minor sources in imposing title V
permitting requirements. See, e.g., CAA section 129(e); 40 CFR 70.3(a)
and 70.3(b)(1), and 40 CFR 60.32e(i). Given that the decision to apply
title V requirements is made in a specific context, we do not share
commenters' concern about the precedent our approach might set for
other situations. We will continue to evaluate each situation on its
own merit. Finally, we do not agree with commenters that this approach
will provide a disincentive to limit emissions because sources will
still be ``capped'' by the emissions limits being promulgated in
today's rule. Neither would progress already achieved in developing
federally enforceable state operating permits be rendered meaningless,
as suggested by some commenters. We anticipate that a source will
likely be able to use the information gathered during the process of
developing a federally enforceable state operating permit (e.g.,
information about its emissions and applicable requirements) in
completing a title V application. Commenters appear to think that
sources will have to start totally anew and without an ability to use
past experience and results. This is neither a realistic nor practical
view of how sources are likely to act.
Commenters opposed to subjecting hazardous waste combustor area
sources to title V had also noted that these sources would be receiving
RCRA permits for the air emissions as well. This argument would have
merit if we choose to promulgate the new standards in both CAA and RCRA
regulations. Since we are promulgating the MACT standards only in the
CAA regulations, however, requirements on air emissions from hazardous
waste combustor area sources would not be included in RCRA
permits.²⁹⁸ Commenters also discount our position in the
NODA about difficulties that would arise if an area source were to move
from one permitting program to another as they make modifications to
their emissions levels that could change their major/area source
determination. They point to our ``once in, always in'' approach to
MACT standards that is stringently applied. Under this approach, once a
MACT standard goes into effect, a major source will always be regulated
under that standard, even if it later decreases its emissions to below
major source levels. This ensures that sources cannot routinely
``flip'' between being regulated or unregulated, which in turn means
that sources would not be moving in and out of the title V permitting
universe. The commenter was correct in raising this to our attention.
We are not relying on this argument to support our decision to subject
hazardous waste combustor area sources to the standards or to title V.
---------------------------------------------------------------------------

\298\ The exception would be, as discussed earlier, cases where
States, at their own choosing, have incorporated the HWC MACT
standards into their State RCRA programs.
---------------------------------------------------------------------------

D. How will Sources Transfer from RCRA to MACT Compliance and Title V
Permitting?
1. In General, How Will this Work?
As discussed in Section A (Placement of Standards and Approach to
Permitting), we are deferring RCRA controls on hazardous waste
combustor air emissions to the part 63 hazardous waste combustor MACT
standards, which are ultimately incorporated into title V permits
issued under the CAA. Promulgation of the new hazardous waste combustor
MACT standards under the CAA does not, however, by itself implement
this deferral or eliminate the need to continue complying with
applicable RCRA requirements--either those in a source's RCRA permit or
in RCRA interim status performance standards. These requirements
include obligations for RCRA permitting (for example, interim status
facilities will continue to be subject to RCRA permitting requirements,
including trial burn planning and testing).
Therefore, today's rule adopts specific provisions that address the
transition from RCRA permitting to the CAA regulatory scheme. As
discussed in Section B.3 (Applicability of RCRA permitting
requirements), the requirements in Secs. 270.19, 270.22, 270.62, and
270.66 do not apply once a source demonstrates compliance with the
standards in part 63 subpart EEE by conducting a comprehensive
performance test and submitting an NOC to the regulatory
agency.²⁹⁹ In this section, we discuss how regulators can
implement the deferral from RCRA to hazardous waste combustor MACT
compliance and title V permitting.
---------------------------------------------------------------------------

\299\ If, however, there is a need to collect information under
Sec. 270.10(k) then the permitting authority may require, on a case-
by-case basis, that facilities use the provisions found in these
sections.
---------------------------------------------------------------------------

a. What Requirements Apply Prior to Compliance Date? You have three
years following promulgation of the MACT standards to achieve
compliance with the emissions standards. However, the rule is effective
shortly after promulgation. During the approximately three years
between the effective date and the compliance date, you will be subject
to applicable requirements for hazardous waste combustor MACT
compliance and title V permitting. For example, there are compliance-
related requirements in 40 CFR part 63 subpart EEE that are separate
from the actual standards for emissions levels, such as those in
Secs. 63.1210(b) and 63.1211(b) for submitting a Notice of Intent to
Comply and a progress report, respectively. Requirements in 40 CFR
parts 70 and 71 for operating permit programs developed under title V
will also apply. These include requirements governing timing for
submitting initial applications, reopenings to include the standards,
and revisions to incorporate applicable requirements into title V
permits. The interface between an NOC and the title V permit has
already been discussed. Consequently, our discussion on implementing
the deferral of RCRA controls focuses on the transition away from RCRA
permits and permit processing once a facility demonstrates compliance
with the standards through a comprehensive performance test and submits
a NOC to the regulatory agency.
Many of the activities undertaken during the three year compliance
period play a role in implementing the transition of RCRA controls to
MACT compliance and title V. For example, some of you may have to make
changes to their design or operations to come into compliance with the
new standards. If you have a RCRA permit, you may need to modify the
RCRA permit to reflect any of these changes before they are actually
made. This may be necessary to remain in compliance with the RCRA
permit while setting the stage for demonstrating compliance with CAA
MACT requirements. We urge you (the source) to seek guidance from your
RCRA permitting authorities as early as possible in this process. As
part of our ``fast track rule'' (see 63 FR 33781, June 19, 1998), we
promulgated a streamlined process in 40 CFR 270.42(j) for modifying the
RCRA permit, so that you can make these necessary changes and begin
operating in accordance with the new limits before the compliance date
arrives. To take advantage of the streamlined process, however, you
must first comply with the Notice of Intent to Comply requirements in
Sec. 63.1210. The Notice of Intent to Comply requirements obligate you
to advertise and conduct an informal meeting with the neighboring
community to discuss plans to comply with the new standards, and to
subsequently provide information about

[[Page 52982]]

these plans to the regulatory agency.³⁰⁰ We anticipate
discussion at this meeting will include modifications to the RCRA
permit that must be processed before you can start upgrading equipment
to meet the emissions limits set by MACT. The goal of these activities
is to ensure that by the end of the three-year compliance period, you
will be in compliance with both the MACT standards and their RCRA
permits or interim status requirements.
---------------------------------------------------------------------------

\300\ The requirements for providing notice of and conducting
the public meeting as part of the Notice of Intent to Comply
provisions are based on the RCRA preapplication meeting requirements
in 40 CFR 124.31.
---------------------------------------------------------------------------

b. What Requirements Apply After Compliance Date? After the
compliance date, a transition period exists during which there will be,
in effect, two sets of standards concerning emissions from hazardous
waste combustors: (1) The MACT standards in 40 CFR part 63; and (2) the
performance standards that are still in the RCRA permit or in the 40
CFR part 265 interim status regulations. During this period, in cases
where operating parameters and limits are addressed by both programs
(MACT and RCRA), you must comply with all applicable parameters and
limits; those which are more stringent will govern. We anticipate that
the MACT standards will be compatible with the RCRA performance
standards, although in some cases the DOC is likely to set narrower or
different operating conditions. Thus, in complying with the MACT
standards, you also will comply with corresponding conditions in the
RCRA permit or in the RCRA interim status regulations. However, at some
sites, certain RCRA permit conditions may be more stringent than the
corresponding MACT standards or may establish independent operating
requirements. Some potential reasons why such a situation would occur
are discussed in the May 2, 1997 Notice of Data Availability (62 FR
21249, 5/2/97). In these situations, you must comply with the more
stringent or more extensive conditions in the RCRA permit.
We also note that there may be situations where it is not clear
whether a RCRA compliance requirement is less stringent than a MACT
requirement. This can occur, for example, when the two compliance
requirements have different averaging periods and different numerical
limits. In this situation, we recommend that the source coordinate with
permitting officials early in the MACT process, perhaps when the source
submits RCRA permit modification pursuant to the fast-track rulemaking,
in order to determine which requirement is more stringent. We believe
the permitting officials should give sources an appropriate level of
flexibility when making this determination.
Our approach of placing the MACT air emission standards for
hazardous waste combustors in 40 CFR part 63 subpart EEE and not
including them, even by reference, in the RCRA regulations means that
the air emissions must ultimately be incorporated into title V permits
issued under the CAA. To completely implement the deferral of RCRA
controls, conditions governing air emissions and related operating
parameters should also be ultimately removed from RCRA permits. (For
the special case of risk-based conditions derived from RCRA omnibus
authority, see earlier discussions.) Similarly, hazardous waste
combustors that are in the process of obtaining RCRA permits will
likely need to have the combustor air emissions and related parameters
transitioned to MACT compliance and title V permits at some point.
We intend to avoid duplication between the CAA and RCRA programs.
We encourage you and regulators to work together to defer permit
conditions governing air emissions and related operating parameters
from RCRA to MACT compliance and title V, and to eliminate any RCRA
provisions that are no longer needed from those permits. As discussed
below, we are adopting a provision in today's final rule to help
permitting authorities accomplish this task in the most streamlined way
possible. The RCRA permits will, of course, retain conditions governing
all other aspects of the hazardous waste combustor unit and the rest of
the facility that continue to be regulated under RCRA (e.g., general
facility standards, corrective action, financial responsibility,
closure, and other hazardous waste management units). Furthermore, if
any risk-based site-specific conditions have been previously included
in the RCRA permit, based either on the BIF metals and/or hydrochloric
acid/chlorine requirements ³⁰¹ or the omnibus authority, the
regulatory authority will need to evaluate those conditions vis-a-vis
the MACT standards and the operating parameters identified in the NOC.
If the MACT-based counterparts do not adequately address the risk in
question, those conditions would need to be retained in the RCRA permit
or included within an appropriate air mechanism. In those limited
cases, sources and permitting agencies may instead agree to identify
the RCRA limit in the title V permit. Since one goal of the title V
program is to clarify a source's compliance obligations, it will be
beneficial, and convenient, to acknowledge the existence of more
stringent limits or operating conditions derived from RCRA authority
for the source in the title V permit, even though the requirements
would not reflect CAA requirements. We strongly encourage Regional,
State, and local permitting authorities to take advantage of this
beneficial option.
---------------------------------------------------------------------------

\301\ The BIF limits for metals under RCRA are based on
different level of site-specific testing and risk analysis (Tier I
through Tier III). It is possible that, if it were based on the more
stringent analysis, a RCRA BIF limit could be more stringent than
the corresponding MACT standard.
---------------------------------------------------------------------------

2. How Will I Make the Transition to CAA Permits?
In the May 1997 NODA, we expressed our intent to rely on the title
V permitting program for implementation of the new standards, and asked
for comments on how and when the transition from RCRA should occur (see
62 FR 24250, May 2, 1997). We are amending the regulations in 40 CFR
part 270 to specify the point at which the RCRA regulatory requirements
for permitting would cease to apply. However, once you have a permit,
you must comply with the conditions in that permit until they are
either removed or they expire. Many commenters expressed an interest in
what happens to conditions in a RCRA permit once the new standards are
published. We received a variety of suggestions, but a common thread
was a request for EPA to lay out a clear path through the permit
transition process. While we recognize the desirability of having a
uniformly defined route for getting from one permit to another, it is
important to provide flexibility to allow a plan that makes the most
sense for the situation at hand. There is not a ``one size fits all''
approach that would be appropriate in all cases. Thus, we are not
prescribing a transition process via regulation, but providing guidance
in the following discussion which we hope will assist regulatory
agencies in determining a route that makes the most sense in a given
situation. Given the level of interest expressed, we will, in the
ensuing discussion, map out a process for implementing the deferral of
air emissions controls from RCRA to MACT compliance and title V
permitting. We address key considerations that should factor into the
decision of how and when to implement the deferral of permit
conditions.³⁰²
---------------------------------------------------------------------------

\302\ Although we are not mandating an approach to transition by
regulation, we are, as discussed in Section 2. How Should RCRA
Permit Be Modified? below, providing a tool in the RCRA permit
modification table in 40 CFR 270.42, Appendix I, that may be used to
assist regulators and sources in effecting the transition.

---------------------------------------------------------------------------

[[Page 52983]]

In identifying key aspects of the transition, we seek the optimal
balance of three basic considerations raised by commenters and other
stakeholders. The considerations are to: (1) Address public perception
issues associated with taking conditions out of a RCRA permit; (2)
minimize the amount of time a source might be potentially subject to
overlapping requirements of RCRA and the CAA (and thus subject to
enforcement under both RCRA and the CAA for the same violation); and
(3) provide flexibility to do what makes the most sense in a given
situation. The first two considerations are primarily factors of time--
when should conditions be removed from the RCRA permit? The third
consideration is more a factor of how--what mechanism should be used
for removing RCRA conditions?
Why do these particular considerations carry such importance? As
for the first, one of the points emphasized in our National Hazardous
Waste Minimization and Combustion Strategy is the importance of
bringing hazardous waste combustors under permits as quickly as
possible. The Strategy has been driving EPA Regions and authorized
States to place their top permitting priority on the hazardous waste
combustor universe. Consequently, the Strategy may have created a
certain perception on behalf of the public about the importance of the
actual permit document. The actual issue we are trying to address here
is more of a concern about a potential break in regulatory coverage of
a source as it transitions from RCRA permitting requirements to the CAA
regulatory scheme.
While it might appear that we are altering the policy expressed in
the Strategy if we allow removal of conditions from a RCRA permit
before the title V permit is in place, it is not the actual permit
document that is of paramount importance. Rather, our focus is and has
been on maintaining a complete and enforceable set of operating
conditions and standards. One of the underlying tenets of the position
taken on permitting in the Combustion Strategy was a commitment to
bring hazardous waste combustors under enforceable controls that
demonstrate compliance with performance standards. Under RCRA, the
permit was the available vehicle to achieve better enforcement of
tighter conditions than exist in interim status.
We remain committed to this underlying tenet. However, the
mechanism for achieving this objective under the CAA is not necessarily
the title V permit. In RCRA, the permitting process provides the
vehicle for the regulatory agency to approve testing protocols
(including estimated operating parameters), to ensure completion of the
testing, and to develop final operating parameters proven to achieve
performance standards. The final RCRA permit is the culmination of
these activities. Under MACT, these activities do not culminate in a
permit, but in a NOC. The development of the NOC is separate from the
development of the title V permit. The title V permitting process is
primarily a vehicle for consolidating in one document all of the
requirements applicable to the source. Conversely, it is the NOC that
contains enforceable operating conditions demonstrated through the
comprehensive performance test to achieve compliance with the hazardous
waste combustor MACT standards (which are generally more stringent than
the RCRA combustion performance standards). Thus, the NOC captures the
intent of the Strategy with regard to ensuring enforceable controls
demonstrated to achieve compliance with relevant standards are in
place.
Another basis for our position on permitting in the Combustion
Strategy is the level of oversight by the regulatory agency during the
permitting process, which is typically greater than that which occurs
during interim status. For example, although BIFs operating under
interim status are required to conduct compliance testing and
subsequently operate under conditions they identify in a certification
of compliance, there are no requirements for the regulatory agency to
review and approve compliance test plans or results. On the other hand,
oversight by the regulatory agency is more intensive during the
permitting process, e.g., through the trial burn planning (including
regulatory approval of the trial burn plan), testing, and development
of permit conditions. Although the process required for interim status
BIFs under RCRA may, at first, seem analogous to the CAA MACT process,
i.e., sources being required to conduct comprehensive performance tests
and subsequently operate under conditions in an NOC, there is a
significant difference. The difference is the level of oversight that
occurs in the MACT process. According to the MACT requirements in 40
CFR 63.1207(e) and 63.1206(b)(3), the regulatory agency must review and
approve the performance test protocol and must make a finding of
compliance based on the test results that are reported in the NOC. The
NOC consequently represents a level of agency oversight that is
actually more analogous to the RCRA permit process than to interim
status procedures.
An additional reason for the importance, under the Combustion
Strategy, of bringing hazardous waste combustors under permits was to
allow for the imposition of additional permit conditions where
necessary to protect human health and the environment. In general,
these conditions are established based on the results of a site-
specific risk assessment and imposed under the RCRA omnibus authority.
This objective will continue to be met even though we are deferring
regulation of hazardous waste combustor air emissions, in general, to
the CAA. Coming into compliance with the more stringent and more
encompassing MACT standards will accomplish part of the Combustion
Strategy's goal of improved protection. For any cases where the
protection afforded by the MACT standards is not sufficient, the RCRA
omnibus authority and RCRA permitting process will continue to be used
to impose additional conditions in the RCRA permit (or, as discussed
earlier, in a title V permit).
With regard to the remaining considerations, we seek here to reduce
duplicative requirements across environmental media programs (i.e., air
emissions under the CAA and RCRA). This objective to reduce duplication
is behind our goal of minimizing the amount of time a source might be
potentially subject to dual permitting and enforcement scenarios. In
order to allow for common sense in implementing environmental
regulations, we need to provide flexibility here to do what makes sense
in a given situation. We have provided this flexibility in today's rule
by not prescribing only one process for transitioning from RCRA to the
CAA.
3. When Should RCRA Permits Be Modified?
We identified two options in the May, 1997, NODA for when
conditions should be ultimately removed from RCRA permits (see 62 FR
24250). Our preferred option at the time is to wait until the source
had completed its comprehensive performance test and the standards had
been included in its title V permit. The alternative option we
identified would be to modify the RCRA permit once the facility submits
the results of its comprehensive performance test.

[[Page 52984]]

Of the comments that spoke to the timing issue, some advocate
waiting for the title V permit, but most opposed this position. The
majority of commenters favor effecting the transition either on the
compliance date, since we had said in the NODA that the pre-NOC would
be due to the regulatory agency on that date ³⁰³ and would
contain enforceable conditions, or upon submittal of the NOC, since it
contains enforceable operating conditions demonstrated to achieve
compliance with the standards. All three of these approaches are
identified in the time line shown in Figure 1. Readers will note that
the time line shows two potential points for the title V permit to be
issued (options 1A and 1B). Option 1A is based on the statutory time
frames for issuing title V permits. Under this option, the title V
permit may be issued prior to the compliance date for the new
standards, but it might only include the standards themselves and a
schedule of compliance. Under option 1B, the operating requirements in
the NOC that actually have been demonstrated to achieve compliance
would be included in the permit.
---------------------------------------------------------------------------

\303\ We are adopting a DOC (previously the pre-NOC) requirement
in today's final rule, but it is amended from how we presented it in
the NODA (as discussed in Part Five, Section IV). Rather than
submitting the DOC to the regulatory agency, a source must maintain
it in their operating record. We encourage source owners and
operators to set up the operating record in an unrestricted location
that is reasonably accessible by the public.
---------------------------------------------------------------------------

We evaluated each of the options in terms of the two timing-related
considerations listed above: addressing the perception issue that stems
from removing conditions from the RCRA permit (which, as discussed
above, is really a concern about a break in regulatory coverage--i.e.,
that there might be a period of time when the source would not have
enforceable controls demonstrated to achieve compliance with stack
emissions standards), and minimizing the amount of time sources would
potentially be subject to the same requirement(s) under both RCRA and
CAA. These considerations may not always be compatible. For example,
one way to address the perception of creating a break in regulatory
coverage would be to continue to place emphasis on the permit, rather
than on the tenet behind the permit (of having enforceable controls
that demonstrate compliance with performance standards). This would
mean waiting to remove conditions from a RCRA permit until a source has
demonstrated compliance with the MACT standards and incorporated the
appropriate combustion operating requirements in its NOC into the title
V permit (i.e., option 1B). However, this approach would maximize the
amount of time the source potentially would be subject to overlapping
requirements under RCRA and the CAA. On the other hand, one way to
address the overlapping requirements consideration would be to allow
removal of conditions from the RCRA permit at the time the standards
are promulgated. But, this would create a time period during which the
source would not have enforceable controls proven to achieve
compliance, which would not address the concern about avoiding a break
in regulatory coverage. Clearly neither of these extremes can provide a
good balance between the two timing-related considerations.

BILLING CODE 6560-50-P

[[Page 52985]]

[GRAPHIC] [TIFF OMITTED] TR30SE99.006

[[Page 52986]]

[GRAPHIC] [TIFF OMITTED] TR30SE99.007

BILLING CODE 6560-50-C

[[Page 52987]]

We evaluated each option to determine which most effectively
balances the relevant issues. Options 1A and 1B focus primarily on
tying the transition timing to title V permitting. Option 2 links the
timing for transition to the DOC (previously called the pre-NOC).
Option 3, which we are recommending be followed, ties transition to
submittal of the NOC.
a. Option 1A. This option is a variation of an option discussed in
the May, 1997, NODA. There we stated, ``The Agency's current thinking
is that the RCRA permit should continue to apply until a facility
completes its comprehensive performance testing and its title V permit
is issued (or its existing title V permit is modified) to include the
MACT standards. The RCRA permit would then be modified to remove the
air emissions limitations which are covered in the title V permit.''
(see 62 FR 24250). Although this description basically applies to
option 1B, the discussion in the NODA might also have been interpreted
to mean that once the standards are in a title V permit, the
corresponding emissions limits should be removed from the RCRA permit.
When reviewing the implementation time line in terms of the statutory
and regulatory time frames governing the title V process, we found that
sources might well have title V permits issued or modified to include
the new standards a year before they ever conduct performance testing.
Although the permit would likely include the standards and a schedule
for complying with the new limits, it would not include any of the key
combustion operating requirements demonstrated in the performance test.
Thus, even though option 1A would seem to address the concern about a
break in coverage because the title V permit would have been issued, in
actuality, the underlying tenet of the Combustion Strategy--that the
source have enforceable operating parameters proven to achieve the new
standards--is not fully addressed.
b. Option 1B. This option calls for the NOC to be incorporated into
title V permits before any conditions could be removed from RCRA
permits. As discussed earlier, this approach would not be consistent
with our goal of minimizing duplication across permitting programs,
even though it was identified as our current thinking in the NODA. As
discussed in the NOC/title V Interface Section, the initial NOC must be
incorporated into the title V permit as a significant permit
modification, which could add another nine months to the transition
period. Moreover, commenters express concern over impacts that existing
delays in title V permitting activities might have. Commenters wrote
that given the tremendous volume of permits to be issued (hazardous
waste combustors being just one small subset) there would be no way to
predict how long it might take regulatory agencies to initially issue
or modify title V permits to include the standards, or to modify
permits to include NOCs, despite time frames set forth in the title V
regulations. We agree that delaying removal of air emissions and
related parameters from RCRA permits until this occurs would
unnecessarily extend the amount of time sources might be subject to
overlapping requirements. As pointed out by commenters, having
overlapping requirements may present technical and administrative
difficulties. Examples of technical difficulties include, but are not
limited to, the potential for conflicting requirements with regard to
testing, monitoring, and compliance certifications. Examples of
administrative difficulties include, but are not limited to, permit
maintenance issues stemming from different permit modification
procedures and appeals procedures.
c. Option 2. Option 2 reflects the time frame suggested by some
commenters for effecting the transition upon submittal of the DOC,
which, under the NODA discussion, would have been due to the regulatory
agency on the compliance date (note: commenters appear to use the terms
``compliance date'' and ``effective date'' interchangeably, but they
are quite different). Basing transition on the DOC was still a viable
option to consider, even with our amended approach of having the source
maintain the DOC in its operating record. The DOC contains enforceable
operating conditions for key combustion parameters that the source
anticipates will achieve compliance with the new standards. Although
the source would have had to comply with other enforceable part 63
requirements by this point (e.g., requirements for the Notice of Intent
to Comply, the progress report, and the performance test plan), this
would be the first point where a source might have overlapping
requirements governing air emissions and related operating parameters--
those in the DOC and those in the RCRA permit. Recommending removal of
RCRA permit conditions at this point would thus minimize the potential
for duplicative requirements. However, we conclude that it would still
not address the perception issue adequately. Specifically, even though
the source is subject to enforceable operating requirements, the source
has not actually demonstrated compliance with the new standards.
d. Option 3. This option reflects the alternative approach we
suggested in the May, 1997, NODA, as well as the preferred option of
the majority of those who submitted comments on the timing issue. Under
this recommended option, a source might well have a title V permit that
addresses the new standards to some extent, even if just by including
the standards themselves and a schedule for compliance. More
importantly, the source will have conducted its comprehensive
performance test, and submitted an NOC containing key operating
parameters demonstrated to actually achieve compliance (and which are
enforceable). Although there would be some time during which a source
might have overlapping requirements (those in its NOC and those in its
RCRA permit), this would be a finite and predictable amount of time.
After considering all the comments, we conclude that option 3 best
meets the dual challenges of ensuring the source is continuously
subject to enforceable controls demonstrated to achieve compliance
while minimizing the time you would be subject to permitting
requirements for, and enforcement of, operating parameters and limits
under both RCRA and the CAA. Therefore, today's rule adopts option 3.
We acknowledge that this approach does not completely eliminate
concerns expressed by some commenters about the potential for
facilities to be subject to dual enforcement mechanisms. Although this
potential may exist during the brief transition period when a source
has enforceable conditions under both CAA and RCRA, we will exercise
enforcement discretion to avoid any duplicative inspections or actions,
and we encourage States to do so as well. If any inspections are
scheduled to occur during the brief transition period (which may be
unlikely given how short this period is), the regulatory agency could
conduct joint inspections by RCRA and CAA enforcement staff. Joint
inspections might help to alleviate some of the potential for any
duplicative efforts, either in terms of individual inspections
targeting the same areas, or enforcement actions being taken under both
RCRA and CAA authorities.
Under Option 3, you would most likely have a title V permit that
addresses the hazardous waste combustor MACT standards to some extent.
We expect that if the permit were issued prior to the comprehensive
performance test and the submittal of the NOC, it would contain the
standards

[[Page 52988]]

themselves, and related requirements in part 63 subpart EEE, such as
the requirements to develop and public notice performance test
protocols, to develop and maintain in its operating record the DOC with
anticipated (and enforceable) operating limits, to conduct the
comprehensive performance test and periodic confirmatory tests, and to
submit the NOC, including the test results, to the regulatory agency.
The public would have had an opportunity to comment on the
requirements in the title V permit as part of the normal CAA
administrative process for issuing permits. Furthermore, the public
would have had other opportunities to be involved in your compliance
planning. For example, under the requirements for the Notice of Intent
to Comply in Sec. 63.1210(b), you would have had to conduct an informal
meeting with the community to discuss how you intend to come into
compliance with the new standards. You also are required in
Sec. 63.1207(e) to provide public notice of the performance test plan,
so the public would have the opportunity to review the detailed testing
protocol that describes how the operating parameters will achieve
compliance.
4. How Should RCRA Permits Be Modified?
Once you have been issued a RCRA permit, you must comply with the
conditions of that permit. Unless the conditions have been written into
the permit with sunset (i.e., automatic expiration) clauses governing
their applicability, conditions remain in effect until the permit is
either modified to remove them or the permit is terminated or expires.
Promulgation of final MACT standards for hazardous waste combustors
does not in itself eliminate your obligation to comply with your RCRA
permit. In the May 1997 NODA, we stated that the RCRA permit would be
modified to remove air emission limitations that are covered under
MACT, but did not elaborate on what modification procedures would be
followed. We solicited comments on how the transition should occur.
Of the commenters that addressed this issue, the recurring theme in
the comments is for EPA to provide a mechanism that would impose
minimal burden on sources and permit writers to process the
modifications. Some express a desire to see the RCRA conditions removed
in some automatic fashion once the MACT standards became effective. A
mechanism for accomplishing this, suggests one commenter, would be to
include a requirement in the final rule that would effect removal of
conditions from all RCRA permits. One commenter suggests adding a new
line item to Appendix I in Sec. 270.42, designated as class 1, to
address the transition to MACT. Another suggests a new line item
designated as class 1 requiring prior agency approval. A third suggests
a new line item designated as class 2.
We do not agree with eliminating conditions from all RCRA permits
as part of a national rulemaking effort (i.e., we do not agree with an
``automatic'' removal), particularly given the existence of authorized
sate programs and state-issued permits. Permits may contain site-
specific conditions developed to address particular situations, e.g.,
conditions based on the results of a site-specific risk assessment. To
ensure that the regulatory agency continues to meet its RCRA obligation
to ensure protection of human health and the environment, these
conditions may need to be evaluated on a case-by-case basis vis-a-vis
the MACT standards before they are removed. If the RCRA risk-based
conditions are more stringent or more extensive than the corresponding
MACT requirements, the conditions must remain in the RCRA permit.
We do agree with commenters that there should be a streamlined
approach to removing conditions from a RCRA permit that are covered by
the hazardous waste combustor MACT regulations at the time an NOC
demonstrating compliance is submitted to the regulatory agency. All
other conditions would, of course, remain in the RCRA permit. Once you
demonstrate compliance with MACT, we consider the transition from RCRA
to be primarily an administrative matter since you will not only be
subject to comparable enforceable requirements under CAA authority, but
also will continue to be subject to any site-specific conditions under
RCRA that are more stringent than MACT. Our intent is not to impose an
additional burden on you or permit writers for a largely administrative
requirement. To this end, we are adding a new line item to the permit
modification table in 40 CFR 270.42, Appendix I, to specifically
address the transition from RCRA to the CAA.
The approach of adding a new line item to the permit modification
table is consistent with the comments we received pursuant to the May
1997 NODA. We agree with the commenter who suggests the new item be
designated as a class 1 modification requiring prior Agency approval.
This classification effectively balances the need to retain some
regulatory oversight of the changes with the goal of minimizing the
amount of time a source will be subject to regulation under both RCRA
and the CAA for essentially the same requirements. A class 1
modification without prior approval, suggests one commenter, would not
be sufficient to accomplish the transition with adequate confidence in
proper regulatory coverage. Even though we consider the deferral to be
an administrative matter, it is important to retain some level of
regulatory oversight prior to effecting the change to provide the
opportunity to address any differences between the two programs. On the
other hand, the administrative exercise of transitioning from RCRA to
the CAA does not warrant the extra measures (and attendant time
commitment) of a class 2 modification procedure.
We are designating the new line item (A.8.) in the Appendix I table
as class 1 requiring prior Agency approval. Thus, the administrative
procedures associated with this mechanism will not be overly
burdensome, yet RCRA permit writers will have an opportunity to confer
with their counterparts in the air program prior to approving the
request to eliminate conditions from the RCRA permit. This allows the
RCRA permit writer to verify that you have completed the comprehensive
performance test and submitted your NOC. In the few situations where
site-specific, risk-based conditions have been incorporated into RCRA
permits, it also provides the RCRA permit writer with the opportunity
to review such conditions vis-a-vis the MACT standards to ensure any
conditions that are more stringent or extensive than those applicable
under MACT are retained in the RCRA permit. The public also would be
informed that the transition from RCRA was being effected because the
modification procedures require a notice to the facility mailing list.
We recommend that the public notice for the RCRA permit modification
also briefly mention that you have completed performance testing under
the CAA, and are operating under enforceable conditions that are at
least as stringent as those being removed from your RCRA permit.
One commenter offered suggestions for preparing the RCRA
modification requests. We found some of these suggestions helpful and
recommend that, to facilitate processing of the RCRA modification
requests, you (1) identify in your modification requests which RCRA
conditions should be removed, and (2) attach your NOC to the requests.
From another perspective, today's approach for removing conditions
from the RCRA permit also may encourage

[[Page 52989]]

you to work closely with the air program to expeditiously resolve any
potential or actual disagreements on the results of the comprehensive
performance test and conditions in the NOC. The RCRA permit writer is
not likely to approve the modification request until he or she has
received confirmation that their air program counterpart is satisfied
with your compliance demonstration under MACT (i.e., that they have
made the finding of compliance based on the test results documented in
the NOC, as discussed in the following paragraph). Thus, you should
continue to be subject to requirements under both RCRA and the CAA
until the differences, if any, are resolved.
We are not including a requirement in either part 63 subpart EEE or
part 270 specifically for the regulatory agency to approve the NOC
before approving the RCRA modification request. We have incorporated
the general provision for making a finding of compliance (see
Sec. 63.6(f)(3)) into the requirements of subpart EEE at
Sec. 63.1206(b)(3). According to these provisions, the regulatory
agency has an obligation to make a finding of compliance with
applicable emissions standards upon obtaining all of the compliance
information, including the written reports of performance test results.
Because of this obligation, air program staff currently review stack
test results that are submitted in NOCs subsequent to performance
testing, and routinely transmit an official letter to you indicating
the acceptability of the test results. Furthermore, if you fail the
comprehensive performance test, there are requirements in part 63
subpart EEE specifying what you must then do. Given this combination of
regulatory obligations and current practices, we see no need to impose
additional requirements governing review of performance test results.
This approach is also consistent with the timing for when permit
requirements are deferred to CAA (see the amended rule language for 40
CFR 270.19, 270.22, 270.62, and 270.66)).
5. How Should Sources in the Process of Obtaining RCRA Permits Be
Switched Over to Title V?
In the initial NPRM and the May, 1997, NODA, we did not
specifically describe, or solicit comment on, permit process issues for
facilities operating under RCRA interim status, or facilities seeking
to renew their RCRA permits (which can occur even after the nominal
permit term has expired). In the above sections, we focused on
implementing the deferral of RCRA controls by determining how and when
to move conditions out of existing RCRA permits. For facilities that do
not yet have RCRA permits, or that need to renew their RCRA permits,
the focus of the discussion shifts to how and when to move nonrisk-
based air emissions considerations out of the RCRA permitting process.
As indicated earlier, RCRA interim status facilities will continue to
be subject to RCRA permitting requirements for air emissions standards
and related operating parameters, including trial burn planning and
testing, until they have demonstrated compliance with the new standards
by conducting a comprehensive performance test and submitting an NOC to
the agency. Facilities in the process of renewing their RCRA permits
will also continue to be subject to RCRA permitting requirements until
the same point.
Again, there is no single approach for moving these two categories
of facilities out of the RCRA permitting process (i.e., for stack air
emissions requirements). The most appropriate route to follow in each
case depends on a host of factors, including, for example: (1) The
status of the facility in the RCRA permitting process at the time this
rule is published; (2) the priorities and schedule of the regulatory
agency; (3) the level of environmental concern at a given site; and (4)
the number of similar facilities in the permitting queue. The
regulatory agency (presumably in coordination with the facility) will
balance all of these factors. In mapping out a site-specific approach,
we are encouraging permitting agencies to give weight to two key
factors. First, we should minimize to the extent practicable the amount
of time a facility would be subject to duplicative requirements between
RCRA and CAA programs. Second, as indicated in Part Five, Section V.B
(Risk Burn/Comprehensive Performance Testing), testing under one
program should not be unnecessarily delayed in order to coordinate with
testing under the other. For example, if a facility is planning to
conduct a RCRA trial burn within a fairly short amount of time after
the rule is promulgated, they generally should not be allowed to delay
the trial burn to coordinate with comprehensive performance testing
under MACT that may not occur for three more years.\304\
---------------------------------------------------------------------------

\304\ There may be a short delay allowed for the purpose of
combining RCRA trial burn and MACT performance test plans. Of
course, even if the timing for the two tests is such that they may
be coordinated, that does not mean that one can simply replace the
other, particularly because test conditions for one may not be
applicable to the other (refer to Section V.B for additional
discussion on this topic).
---------------------------------------------------------------------------

Even though we cannot prescribe a single national approach for the
transition from RCRA permitting for air emissions, we can provide some
other recommendations to help permitting authorities and facility
owners or operators determine a sound approach. In this section, we
walk through some examples, intended as guidance, for transitioning
facilities that are in the process of obtaining or renewing a RCRA
permit. We hope that these examples will also enhance consistency among
the various regulatory agencies.
a. Example 1. Facility has submitted a RCRA permit renewal
application. Some sources, particularly hazardous waste incinerators,
have RCRA permits that are close to expiring. These sources may already
have initiated the renewal process by the time this rule is
promulgated. In these situations, we anticipate the source might need
to modify its current permit to accommodate any upgrades necessary to
comply with the new standards. Facilities may modify RCRA permits that
have been continued under Sec. 270.51 pending final disposition of the
renewal application. Thus, facilities will be able to use the
streamlined permit modification procedures that were promulgated in
Sec. 270.42(j) to effect the necessary changes pending resolution of
their renewal application. Depending on where they are in the renewal
process, the permitting authority may, alternatively, elect to fold the
modifications into the actual renewal process, thereby streamlining
some of the administrative requirements.
Issuance of RCRA hazardous waste combustor permits often takes
several years. If the source and the permitting authority are in the
early stages of renewal, the schedule of permitting activities may not
call for a trial burn to be conducted until sometime close to when the
source would be required to conduct comprehensive performance testing
under MACT. If so, the source may be able to either coordinate the
testing requirements of the two programs, e.g., if a RCRA risk burn is
necessary, or to perform just the comprehensive performance test under
MACT. If, on the other hand, they are further along in the renewal
process, the trial burn might be scheduled for the near future. In this
case, the approach outlined in Example 2 below might be more
appropriate to follow.
Regardless of the approach followed to transition the air emissions
and related operating parameters for the combustion unit to the Air
program, the

[[Page 52990]]

RCRA permit must still be renewed for all other aspects of hazardous
waste management at the facility.
b. Example 2. Permitting authority has approved, or is close to
approving, the RCRA trial burn plan at the time the final MACT
standards are promulgated. Both interim status facilities and those
seeking permit renewal are subject to requirements in Secs. 270.62 and
270.66 to develop and obtain approval for trial burn plans.
Requirements in these sections also call for permitting authorities to
provide public notice of approved (or tentatively approved) trial burn
plans and projected schedules for conducting the burns. We anticipate
that many of the hazardous waste combustors seeking permits who are
subject to this rulemaking will have already had their trial burn plans
approved, or close to being approved, by the time this rule is
promulgated. In such situations, we expect the facility to continue
with the trial burn as planned.
If the burn is successful, we anticipate the permitting authority
will issue a final RCRA permit that covers both the operations of the
hazardous waste combustor unit as well as all other hazardous waste
management activities at the site. We recommend that the permit be
worded flexibly to facilitate transition to title V once the source
subsequently demonstrates compliance with the MACT standards. For
example, conditions in the RCRA permit that would ultimately be covered
under title V might have associated sunset provisions indicating that
the conditions will cease to apply once the combustor unit demonstrates
compliance with the MACT standards. This would ensure that the amount
of time the source might be subject to emissions limits and operating
parameters under both RCRA and the CAA would be minimized. It would
also eliminate the need to engage in a separate permit modification
action to remove the conditions after the MACT compliance
demonstration.
Facilities in this scenario may determine they need to make some
changes to their equipment or operations to meet the new emissions
limits. These facilities will be able to use the streamlined permit
modification procedures that were promulgated in Sec. 270.42(i).
If the trial burn is not successful, we expect permitting
authorities to refer to the RCRA trial burn failure policy (see
Memorandum on Trial Burns, EPA530-F-94-023, July 1994). This policy
includes discussion in the following areas: (1) Taking immediate steps
to restrict operations; (2) initiating procedures for permit denial
(which would be appropriate for interim status or renewal candidates);
(3) initiating proceedings to terminate the permit (which would be
appropriate for proposed new facilities); and (4) authorizing trial
burn retesting after the facility investigates reasons for the failure
and makes changes to address them.
c. Example 3. The permitting authority does not anticipate
approving the trial burn plan, or the trial burn is not scheduled to
occur until after the Notice of Intent to Comply is submitted. As
suggested in the previous example, if a facility is ready to proceed
with a trial burn at the time the final hazardous waste combustor MACT
rule is promulgated, we expect that activities will proceed as planned.
Once the Notice of Intent to Comply is submitted, however, the
regulatory authority will have a better understanding of how and when
the facility intends to comply with the emissions standards, and how
the trial burn would fit in with the MACT compliance demonstration.
Thus, we expect the regulatory authority may wish to decide whether to
separately continue with the trial burn schedule laid out in the RCRA
permitting process or, conversely, coordinate with MACT comprehensive
performance testing, based on a number of considerations, including,
for example: (1) The facility's schedule and planned modifications for
MACT compliance; (2) progress on completing and approving the RCRA
trial burn plan; (3) whether the risk testing that may be necessary
under RCRA is likely to fit in with the MACT performance test schedule;
and (4) whether the facility wants to combine risk testing under RCRA
with the MACT performance test.
Even after a source conducts its comprehensive performance test and
subsequently submits the NOC to the regulatory agency, separate risk
testing might be necessary. For example, if the comprehensive
performance test did not generate sufficient data for a site-specific
risk assessment, a RCRA ``risk burn'' might be required (see discussion
in Part Five, Section V.B.).
E. What Is Meant by Certain Definitions?
When we considered incorporating MACT standards into both RCRA and
CAA regulations, we anticipated some confusion about definitions that
differ between the two programs. In the NPRM, we solicited comments on
our expressed preference not to reconcile these issues on a national
basis. (See 61 FR 17452). Several commenters suggest that EPA reconcile
the issues and clarify definitions. In the final rule, we have made
some changes, as discussed below, to ensure consistency of
interpretation and to minimize uncertainty for facilities seeking to
comply with today's rule. With these changes, we believe that revisions
to the definitions themselves are not necessary.
1. Prior Approval
In the proposed rule, we stated that RCRA and CAA are similar in
that they both require EPA prior approval before construction or
reconstruction of a facility. There were no adverse comments received
regarding this statement. The requirements for obtaining prior approval
are apparently clear under both programs.
We suggested in the proposed rule that readers of part 63 might be
unaware of their obligations under RCRA. Therefore, as proposed, we are
inserting the following note into Sec. 63.1206 Compliance Dates, ``An
owner or operator wishing to commence construction of a hazardous waste
incinerator or hazardous waste-burning equipment for a cement kiln or
lightweight aggregate kiln must first obtain some type of RCRA
authorization, whether it be a RCRA permit, a modification to an
existing RCRA permit, or a change under already existing interim
status. See 40 CFR part 270''. No adverse comments were submitted.
2. 50 Percent Benchmark
As stated in the proposed rule, RCRA and CAA both classify
``reconstruction'' as any modifications of a facility that cost more
than 50 percent of the replacement cost of the facility. However, the
significance of this term is different depending on which statute is
being applied. Two commenters confirmed that the distinction is
critical. Therefore, they concluded that, to avoid confusion, EPA
should defer to the CAA definition of ``reconstruction'' under RCRA
Section 1006(b) because it is the more flexible and appropriate
definition.
The primary concern about the 50 percent benchmark is in relation
to the limit imposed on RCRA interim status facilities for making
modifications. To ensure that this limit would not present a barrier to
making upgrades necessary to comply with MACT, we finalized a revision
to Sec. 270.72(b) to specify that interim status facilities can exceed
the 50 percent limit if necessary to comply with MACT. (See 63 FR
33829, June 19, 1998). Therefore, there is no potential for practical
conflict among the CAA and RCRA regulatory regimes, and no further
amendment or clarification is needed.

[[Page 52991]]

3. Facility Definition
As stated in the NPRM, the definition of ``facility'' differs
between CAA and RCRA. The definition has bearing in determining the
value of the facility with respect to the 50 percent rule on
modifications as discussed above. We proposed that the RCRA definition
should be used for the RCRA application to changes during interim
status, and the CAA definition should be used when determining
applicability of MACT standards to new versus existing sources.
Commenters disagreed with this approach and concluded that EPA should
defer to the CAA definition of facility because it encompasses the
entire operations at a site. We continue to believe that the CAA
definition should apply to CAA requirements and that the RCRA
definition should apply to RCRA requirements, since the definitions are
used for a different purpose under each statute. By clarifying the 50
percent benchmark issue for RCRA interim status facilities as discussed
above, we believe this satisfies commenters' concerns and, thus, it is
not necessary to reconcile the facility definition.
4. No New Eligibility for Interim Status
RCRA bestows interim status on facilities that were in existence on
November 19, 1980, or are in existence on the effective date of
statutory or regulatory changes that render the facility subject to
RCRA permitting requirements. The original RCRA rules for hazardous
waste incinerators and BIFs were finalized in 1980 and 1991,
respectively. Because these rules established the dates on which
incinerators and BIFs were first subject to RCRA permitting
requirements, the effective dates of those rules created the only
opportunity for interim status eligibility. The interim status windows
that occurred in 1980 and 1991 thus are not modified by this rule. The
lone exception is that facilities currently burning only nonhazardous
wastes that become newly listed or identified hazardous waste under
other future rules would still be able, under existing law, to qualify
for interim status (Sec. 270.42(g)).
5. What Constitutes Construction Requiring Approval?
The proposed rule noted that RCRA and CAA both have restrictions
requiring approval prior to construction, but that each statute defines
construction differently. We expressed our intent in the NPRM to retain
the two definitions. In the final rule, we continue to support
retaining the two definitions. Since most facilities currently possess
RCRA and CAA permits, these definitions are already being applied
concurrently with no apparent problems. Consequently, this is the most
practical and least confusing approach for permittees and regulators.

XII. State Authorization

A. What Is the Authority for Today's Rule?
Today's rule is being issued under the joint authority of the Clean
Air Act (CAA), 42 U.S.C. 7401 et seq., and the Resource Conservation
Recovery Act (RCRA), 42 U.S.C. 6924(o), 6924(q) and 6925. The new MACT
air emissions standards are located in 40 CFR part 63. Pursuant to
sections 1006(b) and 3004(a) of RCRA, 42 U.S.C. 6905(b) and 6924(a),
the MACT program will only be carried out under the CAA delegated
program. We strongly encourage States to adopt today's MACT standards
under their CAA statute and to apply for delegation under the CAA if
they do not have section 112 delegation. State implementation of the
MACT portions of this rule through its delegated CAA program will
facilitate coordination between the regulated entity and its State and
reduce duplicative permitting requirements under the CAA and RCRA.
In addition to promulgating the MACT standards, today's rule
modifies the RCRA program in other various respects and States
authorized for the RCRA base program must revise their programs
accordingly. For example, this rule revises the test for determining
whether a facility's waste retains the Bevill exclusion by adding
dioxins/furans to the list of constituents to be analyzed.
B. How Is the Program Delegated Under the Clean Air Act?
States can implement and enforce the new MACT standards through
their delegated 112(l) CAA program and/or by having title V authority.
A State's title V authority is independent of whether it has been
delegated section 112(l) of the CAA.
Section 112(l) of the CAA allows us to approve State rules or
programs to implement and enforce emission standards and other
requirements for air pollutants subject to section 112. Under this
authority, we developed delegation procedures and requirements located
at 40 CFR part 63, subpart EEE, for National Emission Standards for
Hazardous Air Pollutants (NESHAPS) under section 112 of the CAA (see 58
FR 62262, November 26, 1993, as amended, 61 FR 36295, July 10, 1996).
Similar authority for our approval of state operating permit programs
under title V of the CAA is located at 40 CFR part 70 (see 57 FR 32250,
July 21, 1992).
Submission of rules or programs by States under 40 CFR part 63
(section 112) is voluntary. Once a State receives approval from us for
a standard under section 112(l) of the CAA, the State is delegated the
authority to implement and enforce the part 63 standards under the
State's rules and regulations (the approved State standard would be
federally enforceable). States also may apply for a partial 112
program, such that the State is not required to adopt all rules
promulgated in 40 CFR part 63. We will implement the portions of the
112 program not delegated to the State. For example, documents such as
the NOC will be submitted to the Administrator when due, if the State
is not approved for the standards in today's rule.
Under 40 CFR 70.4(a) and section 502(d) of the CAA, States were
required to submit to the Administrator a proposed part 70 (title V)
permitting program by November 15, 1993. If a State did not receive our
approval by November 15, 1995 for its title V program, the title V
program had to be implemented by us in that State. As of today's rule,
all States have approved title V programs.\305\ This means that all
States have the authority to incorporate all MACT standards (changes to
section 112 of the CAA) into the title V permits as permit conditions,
and have the authority to enforce all the terms and conditions of the
title V permits. See 40 CFR 70.4(3)(vii).
---------------------------------------------------------------------------

\305\ Under the CAA, Indian tribes may apply to EPA to be
treated as States and obtain approval of their own Clean Air Act
programs. Section 301(d) of the Clean Air Act, 42 U.S.C. 7601(d);
see also 40 CFR part 49. Tribes may thus become empowered to
implement the section 112 and title V portions of today's rule is
areas where they demonstrate jurisdiction and the capacity to do so.
Currently under RCRA, there is no Tribal authorization for the RCRA
Subtitle C hazardous waste program and thus EPA generally implements
the RCRA portions of today's rule in Indian Country.
EPA has authority to implement the federal operating permits
program 940 CFR part 71) where a State fails to adequately
administer and enforce an approved part 70 program, or where a State
fails to appropriately respond to an EPA objection to a part 70
permit. Additionally, some sources in U.S. Territories, the Outer
Continental Shelf, and Indian Country, are subject, or will soon be
subject, to part 71.
---------------------------------------------------------------------------

The MACT standards are effective upon promulgation of this rule.
Facilities with a remaining permit term of three or more years will be
required to submit title V applications to their permitting authorities
to revise their permits.\306\ States will write the new

[[Page 52992]]

MACT standards into any new, renewed, or revised title V permit and
enforce all terms and conditions in the title V permit. A State's
authority to write and enforce title V permits is independent of its
authority to implement the changes to the MACT standards (changes to
section 112 of the CAA). Therefore, while both we and the State can
enforce the federal MACT standards within a title V permit, until the
State receives approval from us for required changes to section 112 of
the CAA, we will implement the 112 program.
---------------------------------------------------------------------------

\306\ Title V permits are issued for a period not to exceed five
years. See 40 CFR 70.4(b)(3)(iii). You will have three years to come
into compliance with the new MACT standards. If you have fewer than
three years remaining on your title V permit term, our part 70
regulations do not require you to reopen and revise your permit to
incorporate the new MACT standard into the title V permit. See 40
CFR 70.7(f)(1)(i). However, the CAA does allow State programs to
require revisions to your permit to incorporate the new MACT
standard. Therefore, if you have fewer than three years remaining on
your title V permit, you should consult your state permitting
program regulations to determine whether a revision to your permit
is necessary to incorporate the new part 63 MACT standards. If your
are not required to revise your permit to incorporate the new
standard, you must still fully comply with today's standard.
---------------------------------------------------------------------------

C. How Are States Authorized Under RCRA?
Under section 3006(g) of RCRA, enacted as part of the Hazardous and
Solid Waste Amendments (HSWA) of 1984, new requirements imposed by us
as a result of authorities provided by HSWA take effect in authorized
States at the same time as they do in unauthorized States--as long as
the new requirements are more stringent than the requirements a State
is authorized to implement. We implement these new requirements until
the State is authorized for them. After receiving authorization, the
State administers the program in lieu of the Federal government,
although we retain enforcement authority under sections 3008, 3013, and
7003 of RCRA.
Most of the new Federal RCRA requirements in today's final rule are
being promulgated through the HSWA amendments to RCRA. Regulatory
changes based on HSWA authorities are considered promulgated through
HSWA. The following RCRA sections, enacted as part of HSWA, apply to
today's rule: 3004(o) (changes to the MACT standards), 3004(q) (fuel
blending), and 3005 (omnibus). As a part of HSWA, these RCRA provisions
are federally enforceable in an authorized State until the necessary
changes to a State's authorization are approved by us. See RCRA section
3006, 42 U.S.C. 6926. The Agency is adding these requirements to Table
1 in Sec. 271.1(j), which identifies rulemakings that are promulgated
pursuant to HSWA.
In contrast, the change to the permit modification table (Appendix
I to Sec. 270.42) is promulgated through authorities provided to us
prior to HSWA. Therefore, this change does not become effective until
States adopt the revision and become authorized for that revision.
Under RCRA, States that have received authorization to implement
and enforce RCRA regulatory programs are required to review and, if
necessary, to modify their programs when we promulgate changes to the
federal standards that result in the new federal program being more
stringent or broader in scope than the existing federal standards. This
is because under section 3009 of RCRA, States are barred from
implementing requirements that are less stringent than the federal
program. See also 40 CFR 271.21.
In four respects, we consider today's final rule to be more
stringent than current federal RCRA requirements: (1) The added
definitions for dioxins/furans and TEQ (40 CFR 260.10); (2) the
requirement that permits for miscellaneous units must include
appropriate terms and conditions from part 63, subpart EEE standards
(40 CFR 264.601); (3) the establishment of new standards to control
particulate matter (40 CFR 266.105(c)); and (4) the addition of dioxin/
furans as listed potential Products of Incomplete Combustion (PIC) (40
CFR 266.112; Appendix VIII to 40 CFR part 266). Authorized States must
adopt these requirements as part of their State programs and apply to
us for approval of their program revisions. The procedures and
deadlines for State program revisions are set forth in 40 CFR 271.21.
Section 3009 of RCRA allows States to impose standards that are
more stringent or more extensive (i.e., broader) in scope than those in
the Federal program (see also 40 CFR 271.1(i)(1)). Thus, for those
Federal changes that are less stringent, or reduce the scope of the
Federal program, States are not required to modify their programs.
Further, EPA will not implement those provisions promulgated under HSWA
authority that are not more stringent than the previous federal
regulations in States that have been authorized for those previous
federal provisions. EPA will implement these new provisions in States
that are not authorized to implement the previous federal regulations.
In two respects, we consider today's rule to be less stringent than
current federal requirements: (1) The inapplicability of certain
provisions of RCRA once specified part 63, subpart EEE and other
requirements have been met (40 CFR 264.340(b)(1); 265.340(b)(1);
266.100(b)(1), 266.100(d)(1) and (d)(3); 266.100(h); 270.19; 270.22;
270.62; and 270.66); and (2) the provision for RCRA permit
modifications to remove inapplicable RCRA conditions (Appendix I to 40
CFR part 270.42).\307\
---------------------------------------------------------------------------

\307\ States choosing to adopt the other less stringent changes
to RCRA in today's rule also should adopt the change to 40 CFR
270.42. The change to 40 CFR 270.42 provides the RCRA permit
modification procedure to eliminate inapplicable RCRA requirements
once specified part 63, subpart EEE and other requirements have been
met.
---------------------------------------------------------------------------

The rest of the requirements in today's rule, in our view, are
neither more nor less stringent than current regulatory requirements.
They are either reiterations or clarifications of our existing
regulations or policies (40 CFR 264.340(b)(2), 265.340(b)(2),
266.100(b)(2), and 266.101).
Although States must adopt only those requirements that are more
stringent, in the spirit of RCRA section 1006(b), which directs us to
avoid duplicative RCRA and CAA requirements, we strongly urge States to
adopt all aspects of today's final rule (including the clarifying as
well as less stringent sections). The adoption of all portions of
today's final rule by state agencies will ensure clear, consistent
requirements for owners, operators, affected sources, State regulators,
and the public. Pursuant to today's rule, the permitting requirements
will be implemented solely through the CAA title V program. If a RCRA
permitted facility is required to use RCRA risk-based air emissions
standards in addition to the CAA designated technology based standards,
we will exercise our omnibus authority in section 3005 of RCRA to
modify the facility's RCRA permit.\308\ Therefore, we believe that the
standards promulgated today properly implement the goals of sections
3004(o) and (q) of RCRA to ensure the safe and proper management of the
affected combustion units and the goal of section 1006(b) of RCRA to
avoid duplicative and potentially confusing permitting requirements
under two different environmental statutes (RCRA and CAA). For these
reasons, we encourage States to adopt these

[[Page 52993]]

regulations as quickly as their legislative and regulatory processes
will allow.
---------------------------------------------------------------------------

\308\ If a State has a provision in its State air statute or
regulation that is equivalent to the RCRA omnibus authority (RCRA
section 3005(c)), we expect that the State will be able to use its
air authority in pace of its RCRA omnibus authority.
---------------------------------------------------------------------------

Part Six: Miscellaneous Provisions and Issues

I. Does the Waiver of the Particulate Matter Standard or the
Destruction and Removal Efficiency Standard Under the Low Risk Waste
Exemption of the BIF Rule Apply?

Section 266.109 of the current BIF regulation provides a
conditional exemption from the destruction and removal efficiency
standard and the particulate matter standard for low risk wastes. We
proposed to restrict eligibility for the waiver of the particulate
matter standard to BIFs other than cement and lightweight aggregate
kilns because the waiver could supersede the MACT requirements for the
particulate matter standards. We had the same concern for the
destruction and removal efficiency requirements. See 61 FR at 17470.
After reconsidering the issue, we are clarifying that today's MACT
requirements are separately applicable and enforceable and that no
action is needed to ensure that a BIF waiver does not supersede the
MACT requirements. See the discussions in Part Five of today's preamble
regarding integration of the MACT and RCRA standards.

II. What Is the Status of the ``Low Risk Waste'' Exemption?

Section 264.340(b) and (c) exempts certain incinerators from the
RCRA emission standards if the hazardous waste burned contains (or
could reasonably be expected to contain) insignificant concentrations
of Appendix VIII, part 261, hazardous constituents. We proposed that
this ``low risk waste'' provision no longer be applicable incinerators
on the MACT compliance date because a risk-based exemption from
technology-based MACT standards seemed inappropriate. See 61 FR at
17470. After reconsidering the issue, we have determined that no
specific action is necessary because the MACT standards are separately
applicable and enforceable standards. See the discussion in Part Five
of today's preamble regarding integration of the MACT and RCRA
standards.

III. What Concerns Have Been Considered for Shakedown?

In the proposal, we expressed concern that some new units do not
effectively use their allotted 720-hour pre-trial burn shakedown period
or appropriate extensions to correct operational problems. This can
potentially lead to trial burn failures and emission exceedances, which
pose unnecessary risks to human health and the environment. Therefore,
we proposed three shakedown options to enhance regulatory control over
trial burn testing:
(1) Prior to scheduling trial burns, we would require facilities to
provide the Director a minimum showing of operational readiness.
(2) We would require notification of operational readiness prior
to, and following, the shakedown period.
(3) We would provide guidance on how to effectively prepare for a
trial burn. These options were proposed for inclusion under both the
CAA and RCRA regulations, and comments were requested regarding their
usefulness.
A few commenters preferred Option 3 because it would be useful in
determining how to effectively prepare for a trial burn. Regarding
Options 1 and 2, two commenters felt the cost, time, and resources
required for a trial burn already provide adequate financial incentive
to prepare, plan, and conduct trial burns efficiently. Two commenters
felt that Option 3 provided the potential for inequities in
implementation of the guidance by the permit writer. In general, most
commenters agreed that additional regulatory requirements are not
necessary.
In light of the comments, we decided not to adopt any of the
proposed options. We acknowledge that it is in the facility's best
interest to conduct a successful trial burn that most facilities will
properly utilize their shakedown period. However, during the transition
period from RCRA to MACT compliance, we strongly encourage facilities
to properly use their shakedown period to correct operational problems
that pose unnecessary risks to human health and the environment.
Therefore, with the exception of risk burns, we are pursuing the
deferral of RCRA trial burns to the MACT performance test requirements.
A source remains subject to RCRA trial burns during the transition
period to MACT compliance. For facilities where unique considerations
make a SSRA necessary, risk-based permit conditions may result. In such
cases, there likely would need to be conditions for all phases of
operation in the RCRA permit. Thus, start-up and shakedown would still
be an issue for some RCRA combustor facilities given that they would
have to be in compliance with the unique RCRA emission standards even
during startup and shakedown (unless the permit conditions specify
otherwise).

IV. What Are the Management Requirements Prior to Burning?

Today, we are finalizing the proposal to revise 40 CFR 266.101
(``Management prior to burning'') to clarify that fuel blending
activities are regulated under RCRA. See 61 FR at 17474 (April 19,
1996). As described in detail in the proposal, this is already implicit
(and for some units, explicit) in existing rules. Therefore, today's
rule is more an interpretive clarification. See 52 FR 11820 (April 13,
1987). By incorporating the term ``treatment'' into the regulation, we
are clarifying that fuel blending activities that are conducted in
units other than 90-day tanks or containers also are subject to
regulation.
We received two comments expressing concern that this would subject
all fuel blending-related equipment permitting, without allowing for
case-by-case determinations. For example, these commenters believe that
some pre-processing activities conducted by blenders (shredding, drum
crushing, and other physical handling) do not meet the definition of
treatment and should not be subject to permitting standards. However,
we feel that these activities meet the existing definition of
treatment. They are ``processe(s) . . . designed to change the physical
. . . composition of . . . hazardous waste so as to . . . render such
waste amenable for recovery'' via combustion. See 40 CFR 260.10
(definition of ``treatment'').
Moreover, these pre-processing activities should be subject to
permitting requirements. Controls on these activities are necessary to
protect against releases of hazardous constituents to the environment
due to the nature of those operations (e.g., crushing or shredding of
drums containing hazardous wastes, grinding of waste materials, etc.).
See Shell Oil v. EPA, 950 F. 2d 741, 753-56 (D.C. Cir. 1991), which
broadly construes the definition of treatment to assure that the RCRA
goal of cradle-to-grave management of hazardous wastes is satisfied and
that specific types of units remain subject to subtitle C regulation.
For units that do not already meet the definition of a specific unit,
subpart X is available to provide the appropriate standards.

V. Are There Any Conforming Changes to Subpart X?

In today's rule, we are making a conforming change to part 264
subpart X (Sec. 264.601) to make reference to part 63 subpart EEE.
Hazardous waste treatment, storage, and disposal facilities that
are not

[[Page 52994]]

classified under other categories (e.g., tank systems, surface
impoundments, waste piles, incinerators, etc.) are classified as
miscellaneous units and regulated under part 264 subpart X. However,
due to the varying types and designs of miscellaneous units, subpart X
does not include specific performance standards. Instead, subpart X
makes reference to requirements in other sections of the regulations.
Section 264.601 of subpart X states that ``Permit terms and provisions
shall include those requirements of subparts I through O and subparts
AA through CC of this part, part 270, and part 146 that are appropriate
for the miscellaneous unit being permitted .'' This statement directs
the permitting agency to look at the requirements (e.g., performance
standards, operating parameters, monitoring requirements, etc.) from
other sections in the regulations when developing appropriate permit
conditions for miscellaneous units.
In the past, permitting authorities have often looked to the part
264 subpart O regulations for incinerators to develop the appropriate
permit conditions for units such as thermal desorbers and carbon
regeneration units. Since today's rule upgrades the air emission
standards for certain source categories, these new standards also
should be considered when determining the appropriate requirements for
miscellaneous units, most notably those engaged in any type of thermal
operation. Therefore, the language in Sec. 264.601 of subpart X is
being modified to incorporate a reference to part 63 subpart EEE.

VI. What Are the Requirements for Bevill Residues?

A. Dioxin Testing of Bevill Residues
In the proposal, we proposed to add polychlorinated dibenzo-p-
dioxin and polychlorinated dibenzo-furan compounds to appendix VIII of
part 266. Appendix VIII lists those compounds that may be generated as
products of incomplete combustion and that must be included in testing
of Bevill residues conducted pursuant to 40 CFR 266.112. Products of
incomplete combustion can be unburned organic compounds that were
originally present in the waste, thermal decomposition products
resulting from organic constituents in the waste, or compounds
synthesized during or immediately after combustion. We noted in the
proposal that there is a considerable body of evidence to show that
dioxin and furan compounds can be formed in the post-combustion regions
of hazardous waste burning boilers, industrial furnaces, and
incinerators, especially at temperatures between 250-450 deg.C.\309\
\310\ Collected particulate matter in the post-combustion regions of
furnaces can provide sites for adsorption of precursors, formation of
dioxins and furans by surface chlorination of precursors, catalytic
production of chlorine for subsequent chlorination of dioxin and furan
precursors, and de novo synthesis of dioxins and furans. This same
particulate matter may be subsequently managed as excluded Bevill
residue.
---------------------------------------------------------------------------

\309\ USEPA, ``Estimating Exposure to Dioxin-Like Compounds'',
EPA/600/6-88/005Ca, June 1994.
\310\ USEPA, ``Combustion Emissions Technical Resource Document
(CETRED)''. EPA/530/R-94/014, May 1994.
---------------------------------------------------------------------------

No evidence was provided by commenters to show that dioxins and
furans cannot be formed in cooler, post-combustion regions of furnaces
(e.g., ductwork, boiler tubes, heat exchange surfaces, and air
pollution control devices). A few commenters referenced the total
number of nondetects for all of the compounds in the cement kiln dust
database. However, the relevance of this information specifically to
dioxins and furans was unclear. Dioxins and furans have repeatedly been
detected in cement kiln dust, as well as other Bevill residues.\311\
\312\
---------------------------------------------------------------------------

\311\ USEPA, ``Report to Congress on Cement Kiln Dust'', EPA/
530/R-94/001, December 1993.
\312\ USEPA, ``Dioxins/Furans, Metals, Chlorine, Hydrochloric
acid, and Related Testing at a Hazardous Waste-Burning Light-Weight
Aggregate Kiln'', June 1997 Draft Report.
---------------------------------------------------------------------------

The majority of commenters were concerned about implementation
issues. Many felt that the addition of dioxins and furans to part 266
appendix VIII, in conjunction with the proposed requirement for daily
sampling and analysis of Bevill residues, would make Bevill
demonstrations prohibitively expensive. They also noted that the
turnaround time for daily dioxin and furan analyses would delay
compliance demonstrations and result in shortages in storage capacity.
One commenter felt that daily sampling for dioxins and furans is not
warranted because cement kiln dust at their site has already been shown
to meet the proposed Bevill exclusion criteria for dioxins and furans.
None of these arguments directly address our basic premise that dioxin
and furan compounds can be generated in combustion systems, are of
concern to the protection of human health and the environment, and, as
such, should be included in part 266 appendix VIII. Rather, these
comments pertain to issues that are more readily and appropriately
resolved within the context of site-specific Bevill testing plans.
The proposed daily residue test frequency, which was cited most
often as an impediment in conjunction with dioxin and furan analysis,
is not being promulgated as part of today's rule. The rule will leave
maximum flexibility for development of appropriate dioxin and furan
analysis frequencies considering site-specific factors. Most facilities
should be able to substantially limit the number of dioxin and furan
analyses after an initial sampling effort. Most residue test plans rely
on the concentration-based comparisons to F039 nonwastewater levels (40
CFR 266.112(b)(2)) in combination with a phased testing approach. Under
the phased approach, test frequency can be substantially reduced for
those constituents where initial sampling efforts reveal that
concentrations are well below the F039 levels. Of the facilities where
residue testing for dioxins and furans has been performed, we are aware
of only two facilities where dioxins and furans have exceeded the F039
levels. Thus, the burden of higher analytical costs is expected to be
appropriately limited to those few sites with significant dioxin and
furan residue concentrations.
Several commenters pointed out that some Bevill residues (e.g.,
slag from primary smelters) are generated prior to the post-combustion
regions typically associated with dioxin and furan formation. Indeed,
the preamble discussion in the proposal focused exclusively on post-
combustion residues and did not address Bevill-exempt primary smelter
slags. We currently do not have analytical data on dioxins and furans
in smelter slag. However, our current information on dioxin and furan
formation mechanisms suggests that it would be highly unlikely to
expect significant dioxins and furans in smelter slag. Therefore, we
agree that dioxin and furan analyses should be limited to those
residues where there is a reasonable expectation that dioxins and
furans could be present (e.g., post-combustion residues).
Finally, two commenters disagreed with our assertion that dioxins
and furans have been shown, in a national comparison, to be higher in
residues from hazardous waste burning cement kilns than from other
cement kilns. Although this information was included in the proposal as
background, it is not necessary to reconcile various interpretations
regarding national trends for today's rule. The 40 CFR 266.112
provisions are site-specific, and 40 CFR 266.112(b)(1) provides ample
opportunity for you to demonstrate, on a site-specific basis as
necessary, that waste-derived residues are not

[[Page 52995]]

significantly different from normal residues.
After considering all of the comments on the proposal, we are
adding dioxins and furans to part 266 appendix VIII in today's rule. A
notation has been included to clarify that dioxin and furan analyses
are required only for post-combustion residues. Commenters provided no
compelling information to challenge the classification of dioxins and
furans as products of incomplete combustion which can be formed in
post-combustion regions of combustion systems, and the presence of
dioxin and furan compounds in several post-combustion Bevill residues
is clearly documented. Also, the increased use of carbon injection
technology to achieve dioxin and furan stack emissions reductions could
increase dioxin and furan contamination of Bevill residues in the
future. The addition of dioxins and furans to part 266 appendix VIII is
not expected to unduly burden the regulated community because
facilities with dioxins and furans well below exclusion levels should
be able to justify a minimum test frequency.
Dioxins and furans will be listed in part 266 appendix VIII simply
as ``Polychlorinated dibenzo-p-dioxins'' and ``Polychlorinated dibenzo-
furans''. However, the specific form of dioxins and furans that must be
determined analytically will depend on the portion of the two-part test
that is being implemented. If you are performing a comparison with
normal residues pursuant to 40 CFR 266.112(b)(1), specific congeners
and homologues must be measured and converted to TEQ values using the
procedure provided in part 266, appendix IX, section 4.0. We received
no comments regarding this portion of the proposal. If you are
utilizing the concentration-based comparison to the F039 nonwastewater
levels in 40 CFR 268.43 as outlined in 40 CFR 266.112(b)(2), then only
the tetra-, penta-, and hexa-homologues need to be measured (these are
the only homologues with established F039 concentration limits). One
commenter seemed uncertain as to whether the tetra-, penta-, and hexa-
homologue concentrations should be converted to TEQ values. We have
revised the regulatory language to clarify that total concentrations
for each homologue, not TEQs, should be used for the F039 comparisons.
Another commenter objected to the use of F039 levels for the health-
based comparison, noting that the F039 concentrations are technology-
based levels. Our rationale for relying on the F039 concentrations has
been explained previously (see 58 FR at 59598, November 9, 1993) and is
not being revisited in today's rule.
B. Applicability of Part 266 Appendix VIII Products of Incomplete
Combustion List
In the proposal, we noted the confusion regarding whether every
constituent listed on the part 266 appendix VIII list must be included
in residue testing at every facility. We proposed to clarify that the
part 266 appendix VIII list is applicable in its entirety to every
facility.
The only comments received on this issue were objections to our
characterization of this change as a clarification. The commenters felt
this was a substantive change that should not be enforced prior to the
effective date of any final rule establishing the revision as law. The
Agency is proceeding in today's rule to make the part 266 appendix VIII
list applicable in its entirety to every facility by changing the title
of the appendix from ``Potential PICs for Determination of Exclusion of
Waste-Derived Residues'' to ``Organic Compounds for Which Residues Must
Be Analyzed.'' This change is considered a revision to the part 266
regulations effective 30 days after the date of publication of today's
rule. We will not seek to retroactively enforce this provision.

VII. Have There Been Any Changes in Reporting Requirements for
Secondary Lead Smelters?

We proposed that secondary lead smelters subject to MACT standards
for the secondary lead source category not be subject to RCRA air
emission standards. 61 FR at 17474 (April 19, 1996). This exemption
would apply only if a secondary lead smelter processed the type of feed
material we evaluated in promulgating the secondary lead MACT
standards, namely, lead-bearing hazardous wastes containing less than
500 ppm toxic nonmetals and/or hazardous wastes listed in appendix XI
to 40 CFR part 266. Id. at 14475. Secondary lead smelters are presently
not subject to RCRA air emission standards under these circumstances.
See existing Sec. 266.100 (c)(1) and (c)(3). However, they are subject
to certain notification and recordkeeping requirements found in
Sec. 266.100 (c)(1)(I) and (c) (3) and on-going sampling and analysis
requirements in Sec. 266.100 (c)(1)(ii) and Sec. 266.100 (c)(3)(i)(D).
The practical effect of the proposal was to continue to relieve
secondary lead smelters of these administrative requirements.
The proposal was supported by the public commenters. The reason for
the proposal remains. That is, now that secondary lead smelters are
complying with MACT standards for their source category, it is not
necessary for them to be regulated under RCRA also for their air
emissions. 60 FR 29750 (June 23, 1995). For the same reason, it is
unnecessary to have the same level of recordkeeping and other
administrative oversight as when these units were exempt from RCRA air
emission requirements but not yet complying with CAA standards for
hazardous air pollutants. 61 FR at 14474. Consequently, we are
finalizing this portion of the proposal.
Today's rule takes the form of an amendment to the RCRA BIF rule
(new Sec. 266.100 (h)) and indicates that secondary lead smelters are
exempt from all provisions of the BIF rule except for Sec. 266.101,
which contains the restrictions on types of hazardous waste which may
be burned, as described in the first paragraph above. As proposed, a
secondary lead smelter must provide a one-time notice to the Regional
Administrator or State Director identifying each hazardous waste burned
and stating that the facility claims an exemption from other
requirements in the BIF rules. Those secondary lead smelters which have
already notified pursuant to existing regulatory provisions (namely
Sec. 266.100 (c) (1) (i) or Sec. 266.100 (c) (3) (i) (D)) would not
have to renotify.

VIII. What Are the Operator Training and Certification Requirements?

Section 129 of the CAA requires us to develop and promulgate a
program for training and certification of operators of facilities that
burn municipal and medical wastes. We accordingly promulgated operator
training and certification requirements for the operators of municipal
waste combustors (60 FR 65424 (December 19, 1995)) and medical waste
incinerators (62 FR 48348 (September 15, 1997)). At proposal, we
considered similar requirements for hazardous waste combustor operators
also and requested comments on whether: (1) Operator certification
requirements are necessary for hazardous waste combustors, and (2) the
American Society of Mechanical Engineers (ASME) standards (or an
equivalent state certification program) are appropriate and sufficient.
We note that ASME has established a Standard for the Qualification and
Certification of Hazardous Waste Incinerator Operators in collaboration
with the American National Standards Institute (ASME Standard Number
QHO-1-1994) and has been providing certifications since 1996.

[[Page 52996]]

Commenters differed widely on two key issues: (1) Whether such a
training program should be voluntary, mandatory, or even necessary,
considering that RCRA already requires some site-specific training
program (40 CFR 264.16); and (2) whether the certifying agency should
be an independent body like ASME versus an industry organization like
the Cement Kiln Recycling Coalition. Most commenters favored the
establishment of a mandatory operator certification program by an
independent organization that develops consensus standards (e.g., ASME,
American Society for Testing and Materials, or American National
Standards Institute) in order to preserve the integrity of
certification. We agree and note that ASME has already done commendable
work in developing certification programs for operators of municipal
waste combustors, medical waste incinerators, high capacity fossil-fuel
fired plants, and hazardous waste incinerators. Each combustor program
includes defined criteria for certification, including operator
qualifications, recommended training, examination content, minimum
passing grades, and due process. These programs are incorporated (at
least in part) into EPA's combustion regulations to satisfy the CAA
section 129 mandate, and we are extending similar requirements in
today's rule to all hazardous waste combustor operators also. We find
that the concerns about good operator training and certification that
underlie the section 129 requirement for municipal waste combustors and
medical waste incinerators apply as well to those persons charged with
the responsibility for safe handling and burning of hazardous waste.
Some kiln operators and the Cement Kiln Recycling Coalition have
commented that cement and lightweight aggregate kilns are much larger
and more diverse facilities than most hazardous waste incinerators,
that these kilns operate with employee unions that object to additional
outside certification when site-specific training programs are already
in place, and that the ASME certification programs are not pertinent or
applicable to them. We recognize that there are some differences in the
operation of incinerators and cement and lightweight aggregate kilns.
However, these differences do not suggest that operator training and
certification should be abandoned. Rather, they serve to emphasize the
importance of having a rigorous operator training and certification
program in place and having it subject to regulatory agency scrutiny.
In that regard, we are aware of the Cement Kiln Recycling Coalition's
efforts to develop a suitable industry-wide training and certification
program for the kilns. However, the Cement Kiln Recycling Coalition's
efforts to date have not resulted in a final industry-wide set of
standards that can be relied upon in today's rule, and we note that the
current general facility training programs under Sec. 264.16 do not
fully cover the areas that would need to be addressed at facilities
burning hazardous waste. For example, Sec. 264.16 neither identifies
important areas of training with respect to daily operations (such as
hazardous waste and residues handling operations, air pollution control
device operations, troubleshooting, normal start-up and shut-down
procedures, continuous emissions monitoring system operation and
maintenance etc.) nor discriminates among the different categories of
operators. Also, Sec. 264.16 does not specify any operator
certification nor minimum standards for certification, which are needed
to ensure the initial and continual competence of the hazardous waste
combustor facility operators.
We expect that kiln specific programs will be developed in the near
future after complete analysis for consistency, reliability and
conformance with principles of good operating and operator practices
(including training and certification). Today's rule therefore
specifies that each hazardous waste combustor facility must develop an
operator training and certification program. In the case of cement and
lightweight aggregate kilns, the facility must submit its program to
the Agency for approval. The submittal will be evaluated for
completeness, reliability and conformance with appropriate principles
of good operator and operating practices (including training and
certification). If a state-approved certification program becomes
available, the facility's program must conform to that state program.
These are to ensure that sufficient specifics are included in each
facility program. In the case of hazardous waste incinerators, the
facility's program must conform to either a state-approved
certification program or, if none exists, to the ASME certification
program (Standard No. QHO-1-1994). Again, this is to ensure that
sufficient specifics are contained in a facility program.

IX. Why Did the Agency Redesignate Existing Regulations Pertaining to
the Notification of Intent To Comply and Extension of the Compliance
Date?

In today's final rule, we redesignate existing regulations
pertaining to the Notification of Intent to Comply with subpart EEE and
extensions of the compliance date to install pollution prevention or
waste minimization controls to meld them into the new provisions of the
subpart. This ensures that similar topics (e.g., notifications,
compliance requirements) are grouped together in the rule. We also
revise those existing regulations to: (1) Convert the regulatory
language to plain language consistent with the new provisions; (2)
include references to the new provisions; and (3) include references to
the actual effective date of the rule.
We promulgated these regulations as Part 1 of revised standards for
hazardous waste combustors. See 63 FR 33782 (June 19, 1998). We are
promulgating part 2 today, which comprises the emission standards and
compliance requirements. Today's revisions to the existing standards
does not constitute a repromulgation and does not reopen the comment
period for those standards.
We are redesignating the existing regulations as indicated in the
following table:

----------------------------------------------------------------------------------------------------------------
Existing regulation Topic Predesignated regulation
----------------------------------------------------------------------------------------------------------------
Sec. 63.1211(a) and (b)............... Notification requirements for Sec. 63.1210(b) and (c)
the notification of intent to
comply.
Sec. 63.1211(c)....................... Requirements for sources that Sec. 63.1206(a)(2)
do not intend to comply.
Sec. 63.1212.......................... Progress report requirements Sec. 63.1211(b)
for the notification of
intent to comply.
Sec. 63.1213.......................... Certification that must Sec. 63.1212(a)
accompany the notice of
intent to comply.
Sec. 63.1214.......................... Extension of the compliance Sec. 63.1206(a)(1)
date.
Sec. 63.1215.......................... Requirements for sources that Sec. 63.1212(b)
become affected sources after
the effective date of the
emission standards.

[[Page 52997]]

Sec. 63.1216.......................... Extension of the compliance Sec. 63.1213
date to install pollution
prevention or waste
minimization controls.
----------------------------------------------------------------------------------------------------------------

Part Seven: National Assessment of Exposures and Risks

We received many public comments on the risk assessment for the
proposed rule.³¹³ In addition, the risk assessment was peer
reviewed in accordance with EPA guidelines. Many of the commenters
commented on similar topics. These topics included the
representativeness of the HWC facilities modeled, the estimation of
facility emissions, the exposure scenarios evaluated, and the
assessment of risks from mercury. As of result of these comments, we
made significant changes in the risk assessment for the final rule.
Also, new information became available after proposal on food intake
rates for home-produced foods and methods for assessing exposures to
mercury. In addition, EPA issued guidance for use of probabilistic
techniques in risk assessments and a policy for evaluating risks to
children. These were also considered in making revisions to the risk
assessment. A complete discussion of the risk assessment for today's
rule may be found in the background document.³¹⁴
---------------------------------------------------------------------------

\313\ ``Risk Assessment Support to the Development of Technical
Standards for Emissions from Combustion Units Burning Hazardous
Wastes: Background Information Document,'' February, 1996.
\314\ See the background document, ``Human Health and Ecological
Risk Assessment Support to the Development of Technical Standards
for Emissions from Combustion Units Burning Hazardous Wastes:
Background Document--Final Report,'' July, 1999.
---------------------------------------------------------------------------

I. What Changes Were Made to the Risk Methodology?

A. How Were Facilities Selected for Analysis?
The representativeness of the example facilities used in the risk
assessment at proposal was widely questioned by commenters. We analyzed
eleven example facilities for the proposed rule: two commercial
incinerators, two on-site incinerators, two lightweight aggregate
kilns, and five cement kilns.³¹⁵ While these facilities
represented a geographically diverse set of facilities in each source
category, it was not possible to demonstrate in any formal way that the
facilities were representative of the universe of facilities covered by
the rule.
---------------------------------------------------------------------------

\315\ See 61 FR 17370 and ``Risk Assessment Support to the
Development of Technical Standards for Emissions from Combustion
Units Burning Hazardous Wastes: Background Information Document''
(February, 1996).
---------------------------------------------------------------------------

Because of this difficulty, we concluded that the most efficient
approach for assuring the representativeness of the facilities analyzed
was to select a stratified random sample. The number of strata was
determined by the number of categories and subcategories of sources for
which risk information was desired. The final sample of facilities
chosen for analysis includes 66 randomly selected facilities and 10 of
the 11 facilities selected at proposal for a total sample of 76
facilities out of a universe of 165 facilities within the contiguous
United States.³¹⁶ The sample sizes are as follows:
---------------------------------------------------------------------------

\316\ A large on-site incinerator analyzed at proposal that is
undergoing RCRA closure was excluded from the analysis.

Hazardous Waste Combustion Facility Stratum and Sample Sizes
----------------------------------------------------------------------------------------------------------------
High end
Combustion facility category Stratum size Random sample NPRM sample Final sample sampling
size size size probability
--------------------------------------------------------------------------------------------------------\1\-----
Cement Kilns.................... 18 10 5 15 98
Lightweight Aggregate Kilns..... 5 3 2 5 100
Commercial Incinerators:
Including Waste Heat Boilers 20 11 2 13 97
Excluding Waste Heat Boilers 12 7 2 9 95
Large On-Site Incinerators:
Including Waste Heat Boilers 43 17 1 18 94
Excluding Waste Heat Boilers 36 15 0 15 90
Small On-Site Incinerators:
Including Waste Heat Boilers 79 25 0 25 96
Excluding Waste Heat Boilers 65 16 0 16 88
Incinerators With Waste Heat 29 15 1 16 92
Boilers........................
----------------------------------------------------------------------------------------------------------------
\1\ Probability that a facility that lies in the upper 10% of the distribution of risk will be sampled.

For the randomly selected facilities, sample sizes within a given
category were chosen such that the probability of sampling a facility
in the upper ten percent of the distribution of risk would be 90
percent or greater. The probabilities actually achieved range from 88
to 100 percent depending on the size of the original, non-randomly
chosen sample and changes in the sampling frame that occurred during
the random sampling process.³¹⁷
---------------------------------------------------------------------------

\317\ Changes in the sampling frame occurred as a result of
facilities that were missing from the original sampling frame were
misclassified, or were no longer burning hazardous waste and had
begun RCRA closure.
---------------------------------------------------------------------------

We did not target area sources specifically for sampling because
the statutory definition of major sources versus area sources is based
on facility-wide emissions of hazardous air pollutants and such
information was not available at the time the sampling was performed.
Therefore, it was not possible to determine the sampling frame. We
expect that on-site incinerators, both large and small, at large
industrial facilities are major sources rather than area sources.
Because area sources are of interest, we made risk inferences based on
those area source incinerators that could be identified and had
otherwise been

[[Page 52998]]

sampled.³¹⁸ For cement kilns, all area sources were sampled
and used for making such inferences.
---------------------------------------------------------------------------

\318\ Area source incinerators that were identified included
commercial incinerators and on-site incinerators at U.S. Department
of Defense installations.
---------------------------------------------------------------------------

B. How Were Facility Emissions Estimated?
At proposal, we estimated baseline emissions (reflecting current
conditions) for the example facilities from the distribution of stack
gas concentrations for the corresponding category of sources. Both
central tendency and high end emissions estimates were made based on
the 50th and 90th percentiles of the stack gas concentration
distributions. For the purpose of evaluating risks associated with the
proposal, we assumed that facilities emitted at the design level
determined to be necessary to meet the standard, even if this meant an
increase in emissions over baseline. Many commenters thought that using
percentiles to estimate emissions was inappropriate and that site-
specific emissions should be used instead. Commenters also thought that
it was incorrect to project an increase in risk with the proposed
standards (which occurred as a result of allowing emissions to increase
over baseline). We agree with these comments. For the final rule, we
estimated emissions based on site-specific stack gas emission
concentrations and flow rates. Site-specific stack gas concentration
data were used where emissions measurements were available; otherwise,
stack gas concentrations were imputed. For today's rule, we assumed
emissions would remain unchanged from baseline in instances where a
facility's emissions are already below the design level (which is taken
as 70 percent of the MACT standard).³¹⁹ In instances where a
facility's emissions exceed the design level, we determined the
percentage reduction in emissions required to meet the design level. We
then applied this reduction to each chemical constituent to which the
standard applies.
---------------------------------------------------------------------------

\319\ This is also consistent with the assumption made in the
cost and economic analysis that facilities that are currently
emitting below the design level will not need to retrofit using new
control technology.
---------------------------------------------------------------------------

The imputation approach we used in instances where measured data
were not available involves the random selection of emissions
concentrations from a pool of emissions concentrations for other
facilities and test conditions that are believed to be reasonably
representative of the facility in question. For groups of interrelated
constituents (e.g., different dioxin congeners or mercury species),
imputation was carried out for the group of interrelated constituents
taken together rather than each individual constituent separately. We
used the random imputation approach to preserve the variability in
emissions exhibited by the pooled data. Another commonly used approach
for estimating emissions, emissions factors, generally represents
average conditions and does not reflect the variability in emissions
across facilities in a given source category. Because the objective of
the risk assessment is to characterize the distribution of risks across
a given source category, we deemed the use of average emissions to be
inappropriate except where only very limited data are available (i.e.,
for cobalt, copper, and manganese). Although the random imputation
approach may significantly over or under estimate emissions for a given
facility (a problem also inherent in emission factors), we expect that
the distributions of risk across a given source category are better
characterized using random imputation than with an emissions factor
approach or any other approach that does not account for the variation
in emissions from one facility to the next.
Emissions estimates were made for all chemical constituents covered
by the rule for which sufficient data were available, including all
2,3,7,8-chlorine substituted dibenzo(p)dioxins and dibenzofurans,
elemental mercury (Hg⁰), divalent mercury (Hg⁺²),
lead, cadmium, arsenic, beryllium, trivalent chromium
(Cr⁺³), hexavalent chromium (Cr⁺⁶), chlorine, and
hydrogen chloride. In addition, emissions estimates were made for
particulate matter (PM10 and PM2.5) and nine
other metals, three of which (cobalt, copper, and manganese) were not
assessed at proposal but were included in the risk assessment for the
final rule. Chemical-specific emissions estimates could not be made for
organic constituents other than dioxins and furans (e.g., various
products of incomplete combustion) due to the lack of sufficient
emission measurements. We assessed the risks from all constituents for
which chemical-specific emissions estimates could be made, as well as
from particulate matter. A complete discussion of the emissions
estimates used in the risk assessment may be found in the technical
support documents for today's rule.³²⁰
---------------------------------------------------------------------------

\320\ See ``Final technical Support Document for HWC MACT
Standards, Volume V: Emission Estimates and Engineering Costs.''
July, 1999.
---------------------------------------------------------------------------

C. What Receptor Populations Were Evaluated?
The risk assessment at proposal examined risks to individuals
engaged in subsistence activities such as farming and fishing. Some
commenters viewed these types of activities as unlikely to occur and
questioned whether these types of exposures are representative of
actual exposures and risk. Other commenters thought the exposure
pathways included in the analysis did not fully reflect potential
exposures to individuals living a true subsistence lifestyle. We share
the concerns raised by commenters and have refocused the assessment on
non-subsistence receptor populations such as commercial farmers,
recreational anglers, and non-farm residents whose numbers and
locations can be estimated from available census data. At the same
time, we retained the subsistence scenarios and revised them to be more
reflective of a subsistence lifestyle. Although it is not known
precisely how many individuals are engaged in subsistence activities or
exactly where those activities take place, subsistence does occur in
some segments of the U.S. population, and we believe it is important to
evaluate the associated risks.
D. How Were Exposure Factors Determined?
Since the risk assessment at proposal, we have developed new
information on factors that are used to estimate exposures. We obtained
data collected from previously published studies and used the data to
derive exposure factor information, including information for
children.³²¹ In particular, we reanalyzed data collected by
USDA to estimate consumption of home-produced foods, such as meat,
milk, poultry, fish, and eggs. Over half of farm households report
consuming home-produced meats, including nearly 40 percent that report
consumption of home-produced beef. In the Northeast, nearly 40 percent
of farm households report consuming home-produced dairy products, and,
in the Midwest, nearly 20 percent do. The percentage is lower
elsewhere, averaging about 13 percent nationally. Presumably most of
these households are associated with dairy farms. Most farm households
that consume home-produced foods are engaged in farming as an
occupation rather than a means of subsistence.
---------------------------------------------------------------------------

\321\ EPA published the new exposure factor information in the
``Exposure Factors Handbook,'' EPA/600/P-95/002Fb, August, 1997.
---------------------------------------------------------------------------

The data indicate that individual consumption of home-produced
foods is

[[Page 52999]]

higher than consumption of the same foods in the general populace. We
have used the information on home-produced foods to estimate the
exposures to farm households and to households engaged in subsistence
farming. Only the primary food commodity produced on the farm was
assumed to be consumed by farm households. In contrast, a wide variety
of foods was assumed to be produced and consumed by households engaged
in subsistence farming.
E. How Were Risks from Mercury Evaluated?
Commenters viewed the absence of a quantitative assessment of risks
from mercury as a significant failing at proposal. However, a number of
issues related to assessing risks from mercury had not been adequately
resolved at the time of proposal that would have allowed us to proceed
with a quantitative analysis. We have since issued our Mercury Study
Report to Congress, a study that has been subject to extensive peer
review, and the Utility Study Report to Congress.³²²
³²³ With today's rule, we conclude that sufficient technical
basis exists for conducting a quantitative assessment of mercury risks
from hazardous waste combustors. We recognize, however, that
significant uncertainties remain and the results of our mercury
analysis should be interpreted with caution and be used only
qualitatively.
---------------------------------------------------------------------------

\322\ ``Mercury Study Report to Congress, Volume III: Fate and
Transport of Mercury in the Environment,'' U.S. Environmental
Protection Agency, EPA-452/R-97-005, December 1997.
\323\ ``Study of Hazardous Air Pollutant Emissions from Electric
Utility Steam Generating Units--Final Report to Congress,'' U.S.
Environmental Protection Agency, EPA-453/R-98-004a and b, February
1998.
---------------------------------------------------------------------------

Although the mercury analysis that accompanies today's rule is
patterned after the analysis done for the Mercury Study, there are
differences between the two studies in the methods used. The model we
used for evaluating the fate and transport of mercury in lakes is the
same as the IEM-2M model used in the Mercury Study Report to Congress.
However, modifications were made to adapt it for use with rivers and
streams.³²⁴ Both studies used the ISC air dispersion model
for modeling wet deposition of mercury. However, for the Mercury Study
the ISC model was modified to include dry deposition of mercury vapor
whereas, for the current analysis, we used a simplified treatment of
dry vapor deposition. In the Mercury Study, air modeling was carried
out to a distance of 50 kilometers whereas, for the current analysis,
air modeling (and, therefore, the effective size of the modeled
watersheds) was limited to a distance of 20 kilometers. Long-range
transport of mercury emissions (beyond 50 kilometers) was considered in
the Mercury Study but was not included in the current analysis. In the
Mercury Study, a large number of different sources were investigated to
identify whether reductions in anthropogenic or environmental sources
of mercury would reduce the total exposures of mercury to the general
population. The current analysis was designed to assess what reductions
may occur in incremental exposures from specific industrial sources of
mercury to specific individuals rather than what reductions would occur
in total exposures of mercury. Also, the Mercury Study modeled
exposures under varying background assumptions, but the current
analysis did not assess the impact that variable background
concentrations would have on the risk results. In addition, the Mercury
Study received external peer review, whereas we have not conducted an
external peer review of the current analysis.
---------------------------------------------------------------------------

\324\ For a discussion of the mercury surface water model, see
the background document, ``Human Health and Ecological Risk
Assessment Support to the Development of Technical Standards for
Emissions from Combustion Units Burning Hazardous Wastes: Background
Document--Final Report,'' July, 1999.
---------------------------------------------------------------------------

In addition, there are a variety of uncertainties related to the
fate and transport of mercury in the environment, such as the
deposition of mercury emitted to the atmosphere via wet and dry removal
processes, the transport of mercury deposited in upland areas of a
watershed to a body of water, and the disposition of mercury in the
water body itself, including methylation and demethylation processes,
sequestering in the water column and sediments, and uptake in aquatic
organisms. Furthermore, the form of mercury emitted by a given facility
is thought to be a determining factor in the fate and transport of
mercury in the atmosphere. Only limited data are available on the form
of the mercury emitted from hazardous waste combustors. A more complete
discussion of the uncertainties related to the fate and transport of
mercury may be found in the Mercury Study Report to Congress.
Also important to consider is that the reference dose for methyl
mercury represents a ``no-effects'' level that is presumed to be
without appreciable risk. We used an uncertainty factor of 10 to derive
the reference dose for methyl mercury from a benchmark dose that
represents the lower 95% confidence level for the 10% incidence rate of
neurologic abnormalities in children.³²⁵ Therefore, there is
a margin of safety between the reference dose and the level
corresponding to the threshold for adverse effects, as indicated by the
human health data. Furthermore, we applied the reference dose, which
was developed for maternal exposures, to childhood exposures. This
introduces additional uncertainty in the risk estimates for children.
Additional uncertainties associated with assessing individual mercury
risks to nonsubsistence populations and subsistence receptors are
discussed under the ``Human Health Risk Characterization'' section
below.
---------------------------------------------------------------------------

\325\ The uncertainty factor is intended to cover three areas of
uncertainty: Lack of data from a two-generation reproductive assay;
variability in the human population, in particular the wide
variation in the distribution and biological half-life of methyl
mercury; and lack of data on long term sequelae of developmental
effects.
---------------------------------------------------------------------------

We do not know the direction or magnitude of many of the
uncertainties discussed above and did not attempt to quantify the
overall uncertainty of the analysis. Thus, the cumulative impact of
these uncertainties is unknown, and the uncertainties implicit in the
quantitative mercury analysis continue to be sufficiently great so as
to limit its ultimate use for decision-making. Therefore, we have used
the quantitative assessment to make qualitative judgments about the
risks from mercury but have not relied on the quantitative assessment
(nor do we believe it is appropriate) to draw quantitative conclusions
about the risks associated with particular national emissions
standards.
F. How Were Risks From Dioxins Evaluated?
Few changes have been made to the methods used for assessing risk
from dioxins since proposal. Some commenters thought we should modify
the toxicity equivalence factors that are used to characterize the
relative risk from 2,3,7,8-chlorine substituted congeners relative to
that from 2,3,7,8,-tetrachlorodibenzo(p)dioxin. As a matter of policy,
we continue to use the international consensus values that were
published by EPA in 1989. We are aware that revisions to the toxicity
equivalence factors are being considered by the international
scientific community. However, we have not adopted revised values and
continue to use the 1989 toxicity equivalence factors.
We have changed the data being relied upon to characterize the
bioaccumulation of dioxins in fish. Specifically, we believe that the
biota-

[[Page 53000]]

sediment accumulation factors used at proposal, which were derived from
data for the Great Lakes, significantly understate the bioaccumulation
potential in aquatic systems that have recent and ongoing
contamination. Studies in Sweden and elsewhere show that where
contamination is ongoing, biota-sediment accumulation factors may be
higher by as much as an order of magnitude or more relative to the
Great Lakes and other aquatic systems where levels in biota are
influenced primarily by past contamination. For the risk assessment for
today's rule, biota-sediment accumulation factors were derived from
data collected by the Connecticut Department of Environmental
Protection. The Connecticut study, which is discussed in detail in the
dioxin reassessment, involved extensive monitoring of soils, sediments,
and fish near resource recovery facilities operating in the
state.³²⁶ The data show biota-sediment accumulation factors
that are a factor of two to nine times higher (depending on the
individual congener) than those used previously.
---------------------------------------------------------------------------

\326\ ``Estimating Exposure to Dioxin-Like Compounds, Volume
III: Site-Specfic Assessment Procedures, U.S. Environmental
Protection Agency, External Review Draft, EPA/600/6-88/005Cc, June
1994
---------------------------------------------------------------------------

G. How Were Risks from Lead Evaluated?
Risks from exposures to lead were assessed at proposal by comparing
model-predicted lead levels in soil to a health-based soil benchmark
criterion. Commenters pointed out that there are pathways of exposure
other than those related to soils and that we should look at the
overall impact of lead emissions on blood lead levels in children. We
agree with these comments and have modified the risk assessment to
include other pathways of exposure such as inhalation and dietary
exposures, in addition to soil ingestion. The revised assessment
employs the Intake/Exposure Uptake BioKinetic model to assess the
incremental impact of lead intake on blood lead levels in children. The
results of the blood lead modeling are used together with information
on background levels of blood lead in the general population to
estimate the number of children whose blood levels exceed 10 micrograms
per deciliter. Our goal is to reduce children's blood lead to below
this level.
H. What Analytical Framework Was Used To Assess Human Exposures and
Risk?
As a result of the public and peer review comments received on the
risk assessment at proposal, we modified the analysis to focus on the
entire population of persons that are exposed to facility emissions
rather than persons living on a few individual farms and residences. A
study area was defined for each sample facility as the area surrounding
the facility out to a distance of 20 kilometers (or about 12 miles).
All persons residing within the study area were included in the
analysis.³²⁷ The study area was divided up into sixteen (16)
sectors defined by the intersection of rings at two, five, ten and
twenty kilometers and radii extending to the north, south, east, and
west. For each sector, census data were used to estimate the population
of those persons living in farm households by type of farm and the
population of those persons living in non-farm households. Census data
were also used to determine the age of all household members. Four age
groups were delineated: Preschoolers (0 to 5 years), preteens (6 to 11
years), adolescents (12 to 19 years) and adults (20 years and older).
---------------------------------------------------------------------------

\327\ Because the analysis at proposal indicated that exposures
beyond 20 kilometers were well below levels of concern, we did not
consider persons exposed to facility emissions that are transported
beyond 20 kilometers. Also, as discussed elsewhere, the risk
assessment was peer reviewed in accordance with EPA guidelines, and
peer reviewes did not comment that the range of the local scale
study area was insufficient (or recommend that it be increased to 50
or more kilometers).
---------------------------------------------------------------------------

Within each study area, three or four bodies of water were chosen
for analysis based on their proximity to the sample facility and the
likelihood of their being used for recreational purposes, as indicated
by factors such as size and accessibility. Water bodies were also
chosen if they were used to supply drinking water to the surrounding
community. The watershed of each water body was delineated out to a
distance of 20 kilometers from the facility.
We conducted a multi-pathway exposure analysis for all the human
receptors considered in the risk assessment. Household members
regardless of the type of household were assumed to be exposed to
facility emissions through direct inhalation and incidental ingestion
of soil. In addition, in study areas where surface waters are used for
drinking water, household members were also assumed to be exposed
through tap water ingestion. A portion of non-farm households were
assumed to engage in home gardening based on the prevalence of home
gardening in national surveys. Farm households were assumed to consume
the primary food commodity produced on the farm. This contrasts with
the subsistence farmer who was assumed to consume predominantly home-
produced foods, including meat, milk, poultry, fish, and eggs, as well
as fruits and vegetables. For the purpose of characterizing the range
of risks that could result from subsistence farming, it was assumed
that a subsistence farm was located in every sector in a given study
area. A portion of the households in each study area were assumed to
engage in recreational fishing based on the prevalence of recreational
fishing in national surveys. It was assumed that individual
recreational anglers would fish at all of the water bodies delineated
in a given study area. In contrast, households engaged in subsistence
fishing were assumed to consume fish from only a single body of water.
For the purpose of characterizing the range of risks that could result
from subsistence fishing, the assumption was made that every body of
water delineated in a given study area was used for subsistence
fishing.
Air dispersion and deposition modeling were performed for each
study area at all sample facilities using facility-specific information
on stack configuration and emissions, along with site-specific
meteorological data, terrain data (in areas of elevated terrain), and
land use data. Air modeling was conducted to a distance of 20
kilometers. Long-range transport of emissions beyond this distance was
not considered. Bioaccumulation in the terrestrial food chain was
modeled from estimates of deposition and uptake in plants and
subsequent uptake in agricultural livestock from consumption of forage
and silage. Bioaccumulation in the aquatic food chain was modeled from
estimates of deposition to watershed soils (and subsequent soil erosion
and runoff) and direct deposition to water bodies and subsequent uptake
in fish. Surface water modeling was conducted for each body of water
using site-specific information relative to watershed size, surface
runoff, soil erosion, water body size, and dilution flow.
Exposure modeling was performed using central tendency exposure
factors (e.g., duration of exposure and daily food intake) for all
receptor populations. As noted below, an exposure variability analysis
was also performed for selected constituents and receptor populations
using exposure factor distributions. Exposure pathways varied depending
on the particular human receptor and the types of activities that lead
to human exposures. Age-specific rates of mean daily food intake and
media contact rates, in conjunction with sector-specific media
concentrations and concentrations in food, were used

[[Page 53001]]

to calculate the total (administered or potential) dose from all
exposure pathways combined. Lifetime average daily dose was used as the
exposure metric for assessing cancer risk and average daily dose
(reflecting less than lifetime exposure) was used for assessing risks
of non-cancer effects.
We estimated the risk of developing cancer from the estimated
lifetime average daily dose and the slope of the dose-response curve. A
cancer slope factor is derived from either human or animal data and is
taken as the upper bound on the slope of the dose-response curve in the
low-dose region, generally assumed to be linear, expressed as a
lifetime excess cancer risk per unit exposure. Total carcinogenic risk
was determined for each receptor population assuming additivity. The
same approach was used for estimating cancer risks in both adults and
children. This is also the same approach we used at proposal for
estimating lifetime cancer risks stemming from childhood exposures.
However, individuals exposed to carcinogens in the first few years of
life may be at increased risk of developing cancer. For this reason, we
recognize that significant uncertainties and unknowns exist regarding
the estimation of lifetime cancer risks in children. Although the risk
assessment at proposal was externally peer reviewed, EPA's charge to
the peer review panel did not specifically identify the issue of cancer
risk in children and the peer review panel did not address it.
To characterize the potential risk of non-cancer effects, we
compared the average daily dose (reflecting less than lifetime
exposure) to a reference dose and expressed the result as a ratio or
hazard quotient. The reference dose is an estimate of a daily exposure
to the human population, including sensitive subgroups, that is likely
to be without an appreciable risk of deleterious effects during a
lifetime. The hazard quotient, by indicating how close the average
daily dose is to the reference dose, is a measure of relative risk.
However, the hazard quotient is not an absolute measure of risk. For
inhalation exposures, we compared modeled air concentrations to a
reference concentration and expressed the result as a ratio or
inhalation hazard quotient. The reference concentration is an estimate
of a concentration in air that is likely to be without an appreciable
risk of deleterious effects in the human population, including
sensitive subgroups, from continuous exposures over a lifetime. In
addition, inhalation and ingestion hazard indices were generated for
each receptor population by adding the constituent-specific hazard
quotients by route of exposure. The hazard index is an indicator of the
potential for risk from exposures to chemical mixtures.
For dioxins, we used a margin of exposure approach to assess the
potential risks of non-cancer effects. The average daily dose, in terms
of 2,3,7,8-TCDD toxicity equivalents (TEQ), was compared to background
TEQ exposures in the general population and expressed as a ratio or
incremental margin of exposure. An incremental margin of exposure was
generated for infants exposed through intake of breast milk and for
other age groups exposed through dietary intake and other pathways of
exposure. For lead, we characterized the risk of adverse effects in
children by modeling body burden levels in blood that result from
intake of lead in the diet, direct inhalation, and incidental soil
ingestion and comparing these levels to levels at which community-wide
efforts aimed at prevention of elevated blood levels are indicated.
Distributions of individual risk were generated for a given
category of sources by weighting the individual risks using sector-
specific population weights and facility-specific sampling weights.
Such distributions, which were derived using central tendency exposure
factors, were generated for all constituents and receptor populations.
In addition, for those receptor populations and chemical constituents
that exhibited risks within an order of magnitude of a potential level
of concern (using central tendency exposure factors), we performed an
exposure variability analysis. Normalized, age-specific distributions
of food intake and exposure duration were used to adjust the risk
estimates to generate a distribution of risks in each sector. For
children, food intake changes significantly with age, which can affect
the lifetime average daily dose. To adjust for this, a life table
analysis was conducted in which individuals were followed over the
duration of exposure to arrive at an age adjustment factor. The
individual sector distributions were combined for a given source
category using Monte Carlo sampling and the appropriate sector-specific
population weights and facility-specific sampling weights.
Estimates of population risk, or the incidence of health effects in
the exposed population, were made for selected receptor populations and
chemical constituents. Local excess cancer incidence was estimated from
the mean individual risk for a given sector and the number of persons
who reside in a sector. These sector-specific cancer incidence rates
were then adjusted using facility-specific sampling weights and summed
for a given category of sources. Cancer incidence associated with the
consumption of dioxin contaminated beef, pork, and milk by the general
population was estimated at the sector level from the number of dairy
cattle and the number of beef cattle and hogs slaughtered annually,
adjusted using facility-specific sampling weights, and summed by source
category. Excess incidence of lead poisoning in children (over and
above background) was estimated at the sector level from the intake of
lead in the diet, direct inhalation, and incidental soil ingestion,
adjusted using facility-specific sampling weights, and summed.
Generally speaking, incidence rates for non-cancer effects can be
estimated from the number of persons exposed above the reference dose
(i.e., the number of exceedances) and the annual turnover in the
exposed population. However, non-cancer incidence rates of interest,
such as the incidence of exceedances of the methyl mercury reference
dose from consumption of freshwater fish, could not be estimated due to
the difficulty in determining the number and frequency of visits made
by recreational anglers to a given body of water. However, by making
certain assumptions, it was possible to make an estimate of the portion
of recreational anglers who consume fish from local water bodies that
may be at risk.³²⁸
---------------------------------------------------------------------------

\328\ The assumption is that fishing activity typical of
recreational fishing takes place only at the particular water bodies
delineated in the analysis.
---------------------------------------------------------------------------

Due to concerns of commenters about the representativeness of the
risk assessment, we also made estimates of confidence intervals about
the risk estimates. Estimation of confidence intervals was made
possible by virtue of the sampling design used for facility selection.
The confidence intervals quantify the magnitude of the uncertainty of
the risk estimates associated with sampling error only. We emphasize
that the confidence intervals do not reflect other sources of
uncertainty, which may be of considerably greater magnitude.
In addition to the risk estimates for individual chemical
constituents, we estimated the incidence of excess mortality and
morbidity associated with particulate matter emissions. Mortality and
morbidity estimates were made for children and the elderly, as well as
the general population, using concentration-response functions derived
from human epidemiological studies. Incidence rates

[[Page 53002]]

in a given sector were estimated from the size of the exposed
population, including susceptible populations such as children and the
elderly, and either annual mean PM10 and PM2.5
concentrations or distributions of daily PM10 and
PM2.5 concentrations. Morbidity effects include respiratory
and cardiovascular illnesses requiring hospitalization, as well as
other illnesses not requiring hospitalization, such as acute and
chronic bronchitis, acute upper and lower respiratory symptoms, and
asthmatic attacks. As with other incidence estimates, sector-specific
incidence rates were adjusted using facility-specific sampling weights
and summed for a given source category.
I. What Analytical Framework Was Used to Assess Ecological Risk?
Public comments on the ecological assessment at proposal expressed
the view that we should expand the assessment beyond water quality
criteria. We agree with these commenters and have extended the
ecological analysis to include the use of soil and sediment criteria,
in addition to water quality criteria. Also, the analysis was expanded
to include additional metals that are of ecological concern, such as
mercury and copper.
The ecological assessment represents a screening level analysis
that uses media-specific ecological criteria thought to be protective
of a range of ecological receptors. Modeled surface water
concentrations were compared to water quality criteria protective of
aquatic life, such as algae, fish, and aquatic invertebrates, as well
as piscivorous wildlife. Similarly, modeled soil concentrations were
compared to soil criteria protective of the terrestrial soil community,
as well as terrestrial plants and mammalian and avian wildlife. Modeled
sediment concentrations were compared to sediment criteria protective
of the benthic aquatic community. As a screening level analysis, we did
not attempt to determine whether the specific ecological receptors upon
which the media-specific criteria are based are actually present at a
given site. Furthermore, we did not ascertain the occurrence of
threatened or endangered species at individual sites. However, the
ecological receptors upon which the media-specific criteria are based
are commonly occurring species and may not be any less sensitive than
other species and may be more sensitive than some, including perhaps
threatened or endangered species.³²⁹
---------------------------------------------------------------------------

\329\ Multiple ecological criteria were available for most
constituents and the lowest criteria were used to establish the
media-specific values that were in the eco-analysis. In addition,
ecotoxicological benchmarks for mammals and birds were typically
derived from studies involving measures of reproductive success.
---------------------------------------------------------------------------

II. How Were Human Health Risks Characterized?

This section describes the conclusions of the human health risk
assessment. For a full discussion of the methodology and the results of
the assessment, see the background document for today's
rule.³³⁰
---------------------------------------------------------------------------

\330\ ``Human Health and Ecological Risk Assessment Support to
the Development of Technical Standards for Emissions from Combustion
Units Burning Hazardous Wastes: Background Document--Final Report,''
July 1999.
---------------------------------------------------------------------------

A. What Potential Health Hazards Were Evaluated?
This section summarizes the potential health hazards from exposures
to emissions from hazardous waste combustors, in particular the human
health hazards associated with the chemical constituents evaluated in
the risk assessment, including dioxins, mercury, lead, other metals,
hydrogen chloride and chlorine, and particulate matter.
1. Dioxins
A large body of evidence demonstrates that chlorinated
dibenzo(p)dioxins and dibenzofurans can have a wide variety of health
effects, ranging from cancer to various developmental, reproductive and
immunological effects. Dioxins are persistent and highly
bioaccumulative in the environment and most human exposures occur
through consumption of foods derived from animal products such as meat,
milk, fish, poultry, and eggs. In 1985, we developed a carcinogenic
slope factor for 2,3,7,8-TCDD of 1.56e-4 per picogram per kilogram body
weight per day.³³¹ The slope factor represents the 95
percent upper confidence limit estimate of the lifetime excess cancer
risk. Re-analysis of data from laboratory animals and cancer in humans
lends support to the slope factor derived in 1985, and we continue to
use the 1985 estimate pending completion of our dioxin
reassessment.³³² ³³³
---------------------------------------------------------------------------

\331\ USEPA, ``Health Assessment Document for Polychlorinated
Dibenzo-p-Dioxins,'' EPA/600/8-84-014F, September 1985.
\332\ USEPA, ``Health Assessment Document for 2,3,7,8-
Tetrachlorodibenzo-p-Dioxin (TCDD) and Related Compounds,'' External
Review Draft, EPA/600/BP-92/001b, June 1994.
\333\ USEPA, ``Dose Response Modeling of 2,3,7,8-TCDD,''
Workshop Review Draft, EPA/600/P-92/100C8, January 1997.
---------------------------------------------------------------------------

For non-cancer effects, we believe it is inappropriate to develop a
reference dose, or level which is without appreciable risk, using
standard uncertainty factors. This is due to the high levels of
background exposures in the general population and the low levels at
which effects have been seen in laboratory animals. Instead, we have
chosen to use a margin of exposure approach in which the average daily
dose from a given facility is compared to the average daily dose in the
general population. The ratio of the two represents the incremental
margin of exposure and, as such, measures the relative increase in
exposures over background.
2. Mercury
The most bioavailable form of mercury is methyl mercury, and most
human exposures to methyl mercury occur through consumption of fish.
Methyl mercury is known to cause neurological and developmental effects
in humans at low levels. The most susceptible human population is
thought to be developing fetuses. We have developed a reference dose
for methyl mercury of 0.1 microgram per kilogram body weight per day
that is presumed to be protective of the most sensitive human
populations.³³⁴ The reference dose is based on neurotoxic
effects observed in children exposed in utero. Although epidemiological
studies in fish-eating populations are ongoing, we believe that the
reference dose is the best estimate at the present time of a daily
exposure that is likely to be without an appreciable risk of
deleterious effects. However, because it was derived from maternal
exposures, application of the reference dose to assess children's
exposures carries with it additional uncertainty beyond that otherwise
related to the data and methods used for its development.
---------------------------------------------------------------------------

\334\ USEPA, ``Mercury Study Report to Congress,'' EPA-452/R-97-
007, December 1997.
---------------------------------------------------------------------------

3. Lead
Exposures to lead in humans are associated with toxic effects in
the nervous system at low doses and at higher doses in the kidneys and
cardiovascular system. Infants and children are particularly
susceptible to

[[Page 53003]]

the effects of lead due to behavioral characteristics such as mouthing
behavior, heightened absorption in the respiratory and gastrointestinal
tracts, and the intrinsic sensitivity of developing organ systems.
Symptoms of neurotoxicity include impairment in psychomotor, auditory,
and cognitive function. These effects extend down to levels in blood of
at least 10 micrograms lead per deciliter. Impairment of intellectual
development, as measured by standardized tests, is thought to occur at
levels below 10 micrograms per deciliter. Maternal lead exposure has
been shown to be a risk factor in premature infant mortality, lead
being associated with reduced birth weight and decreases in gestational
age. Lead has also been associated with hypertension in both men and
women and, as such, may be a risk factor for coronary disease, stroke,
and premature mortality. Although dose-response relationships have been
developed between blood lead levels and many of these health effects,
EPA has not applied the relationships in the HWC risk analysis due to
uncertainties related to the relatively small changes in blood lead
expected to occur as a consequence of the MACT standards and the
uncertain significance of any health benefits that might be attributed
to such changes. Instead, our characterization of risks from lead
focuses on the reductions in blood levels themselves and EPA's goal of
reducing blood lead in children to below 10 micrograms per deciliter.
4. Other Metals
Metals that pose a risk for cancer include arsenic, cadmium, and
chromium. Human epidemiological studies have shown an increase in lung
cancer from inhalation exposures to arsenic, primarily in
occupationally exposed individuals, and multiple internal cancers (such
as liver, lung, kidney, and bladder), as well as skin cancer, from
exposures to arsenic through drinking water. Human epidemiological
studies have also shown an association between exposures to cadmium and
lung cancer in occupational settings. These studies have been confirmed
by animal studies which have shown significant increases in lung tumors
from inhalation exposures to cadmium. However, cadmium administered
orally has shown no evidence of carcinogenic response. A strong
association between occupational exposures to chromium and lung cancer
has been found in multiple studies. Although workers were exposed to
both trivalent and hexavalent chromium, animal studies have shown that
only hexavalent chromium is carcinogenic. There have been no studies
that have reported that either hexavalent or trivalent chromium is
carcinogenic by the oral route of exposure.
Other metals may pose a risk of noncancer effects. For example, in
animal studies thallium has been shown to have ocular, neurological,
and dermatological effects and effects on blood chemistry and the
reproductive system. Signs and symptoms of similar and other effects
have been observed in occupational studies of thallium exposures.
5. Hydrogen Chloride
Data on the effects of low-level inhalation exposures to hydrogen
chloride are limited to studies in laboratory animals. Based on a
lifetime study in rats which showed histopathological changes in the
nasal mucosa, larynx, and trachea associated with exposures to hydrogen
chloride, we estimated a reference concentration of 0.02
milligrams per cubic meter. The reference concentration was derived
from a human equivalent lowest observed adverse effects level of 6
milligrams per cubic meter using an uncertainty factor of 300 to
account for extrapolation from a lowest observed adverse effects level
to a no observed adverse effects level, as well as extrapolation from
animals to humans (including sensitive individuals).
6. Chlorine
Chlorine gas is a potent irritant of the eyes and respiratory
system. Based on a lifetime study in rats and mice which showed
histopathological changes affecting all airway tissues in the nose, we
derived an interim chronic toxicity benchmark for chlorine gas of 0.001
milligrams per cubic meter. This value was derived from a human
equivalent no observed adverse effects level of 0.04 milligrams per
cubic meter and an uncertainty factor of 30 to account for
extrapolation from animals to humans (including sensitive individuals).
The human equivalent no observed adverse effects level from this study
is also supported by a year-long study in monkeys.³³⁵
---------------------------------------------------------------------------

\335\ For a complete description of the derivation of the
chronic toxicity benchmark for chlorine, see the background
document, ``Human Health and Ecological Risk Assessment Support to
the Development of Technical Standards for Emissions from Combustion
Units Burning Hazardous Wastes: Background--Final Report,'' July,
1999.
---------------------------------------------------------------------------

B. What Are the Health Risks to Individuals Residing Near HWC
Facilities?
In this section, we address risks to populations that could be
enumerated using estimation methods based on U.S. Census data and
Census of Agriculture data. Estimates of the population of persons
residing within 20 kilometers of hazardous waste combustion facilities
were made for beef, dairy, produce, and pork farming households and for
non-farm households. The number of home gardeners was estimated using
national survey data on the portion of households that engage in home
gardening. Estimates were made for each of four different age groups.
In addition, population estimates were made for recreational anglers
age 16 and older based on U.S. Fish and Wildlife Service survey data on
recreational fishing and hunting.³³⁶
---------------------------------------------------------------------------

\336\ However, it was not possible to determine the number of
recreational anglers that fish specifically at water bodies located
in the vicinity of hazardous waste combustion facilities, such as
those that were selected for modeling analyses.
---------------------------------------------------------------------------

The risks to individuals of carcinogenic effects are expressed as
the estimated increase in the probability that an individual will
develop cancer over a lifetime. For non-cancer effects, risks are
expressed as a hazard quotient, which is the ratio of an estimate of an
individual's exposure to a health benchmark thought to be without
appreciable risk. Both cancer and non-cancer risks are summarized in
terms of percentiles of the national distribution of risks to
individuals across a combustor category. High end risks are represented
by the 90th to 99th percentiles of the distribution. Distributions for
only the most highly exposed receptor populations are discussed here.
The most highly exposed population varies depending on the particular
chemical constituent, its fate and transport in the environment, and
the pathways that lead to human exposures. Also, 90 percent confidence
limits are estimated for each percentile. The size of the confidence
interval reflects sampling error which is introduced by not sampling
all the facilities in a given category of sources.\337\ In some
instances, estimates of the 90 percent confidence limits could not be
made either because there were too few data points or there was
insufficient spread in the data. For lightweight aggregate kilns, there
is no sampling error because the sample included all known

[[Page 53004]]

hazardous waste burning lightweight aggregate kilns.
---------------------------------------------------------------------------

\337\ A 90 percent confidence interval indicates that there is a
10 percent chance that the actual value could lie outside the
interval indicated, either higher or lower.
---------------------------------------------------------------------------

1. Dioxins
For dioxins, our analysis shows that the most exposed population is
children of dairy farmers who consume home-produced milk. High
exposures were estimated for this population due to the relatively high
consumption of milk by households that consume home-produced milk, the
relatively high intake of milk by children compared to other age
groups, and the tendency of chlorinated dioxins and furans to
bioaccumulate in milk fat. A distribution of cancer risks for dioxins
was generated which reflects variability in individual exposures due to
site-specific differences in dioxin emissions, location of exposure,
and other factors, as well as differences between individuals in
exposure factors such as the length of exposure and the amount of milk
consumed.
As a result of today's rule, we project that high end lifetime
excess cancer risks will be reduced in this population from 2 in
100,000 (99th percentile) for both lightweight aggregate kilns and
incinerators with waste heat recovery boilers to below one in one
million (99th percentile) for lightweight aggregate kilns and 1 in one
million (99th percentile, 90 percent upper confidence limit of 2 in one
million) for incinerators with waste heat recovery boilers. For cement
kilns, high end lifetime excess cancer risks are reduced only slightly,
from 7 in one million (99th percentile) to 5 in one million (99th
percentile). These reductions, which represent the reduction in the
increment of exposure that results from dioxin emissions from hazardous
waste combustors, are relatively small in relation to background
exposures to dioxins generally. Considering that the number of
individuals in the affected population is relatively small, only a few
individuals may benefit from such reductions.
We also project that the incremental margin of exposure relative to
background will be reduced in the same population from 0.2 (99th
percentile for lightweight aggregate kilns) and 0.3 (99th percentile
for incinerators with waste heat recovery boilers, 90 percent upper
confidence limit of 0.5) to below 0.1 across all categories of
combustors. Therefore, the risks associated with non-cancer effects
from hazardous waste combustors are an order of magnitude or more lower
than any (unknown and unquantifiable) risks that may be attributable to
background exposures.
Unlike the distribution of cancer risks, the distribution of the
margin of exposure reflects only site-to-site differences and does not
reflect differences between individuals in the amount of milk consumed.
Therefore, the exposures at the upper percentiles are likely to be
underestimated.³³⁸ Additional uncertainty is introduced
because background exposures to dioxins in children have not been well
characterized.
---------------------------------------------------------------------------

\338\ The precise extent of underestimation at the upper
percentiles associated with variability in milk consumption is
unknown but is expected to be a factor of two.
---------------------------------------------------------------------------

Other uncertainties include milk consumption rates and the
limitations of the data available to assess consumption of home-
produced milk. In addition, there are a variety of uncertainties
related to the fate and transport of dioxins in the environment,
including partitioning behavior into vapor and particle phases
following release to the atmosphere and subsequent deposition via
various wet and dry removal processes, uptake in plants such as forage
and silage used by dairy cows for grazing and feeding, and the factors
which affect the disposition of dioxins in dairy cattle and the extent
of bioaccumulation in cow's milk.
2. Mercury
For mercury, our analysis shows that the most exposed population is
recreational anglers and their families who consume recreationally-
caught freshwater fish. This is because methyl mercury is readily
formed in aquatic ecosystems and bioaccumulates in fish. Children have
the highest exposures due to their higher consumption of fish, relative
to body weight, compared to adults. Risks from exposures to methyl
mercury are expressed here in terms of a hazard quotient, which is
defined as the ratio of the modeled average daily dose to our reference
dose. Although the reference dose was developed to be protective of
exposures in utero, we applied the reference dose not just to maternal
exposures but also to non-maternal adult and childhood exposures based
on the presumption that the reference dose should be protective of
neurological and developmental effects in these populations as well.
A distribution of hazard quotients was generated that reflects
variability in individual exposures due to site-specific differences in
mercury emissions, location of water bodies, and other factors, as well
as differences between individuals in the amount of fish consumed.
Other factors, such as water body-specific differences in the extent of
methylation of inorganic mercury and the age and species of fish
consumed were not reflected in the risk distribution. However, it is
unclear what effect such factors would have on the distribution given
the high degree of variability that is attributable to the factors that
were considered in our analysis.
The results of our quantitative analysis for mercury are as
follows. For cement kilns, we project that high end hazard quotients in
adults will be reduced from a range of 0.09 to 0.4 (90th percentile,
upper confidence limit of 0.1, and 99th percentile, respectively) at
baseline to a range from 0.06 to 0.2 under today's rule (90th
percentile, upper confidence limit of 0.08, and 99th percentile,
respectively). In children, high end hazard quotients are projected to
be reduced from a range of 0.2 to 0.8 (90th percentile, upper
confidence limit of 0.3, and 99th percentile, respectively) at baseline
to a range of 0.2 to 0.6 under today's rule (90th percentile, upper
confidence limit of 0.2, and 99th percentile, respectively). For
lightweight aggregate kilns, high end hazard quotients in both adults
and children are below 0.1 at baseline and under today's rule. For
incinerators, high end hazard quotients are below 0.01 in adults and
below 0.1 in children at baseline and under today's rule. Taken
together, these results appear to suggest that risks from mercury
emissions (on an incremental basis) are likely to be small, although we
cannot be certain of this for the reasons discussed below.
The risk results for mercury are subject to a considerable degree
of uncertainty. In addition to the uncertainties discussed above in
``Overview of Methodology--Mercury'', there are other uncertainties
when assessing individual mercury risks to nonsubsistence populations.
In order to assess exposures to mercury emissions, we assumed that
recreational anglers fish only at the water bodies within a given study
area that were selected for modeling (and at no other water bodies) and
that the extent of fishing activity at a given water body is related to
the size of the water body.³³⁹ As a result, in those
situations where relatively low fish concentrations were modeled (and
particularly if the water body was relatively large), a large portion
of fish were assumed to have relatively low levels of mercury
contamination and, therefore, recreational anglers who consume
relatively large amounts of recreationally-caught fish were estimated
to have relatively low levels

[[Page 53005]]

of exposure. In reality, some portion of the fish consumed by
recreational anglers is likely to be contaminated with mercury at
levels typical of background conditions. The effect of such background
exposures is to increase actual exposures, except perhaps at the high
end of the exposure distribution.³⁴⁰
---------------------------------------------------------------------------

\339\ Ideally, detailed information on the fishing activities of
individual anglers, including the size of the catch taken from
individual locations, would be used to better assess exposures from
consumption of recreationally-caught fish.
\340\ We have previously estimated that median exposures to
methyl mercury in the general population from seafood consumption
are in the range of 0.01 to 0.03 g/kg BW/day (Mercury Study
Report to Congress, December 1997). These exposures correspond to
hazard quotients of 0.1 to 0.3, values which (except for cement
kilns) are higher than the 90th to 99th percentile hazard quotients
estimated here for incremental exposures among recreational anglers.
---------------------------------------------------------------------------

We believe that the uncertainties implicit in the quantitative
mercury analysis continue to be sufficiently great so as to limit its
ultimate use for decision-making. Therefore, we have used the
quantitative analysis to make qualitative judgments about the risks
from mercury but have not relied on the quantitative analysis (nor do
we believe it is appropriate) to draw quantitative conclusions about
the risks associated with the MACT standards.
3. Lead
For lead, children are the population of primary concern for
several reasons, including behavioral factors, absorption, and the
susceptibility of the nervous system during a child's development. We
have chosen to use blood lead level as the exposure metric, consistent
with the U.S. Centers for Disease Control criteria for initiating
intervention efforts. Lead exposures occur through a variety of
pathways, including inhalation, incidental ingestion of soil and
household dust, and dietary intake. Our analysis indicates that the
population having the highest exposures are children who consume home-
produced fruits and vegetables. However, children who do not consume
home-produced foods also have relatively high exposures due to
incidental ingestion of soil and household dust.
Blood lead distributions were generated that represent incremental
exposures to lead emissions from hazardous waste combustors. These
distributions reflect variability in individual exposures due to site-
specific differences in lead emissions, location of exposure, and other
factors, as well as differences between individual children in behavior
patterns, absorption, and other pharmacokinetic factors. The IEUBK
model that was used to estimate blood lead levels considers inter-
individual variability in behavior related to lead exposure, such as
mouthing activity. However, the model does not explicitly consider
variability for the specific dietary pathways assessed for children of
home gardeners, that is, consumption of home-produced fruits and
vegetables. Therefore, the blood lead distributions may not fully
reflect inter-individual variability that results from such individual
differences.
Modeled blood lead (PbB) levels can be compared with background
exposures in the same age group (children ages 0 to 5 years) in the
general population. The median blood lead level in children in the
general population is 2.7 micrograms per deciliter (g/dL), and
4.4 and 1.3 percent of children have blood lead levels that exceed 10
and 15 g/dL, the levels at which community wide prevention and
individual intervention efforts, respectively, are
recommended.³⁴¹ However, the percentages vary widely
depending on such factors as race, ethnicity, income, and age of the
housing units occupied. Children whose blood lead levels are already
elevated are the most susceptible to further increases in blood lead
levels.
---------------------------------------------------------------------------

\341\ Data from the Centers for Disease Control's National
Health and Nutrition Examination survey (NHANES III, phase 2)
conducted from October 1991 to September 1994.
---------------------------------------------------------------------------

As a result of today's rule, we project that high end (90th to 99th
percentile) incremental blood lead (PbB) levels in children will
decrease from 0.24 to 0.50 micrograms per deciliter to 0.02 to 0.03
g/dL for cement kilns. For incinerators, incremental PbB
levels are projected to decrease from 0.6 to 1.2 g/dL (90th to
99th percentile) to 0.02 to 0.03 g/dL. For lightweight
aggregate kilns, incremental PbB levels are projected to decrease from
0.02 to 0.03 g/dL (90th to 99th percentile) to less than 0.01
g/dL under the MACT standards. Although these reductions in
incremental exposures represent only a fraction of the PbB level of
concern (10 g/dL), they can be significant in children with
PbB levels that are already elevated from exposures to other sources of
lead. In addition, there is evidence that effects on the neurological
development of children may occur at blood lead levels so low as to be
essentially without a threshold. Under the MACT standards, blood lead
levels attributable to HWCs will be one percent or less of background
levels typical of children in the general population.
4. Other Metals
We assessed both direct and indirect human exposures to a dozen
different metals in addition to mercury. Exposures to non-mercury
metals are generally quite low. Under today's rule, we project that
lifetime excess cancer risks from exposures to carcinogenic metals
(i.e., arsenic) will be below 1 in 10 million for all source
categories. Hazard quotients for all source categories are projected to
be at or below 0.01 (99th percentile) for all non-mercury metals under
the MACT standards. These risks reflect variability in individual
exposures due to site-specific differences in emissions, location of
exposure, and other factors. However, the risks do not reflect
differences between individuals in exposure factors such as the length
of exposure and the amount of food ingested. Therefore, we may have
underestimated risks at the upper percentiles of the
distribution.³⁴² A full exposure factor variability analysis
was not carried out because the risks using mean exposure factors are
comparatively low. Risks from exposure to metals are also subject to
uncertainty related to modeling of fate and transport in the
environment such as deposition of airborne metals to soils, forage, and
silage and subsequent uptake in farm animals.
---------------------------------------------------------------------------

\342\ For dioxins, inclusion of exposure factor variability
increased the risk of cancer at the upper (90th to 99th) percentiles
by less than a factor of two to a factor of five. However, the
effect on the distribution of risks could differ for metals
depending on the health effect of concern (i.e., cancer versus non-
cancer), the pathway of exposure, and relative differences in the
site-to-site variability of emissions.
---------------------------------------------------------------------------

5. Inhalation Carcinogens
We also assessed the combined cancer risk associated with
inhalation exposures to all inhalation carcinogens, assuming additivity
of the risks from individual compounds. The populations that have the
highest inhalation exposures are adult farm or non-farm residents.
Adults have the longest exposure duration relative to other age groups
and adult farmers have less mobility and, therefore, longer durations
of exposure than non-farm residents. However, depending on the location
of farms and non-farm households, adult non-farm residents can have
lifetime average exposures that are as high as adult farm residents.
Under today's rule, we project that lifetime excess cancer risks
from inhalation exposures will be below 1 in 10 million for all source
categories. The risks for inhalation carcinogens reflect variability in
individual exposures due to site-specific differences in metals
emissions, location of exposure, and other factors. However, they do
not reflect differences between individuals

[[Page 53006]]

in the length of exposure or other exposure factors. Therefore, we may
have underestimated risks at the upper percentiles of the
distribution.³⁴³ A full exposure factor variability analysis
was not carried out for inhalation carcinogens because the risks using
mean exposure factors are comparatively low.
---------------------------------------------------------------------------

\343\ The precise extent of underestimation at the upper
percentiles associated with variability in the duration of exposure
is unknown but is expected to be a factor of three or less.
---------------------------------------------------------------------------

Estimates of inhalation risks are subject to a number of
uncertainties. Individuals spend a majority of their time indoors and
it is uncertain how representative modeled, outdoor, ambient air
concentrations are of concentrations indoors. Also, the daily
activities of individuals living in the vicinity of a given facility
will tend to moderate actual exposures compared to modeled exposures at
a fixed location. Meteorological information was generally obtained
from locations well removed from modeled facilities and, therefore, may
not be representative of conditions in the immediate vicinity of the
stack. Limited information was available on the size of structures
located near or adjacent to stacks at the modeled facilities. Building
downwash, that can result from the presence of such structures, may
significantly increase ground-level ambient air concentrations,
particularly at locations that are relatively close to the point of
release. In addition, the effect of elevated terrain was only
considered when the terrain rose above the height of the stack.
However, elevated terrain below stack height can lead to an increase in
ground-level concentrations depending on the distance from the stack.
Nevertheless, our projections of inhalation cancer risks are
sufficiently low that we do not believe the uncertainties introduced by
these factors impacts our conclusion that these risks are relatively
low.
6. Other Inhalation Exposures
Of the compounds we evaluated that are not carcinogenic, the
highest inhalation exposures are for hydrogen chloride and chlorine
gas. We express the risks from these in terms of an
inhalation hazard quotient, which is defined as the ratio of the
modeled air concentration to our reference concentration. The receptor
population with the highest inhalation hazard quotients is variable and
depends on site-to-site differences in the location of farm and non-
farm households and differences in emissions. A distribution of hazard
quotients was generated that reflects variability in individual
exposures due to site-specific differences in chlorine emissions,
location of exposure, and other factors. However, the distribution does
not reflect individual differences in activity patterns or breathing
rates.³⁴⁴ Also, because the reference concentration is
intended to be protective of long-term, chronic exposures over a
lifetime, the distribution does not reflect temporal variations in
exposure.³⁴⁵
---------------------------------------------------------------------------

\344\ Differences in breathing rates are not considered because
the exposure factors used in deriving the reference concentration
are fixed.
\345\ Although short-term exposures to hydrogen chloride and
chlorine gas resulting from routine releases can be significantly
higher than long-term exposures, we do not believe that such
exposures are high enough to pose a health concern because the
threshold for acute effects is quite high in comparison to that for
chronic effects.
---------------------------------------------------------------------------

Under today's rule, we project that inhalation hazard quotients
will be at or below 0.01 for both hydrogen chloride and chlorine gas
for all source categories. The same uncertainties related to indoor
versus outdoor concentrations and atmospheric dispersion modeling are
also applicable to hydrogen chloride and chlorine. However, our
projections of non-cancer inhalation risks are sufficiently low that we
do not believe the uncertainties impact our conclusion that these risks
are relatively low.
C. What Are the Potential Health Risks to Highly Exposed Individuals?
We also assessed exposures to individuals that could be more highly
exposed than the populations that could be characterized using census
data. These include persons engaged in subsistence activities such as
farming and fishing. Although the frequency of these activities is
unknown, such activities do occur in some segments of the U.S.
population, and we believe that it is important to evaluate risks
associated with such activities. In addition, risks associated with
subsistence farming place a bound on potential risks to farmers who
raise more than one type of livestock. Information on the numbers of
farms that produce more than one food commodity (e.g., beef and milk)
is not available from the U.S. Census of Agriculture. Therefore, in
assessing risks to farm populations, we may have underestimated the
risks to farmers and their families that consume more than one type of
home-produced food commodity.
We assumed that subsistence farmers obtain substantially all of
their dietary intake from home-produced foods, including meats, milk,
poultry, fish, and fruits and vegetables. We used data on the mean rate
of consumption of home-produced foods in households that consume home-
produced foods to estimate the average daily intakes from subsistence
farming. For subsistence fishing, we used data on the mean rate of fish
consumption among Native American tribes that rely on fish for a major
part of their dietary intake.
We do not have specific information on the existence or location of
subsistence farms or water bodies used for subsistence fishing at sites
where hazardous waste combustors are located. Therefore, we
hypothetically assumed that subsistence farming does occur at each of
the modeled facilities and, furthermore, that it occurs within each of
the sixteen sectors within a study area. We also assumed that
subsistence fishing takes places at each of the modeled water bodies.
The results of the analysis are summarized in the form of frequency
distributions of individual risk. The distributions must be interpreted
in relation to the frequency of the modeled scenarios and not the
likelihood of such exposures actually occurring.³⁴⁶
---------------------------------------------------------------------------

\346\ Moreover, the modeled scenarios cannot be considered
equally probable because the sectors in which farms were located are
of unequal area, being much smaller closer to a facility and much
larger farther away and because any particular sector may be more or
less likely to support farming activities depending on soils,
precipitation, existing land uses, and other conditions. Similarly,
the modeled water bodies may be more or less likely to support
intensive fishing activity depending on their size, productivity,
and other characteristics.
---------------------------------------------------------------------------

The risk results for subsistence receptors are highly uncertain,
primarily due to the lack of information on the location of subsistence
farms (or even the occurrence of subsistence farms within the study
area of a given facility) and the assumption that individuals engaged
in subsistence farming obtain essentially their entire dietary intake
from home-produced foods.
1. Dioxins
Under today's rule, we project that lifetime excess cancer risks
from dioxin exposures associated with subsistence farming will be below
1 in 100,000 for all categories of combustors, with the exception of
cement kilns at the lowest frequency of occurrence. The lifetime excess
cancer risk for cement kilns is estimated to be 2 in 100,000 at a
frequency of 1 percent. This indicates that only 1 in 100 sectors are
expected to have risks of this magnitude or greater, assuming that
subsistence farms are located in all sectors at all hazardous waste
burning cement kilns. However, because the sectors increase in size
with increasing distance, the probability that a subsistence farm would
be exposed to

[[Page 53007]]

this level of risk is probably considerably less than 1 percent.
We project that the incremental margin of exposure relative to
background will be reduced to 0.1 or below for incinerators under
today's rule except at the lowest frequency of occurrence (i.e., 1
percent) for which a margin of exposure of 0.2 is projected. However,
the incremental margins of exposure for cement kilns and lightweight
aggregate kilns are projected to remain above 0.1 at a frequency of 10
percent or greater (ranging up to 0.2 at a frequency of 5 percent for
lightweight aggregate kilns and 0.7 at a frequency of 1 percent for
cement kilns). This indicates that more than 1 in 10 sectors are
expected to have risks associated with non-cancer effects that are
within an order of magnitude of any (unknown and unquantifiable) risks
that may be attributable to background exposures. However, for the
reasons stated previously, the probability that a subsistence farm
would be exposed to this level of risk is probably considerably lower
than indicated by the number of sectors.
Under today's rule, we project lifetime excess cancer risks from
dioxin exposures associated with subsistence fishing will be below 1 in
one million for incinerators and lightweight aggregate kilns. For
cement kilns, high end cancer risks under today's rule range from 3 in
one million to 4 in one million (at frequencies of 10 and 5 percent,
respectively) in adults and from 2 in one million to 4 in one million
(at frequencies of 10 and 5 percent, respectively) in children (6 to 11
years of age). We project that the incremental margin of exposure
relative to background will be below 0.1 for subsistence fishing for
both children and adults for all categories of combustors under today's
rule.
2. Metals
Our analysis indicates that the highest risks from metals (other
than mercury) are from arsenic, thallium, and lead. Under today's rule,
we project that lifetime excess cancer risks from arsenic exposures
associated with subsistence farming will be below 1 in one million for
all source categories. Hazard quotients for thallium are projected to
be at or below 0.01 (99th percentile) under today's rule, except for
cement kilns. For cement kilns, hazard quotients for thallium are
projected to range from 0.03 to 0.4 (90th to 99th percentiles).
Incremental blood lead levels are projected to be at or below 0.03
g/dL for all source categories under today's rule. Blood lead
at these levels are about one percent of background levels typical of
children in the general population.
3. Mercury
From the results of our quantitative analysis we project that,
under today's rule, hazard quotients for incremental exposures to
mercury associated with subsistence fishing will be at or below 1 in
both adults and children. These results apply to incinerators,
lightweight aggregate kilns, and cement kilns at the very lowest
frequency of occurrence that was analyzed (i.e., 1 percent).
The risk results for mercury are subject to a considerable degree
of uncertainty. In addition to the uncertainties discussed above in
``Overview of Methodology--Mercury'', there are other uncertainties
when assessing individual mercury risks to subsistence receptors. We
assumed that individuals engaged in subsistence fishing obtain all the
fish they consume from a single water body. To the extent that
individuals may fish at more than one water body, the effect of this
assumption may be to exaggerate the risk from water bodies having
relatively high modeled fish concentrations.
The uncertainties implicit in the quantitative mercury analysis
continue to be sufficiently great so as to limit its ultimate use for
decision-making. Therefore, we have used the quantitative analysis to
make qualitative judgments about the risks from mercury but have not
relied on the quantitative analysis (nor do we believe it is
appropriate) to draw quantitative conclusions about the risks
associated with the MACT standards.
D. What Is the Incidence of Adverse Health Effects in the Population?
We estimated the overall risk to human receptor populations for
those chemical constituents that posed the highest individual risks and
whose populations could be enumerated. These included excess cancer
incidence in the general population from the consumption of
agricultural commodities produced in the vicinity of hazardous waste
burning facilities, excess cancer incidence in the local population,
and excess incidence of children with elevated blood lead levels. In
addition, we estimated the avoided incidence of mortality and morbidity
in the local population associated with reductions in exposures to
particulate matter emissions.³⁴⁷ Incidence is generally
expressed in terms of the annual number of new cases of disease in the
exposed population. However, for diseases such as cancer which have a
long latency period, the annual incidence represents the lifetime
incidence associated with an exposure of one year. For diseases with
recurring symptoms, the annual incidence represents the number of
episodes of disease over a year's time.
---------------------------------------------------------------------------

\347\ Excess incidence refers to the incidence of disease beyond
that which would otherwise be observed in the population, absent
exposures to the sources in question. Avoided incidence is the
reduction in incidence of disease in the population that would be
expected from a reduction in exposures to the sources in question.
---------------------------------------------------------------------------

1. Cancer Risk in the General Population
Agricultural commodities produced in the vicinity of hazardous
waste combustors may be consumed by the general population (i.e.,
individuals who reside outside the study area). Commodities such as
meat and milk may be contaminated with dioxins and, therefore, pose
some risk to individuals that consume them. We estimated the amount of
``diet accessible'' dioxin in meat and milk produced at hazardous waste
combustors that would be consumed by the general population and
estimated the number of additional cancer cases that could result from
such exposures. The approach is predicated on the assumption that
cancer risks follow a linear, no-threshold model in the low dose
region.
Our agricultural commodity analysis indicates that, as a result of
today's rule, annual excess cancer incidence in the general population
will be reduced from 0.5 cases per year (90 percent confidence
interval, 0.4 to 0.6) to 0.1 cases per year (90 percent confidence
interval, 0.1 to 0.2). Most of the risk is associated with the
consumption of milk and other dairy products. The combustor categories
that contribute most to the reduction are incinerators with waste heat
recovery boilers and lightweight aggregate kilns.
2. Cancer Risk in the Local Population
Individuals that live and work in the vicinity of hazardous waste
combustors are exposed to a number of compounds that are carcinogenic
by oral or inhalation routes of exposure or both. These include dioxin,
arsenic, beryllium, cadmium, chromium, and nickel. We estimated the
annual cancer incidence in each of the enumerated receptor populations
based on the mean individual risk in each sector and sector-specific
population estimates. The resulting incidence estimates were weighted
using facility-specific sampling weights and summed.
Our analysis of cancer risks in the local population indicates
that, as a result of today's rule, annual excess

[[Page 53008]]

cancer incidence will be reduced from 0.1 cases per year (90 percent
confidence interval, 0.08 to 0.2) to 0.02 cases per year (90 percent
confidence interval, 0.01 to 0.03). Nearly all of the risk reduction,
which occurs almost entirely among non-farm residents, is attributable
to incinerators and results mainly from reductions in emissions of
metals, primarily arsenic, cadmium, and chromium.
3. Risks From Lead Emissions
Children that live near hazardous waste combustor are exposed to
lead emissions through the diet and through inhalation and incidental
soil ingestion. Children that already have elevated blood lead levels
may have their levels further increased as a result of such exposures,
some of whom may have their blood lead levels increased beyond 10
g/dL. We estimated the increase, or excess incidence, of
elevated blood levels above 10 g/dL by estimating the number
of children in each sector with blood lead levels above 10 g/
dL as a result of background exposure and subtracting that from the
number of children above 10 g/dL as a result of both
background exposure and incremental exposures from hazardous waste
combustors. This estimate represents the annual rate of increase in the
number of children with elevated blood lead beyond background.
Our analysis indicates that, as a result of today's rule, the
excess incidence of elevated blood lead will be reduced from 7 cases
per year to less than 0.1 cases per year. The reduction is primarily
attributable to incinerators, although a small reduction (0.4 cases per
year) is attributable to cement kilns. These reductions occur entirely
among non-farm residents. Children of minority and low income
households generally have higher background exposures to lead and are
more likely to have blood levels elevated above 10 g/dL than
children from other demographic groups and, therefore, are more likely
to benefit from reductions in lead exposures. However, our analysis did
not consider the influence of such socioeconomic factors. For this
reason, we believe that we may have underestimated the reductions in
excess incidence of elevated blood lead levels, including potential
reductions attributable to cement kilns and lightweight aggregate
kilns.
4. Risks From Emissions of Particulate Matter
Human epidemiological studies have demonstrated a correlation
between community morbidity and mortality and ambient levels of
particulate matter, particularly fine particulate matter (below 2.5 or
10 microns in diameter, depending on the study), across a wide variety
of geographic settings. Lower particulate matter is associated with
lower mortality, lower rates of hospital admissions, and a lower
incidence of respiratory disease. Concentration-response functions for
various health endpoints have been derived from these studies, and we
used these functions to estimate the reduction in the incidence of
mortality and morbidity associated with a reduction in emissions of
particulate matter.
Our analysis indicates that, as a result of today's rule, there
will be between 1 and 4 fewer premature mortalities per year associated
with particulate matter emissions (depending on which study is used).
In addition, we project there will be 6 fewer hospitalizations, 25
fewer cases of chronic bronchitis, 180 fewer cases of lower respiratory
disease, per year.
The mortality estimates are subject to some uncertainty due to the
fact that the lower estimate (which is derived from long-term studies)
assumes no threshold for effects and the upper estimate (which is
derived from short-term studies) may include mortalities that are
premature by as little as a few days. The no threshold assumption may
be appropriate, however, considering that the reduction in mortality is
projected to occur entirely from incinerators, especially on-site
incinerators. Such incinerators are located at manufacturing facilities
that are likely to have other particulate matter emissions and both on-
site, and commercial incinerators are typically located in industrial
areas where there may be many other sources of particulate matter
emissions, resulting in ambient particulate matter levels that are well
above any threshold. Also, because the particulate matter modeling was
conducted to 20 rather than 50 kilometers, the inhalation risks may be
understated, especially from PM that is 2.5 microns in diameter and
smaller which can be transported over long distances from HWCs.

III. What Is the Potential for Adverse Ecological Effects?

The ecological assessment is based on a screening level analysis in
which model-estimated media concentrations are compared to media-
specific ecotoxicological criteria that are protective of multiple
ecological receptors. The analysis used an ecological hazard quotient
as the metric for assessing ecological risk. The ecological hazard
quotient is the ratio of the model-estimated media concentration to the
ecotoxicological criterion. Hazard quotients above 1 suggest that a
potential for adverse ecological effects may exist. Ecotoxicological
criteria for soils, surface waters, and sediments were used in the
analysis. Ecotoxicological criteria for soil are intended to be broadly
protective of terrestrial ecosystems, including the soil community,
terrestrial plants, and consumers such as mammals and birds.
Ecotoxicological criteria for surface water are intended to be
protective of the aquatic community, including fish and aquatic
invertebrates, primary producers such as algae and aquatic plants, and
fish-eating mammals and birds. Sediment criteria are intended to be
protective of the benthic community. The analysis was conducted for
dioxins, mercury, and fourteen other metals. Only the results for
dioxins and mercury are discussed here. Very low or no potential for
ecological risk was found for the other metals.³⁴⁸ For a
full discussion of the ecological assessment, see the background
document for today's rule.³⁴⁹
---------------------------------------------------------------------------

\348\ Although minor exceedances of the ecotoxicological
criteria for lead were noted for incinerators, the exceedances were
eliminated under today's rule.
\349\ ``Human Health and Ecological Risk Assessment Support to
the Development of Technical Standards for Emissions from Combustion
Units Burning Hazardous Wastes: Background Document--Final Report,''
July, 1999.
---------------------------------------------------------------------------

A. Dioxins
A variation on the general screening level approach was used for
assessing ecological risks from dioxins in surface water. Rather than
basing the assessment on ambient water quality criteria for the
protection of wildlife, ecotoxicological benchmarks for 2,3,7,8-
tetrachlorodibenzo(p)dioxin (TCDD) for fish-eating birds and mammals
(i.e., no observed adverse effects levels) were used to make a direct
comparison with estimated intakes of dioxins in fish in terms of
2,3,7,8-TCDD toxicity equivalents (TEQ). This approach accounts for the
different rates of bioaccumulation of the various 2,3,7,8
dibenzo(p)dioxin and dibenzofuran congeners and avoids the conservatism
of comparing an ambient water quality criterion for 2,3,7,8-TCDD to
model-estimated water concentrations in terms of 2,3,7,8-TCDD TEQs. The
results of our analysis indicate no exceedances of the ecotoxicological
benchmarks for 2,3,7,8-TCDD for any category of hazardous waste
combustors. One limitation of the ecological assessment for dioxins is
that water quality criteria for the protection of aquatic life are not

[[Page 53009]]

available. However, fish and aquatic invertebrates are generally less
sensitive to dioxins than mammals and birds.
For assessing the potential for ecological risk in terrestrial
ecosystems, soil criteria developed for 2,3,7,8-TCDD for the protection
of mammals and birds were compared to model-estimated soil
concentrations in terms of 2,3,7,8-TCDD TEQs. Because the more highly
chlorinated 2,3,7,8 dibenzo(p)dioxin and dibenzofuran congeners are
expected to bioaccumulate in prey species more slowly than 2,3,7,8-
TCDD, the potential for ecological risk is likely to be overstated. Our
analysis indicates that, at baseline, less than one percent of the
study areas surrounding hazardous waste combustors have the potential
for ecological risk from dioxins in soil. Under today's rule, we
project no exceedances of the ecotoxicological criteria for dioxins in
soil. The soil ecotoxicological criterion for dioxins is derived from
studies of reproductive and developmental effects in mammals. Potential
impacts to terrestrial plant and soil communities could not be
evaluated due to a lack of sufficient ecological toxicity data.
However, vertebrates such as mammals and birds are known to be more
sensitive to dioxin exposure than invertebrates. Therefore, we consider
the potential for risk to invertebrate receptors to be low.
B. Mercury
The ecological assessment of mercury is based on water quality
criteria for the protection of wildlife that were developed for the
Mercury Study Report to Congress. The assessment used the lowest of the
available water quality criteria for individual fish-eating avian and
mammalian wildlife species. The frequency distribution of ecological
hazard quotients for total mercury indicates the potential for adverse
ecological effects for cement kilns. Our analysis indicates that, for
cement kilns, exceedances of the ecotoxicological criteria for total
mercury may occur over 40 percent of study area surface waters at
baseline. Above a hazard quotient of 1, the frequency of exceedances
drops off quickly, with hazard quotients above 2 occurring at a
frequency of 1 percent. The ecological hazard quotients remain
essentially unchanged under today's rule. However, we project no
exceedances of the ecotoxicological criteria for methyl mercury.
Because methyl mercury is the form of mercury that is of greatest
concern for fish-eating mammals and birds, the lack of exceedances
suggests that the potential for ecological effects is relatively low.
Our analysis also suggests relatively low potential for ecological
effects for incinerators and lightweight aggregate kilns. Although our
analysis indicates that exceedances of the ecotoxicological criteria
for total mercury may occur over 22 percent of study area surface
waters for lightweight aggregate kilns and 6 percent for incinerators
at baseline, these are reduced to no exceedances and less than 1
percent, respectively, under today's rule. Moreover, we project no
exceedances of the ecotoxicological criteria for methyl mercury. The
significance of these results must be judged in the context of the
considerable uncertainties related to the fate and transport of mercury
in the environment, as discussed elsewhere in today's notice, the
presence of background levels of mercury, and the level of protection
afforded by the underlying ecotoxicological criteria.
For soils, our analysis indicates that less than one percent of the
study areas surrounding hazardous waste combustors have the potential
for ecological risk at baseline. Under today's rule, we project no
exceedances of the ecotoxicological criteria for mercury for
incinerators and lightweight aggregate kilns. For cement kilns, we
project exceedances at a frequency of much less than one percent. The
soil ecotoxicological criterion for mercury is derived from studies of
the reproductive capacity of earthworms. Although earthworms serve a
vital function in the soil community, given the redundancy and
abundance of soil organisms and the low frequency of exceedances, we
believe that adverse impacts to the terrestrial ecosystem, including
higher trophic levels such as terrestrial mammals, are unlikely.
As a screening level analysis, the ecological assessment is subject
to a number of limitations. The analysis assumes the occurrence of the
ecological receptors on which the ecotoxicological criteria are based
in all modeled sectors and water bodies. Although the ecological
receptors included in the analysis are commonly occurring species, they
may not be present in the same locations at which exceedances are
predicted due to a lack of suitable habitat or other factors.
Furthermore, the range of predator and prey species may exceed the
spatial extent of the estimated exceedances. Many primary and secondary
consumers are opportunistic feeders with substantial variability in
both the type of food items consumed as well as the seasonal patterns
of feeding and foraging. These behaviors can be expected to moderate
exposures to chemical contaminants and reduce the potential for risk.
On the other hand, gaps exist in the ecotoxicological data base such
that not all combinations of chemical constituents and ecological
receptors could be evaluated. In addition, media concentrations could
not be estimated for all habitats that may be important to ecological
receptors, such as wetlands. Also, our analysis did not consider the
possible impact of background concentrations. Therefore, although as a
screening level analysis the ecological assessment has a tendency
toward conservatism, we cannot say for certain that no potential exists
for ecological risks that fall beyond the scope of the assessment.

Part Eight: Analytical and Regulatory Requirements

I. Executive Order 12866: Regulatory Planning and Review (58 FR 51735)

Is This a Significant Regulatory Action?
Under Executive Order 12866 (58 FR 51735, October 4, 1993), we must
determine whether a regulatory action is ``significant'' and,
therefore, subject to OMB review and the requirements of the Executive
Order. The Order defines ``significant regulatory action'' as one that
is likely to result in a rule that may:
(1) Have an annual effect on the economy of $100 million or more,
adversely affect in a material way the economy, a sector of the
economy, productivity, competition, jobs, the environment, public
health or safety, or state, local, or tribal governments or
communities;
(2) Create a serious inconsistency or otherwise interfere with an
action taken or planned by another agency;
(3) Materially alter the budgetary impact of entitlement, grants,
user fees, or loan programs or the rights and obligations of recipients
thereof; or
(4) Raise novel legal or policy issues arising out of legal
mandates, the President's priorities, or the principles set forth in
this Executive Order.
Under the terms of Executive Order 12866, we have reviewed today's
rule and determined that it does not represent an ``economically
significant'' regulatory action, as defined under point one above. The
aggregate annualized social costs for this rule are under $100 million
(ranging from $50 to $63 million for the final standards). However, it
has been determined that this rule is a ``significant regulatory
action'' because it may raise novel legal or policy issues (point four
above). As such, this action was submitted to OMB for review. Changes
made in response to OMB suggestions or recommendations will be
documented in the public record.

[[Page 53010]]

We have prepared economic support materials for today's final
action. These documents are entitled: Assessment of the Potential
Costs, Benefits, and Other Impacts of the Hazardous Waste Combustion
MACT Standards--Final Rule, and, Addendum To The Assessment of the
Potential Costs, Benefits, and Other Impacts of the Hazardous Waste
Combustion MACT Standards--Final Rule. The Addendum and Assessment
documents were designed to adhere to analytical requirements
established under the Executive Order, and corresponding Agency and OMB
guidance; subject to data, analytical, and resource limitations.
This part of the Preamble is organized as follows: I. Executive
Order 12866 (as addressed above), II. What Activities have Led to
Today's Rule?--presenting a summary of the analytical methodology and
findings from the 1996 RIA for the proposed action, and, a summary of
substantive peer review and public stakeholder comments on this
document, with Agency responses, III. Why is Today's Rule Needed?--
justifying the need for Federal intervention, IV. What Were The
Regulatory Options?--presenting a brief discussion of the scope of
alternative regulatory options examined, V. What Are the Potential
Costs and Benefits of Today's Rule?--summarizing methodology and
findings from the final Assessment document, VI. What Considerations
Were Given to Issues Like Equity and Children's Health?, VII. Is
Today's Rule Cost-Effective?, VIII. How Do the Costs of Today's Rule
Compare to the Benefits?, IX. What Consideration Was Given to Small
Businesses? X. Were Derived Air Quality and Non-Air Impacts Considered?
XI. Is Today's Rule Subject to Congressional Review?, XII. How is the
Paperwork Reduction Act Considered in Today's Rule?, XIII. Was the
National Technology Transfer and Advancement Act Considered?, and, XIV.
Were Tribal Government Issues Considered? (Executive Order 13084).
The RCRA docket established for today's final rulemaking maintains
a copy of the complete final Assessment and Addendum documents for
public review. Readers interested in these economic support materials
are strongly encouraged to read both documents to ensure full
understanding of the methodology, data, findings, and limitations of
the analysis.

II. What Activities Have Led to Today's Rule?

In May of 1993, we introduced a draft Waste Minimization and
Combustion Strategy designed to reduce reliance on the combustion of
hazardous waste and encourage reduced generation of these wastes. Among
the key objectives of the strategy was the reduction of health and
ecological risks posed by the combustion of hazardous wastes. As part
of this strategy, we initiated the development of MACT emissions
standards for hazardous waste combustion facilities.
On April 19, 1996, we published the proposal, which included
revisions to standards for hazardous waste incinerators and hazardous
waste burning cement kilns and lightweight aggregate kilns. These
proposed MACT standards were designed to address a variety of hazardous
air pollutants, including dioxins/furans, mercury, semivolatile and low
volatile metals, and chlorine. We also proposed to use emissions of
carbon monoxide and hydrocarbons as surrogates for products of
incomplete combustion.
A. What Analyses Were Completed for the Proposal?
We completed an economic analysis in support of the proposal. This
Regulatory Impact Assessment (RIA), examined and compared the costs and
benefits of the proposed standards, as required under Executive Order
12866. Industry economic impacts, environmental justice, waste
minimization incentives, and other impacts were also examined. This RIA
also fulfilled the requirements of the Regulatory Flexibility Act by
evaluating the effects of regulations on small entities. This document,
Regulatory Impact Assessment for Proposed Hazardous Waste Combustion
MACT Standards (November 13, 1995), Appendices (November 13, 1995), and
two Addenda (November 13, 1995 and February 12, 1996) are available in
the docket established for the proposed action.
Throughout the development of the proposal, we considered many
alternative regulatory options. A full discussion of the methodology
and findings of all options considered is in the Regulatory Impact
Assessment (RIA). Only the floor option and our preferred option (i.e.,
the floor option and beyond-the-floor options for selected hazardous
air pollutants) are discussed in this summary.
1. Costs
To develop industry compliance cost estimates, we categorized or
modeled combustion units based on source category and size and
estimated engineering costs for the air pollution control devices
needed to achieve the proposed standards. Based on current emissions
and air pollution control device information, we developed assumptions
regarding the type of upgrades that units would require. This ``model
plants'' engineering cost analysis was used because our data were
limited.
Total annual compliance cost estimates for the floor option and the
beyond-the-floor standards ranged from $93 million to $136 million,
respectively, beyond the baseline. For the floor option, on-site
incinerators represented 55 percent of total nationwide costs, cement
kilns represented 29 percent, commercial incinerators represented 14
percent, and lightweight aggregate kilns represented 2 percent. Of the
total beyond-the-floor costs, on-site incinerators represented 50
percent, cement kilns represented 32 percent, commercial incinerators
represented 15 percent, and lightweight aggregate kilns represented 3
percent. For the incremental impacts of going from the floor to beyond-
the-floor, lightweight aggregate kilns were projected to experience a
100 percent increase in compliance costs, cement kilns would experience
a 63 percent increase, commercial incinerators and on'site
incinerators, at 54 and 34 percent, respectively. Overall, compliance
costs associated with the proposed action were projected to result in
significant economic impacts to the combustion industry.
The RIA also examined average total annual compliance costs per
combustion unit. This indicator was designed to assess the relative
impact of the rule on each facility type in the combustion universe.
Findings projected that cement kilns were likely to incur the greatest
average incremental cost per unit, totaling $770,000 annually at the
floor and $1.1 million annually for the proposed beyond-the-floor
standards. The costs for LWAKs ranged from $490,000 to $825,000. The
costs for on-site incinerators ranged from $340,000 to $486,000. The
costs for commercial incinerators ranged from $493,000 to $730,000.
These costs assume no market exits. Once market exit occurs, average
per unit costs may be significantly lower, particularly for on-site
incinerators.
The analysis also examined the floor and proposed beyond-the-floor
impacts on a per ton basis. In the baseline, average prices charged to
burn hazardous waste were estimated to be $178 per ton for cement
kilns, $188 per ton for lightweight aggregate kilns, $646 per ton for
commercial incinerators, and $580 per ton for on-site incinerators
(approximate internal transfer price).

[[Page 53011]]

Baseline burn costs (before consolidation) for these facilities were
found to average $104 per ton for cement kilns, $194 per ton for
lightweight aggregate kilns, $806 per ton for commercial incinerators,
and $28,460 per ton for on-site incinerators. ³⁵⁰
Incremental compliance costs at the floor and proposed BTF levels were
estimated to be $23 to $31 per ton for commercial incinerators, $40 to
$50 per ton for cement kilns, $39 to $56 per ton for lightweight
aggregate kilns, and $47 to $57 per ton for on-site incinerators.
---------------------------------------------------------------------------

\350\ Baseline costs were calculated by identifying all costs of
hazardous waste burning. For commercial incinerators and on-site
incinerators, all costs of construction, operation and maintenance
are included. This also includes RCRA permits and existing air
pollution control devices. The costs for on-site burners are
extremely high because the costs are distributed across the small
amount of hazardous waste burned. For cement kilns and lightweight
aggregate kilns, only the incremental costs of with burning
hazardous waste are included (e.g., permits). The cost of the actual
units (which are primarily for producing cement or aggregate) are
not included in the baseline.
---------------------------------------------------------------------------

From comparison of these prices and baseline burn costs, some high-
cost facilities, especially commercial and on-site incinerators,
appeared to be burning below break-even levels. The incremental
compliance costs of the proposal would make these facilities even less
competitive. The RIA estimated that, of the facilities which are
currently burning hazardous waste, three cement kilns, two lightweight
aggregate kilns, six commercial incinerators, and eighty-two on-site
incinerators would likely stop burning hazardous waste over the long
term. These were incremental to projected baseline market exits
estimated at the time of proposal. Most of the facilities that exit the
market were ones that burned smaller amounts of hazardous waste.
We also conducted a generalized cost effectiveness analysis for the
proposal. We found that the cost per hazardous air pollutant is often
difficult to estimate because the air pollution control devices often
control more than one pollutant. Therefore, it was not feasible to
estimate precise costs per pollutant. Once the compliance expenditures
had been estimated, the total mass emission reduction achieved when
facilities comply with the standards option was estimated. The total
incremental cost per incremental reduction in pollutant emissions was
then estimated. Considering all facilities together, dioxin, mercury,
and metals costs per unit reduction are quite high because small
amounts of the dioxin and metals are released into the environment. For
other pollutants, expenditures per ton are much lower. Please refer to
the November 13, 1995 draft RIA for a complete discussion of the
methodology and findings.
2. Benefits
Our evaluation showed that background levels of dioxin in beef,
milk, pork, chicken, and eggs were approximately 0.50, 0.07, 0.30,
0.20, and 0.10 parts per trillion fresh weight, respectively, on a
toxicity equivalent (TEQ) basis. These background levels and
information on food consumption were then used to estimate dietary
intake in the general population. That estimate was 120 picograms TEQ
per day. We also collected background data on dioxins in fish, taken
from 388 locations nationwide. At 89 percent of the locations, fish
contained detectable levels of at least two of the dioxin and furan
compounds for which analyses were conducted. We then estimated total
dioxin emissions from hazardous waste combustors at 0.94 kg TEQ per
year. This represented about 9 percent of total anthropogenic emissions
of dioxins in the U.S. at the time. The dioxin estimates have been
revised since then.
While no one-to-one relationship between emissions and risk exists,
it was inferred that hazardous waste-burning sources were likely to
contribute significantly to dioxin levels in foods. In the proposal, we
estimated that these dioxin emissions would be reduced to 0.07 kg TEQ
per year at the floor levels and to 0.01 kg TEQ per year at the beyond
the floor levels. We estimated this to result in decreases of
approximately 8 and 9 percent in total estimated anthropogenic U.S.
emissions, respectively. Our position at proposal was that reductions
in these emissions, in conjunction with reductions from other dioxin-
emitting sources, would help reduce dioxin levels in foods over time
and, therefore, reduce the likelihood of adverse health effects,
including cancer.
Mercury is a concern in both occupational and environmental
settings. Human exposures to methyl mercury occur primarily from
ingestion of fish. Mercury contamination results in routine fish
consumption bans or advisories in over two thirds of the States. At the
proposal, we estimated a safe exposure level to methyl mercury (the
reference dose) at 0.0001 mg per kg per day. We collected data on
chemical residues in fish from 388 locations nationwide and found that
fish contained detectable levels of mercury at 92 percent of the
locations. Similar results have been obtained in other studies,
strongly suggesting that long-range atmospheric transport and
deposition of anthropogenic emissions is occurring. Our research found
that, for persons who eat significant amounts of freshwater fish,
exposures to mercury may be significant compared to the threshold at
which effects may occur in susceptible individuals.
Our estimates for the proposal indicated that hazardous waste
combustors emitted a total of 10.1 Mg of mercury per year, representing
about 4 percent of the U.S. anthropogenic total. Implementation of the
floor levels were estimated to reduce mercury emissions from all
hazardous waste-burning sources to 3.3 Mg per year. The proposed
beyond-the-floor levels would drop this to an estimated 2.0 Mg per
year. Such reductions were estimated to lower total anthropogenic U.S.
emissions by approximately 3 percent. Reductions in these mercury
emissions, in conjunction with the Agency's efforts to reduce emissions
from other mercury-emitting sources, would help diminish mercury levels
in fish over time and, therefore, reduce the likelihood of adverse
health effects occurring in fish-consuming populations.
Other benefits we investigated for the proposal included ecological
benefits, property value benefits, soiling and material damage,
aesthetic damages, and recreational and commercial fishing impacts.
Overall, the analysis of the ecological risk suggested that water
quality criteria may be exceeded only in small watersheds located near
waste combustion facilities. Furthermore, such exceedances would occur
only when assuming very high emissions. The preliminary analysis for
the proposal indicated that property value impacts may be very
significant because of emission reductions from hazardous waste
combustion facilities. A detailed review of this analysis, as well as
other benefits (e.g., avoided clean-up as result of reduced particulate
matter releases), is presented in chapter 5 of the November 13, 1995
Regulatory Impact Assessment.
3. Other Regulatory Issues
We also examined other issues associated with the proposal. These
included environmental justice, unfunded federal mandates, regulatory
takings, and waste minimization.
a. Environmental Justice. We completed an analysis of demographic
characteristics of populations near cement plants and commercial
hazardous waste incinerators and compared them to county and state
populations. This analysis focused on spatial relationships between
these

[[Page 53012]]

facilities and the adjacent minority and low income populations. The
study did not describe the actual health status of these populations
nor how their health might be affected in proximity to hazardous waste
facilities. Results indicated that 27 percent of all cement plants and
37 percent of the sample of incinerators had minority percentages
within a one mile radius which exceed the corresponding county minority
percentages by more than five percentage points. Eighteen percent of
all cement plants and 36 percent of the sample of incinerators had
poverty percentages which exceed the county poverty percentages by more
than five percentage points. Please see chapter seven of the November
13, 1995 RIA for a full discussion of the environmental justice
methodology and findings conducted for the proposal.
b. Unfunded Federal Mandates. Our analysis of compliance with the
Unfunded Mandates Reform Act (UMRA) of 1995 found that the proposal
contained no State, local, tribal government, or private sector Federal
mandates as defined under the regulatory provisions of Title II of
UMRA. We concluded that the rule implements requirements specifically
set forth by Congress, as stated in the CAA and RCRA. The proposed
standards were not projected to result in mandated annualized costs of
$100 million or more to any state, local, or tribal government.
Furthermore, the proposed standards would not significantly or uniquely
affect small governments.
c. Regulatory Takings. We found no indication that the proposed
MACT standards would be considered a taking, as defined by legislation
being considered by Congress at the time. Property would not be
physically invaded or taken for public use without the consent of the
owner. Also, the proposed standards would not deprive property owners
of economically beneficial or productive use of their property or
reduce the property's value.
d. Incentives for Waste Minimization and Pollution Prevention. We
briefly examined the potential for waste minimization in the proposal.
Preliminary results suggested that generators have a number of options
for reducing or eliminating waste. To evaluate whether facilities would
adopt applicable waste minimization measures, a simplified pay back
analysis was used. Using information on per-facility capital costs for
each technology, we estimated the time period required for the cost of
the waste minimization measure to be returned in reduced combustion
expenditures. Our assessment of waste minimization found that
approximately 630,000 tons of waste may be amenable to waste
minimization. For a complete description of the analysis please see the
November 13, 1995 Regulatory Impact Assessment.
4. Small Entity Impacts
The Regulatory Flexibility Act (RFA) of 1980 requires Federal
agencies to consider impacts on small entities throughout the
regulatory process. Section 603 of the RFA calls for an initial
screening analysis to determine whether small entities will be
adversely affected by the regulation. If affected small entities are
identified, regulatory alternatives must be considered to mitigate the
potential impacts. Small entities, as described by the Act, are only
those ``businesses, organizations, and governmental jurisdictions
subject to regulation.'' We used information from Dunn & Bradstreet,
the American Business Directory, and other sources to identify small
businesses. Based on the number of employees and annual sales
information, we identified eleven firms which might be considered
directly affected small entities. We found that directly affected small
entities were unlikely to be significantly affected and that over one-
third of those that were considered small, while having a relatively
small number of employees, had annual sales in excess of $50 million
per year. Also, small entities impacted by the proposal were found to
be those that burn very little waste and hence face very high cost per
ton burned. These facilities were expected to discontinue burning
hazardous waste rather than complying with the proposal. These costs of
discontinuing waste burning would not be so high as to be a significant
impact. Thus, we found that the proposal may, at most, have a minor
impact on a limited number of affected small businesses.
B. What Major Comments Were Received on the Proposal RIA?
The November 13, 1995 Regulatory Impact Assessment (RIA) received
comment from many concerned stakeholders. We also conducted a formal
peer review of the RIA. We appreciate all comments received and
incorporated many of the suggestions into the final Assessment document
to improve the analysis. A summary of the key issues presented by
stakeholders and the peer reviewers is presented below, along with our
responses. You are requested to review the complete documents: Comment
Response Document--Addressing The Public Comments Received On:
Regulatory Impact Assessment for Proposed Hazardous Waste Combustion
MACT Standards, Draft, November 13, 1995, and, Peer Review Response
Document--Addressing The Peer Review Received On: Regulatory Impact
Assessment for Proposed Hazardous Waste Combustion MACT Standards,
Draft, November 13, 1995. These documents, available in the RCRA docket
established for today's action, present complete responses to all
substantive comments received on the 1995 RIA.
1. Public Comments
We received several general comments on the accuracy of the
baseline and compliance costs applied in the RIA. Several commenters
suggested that we revise baseline and compliance costs to improve their
accuracy, which we did. Instead of using a model plant approach for
assigning compliance and baseline costs to modeled combustion
facilities, costs for today's rule have been estimated using combustion
system-specific parameters including gas flow rate, baseline emissions,
air pollution control devices currently in place, total chlorine in
feed, stack moisture, and temperature at the inlet to the air pollution
control device. These system-specific baseline and compliance costs
allow for greater accuracy in estimating national costs and predicting
which facilities are likely to stop burning hazardous waste. Also, the
baseline costs include clinker production penalties at cement kilns and
use updated incinerator capital costs, labor requirements, and ash
disposal costs.
Various commenters were concerned that the consolidation routine in
the economic modeling was unrealistic. For the final economic
assessment, we revised the consolidation routine to incorporate
capacity constraints that affect the ability of combustion facilities
to consolidate wastes into fewer systems at a given facility. Maximum
capacity rates (tons per year) were derived by using the feed rates in
OSW's database (pounds per year) and assuming 8,000 hours per year of
operation. Wastes are assumed to be consolidated into fewer combustion
systems at a single facility to the extent that the capacity
constraints allow the systems to absorb the displaced hazardous wastes.
Many commenters felt that the waste minimization analysis of the
1995 RIA was unrealistic and overestimated gains. They suggested that
the waste minimization analysis be improved to reflect other
constraints faced by waste generators. For the 1999 Assessment, we
conducted an expanded and significantly improved analysis of waste

[[Page 53013]]

minimization alternatives, using a more detailed decision framework for
evaluating waste minimization investment decisions. This framework
attempts to capture the full inventory of costs, savings, and revenues,
including indirect, less tangible items typically omitted from waste
minimization analysis, such as liability and corporate image. For each
alternative that was identified as viable for currently combusted waste
streams, cost curves were developed for a range of waste quantities, as
cost varies by waste quantity. These cost curves were then used to
determine whether a waste generator would shift from combustion to
waste minimization alternatives as combustion prices rise.
Some commenters suggested that we model waste markets to reflect
segmentation across waste types, instead of simply applying different
prices for kilns and incinerators. In response, we have developed a
revised pricing approach that covers seven categories of waste types
and prices. The economic model used for the 1999 Assessment
incorporates these seven different waste types and prices. Waste
management prices depend on several factors: Waste form (solid/liquid/
sludge), heat content, method of delivery (e.g., bulk versus drum), and
contamination level (e.g., metals or chlorine content). In addition,
regulatory constraints (e.g., prohibitions against burning certain
types of wastes) and technical constraints (e.g., adverse effects of
certain waste streams on cement product quality) also influence
combustion prices. Although data limitations prevent the inclusion of
all factors, the information on heat content and constituent
concentrations from EPA's National Hazardous Waste Constituent Survey
(NHWCS) allowed us to enhance the characterization of combusted waste.
A few commenters indicated that the baseline costs of waste burning
for cement kilns should include the shared joint costs of cement
production. We do not include cement production costs in the costs of
waste burning because they are not part of the incremental costs
introduced by hazardous waste burning at kilns. We believe this
assumption is appropriate, given that cement production is the
principal activity of cement kilns that burn hazardous waste.
Furthermore, that same kiln would be required in the production of
cement regardless of hazardous waste combustion activities. We did,
however, evaluate whether some of the more economical marginal kilns
may be covering cement production costs with hazardous waste burning
revenues. These findings are reported in the 1999 Assessment document.
Some were concerned that shutdown costs and environmental risks
associated with combustion facility closures were not accounted for in
the 1995 economic analysis. We found that many of the facilities that
are expected to close are those that are were operating significantly
below capacity in the baseline. This suggests that such facilities may
not have been fully recovering their capital costs and are likely to
close, even in the absence of the MACT standards. Therefore, while
closure is not costless, closure costs attributable directly to the
MACT standards are likely to be relatively small. With regard to
increased risks from transportation of hazardous wastes, the
incremental health risks will be minimal since these facilities are
burning small quantities of waste. In fact, we estimate that less than
1.5 percent of the wastes currently burned at combustion facilities
will be reallocated due to facility closure. Moreover, spills and other
accidents caused by trucking hazardous waste (the most common means of
shipment for hazardous materials) generally are considered low-
probability events, especially relative to the total number of
accidents occurring within transportation overall.
Some commenters felt that potential impacts on generators and fuel
blenders were not adequately addressed. In the 1995 RIA, we considered
these costs and determined that hazardous waste generators and fuel
blenders would likely see price increases for combusted waste streams,
though the magnitude of the price increase will depend on the type of
waste and the non-combustion waste management alternatives available
for that waste type. The price increase faced by generators was
estimated at 10 percent of market prices.
The major hazardous waste burning sectors frequently presented
alternative views regarding various key waste burning issues. These
included: Facility market exits, revenues, impacts resulting from waste
feedrate modifications, impacts from alternative fuel usage, price
impacts, and available practical capacity. We have reviewed and
evaluated the substantiative information submitted by all concerned
stakeholders and believe our final Assessment and Addendum documents
reflect a fair and balanced representation of baseline conditions and
post-rule incremental economic impacts.
2. Peer Review
The peer reviewers suggested that we clarify the aims, objectives,
and organizing principles for the 1995 RIA. They stated that, while the
1995 RIA generally meets the requirements set forth by OMB's Guidance
regarding the economic analysis of federal regulations under Executive
Order 12866, the RIA would be substantially improved if it fully
conformed with OMB's Guidance, especially with regard to organization
and statement of objectives. For the 1999 Assessment, we have tried to
restructure the document to be more in line with OMB's 1996 Guidance
for conducting Economic Analysis of Federal Regulations Under Executive
Order 12866. The 1999 Assessment includes the following elements in the
first chapter to address concerns of the reviewers: the objectives of
the Economic Assessment, the analytical requirements the document
fulfills, the rationale for regulatory action, an examination of
alternative regulatory options, the anticipated effect of the MACT
standards, and the analytic approach and organization for the
subsequent chapters.
The peer reviewers also suggested that the compliance costs need to
be clearly distinguished from social costs, as defined by the theory of
applied welfare economics. For the 1999 Assessment, we have been
careful to clarify the difference between compliance costs and social
costs and explain how the rule will likely affect producers and
consumers. The final Assessment explicitly lays out the economic
framework for the social cost analysis and distinguishes these from
compliance cost estimates. The hazardous waste combustion market is
diverse, dynamic, and segmented. Because data are not adequate to
support a full econometric analysis at this level of complexity, we
have applied a simplified approach that brackets the welfare loss
attributable to today's rule. This approach bounds potential economic
welfare losses by considering two scenarios: (1) Compliance costs
assuming no market adjustments (the upper bound) and (2) market
adjusted compliance costs (the lower bound).
The peer reviewers also suggested that the benefits analysis was
not fully responsive to the requirements of Executive Order 12866. For
the 1999 Assessment, we have applied results from an extensive multi-
pathway risk assessment to develop human health and ecological benefit
estimates. For the human health analysis, benefits are estimated from
cancer and noncancer

[[Page 53014]]

risk reductions. Cancer risk reduction estimates are monetized by
applying the value of a statistical life (VSL) to the risk reduction
expected to result from the MACT standards. Monetary values are
assigned to noncancer benefits using a direct-cost approach which
focuses on the expenditures averted by decreasing the occurrence of an
illness or other health effect. Ecological benefits are also included
in the 1999 Assessment.
The peer reviewers suggested that easily burned waste streams would
command lower prices and that this should be reflected in the economic
modeling. They also indicated that certain combustion sectors may only
handle these easy-to-burn waste types and that this should be reflected
in baseline costs for these combustors. The pricing approach used in
the 1999 Assessment assigns different prices to different types of
wastes. Waste management prices depend on several factors, which
include: waste form (solid/liquid/sludge), heat content, method of
delivery (e.g., bulk versus drum), and contamination level (e.g.,
metals or chlorine content). In addition, regulatory constraints (e.g.,
prohibitions against burning certain types of wastes) and technical
constraints (e.g., adverse effects of certain waste streams on cement
product quality) also influence combustion prices. Although data
limitations prevent us from accounting for all factors, the information
on heat content and constituent concentrations from EPA's National
Hazardous Waste Constituent Survey (NHWCS) allowed us to enhance the
characterization of combusted waste. In addition to pricing
refinements, the 1999 Assessment adjusts baseline costs to reflect
differences in the performance and capabilities across combustion
systems.
The peer reviewers were also concerned that the 1995 RIA applied
outdated data in the analysis. The most recent available data were used
in the 1995 RIA. The 1999 Assessment, once again, applies the most
recently available, and verified data.
The peer reviewers suggested that fully-loaded cost-per-ton
estimates should be provided for each waste minimization alternative so
that these could be compared with combustion prices. For the 1999
Assessment, we conducted an expanded and significantly improved
analysis of waste minimization alternatives. This analysis used a more
detailed decision framework for evaluating waste minimization
investment decisions that captures the full inventory of costs,
savings, and revenues, including indirect, less tangible items
typically omitted from waste minimization analysis, such as liability
and corporate image. For each viable waste minimization alternative for
currently combusted waste streams, cost curves were developed for a
range of waste quantities because cost varies by waste quantity. These
cost curves were then used to determine whether a waste generator would
shift from combustion to waste minimization alternatives as combustion
prices rise.

III. Why Is Today's Rule Needed?

Today's rule will reduce the level of several hazardous air
pollutants, including dioxins and furans, mercury, semi-volatile and
low volatile metals, and chlorine gas. Carbon monoxide, hydrocarbons,
and particulate matter will also be reduced. Most hazardous waste
combustion facilities are currently operating with some air pollution
control devices in place. However, existing pollutants from these
facilities are still emitted at levels found to result in risks to
human health and the environment. Human exposure to these combustion
air toxics occurs both directly and indirectly and leads to cancer,
respiratory diseases, and possibly developmental abnormalities. A
preliminary screening analysis suggests that ecosystems are also at
risk from these air pollutants.
The hazardous waste combustion industry operates in a dynamic
market. Several combustion facilities and systems have closed or
consolidated over the past several years and this trend is likely to
continue. These closures and consolidations may lead to reduced air
pollution, in the aggregate, from hazardous waste facilities. However,
the ongoing demand for hazardous waste combustion services will
ultimately result in a steady equilibrium as the market adjusts over
the long-term. We therefore expect that air pollution problems from
these facilities, and the corresponding threats to human health and
ecological receptors, will continue if today's rule were not
implemented.
The market has generally failed to correct the air pollution
problems resulting from the combustion of hazardous wastes. This has
occurred for several reasons. First, there exists no natural market
incentive for hazardous waste combustion facilities to incur additional
costs implementing control measures because the individuals and
entities who bear the negative human health and ecological impacts
associated with these actions have no direct control over waste burning
decisions. This may be referred to as an environmental externality,
where the private industry costs of combustion do not fully reflect the
human health and environmental costs of hazardous waste combustion.
Second, the parties injured by the combusted pollutants are not likely
to have the resources or technological expertise to seek compensation
from the damaging entity (combustion facility) through legal or other
means. Finally, emissions from hazardous waste combustion facilities
directly affect a ``public good,'' the air. Improved air quality
benefits human health and the environment. These benefits cannot be
limited to just those who pay for reduced pollution. The absence of
government intervention, therefore, will result in a free market that
does not provide the socially optimal quantity and quality of public
goods, such as air.
We recognize the need for federal regulation as the optimal means
of correcting market failures leading to the negative environmental
externalities resulting from the combustion of hazardous waste. The
complex nature of the pollutants, waste feeds, waste generators, and
the diverse nature of the combustion market would limit the
effectiveness of a non-regulatory approach such as taxes, fees, or an
educational-outreach program. Furthermore, requirements for MACT
standards under the Clean Air Act, as mandated by Congress, has
compelled us to select today's regulatory approach.

IV. What Were the Regulatory Options?

We carefully assembled and evaluated all data and relevant
information acquired since the proposal. We considered several
alternative MACT options since the proposal, ultimately leading to
today's rule. Please refer to Part Four of this preamble for more
detail on option development and the specific approach and methodology
used in developing the final standards. This section of today's
preamble briefly discusses and assesses the final regulatory levels and
two primary options. The final regulatory levels, as discussed in Part
Four, establish a combination of floor and beyond-the-floor standards
for the pollutants of concern. Of the options analyzed, one addresses a
floor only scenario and the other examines beyond-the-floor levels for
dioxins/furans and mercury, based on activated carbon injection (ACI).
The reader may wish to examine the Assessment document for a complete
discussion of the analytical methodology, costs, benefits, and other
projected impacts of today's rule and options. This Assessment document
is available in the RCRA docket for today's rule.

[[Page 53015]]

V. What Are the Potential Costs and Benefits of Today's Rule?

A. Introduction
The value of any regulatory policy is traditionally measured by the
net change in social welfare that it generates. Our economic assessment
for today's rule evaluates costs, benefits, economic impacts, and other
impacts such as environmental justice, children's health, unfunded
mandates, waste minimization incentives, and small entity impacts. To
conduct this analysis, we examined the current combustion market and
practices, developed and implemented a methodology for examining
compliance and social costs, applied an economic model to analyze
industry economic impacts, quantified (and, where possible, monetized)
benefits, and followed appropriate guidelines and procedures for
examining equity considerations, children's health, and other impacts.
The data we used in this analysis were the most recently available at
the time of the analysis. Data verification, relevance, and public
disclosure issues prevented us from incorporating data from certain
sources. Furthermore, because our data were limited, the estimated
findings from these analyses should be viewed as national, not site
specific impacts.
B. Combustion Market Overview
The hazardous waste industry comprises three key segments:
hazardous waste generators, fuel blenders and intermediaries, and
hazardous waste incinerators. Hazardous waste is combusted at three
main types of facilities: Commercial incinerators, on-site
incinerators, and waste burning kilns (cement kilns and lightweight
aggregate kilns). Commercial incinerators are generally larger in size
and designed to manage virtually all types of solids, as well as liquid
wastes. On-site incinerators are more often designed as liquid-
injection systems that handle liquids and pumpable solids. Waste
burning kilns burn hazardous wastes to generate heat and power for
their manufacturing processes.
As of the date of our analysis, 172 combustion facilities are
permitted to burn hazardous waste in the United States. On-site
incinerators (private and government) represent 129 facilities (or 75
percent of this total), commercial incinerators represent 20
facilities, cement kilns represent 18 facilities, and lightweight
aggregate kilns represent five facilities. A facility may have one or
more combustion systems. Companies that generate large quantities of
uniform hazardous wastes generally find it more economical and
efficient to combust these wastes on-site using their own noncommercial
systems. Commercial incineration facilities manage a wide range of
waste streams generated in small to medium quantities by diverse
industries. Cement kilns and lightweight aggregate kilns derive heat
and energy by combining clean burning (solvents and organics) high-Btu
liquid hazardous wastes with conventional fuels. The EPA Biennial
Reporting System (BRS) reports a total demand for all combusted
hazardous waste, across all three types of facilities, at nearly 3.3
million tons in 1995.
Most of the waste managed by combustion comes from a relatively
narrow set of industries. The entire chemical industry in 1995
generated 74 percent of all combusted waste. Within this sector, the
organic chemicals subsector was the largest source of waste sent to
combustion, providing about 32 percent of all combusted waste. The
pesticide and agricultural chemical industry generated 12 percent of
the total. No other single sector generated more than 10 percent of the
total.
Regulatory requirements, liability concerns, and economics
influence the demand for combustion services. Regulatory forces
influence the demand for combustion by mandating certain hazardous
waste treatment standards (land disposal restriction requirements,
etc.). Liability concerns of waste generators affect combustion demand
because combustion, by destroying organic wastes, greatly reduces the
risk of future environmental problems. Finally, if alternative waste
management options are more expensive, hazardous waste generators will
likely choose to send their wastes to combustion facilities in order to
increase their overall profitability.
Throughout much of the 1980s, hazardous waste combustors enjoyed a
strong competitive position and generally maintained a high level of
profitability. During this period, EPA regulations requiring combustion
greatly expanded the waste tonnage for this market. In addition,
federal permitting requirements, as well as powerful local opposition
to siting of new incinerators, constrained the entry of new combustion
systems. As a result, combustion prices rose steadily, ultimately
reaching record levels in 1987. The high profits of the late 1980s
induced many firms to enter the market, in spite of the difficulties
and delays anticipated in the permitting and siting process. Hazardous
waste markets have changed significantly since the late 1980s. In the
early 1990s, substantial overcapacity resulted in fierce competition,
declining prices, poor financial performance, numerous project
cancellations, and some facility closures. Since the mid 1990s, several
additional combustion facilities have closed, while many of those that
have remained open have consolidated their operations. There still
remains significant overcapacity throughout the hazardous waste
combustion industry.
C. Baseline Specification
Proper and consistent baseline specification is vital to the
accurate assessment of incremental costs, benefits, and other economic
impacts associated with today's rule. The baseline essentially
describes the world absent today's rule. The incremental impacts of
today's rule are evaluated by predicting post MACT compliance responses
with respect to the baseline. The baseline, as applied in this
analysis, is the point at which today's rule is promulgated. We
recognize that the baseline should not simply describe a point in time,
but rather should describe the state of the world over time, absent
today's rule. The Assessment describes the data sources used in
specifying the baseline and examines how each of these factors are
likely to change over time in the absence of today's rule. Finally,
because this analysis precedes final rule promulgation, data sources
used to determine the baseline will necessarily predate the point of
rule promulgation. A full discussion of baseline specification is
presented in the Assessment document for today's rule.
D. Analytical Methodology and Findings--Engineering Compliance Cost
Analysis
The total compliance costs for existing hazardous waste combustion
facilities are developed using engineering models that assign pollution
control measures and costs to each modeled combustion system. The
engineering model also incorporates other compliance costs, such as
monitoring requirements, permit modifications, sampling and analyses,
and other recordkeeping and reporting requirements. We applied the same
basic approach in developing compliance costs for new sources as was
used for existing sources. Please see the Assessment document for a
complete discussion of the analytical methodology applied for existing
and new facilities.
Compliance costs presented in this section are based on a static
analysis assuming no market adjustments.

[[Page 53016]]

Results from this static analysis should therefore be considered
``high-end'' estimates. The engineering compliance cost analysis
reveals that each combustion system will likely comply with the final
standards through a different combination of pollution control
measures. This is likely to result in widely diverse per system
compliance costs across combustion sectors. The average annualized per
system costs, across all sectors, are projected to range from about
$0.16 to $0.72 million for compliance with the final standards. Per
system costs at the floor are estimated to range from $0.16 to $0.68
million, while these costs under the beyond-the-floor activated carbon
injection (ACI) option would range from $0.36 to $0.99 million. Cement
kilns were generally found to experience the highest per system
compliance costs, while the commercial and on-site incinerators would
generally experience the lowest per system costs. The compliance costs
per ton of hazardous waste burned are projected to increase from 31 to
41 percent for cement kilns and about 35 percent for lightweight
aggregate kilns. The increase for commercial incinerators is estimated
at 20 percent of the baseline burn costs. The regulated community is
also likely to experience some cost savings as a result of the
streamlined administrative procedures established through today's final
rule.
The compliance cost analysis contains a variety of uncertainties.
The most significant include: The limited availability of emissions
data upon which engineering controls are based, lack of baseline air
pollution control device data for a number of facilities, and the
difficulty in determining the extent to which feed control may be used
as a feasible alternative method of compliance. While uncertainties are
acknowledged, we do not believe that the above data limitations
significantly bias the results either upward or downward.
In addition to costs incurred by the private sector, today's rule
is also likely to result in incremental costs and savings to government
regulatory entities at different levels as they administer and enforce
the new emissions standards and related requirements. EPA Regional
offices, state agencies, as well as some local agencies may incur some
combination of incremental costs associated with permitting.
Modifications of the permitting process related to Clean Air Act
provisions could cost governmental entities, nationwide, approximately
$330,000 per year. Potential government activities could also include
the state rulemaking efforts necessary for agencies to modify their
RCRA permitting processes as part of the ``Fast-Track'' provisions.
State rulemakings and authorization of the modified procedures could
cost states between $500,000 and $700,000, nationwide. Streamlined RCRA
permit modification procedures may also result in aggregate savings
ranging from $0.4 to $2.1 million. Overall economic impacts on
particular governmental regulatory entities will depend on a variety of
factors that are difficult to characterize with precision. Furthermore,
economic impacts associated with governmental activities will differ in
the way in which a particular governmental entity may choose to
implement the requirements.
E. Analytical Methodology and Findings--Social Cost Analysis
We examined social cost impacts potentially associated with today's
rule. Total social costs include the value of resources used to comply
with the standards by the private sector, the value of resources used
to administer the regulation by the government, and the value of output
lost due to shifts of resources to less productive uses. To evaluate
these shifts in resources and changes in output requires predicting
changes in behavior by all affected parties in response to the
regulation, including responses of directly-affected entities, as well
as indirectly-affected private parties.
For this analysis, social costs are grouped into two categories:
economic welfare (changes in consumer and producer surplus), and
government administrative costs. The economic welfare analysis
conducted for today's rule uses a simplified partial equilibrium
approach to estimate social costs. In this analysis, changes in
economic welfare are measured by summing the changes in consumer and
producer surplus. This simplified approach bounds potential economic
welfare losses associated with the rule by considering two scenarios:
Compliance costs assuming no market adjustments, and market adjusted
compliance costs.
Social costs presented in this section assume market adjustments.
Under this scenario, increased compliance costs are examined in the
context of likely incentives combustion facilities would have to
continue burning hazardous wastes and the competitive balance in
different combustion sectors. Furthermore, combustion facilities are
likely to try to recover these increased costs by charging higher
prices to generators and fuel blenders. This scenario estimates market
adjusted compliance costs by assessing baseline profitability,
profitability post-rule using different price increase scenarios, and
waste management alternatives in order to help predict combustion price
increases.
Overall, the difference in aggregate compliance costs for all
sectors of the existing regulated community to meet any of the examined
scenarios is not substantial. Total annualized market adjusted costs
for all sectors are estimated to range from $44 to $50 million under
the floor option. Under the beyond-the-floor (ACI) option, these costs
are estimated to range from $98 to $107 million. For all sectors to
meet the final standards, our best estimate of total annualized costs
ranges from $50 to $63 million, depending upon level of price pass-
through. All cost estimates are incremental to the baseline. These
estimates, however, are not incremental to any mutual requirements
potentially associated with cement kilns meeting standards established
under the nonhazardous waste burner cement kiln rule.
Cement kilns ($17-24 million) and private on-site incinerators
($20-24 million) make up about 76 percent of aggregate national costs
under the final standards. For cement kilns, this is due primarily to
the high costs per system. For private on-site incinerators, the high
costs are primarily due to the large number of combustion systems.
Total costs are less for commercial incinerators ($5-6 million, or 10
percent) because of lower costs per system relative to cement kilns and
due to the limited number of commercial units relative to on-site
incinerators. Lightweight aggregate kilns ($3 million) represent about
5 to 6 percent of the total costs, due primarily to the limited number
of units. Government on-site units make up the remainder.
F. Analytical Methodology and Findings--Economic Impact Analysis
Various market adjustments are expected in response to the
increased costs of hazardous waste combustion associated with today's
rule. Economic impacts may be measured through numerous factors. This
analysis examines market exit estimates, waste reallocations,
employment impacts, combustion price increases, industry impacts, and
the multirule or joint impacts analysis. Economic impacts presented in
this section are distinct from the social costs analysis, which
represents only the monetary value of market disturbances.

[[Page 53017]]

1. Market Exit Estimates
The hazardous waste combustion industry operates in a dynamic
market, with a number of systems/facilities projected to exit the
hazardous waste burning market under baseline conditions (see Section
V. B of this Part). As a result, this analysis presents market exit
estimates expected to result under the baseline, as well as from
today's rule. This approach is developed in an effort to present a more
accurate estimate of ``real-world'' incremental impacts resulting from
the final standards. Market exit estimates are derived from a breakeven
analysis designed to determine system and facility viability. This
analysis is subject to several assumptions, including: engineering cost
data on the baseline costs of waste burning, cost estimates for
pollution control devices, prices for combustion services, and
assumptions about the waste quantities burned at these facilities. It
is important to note that, for most sectors, exiting the hazardous
waste combustion market is not equivalent to closing a plant. (Actual
plant closure would only be expected in the case of an exit from the
hazardous waste combustion market of a commercial incinerator closing
all its systems.)
A relatively small percentage of facilities (including no
lightweight aggregate kilns) are projected to stop burning hazardous
waste as a result of the incremental requirements associated with
today's rule. Those facilities that do exit were found to be marginally
profitable in the baseline, burning low quantities of hazardous waste.
The economic model post-consolidation results indicate that, in
response to today's rule, the following number of combustion facilities
are expected to cease burning hazardous waste in the short term: Cement
kilns, zero out of 18 facilities; lightweight aggregate kilns, zero out
of five facilities; commercial incinerators, zero out of 20 facilities;
and private on-site incinerators, 16 out of 111 facilities.
The number of anticipated market exits increases in the long term
due to the necessity of recovering the capital costs of combustion.
However, because this also holds true in the baseline, an increased
number of projected long-term baseline market exits may, in some cases,
actually decrease the number of incremental long-term exits. There
remain zero incremental market exits for LWAKs and commercial
incinerators over the long-term. Incremental market exits for cement
kilns, however, increase from zero in the short-term to up to two over
the long-term. Incremental market exits for private on-site
incinerators decline from 16 in the short-term to 13 over the long-
term. This is due to a 62 percent increase in baseline market exits
from the short-term to the long-term.
2. Quantity of Waste Reallocated
Combustion systems that can no longer cover costs (i.e., those
below the dynamic breakeven quantity) are projected to stop burning
hazardous waste. Hazardous wastes from these systems will likely be
reallocated to other viable combustion systems at the same facility if
there is sufficient capacity, alternative combustion facilities that
continue burning, or waste management alternatives (e.g., solvent
reclamation). Because combustion is likely to remain the lowest cost
option, we expect most reallocated wastes will continue to be managed
at combustion facilities.
The economic model indicates that, in response to today's rule,
between 14,000 to 42,000 tons of currently burned hazardous waste could
be reallocated to other facilities or waste management alternatives.
This estimate represents between 0.4 and 1.3 percent of the total
quantity of combusted hazardous wastes and is incremental to projected
long-term baseline reallocations of approximately 100,000 tons.
Currently, there is more than adequate capacity within the remaining
sources of the combustion market to accommodate this reallocated waste,
even at the high-end estimate.
3. Employment Impacts
Today's rule is likely to cause employment shifts across all of the
hazardous waste combustion sectors. These shifts will occur as specific
combustion facilities find it no longer economically feasible to keep
all of their systems running, or to stay in the hazardous waste market
at all. When this occurs, workers at these locations may lose their
jobs. At the same time, the rule may result in employment gains, as new
purchases of pollution control equipment stimulate additional hiring in
the pollution control manufacturing sector and as additional staff are
required at combustion facilities for various compliance activities.
a. Employment Impacts--Losses. Primary employment losses in the
combustion industry are likely to occur when combustion systems
consolidate the waste they are burning into fewer systems or when a
facility exits the hazardous waste combustion market altogether.
Operation and maintenance labor hours are expected to be reduced for
each system that stops burning hazardous waste. For each facility that
completely exits the market, employment losses will likely also include
supervisory and administrative labor.
Total incremental employment dislocations potentially resulting
from the final standards range from approximately 100 to 230 full-time-
equivalent (FTE) jobs under the floor and the recommended options.
Under the beyond-the-floor (ACI) option the high-end estimate of
employment dislocations increases by almost 9 percent to approximately
250 FTEs. Among the different sectors, on-site incinerators are
responsible for most of the total estimated number of job losses. Their
significant share of the losses is a function of both the large number
of on-site incinerators in the universe as well as the relatively high
number of expected exits within this sector. Cement kilns are
responsible for the second largest number of expected employment losses
due to the number of systems that consolidate waste-burning at these
facilities.
b. Employment Impacts--Gains. In addition to employment losses,
today's rule will also lead to job gains as firms invest to comply with
the various requirements of the rule and add additional operation and
maintenance personnel for the new pollution equipment and other
compliance activities, such as new reporting and record keeping
requirements.
The total annual employment gains (without particulate matter
continuous emission monitors) associated with the floor and recommended
final standards are approximately 300 FTEs. The beyond-the-floor (ACI)
option may increase the high-end employment gain estimate to as much as
620 FTEs. About one-third to one-half of all estimated job gains are
projected to occur in the pollution control equipment industry. The
remaining job gains will occur at the combustion facilities as
additional personnel are hired for operation and maintenance and
permitting requirements.
While it may appear that this analysis suggests overall net job
creation under particular options and within particular combustion
sectors, such a conclusion would be inappropriate. Because the gains
and losses occur in different sectors of the economy, they should not
be added together. Doing so would mask important distributional effects
of the rule. In addition, the employment gain estimates reflect within
sector impacts only and therefore do not account for job displacement
across sectors as

[[Page 53018]]

investment funds are diverted from other areas of the larger economy.
4. Combustion Price Increases
All combustion facilities that remain in operation will experience
increased operational costs under today's rule. To protect their
profits, each facility will have an incentive to pass these increased
costs on to their customers (generators and blenders) in the form of
higher combustion prices. Generators and blenders are expected to pay
these higher prices unless they have less expensive waste management
alternatives.
Under the theory of market price adjustments, as applied in the
economic model, waste would be sent to the least expensive alternatives
first, all else being equal. At the same time, prices would rise to the
point at which all demand for waste management is met. In theory, the
last tons would be managed by substituting non-combustion or waste
minimization alternatives. The most efficient waste management
substitute for these wastes would cap price increases, resulting in a
new market price. Combustion facilities, in turn, would each set their
prices at this market price in order to maximize profits. Less
efficient waste management scenarios may earn just enough to stay in
business over the short term, but would not recover capital costs.
Combustion systems operating above the market price would lower their
prices or exit the market. In reality, the hazardous waste combustion
marketplace is very complex, and the determination of an adjusted
market price would be an ongoing process affected by numerous factors,
including price differentials among regions, waste stream types, and
generators.
Available economic data on the cost of waste management
alternatives for combusted hazardous waste, including source reduction
and other waste minimization options, are not precise enough to allow
for an accurate estimate of the maximum price increase that combustors
may pass through to generators and fuel blenders. However, available
data do indicate that the demand for hazardous waste combustion is
relatively inelastic and that combustion facilities are likely to pass
through approximately 75 percent of compliance costs in the least-cost
sector. High-cost sectors, however, may pass through less than the 75
percent estimate. We also analyzed a 25 percent price pass through
scenario. Under the recommended final standards, the weighted average
combustion price per ton is projected to increase anywhere from about
0.5 to 11 percent, depending upon sector and scenario. Prices were
found to increase by as much as 25 percent under the beyond-the-floor
(ACI) option.
5. Industry Profits
Hazardous waste-burning profits for all combustion sectors, on
average, are expected to decline post-rule. This decline, however, will
not be consistent across sectors. Hazardous waste-burning profits for
cement kilns are projected to decrease by no more than 10 percent,
while profits for commercial incinerators would decrease by no more
than 2 percent. These profit margin estimates are based on a simple
calculation that subtracts projected operating costs from revenues.
These estimates provide relative measures of profit changes and should
not be used to predict absolute profit margins in these industries.
Compliance costs associated with meeting today's rule are estimated
to represent less than 2 percent of the pollution control expenditures
in industries that contain facilities with on-site incinerators. For
cement kilns, however, compliance costs are expected to increase total
pollution control expenditures by no more than 60 percent at waste-
burning facilities.
To comply with today's rule, many facilities will need to purchase
additional pollution control equipment. From the perspective of the
pollution control industry, these expenditures will translate into
additional revenues and profits. Total profits for the air pollution
control industry are likely to increase as a result of today's rule.
6. National-Level Joint Economic Impacts
Analyzing national-level economic impacts in a market context
provides an opportunity to assess the distributional effects on cement
producers, lightweight aggregate kilns, and commercial incinerators. As
a supplement to today's analysis, we used the model developed for the
Portland Cement MACT rulemaking to estimate national-level economic
impacts of today's Hazardous Waste Combustion (HWC) MACT rule in an
interactive market context. This analysis was conducted to estimate
joint impacts of today's rule in conjunction with the Portland Cement
MACT rule and the Cement Kiln Dust rule. The Portland Cement MACT model
incorporates compliance costs for each affected cement kiln,
lightweight aggregate kiln, and commercial incinerator and then
projects national level impacts associated with these facilities and
for the general Portland cement market. On-site incinerators were not
included in this analysis because they do not generally compete in the
commercial hazardous waste combustion market. Results from this
analysis are separated into three categories: Market-, industry-, and
social-level impacts associated with imposition of the recommended
final standards and the two HWC MACT options (floor and beyond-the-
floor (ACI)).
Joint national-level economic impact results combining the HWC MACT
options with the Portland Cement MACT and Cement Kiln Dust Rule are
summarized in this section. Market, industry, and social cost impacts
are discussed. This analysis assumes simultaneous implementation of all
three rules.
Market-level impacts for this joint scenario, assuming the floor
option, result in increased costs of cement production and burning
hazardous waste at affected cement kilns. The national market price of
Portland cement is projected to increase by about 2.0 percent, while
domestic production would decline by about 4.0 percent. Market impacts
for the joint scenario with the recommended final standards and the
beyond-the-floor (ACI) option were found to be generally equivalent to
results under the floor option. The extent to which domestic cement
producers face competition from foreign cement imports will limit the
degree of domestic price increases. Furthermore, the U.S. cement market
is regionally specific. While nationwide average market price and
production impacts are estimated to be relatively minor, producers in
selected regions may experience significant revenue and production
impacts, either positive or negative.
Under the joint scenario with the floor option, the market prices
for both liquid and solid hazardous waste incineration are projected to
increase by about 8.6 percent and 1.4 percent, respectively. The price
change for liquids is higher than that observed for the floor only,
while the price change for solids is virtually the same. For cement
kilns, the increased costs associated with all three regulations,
combined with their reductions in cement production, is projected to
cause their supply of hazardous waste incineration services to fall by
around 11.0 percent for both liquids and solids. In response to the
regulatory costs, lightweight aggregate kilns also reduce their supply
of liquid hazardous waste incineration by around 9.0 percent. For
commercial incinerators, the supply of hazardous waste incineration
increases by nearly 6.0 percent for liquids and close to 3.0

[[Page 53019]]

percent for solids. The market impacts for the joint scenario, using
the recommended final standards and the beyond-the-floor (ACI)
alternative, were found to be similar to those for the floor option.
One exception is the market price for liquids, which increases by a
greater percentage under the joint scenario with the beyond-the-floor
(ACI) alternative. This results in a greater reduction in liquid
hazardous waste burned at cement kilns and lesser decreases in liquids
incinerated at commercial incinerators.
Industry-level impacts under the joint impacts scenario with the
floor option indicate that Portland cement plants may see total gross
revenues decline by nearly 3.0 percent from their current baseline.
This decline in total revenue results from foregone revenues associated
with producing less Portland cement and lost revenues from burning
hazardous waste. The total net costs for these cement plants are also
projected to decrease, reflecting the increase in costs associated with
burning hazardous waste, plus the increase in cement kiln dust
management costs, and the decrease in costs associated with producing
less cement. The net result, indicates a decline in aggregate
nationwide earnings before interest and taxes (EBIT) of about 5.5
percent from the current baseline. Lightweight aggregate kilns are also
projected to incur a decline in hazardous waste-related EBIT of about
5.5 percent. Alternatively, as a group, the commercial incinerators are
expected to experience a net gain of around 11.0 percent in annual
earnings under this joint scenario with the floor option. These joint
industry-level impacts on EBIT indicate a similar pattern across each
regulatory scenario, except for lightweight aggregate kilns under the
beyond-the-floor (ACI) option, where EBIT declines by nearly 14.0
percent. Industry-level impacts under the joint impact analysis also
includes estimates of plant or system closures. The joint analysis
under each hazardous waste combustion scenario indicates that three
cement plants and 14 to 15 kilns may cease production. Furthermore,
five cement kilns are projected to stop burning hazardous waste. The
analysis also indicates that one lightweight aggregate kiln may
discontinue burning hazardous waste and one to two commercial
incinerators may close operations and stop burning hazardous waste with
the joint implementation of all three rules. These market exit
estimates include projected baseline closures.
Social-level impacts, or social costs, under the joint scenarios
indicate that, for both Portland cement and hazardous waste
incineration services, consumers are worse off due to the increase in
prices and reductions in consumption. For producers of Portland cement
and incineration services, cement kilns and lightweight aggregate kilns
are worse off (on a nationwide basis) due to the decline in market
share, while commercial incinerators are better off due to the increase
in prices and market share.
Refer to the final Assessment document and appendices for a
complete discussion of joint impacts.
G. Analytical Methodology and Findings--Benefits Assessment
This section discusses the benefits assessment for today's rule.
Results from our multi-pathway human health and ecological risk
assessment are used to evaluate incremental benefits to society of
emission reductions at hazardous waste combustion
facilities.³⁵¹ Total monetized benefits are estimated at
$19.2 million. This section also summarizes how today's rule may lead
to changes in the types and quantities of wastes generated and managed
at combustion facilities through increased waste minimization.
---------------------------------------------------------------------------

\351\ The RIA for the proposal included results from a screening
analysis designed to assess the potential magnitude of property
value benefits caused by the MACT standards. This analysis is not
included in the Economic Assessment for the Final Rule due to
limitations of the benefits transfer approach and because property
value benefits likely overlap with human health and ecological
benefits. Including property value benefits would result in double-
counting.
---------------------------------------------------------------------------

1. Human Health and Ecological Benefits
a. Risk Assessment Overview. The basis for the benefits assessment
is our multi-pathway risk assessment model. This model estimates
baseline risks from hazardous waste combustion emissions, as well as
expected risks after today's rule is implemented. The model examines
both inhalation and ingestion pathways to estimate human health risks.
A less detailed screening-level analysis is used to identify the
potential for ecological risks. The risk assessment is carried out for
the regulatory baseline (no regulation), the final recommended
standards, and the two MACT options (floor and beyond-the-floor (ACI)).
The assessment uses a case study approach in which 76 hazardous waste
combustion facilities and their site-specific land uses and
environmental settings are characterized. The randomly selected
facilities in the study include 43 on-site incinerators, 13 commercial
incinerators, 15 cement kilns, and five lightweight aggregate kilns.
The pollutants analyzed in the risk assessment are dioxins and
furans, selected metals, particulate matter, chlorine, and hydrogen
chloride. The metals modeled in the analysis include antimony, arsenic,
barium, beryllium, cadmium, chromium, copper, cobalt, lead, manganese,
mercury, nickel, selenium, silver, and thallium. The fate and transport
of the emissions of these pollutants is modeled to arrive at
concentrations in air, soil, surface water, and sediments. To assess
human health risks, these concentrations can be converted to estimated
doses to the exposed populations using exposure factors such as
inhalation and ingestion rates. These doses are then used to calculate
cancer and noncancer risks, if the appropriate health benchmarks are
available. To assess potential ecological risks, soil, surface water
and sediment concentrations are compared with eco-toxicological
criteria representing protective screening values for ecological risks.
Because these criteria are based on de minimis ecological effects and
thus represent conservative values, an exceedance of the eco-
toxicological criteria does not necessarily indicate ecological
damages. It simply suggests that potential damages cannot be ruled out.
To characterize the cancer and noncancer risks to the populations
listed above, the risk assessment breaks down the area surrounding each
modeled combustion facility into 16 polar grid sectors. For each polar
grid sector, risk estimates can be developed for different age groups
and receptor populations (e.g., 0 to 5 year old children of subsistence
fishers). This approach is used because geographic and demographic
differences across polar grid sectors leads to sectoral variation in
individual risks. Thus, individual risk results are aggregated across
sectors to generate the distribution of risk to individuals in the
affected area. An additional Monte Carlo analysis was conducted to
incorporate variability in other exposure factors such as inhalation
and ingestion rates for three scenarios that were thought to comprise
the majority of the risk to the study area population. These scenarios
address cancer risk from dioxin exposure to beef and dairy farms and
noncancer risk from methyl mercury exposure to recreational anglers.
b. Human Health Benefits--Methodology. Human health benefits are
assessed by identifying those pollutants for which emission reductions
are expected to result in improvements to human health or the

[[Page 53020]]

environment. The relevant results from the risk assessment for the
pollutants of concern are then examined, focusing on population risk
results based on central tendency exposure parameters. The risk
assessment data are expressed as indicators of potential benefits, such
as reduced cancer incidence or reduced potential for developing
particular illnesses or abnormalities. Where possible, monetary values
are assigned to these benefits using a benefits transfer approach.
To assign monetary values to cancer risk reduction estimates, we
apply the value of a statistical life to the risk reduction expected to
result from the MACT standards. The value of a statistical life is
based on an individual's willingness to pay to reduce a risk of
premature death or their willingness to accept increases in mortality
risk. Because there are many different estimates of value of a
statistical life in the economic literature, we estimate the reduced
mortality benefits using a range of value of a statistical life
estimates from 26 policy-relevant value-of-life studies. The estimated
value of a statistical life figures from these studies range from $0.7
million to $15.9 million (adjusted to 1996 dollars), with a mean value
of $5.6 million. The expected number of annual premature statistical
deaths avoided are multiplied by the value of a statistical life
estimate to determine the estimated monetary value of the mortality
risk reductions.
A variety of approaches are used to evaluate the benefits
associated with noncancer risk reductions. For particulate matter, both
morbidity and mortality benefits are estimated. Particulate matter is
the only non-carcinogen in the risk assessment for which there is
sufficient dose-response information to estimate numbers of cases of
disease and deaths from exposures. For lead and mercury, upper bound
estimates of the population at risk are used. This is because
information is only available on the potential of an adverse effect,
with no estimates available on the likelihood of these effects.
We assign monetary values to noncancer benefits using a direct cost
approach which focuses on the expenditures averted, and the opportunity
cost of time spent in the hospital, by decreasing the occurrence of an
illness or other health effect. While the willingness to pay approach
used for valuing the cancer risk reductions is conceptually superior to
the direct cost approach, measurement difficulties, such as estimating
the severity of various illnesses, precludes us from using this
approach here. Direct cost measures are expected to understate true
benefits because they do not include cost of pain, suffering, and time
lost. On the other hand, because we use upper bound estimates of the
population at risk, we cannot conclude that the results are biased in
one direction or the other.
c. Human Health Benefits--Results. Human health benefits are
expected from both cancer and noncancer risk reductions. Less than one
cancer case per year is expected to be avoided due to reduced emissions
from combustion facilities. The majority of the cancer risk reductions
are linked to consumption of dioxin-contaminated agricultural products
exported beyond the boundaries of the study area. Less than one-third
of the cancer risk reductions occur in local populations living near
combustion facilities. Cancer risks for local populations are
attributed primarily to reductions in arsenic and chromium emissions.
These pollutants account for almost 85 percent of total local cancer
incidences in the baseline. By applying value of a statistical life
estimates to these cases, the total annual cancer risk reductions
(benefits) in going from the baseline to the final standards, are
valued at between $0.13 and $9.9 million, with a best estimate of
approximately $2.02 million.
Across all receptor populations, individual cancer risks are
greatest for subsistence farmers. Dioxin is the primary pollutant that
drives the cancer risk for this sensitive receptor population. A lack
of population data prevented us from quantifying benefits for this sub-
population. It is possible, however, to characterize the reduction in
risk from baseline to implementation of today's rule. With the
exception of one particular scenario, the cancer risk for all
subsistence farmers is reduced to below levels of concern after
implementation of today's rule. Today's rule is also expected to result
in lower cancer risks for children of subsistence farmers.
Most of the noncancer human health benefits from today's rule come
from reductions in particulate matter. Some additional noncancer
benefits come from reduced blood lead levels in children living near
combustion facilities. Total annual noncancer benefits from
quantifiable sources are valued at between $9.85 and $73.8 million,
with a best estimate of about $17.2 million. Uncertainties implicit in
the quantitative mercury analysis continue to be sufficiently great so
as to limit its ultimate use in the monetization of noncancer benefits.
Please review the Addendum and chapter six of the Assessment document
for a complete discussion of human health benefits resulting from
today's rule.
d. Ecological Benefits--Methodology. Ecological benefits are based
on a screening analysis for ecological risks that compares soil,
surface water, and sediment concentrations with eco-toxicological
criteria based on de minimis thresholds for ecological effects. Because
these criteria represent conservative values, an exceedance of the eco-
toxicological criteria only indicates the potential for adverse
ecological effects and does not necessarily indicate ecological
damages. For this reason, benefits of avoiding adverse ecological
impacts are discussed only in qualitative terms.
The basic approach for determining whether ecosystems or biota are
potentially at risk consists of five steps: (1) Identify susceptible
ecological receptors that represent relatively common species and
communities of wildlife, (2) develop eco-toxicological criteria for
receptors that represent acceptable pollutant concentrations, (3)
estimate baseline and post-rule pollutant concentrations in sediments,
soils, and surface waters of the study areas, (4) for each land area or
water body modeled, compare the modeled media concentrations to
ecologically protective levels to estimate eco-toxicological hazard
quotients, and (5) total the land and water areas containing hazard
quotients exceeding one and compare this number for the baseline and
post-rule scenario. The reduction in the land and water area
potentially at risk indicates a potential for avoiding adverse
ecological impacts. Monetary values are not assigned to these potential
benefits.
e. Ecological Benefits--Results. Ecological benefits are
attributable primarily to reductions in dioxin and mercury for
terrestrial ecosystems. For these ecosystems, hazard quotients are
reduced to acceptable levels for approximately 115 to 150 square
kilometers of land located within 20 kilometers of all combustion
facilities. Ecological benefits associated with freshwater aquatic
ecosystems are attributable to reductions in lead, with hazard
quotients reduced to acceptable levels for approximately 35 to 40
square kilometers of these surface waters. These reductions of
ecological risk criteria below levels of concern only indicates a
potential for ecological improvement.
2. Waste Minimization Benefits
While many facilities may implement end-of-pipe controls such as
fabric

[[Page 53021]]

filters and high-energy scrubbers to achieve MACT control, emission
reductions may also be accomplished by reducing the volume or toxicity
of wastes currently combusted. In addition, generators may also
consider waste management alternatives such as solvent recycling. For
purposes of this analysis, these types of responses will be referred to
as ``waste minimization.'' This section summarizes the potential waste
minimization benefits resulting from implementation of today's rule.
As today's rule is implemented, the costs of burning hazardous
waste will increase, resulting in market incentives for greater waste
minimization. To predict the quantity of waste that could be
reallocated from combustion to waste minimization due to economic
considerations, we conducted a comprehensive waste minimization
analysis that considered in-process recycling, out-of-process
recycling, and source reduction. The objective of the analysis was to
predict the quantity of hazardous wastes that may be reallocated to
these waste minimization alternatives under different combustion price
increase scenarios.
Overall, the analysis shows that a variety of waste minimization
alternatives are available for managing those hazardous waste streams
that are currently combusted. The quantity projected to be reallocated
from combustion to waste minimization alternatives, however, depends
upon the expected price increase for combustion services. At potential
price increases ranging from $10 to $20 per ton, as much as 240,000
tons of hazardous waste may be reallocated from combustion to waste
minimization alternatives. This represents approximately 7 percent of
the total quantity of hazardous waste currently combusted.

VI. What Considerations Were Given to Issues Like Equity and Children's
Health?

By applicable statute and executive order, we are required to
complete an analysis of today's rule with regard to equity
considerations and other regulatory concerns. This section assesses the
potential impacts of today's rule as it relates to environmental
justice, children's health issues, and unfunded federal mandates. Small
entity impacts are examined in a separate section.
A. Executive Order 12898, ``Federal Actions To Address Environmental
Justice in Minority Populations and Low-Income Populations'' (February
11, 1994)
This Order is designed to address the environmental and human
health conditions of minority and low-income populations. To comply
with the Executive Order, we have assessed whether today's rule may
have disproportionate effects on minority populations or low-income
populations. We have analyzed demographic data presented in the reports
``Race, Ethnicity, and Poverty Status of the Populations Living Near
Cement Plants in the United States'' (EPA, August 1994) and ``Race,
Ethnicity, and Poverty Status of the Populations Living Near Hazardous
Waste Incinerators in the United States'' (EPA, October 1994). These
reports examine the number of low-income and minority individuals
living near a relatively large sample of cement kilns and hazardous
waste incinerators and provide county, state, and national population
percentages for various sub-populations. The demographic data in these
reports provide several important findings when examined in conjunction
with the risk reductions projected from today's rule.
We find that combustion facilities, in general, are not located in
areas with disproportionately high minority and low-income populations.
However, there is evidence that hazardous waste burning cement kilns
are somewhat more likely to be located in areas that have relatively
higher low-income populations. Furthermore, there are a small number of
commercial hazardous waste incinerators located in highly urbanized
areas where there is a disproportionately high concentration of
minorities and low-income populations within one and five mile radii.
The reduced emissions at these facilities due to today's rule could
represent meaningful environmental and health improvements for these
populations. Overall, today's rule should not result in any adverse
environmental or health effects on minority or low-income populations.
Any impacts on these populations are likely to be positive due to the
reduction in emissions from combustion facilities near minority and
low-income population groups. The Assessment document available in the
RCRA docket established for today's rule presents the full
Environmental Justice Analysis.
B. Executive Order 13045: Protection of Children From Environmental
Health Risks and Safety Risks (62 FR 19885, April 23, 1997)
Executive Order 13045: ``Protection of Children from Environmental
Health Risks and Safety Risks'' (62 FR 19885, April 23, 1997) applies
to any rule that: (1) Is determined to be ``economically significant''
as defined under E.O. 12866, and (2) concerns an environmental health
or safety risk that EPA has reason to believe may have a
disproportionate effect on children. If the regulatory action meets
both criteria, the Agency must evaluate the environmental health or
safety effects of the planned rule on children, and explain why the
planned regulation is preferable to other potentially effective and
reasonably feasible alternatives considered by the Agency.
Today's final rule is not subject to the Executive Order because it
is not economically significant as defined under point one of the
Order, and because the Agency does not have reason to believe the
environmental health or safety risks addressed by this action present a
disproportionate risk to children.
The topic of environmental threats to children's health is growing
in regulatory importance as scientists, policy makers, and village
members continue to recognize the extent to which children are
particularly vulnerable to environmental hazards. Recent EPA actions
including today's rule, are in the forefront of addressing
environmental threats to the health of children. The risk assessment
conducted in support of today's rule indicates that children are the
beneficiaries of much of the reduction in potential illnesses and other
adverse effects associated with combustion facility emissions. The risk
assessment used a multi-pathway and multi-constituent evaluation in
order to examine potential effects of combined exposures on children.
Setting environmental standards that address combined exposures and
that are protective of the heightened risks faced by children are both
goals named within EPA's ``National Agenda to Protect Children's Health
from Environmental Threats.'' Areas for potential reductions in risks
and related health effects that were identified by the risk assessment
are all targeted as priority issues within EPA's September 1996 report,
Environmental Health Threats to Children.
A few significant physiological characteristics are largely
responsible for children's increased susceptibility to

[[Page 53022]]

environmental hazards. First, children eat proportionately more food,
drink proportionately more fluids, and breathe more air per pound of
body weight than do adults. As a result, children potentially
experience greater levels of exposure to environmental threats than do
adults. Second, because children's bodies are still in the process of
development, their immune systems, neurological systems, and other
immature organs can be more easily and considerably affected by
environmental hazards. The connection between these physical
characteristics and children's susceptibility to environmental threats
are reflected in the higher baseline risk levels for children living
near hazardous waste combustion facilities. The risk assessment
addresses threats to children's health associated with hazardous waste
combustion by evaluating reductions in risk for children as well as for
adults and the population overall. For all exposed sub-populations, the
assessment evaluated risks to four different age groups: 0 to 5 years,
6 to 11 years, 12 to 19 years, and adults over 20 years. Where
possible, the risk assessment has provided both population and
individual risk results for children. Both cancer and noncancer risks
are examined across the age groups of children, focusing on the most
susceptible sub-populations. The combined effects of several
carcinogens, one of the goals named within the Agency's ``National
Agenda to Protect Children's Health from Environmental Threats,'' were
examined.
The key findings from the risk assessment indicate that children do
not face significant cancer risks from hazardous waste combustion
emissions. Only in the case of children of subsistence farmers do
baseline cancer risks exceed 1 x 10^-5 for the most highly
exposed children. Implementation of the final standards would reduce
these risks below levels of concern ³⁵².
---------------------------------------------------------------------------

\352\ Also, the analysis used the same approach to estimate
cancer risks in both adults and children. However, individuals
exposed to carcinogens in the first few years of life may be at
increased risk of developing cancer. For this reason, we recognize
that significant uncertainties and unknowns exist regarding the
estimation of lifetime cancer risks in children. We also note that
this analysis of cancer risks in children has not been externally
peer reviewed.
---------------------------------------------------------------------------

The analysis also found that much of the noncancer risk reductions
resulting from implementation of today's rule may benefit children
specifically. These are projected as a result of lower exposures to
mercury, lead, and particulate matter, three types of pollutants
addressed in the noncancer risk reductions which primarily affect
children. Mercury emission reductions may reduce risks of developmental
abnormalities in potential future offspring of recreational anglers and
subsistence fishermen. In addition, particulate matter reductions may
prevent some asthma attacks affecting children, but these benefits have
not been quantified. Finally, reduced lead exposures for children are
expected from today's rule. This benefit may help prevent cognitive and
nervous system developmental abnormalities for children of the most
highly exposed sub-populations, including subsistence fishermen and
beef and dairy farmers. Analytical and data limitations prevented
reasonable monetization of these findings.
C. Unfunded Mandates Reform Act of 1995 (UMRA) (Pub. L. 104-4)
Executive Order 12875, ``Enhancing the Intergovernmental
Partnership'' (October 26, 1993), calls on federal agencies to provide
a statement supporting the need to issue any regulation containing an
unfunded federal mandate and describing prior consultation with
representatives of affected state, local, and tribal governments.
Signed into law on March 22, 1995, the Unfunded Mandates Reform Act
(UMRA) supersedes Executive Order 12875, reiterating the previously
established directives while also imposing additional requirements for
federal agencies issuing any regulation containing an unfunded mandate.
Today's rule is not subject to the requirements of sections 202,
204 and 205 of UMRA. In general, a rule is subject to the requirements
of these sections if it contains ``Federal mandates'' that may result
in the expenditure by State, local, and tribal governments, in the
aggregate, or by the private sector, of $100 million or more in any one
year. Today's final rule does not result in $100 million or more in
expenditures. The aggregate annualized social costs for today's rule
are projected to range from $50 to $63 million under the final
standards.
For rules that are subject to the requirements of these sections,
key requirements include a written statement with an analysis of
benefits and costs; input from state, local and tribal governments; and
selection of the least burdensome option (if allowed by law) or an
explanation for the option selected. We recognize the potential for
aggregate one-time capital expenditures to exceed $100 million in any
one year should various industry sectors choose not to amortize capital
expenditures. Under this scenario, the Assessment document for today's
rule meets analytical requirements established under UMRA.
Today's rule is not subject to the requirements of section 203 of
UMRA. Section 203 requires agencies to develop a small government
Agency plan before establishing any regulatory requirements that may
significantly or uniquely affect small governments, including tribal
governments. EPA has determined that this rule will not significantly
or uniquely affect small governments. The small entity impacts
analysis, presented in Appendix G of the final Assessment, found that
no hazardous waste combustion units are owned by small governments.
Finally, because we are issuing today's rule under the statutory
authority of the Clean Air Act, the rule should be exempt from all
relevant requirements of the UMRA. In addition, compliance with the
rule is voluntary for nonfederal governmental entities since state and
local agencies choose whether or not to apply to EPA for the permitting
authority necessary to implement today's rule.

VII. Is Today's Rule Cost Effective?

We have developed a cost-effectiveness measure that examines cost
per unit reduction of emissions for each hazardous air pollutant,
pollutant group, or surrogate. Cost-effectiveness measures are useful
for comparing across different air pollution regulations. Moreover, we
have typically used cost-effectiveness measures (defined as ``dollar-
per-unit of pollutant removed'') to assess the decision to go beyond-
the-floor for MACT standards.
Developing cost-effectiveness estimates for individual air
pollutants assists us in making beyond-the-floor decisions for
individual pollutants. The two analytic components of the individual
cost-effectiveness analysis are: (1) Estimates of emission control
expenditures per air pollutant for each regulatory option, and (2)
estimates of emission reductions under each regulatory option.
Individual cost-effectiveness measures for each MACT option are
calculated as follows:
HWC MACT Floor--Costs and emission reductions are
incremental to the baseline,
HWC MACT Final Standards--Costs and emission reductions
are incremental to the MACT Floor, and
Beyond-the-Floor--Activated Carbon Injection (ACI) MACT--
Costs and emission reductions are incremental to the MACT Floor.
Single-level cost-effectiveness results across all HWC MACT options
range

[[Page 53023]]

from seven hundred dollars to $34.3 million per megagram reduced for
all pollutants, individually, except dioxin. Dioxin control ranges from
$25,000 to $903,000 per gram reduced. Dioxin control for incinerators
to meet the floor standard is estimated at $903,000 per gram, with an
additional $368,000 per gram to go from the floor to the final BTF TEQ
standard. The control of SVM emitted from cement kilns is estimated to
cost $67,000 per megagram from the baseline to the floor. Moving from
the floor standard to the final BTF SVM standard for cement kilns is
estimated to cost $502,000 per megagram. These results indicate that
the more highly toxic pollutants such as dioxin are often much more
expensive to control on a per-gram basis.
We did not apply cost-effectiveness alone in establishing beyond-
the-floor levels for selected constituents regulated under the final
HWC MACT standards. Several other measurement factors were incorporated
into the beyond-the-floor decision, including: health benefits
(especially those for children), regulatory precedent, cost-
effectiveness of other MACT standards, and reliability of baseline
data.
The method for calculating cost-effectiveness makes several
simplifying assumptions. The two most important address the metrics
employed for measuring cost-effectiveness and the actual methodology
used to estimate the cost and emission reduction figures. Alternative
measurement criteria for different constituents may lead to perceived
distortions in scope. The cost-effectiveness methodology assumes that
all facilities continue operating and install pollution control
equipment or implement feed reductions to comply with the MACT
standards. Both of these limiting assumptions may lead to overstatement
or understatement of results. Other limitations that will influence
these cost-effectiveness estimates include: (1) The feed control
costing approach, which may lead to the overstatement of expenditures
per pollutant due to the assumption of upper-bound cost estimates, (2)
apportionment of costs, which are currently assigned according to the
percentage reduction required to meet the standard for each pollutant
controlled by the device, and (3) the assumption that units control
emissions to the 70 percent design level.

VIII. How Do the Costs of Today's Rule Compare to the Benefits?

Comparing overall costs and benefits may help provide an assessment
of this rule's overall efficiency and impacts on society. This section
compares the total social costs of today's rule with its total
monetized and nonmonetized benefits. The total annual monetized
benefits of today's rule are estimated at $19.2 million (undiscounted)
for the recommended final standards. These monetized benefits, however,
may represent only a subset of potential avoided health effects, both
cancer and noncancer cases. In comparison, the total annualized social
costs of the rule are projected to range from $50 to $63 million.
Social costs also include government administrative costs.
Across regulatory options, costs exceed monetized benefits more
than two-fold. However, today's rule is expected to provide benefits
that cannot be readily expressed in monetary terms. These benefits
include health benefits to sensitive sub-populations such as
subsistence anglers and improvements to terrestrial and aquatic
ecological systems. When these benefits are taken into account, along
with equity-enhancing effects such as environmental justice and impacts
on children's health, the benefit-cost comparison becomes more complex
but also more favorable. Consequently, the final regulatory decision
becomes a policy judgment which takes into account efficiency as well
as equity concerns and the positive direction of real, but
unquantifiable, benefits.

IX. What Consideration Was Given to Small Businesses?

A. Regulatory Flexibility Act (RFA) as amended by the Small Business
Regulatory Enforcement Fairness Act of 1996 (SBREFA), 5 USC 601 et seq.
This Act generally requires an agency to prepare a regulatory
flexibility analysis of any rule subject to notice and comment
rulemaking requirements under the Administrative Procedure Act or any
other statute unless the agency certifies that the rule will not have a
significant economic impact on a substantial number of small entities.
Small entities include small businesses, small not-for-profit
enterprises, and small governmental jurisdictions.
We have determined that hazardous waste combustion facilities are
not owned by small entities (local governments, tribes, etc.) other
than businesses. Therefore, only businesses were analyzed. For the
purposes of the impact analyses, small entity is defined either by the
number of employees or by the dollar amount of sales. The level at
which a business is considered small is determined for each Standard
Industrial Classification (SIC) code by the Small Business
Administration.³⁵³
---------------------------------------------------------------------------

\353\ SIC codes are used rather than the new NAICS codes because
waste generator, blender, and combustor data were only available
according to SIC code. However, a general conversion table
containing NAICS codes for each reported SIC code is presented in
the Assessment document.
---------------------------------------------------------------------------

Affected individual waste combustors (incinerators, cement kilns,
and lightweight aggregate kilns) will bear the impacts of today's rule.
These units will incur direct economic impacts as a result of today's
rule. While not required under the Act and guidelines, we have also
examined potential secondary impacts on small business units
potentially affected by today's rule, such as hazardous waste
generators and fuel blenders. Although hazardous waste combustors are
the only group that would bear direct economic impacts from today's
rule, this ``secondary impacts'' analysis was conducted because we
assume that some portion of the burden would be passed on to customers
of combustion facilities through price increases. This section
describes the small entity analysis we conducted in support of today's
rule.
B. Analytical Methodology
For combustors and blenders, we conducted facility-by-facility
analyses of small businesses. We examined company data on employment
and sales and then compared these data to statutory small business
thresholds based on employment or annual sales, as defined for its
industry by the Small Business Administration in 13 CFR part 121.
Combustion or blender units where the facility or parent company data
fell below the small business thresholds were classified as small
businesses. The analysis was more complex for generators, however,
because the rule may indirectly affect more than 11,000 generators.
Given the large number of generators who would be affected by today's
rule, it was necessary to conduct an initial, broad screening analysis
to identify small business generators that might face significant
secondary impacts. This screening analysis involved assigning each
facility to an industry group, identifying industry groups that are
dominated by small businesses, and then assuming that all generators in
those small business dominated industries are small. Further analyses
were then conducted on these groups or specific facilities.
We next compiled compliance cost data in an effort to establish a
threshold for measuring ``significant economic impact.'' This threshold
was set where compliance costs exceed one percent of

[[Page 53024]]

facility gross sales. If costs do not exceed one percent of sales, then
the regulation is unlikely to have a significant economic impact on
small businesses within the category examined. Finally, we examined
whether the significant economic impact (if any) would be borne by a
``substantial number'' of small businesses. If the regulation results
in required compliance costs exceeding one percent of gross sales for
more than 100 small businesses or 20 percent of all small businesses
within the industry category examined, then the ``substantial number''
threshold is exceeded.
The cost of compliance with the new standards will determine the
severity of impacts on small businesses. The costs to combustors used
in this analysis coincide with the 70 percent engineering standard
analyzed in the full economic assessment. The price increases
experienced by generators and blenders were calculated on a per ton
basis of waste shipped using 25 and 75 percent price pass-through
scenarios. The price impacts were assumed to be uniform across facility
types, with both generators and blenders experiencing the price pass-
through effect. In practice, this pass through would likely be split
between the two, depending on market factors. Note that the impacts
from these price increases are indirect effects, as only hazardous
waste combustors bear direct economic impact of today's rule.
C. Results--Direct Impacts
Only six facilities, out of the total universe of 172 hazardous
waste combustion facilities, met the definition of small businesses. Of
these six, two were found to experience annual compliance costs
exceeding one percent of sales. Both of these facilities are owned by a
common parent that qualifies as a small business. Therefore, this final
rule affects a very limited number of small business combustors and has
effects of greater than one percent on only two of these facilities
(one business).
While the significant economic impact threshold was exceeded for
two facilities (one corporation), these impacts do not extend to a
substantial number of small entities. With just two facilities
exceeding the one percent threshold, neither a substantial number of
facilities nor a substantial fraction of an affected industry would
face these impacts. After considering the economic impacts of today's
final rule on small entities, I certify that this action will not have
a significant economic impact on a substantial number of small
entities.
Although this final rule will not have a significant economic
impact on a substantial number of directly impacted small entities, EPA
nonetheless has assessed the potential of this rule to adversely impact
small entities subject to the rule.
D. Results--Indirect Impacts
Direct impacts of the rule extend only to combustors of hazardous
waste. To supplement our analysis, indirect impacts on generators and
blenders were also examined. We understand that some portion of the
combustor's compliance costs would most likely be passed on to
generators and blenders, and we have made an effort to analyze these
impacts in the spirit of the legislation.
We found that indirect economic effects on generators would not
impose a significant impact on a substantial number of small
generators. Under both price pass-through scenarios (25 and 75
percent), some generators exceeded the one percent cost as percentage
of sales threshold for ``significant impacts.'' In no case, however,
was the ``substantial number'' threshold exceeded. Under the 25 percent
pass-through scenario, 18 generators had a cost as percentage of sales
greater than one percent, but that accounts for only 0.85 percent of
all small business generators. While the impact threshold was exceeded
by 58 generators in the 75 percent pass through scenario, this is still
less than the 100 entity threshold established for a substantial
number. You should note that the sales thresholds were selected
conservatively as the average sales for the smallest establishments in
the SIC code.
Like generators, blenders do not incur direct costs as a result of
the rule. However, they may bear a portion of its impact indirectly as
costs are passed through from combustors. A total of 21 small business
blenders were identified. Depending on the pass-through assumption,
between six and 14 blenders exceed the significant impact threshold.
Impacts for some of these facilities were found to represent a
significant share of their annual gross sales.
Under the 25 percent price pass-through scenario, the number of
blenders exceeding the cost as percentage of sales threshold do not
represent a substantial number of facilities, either in absolute number
or as a percentage of total blenders. Under the 75 percent scenario,
however, the 14 establishments with cost as percentage of sales greater
than one percent represent just over 20 percent of the 67 blenders
identified for this analysis. In a few cases, the cost as percentage of
sales could exceed 10 percent.
E. Key Assumptions and Limitations
This analysis was based on several simplifying assumptions. Four
key assumptions may have the most significant impact on findings.
First, not all small generators may be captured in our analysis of
small business dominated industries. This exclusion may be offset by
the fact that some generators who are not small may be incorporated in
the small business dominated industries. Second, to calculate the
benchmark sales for generators, we used average sales by four-digit SIC
code for firms with fewer than 20 employees. This may understate
economic impacts for the smallest firms in the industry while
overstating impacts for larger firms. Third, compliance costs were
assumed to be passed through almost completely to the shipper of the
waste. This may overstate the impact on generators and blenders.
Finally, we assumed that all waste currently managed by combustion
continues to be disposed of in this manner. Impacts on combustors,
generators, and blenders may be overstated if waste minimization or
other lower cost alternatives are available.
Results from this report should also be evaluated within the
context of some key analytical limitations. For example, in recent
years there has been significant volatility in market behavior and
pricing practices in the hazardous waste combustion industry.
Furthermore, combustion prices have experienced a general downward tend
since 1985 as a result of overcapacity in the market and slow growth in
the generation of hazardous waste. Accounting for this price trend, the
increase expected under today's rule may affect generators and blenders
less significantly than anticipated. Finally, many hazardous waste
generators may be more concerned about other aspects of waste
management than with prices.

X. Were Derived Air Quality and Non-Air Impacts Considered?

The final Combustion MACT standards are projected to result in the
reallocation and diversion of relatively small amounts of hazardous
waste resulting in an unspecified increase in the level of fossil fuel
substitution. This substitution with nonhazardous waste fuel sources
may result in marginal increases in the annual number of mining and
transport injuries, in addition to potential increased emissions of
criteria pollutants (SOx, NOx, and
CO2). We recognize these

[[Page 53025]]

concerns but feel any potential non-air impacts are largely addressed
through alternative regulatory or market scenarios. First, some of the
hazardous waste reallocated from current combustors will likely be sent
to other waste-burning facilities, thereby off-setting primary or
supplementary fossil fuel usage. Even if fossil fuel burning does
increase to some degree, these SO2 and NOx
emissions are expected to be regulated under existing standards, e.g.,
criteria pollutant emissions are currently addressed by the Clean Air
Act. Finally, we find that even if fossil fuel use is increased, the
risks to miners (primarily coal miners) are voluntary risks. Miners are
compensated for these increased risks through wage premiums established
in response to market dynamics and recurrent negotiations between union
and corporate representatives.
While the primary environmental impact of the MACT standards are
improvements in air quality resulting from emissions reductions at
combustion facilities, other non-air environmental impacts also result
from the rule. Namely, use of some air pollution control equipment and
shifts in waste burning result in increased water, solid waste, and
energy impacts. We did not assess the monetary costs of these impacts
because we expect the incremental costs will be small relative to the
total compliance costs of the rule. You are requested to review the
Addendum prepared in support of today's final rule for an expanded
discussion of these impacts.

XI. The Congressional Review Act (5 U.S.C. 801 et seq., as Added by the
Small Business Regulatory Enforcement Fairness Act of 1996)

Is Today's Rule Subject to Congressional Review?
The Congressional Review Act, 5 U.S.C. 801 et seq., as added by the
Small Business Regulatory Enforcement Fairness Act of 1996, generally
provides that before a rule may take effect, the agency promulgating
the rule must submit a rule report, which includes a copy of the rule,
to each House of the Congress and to the Comptroller General of the
United States. EPA will submit a report containing this rule and other
required information to the U.S. Senate, the U.S. House of
Representatives, and the Comptroller General of the United States prior
to publication of the rule in the Federal Register. A Major rule cannot
take effect until 60 days after it is published in the Federal
Register. This action is not a ``major rule'' as defined by 5 U.S.C.
804(2). This rule will be effective September 30, 1999.

XII. Paperwork Reduction Act (PRA), 5 U.S.C. 3501-3520

How Is the Paperwork Reduction Act Considered in Today's Rule?
The Office of Management and Budget (OMB) has approved the
information collection requirements (ICR) contained in this rule under
the provisions of the Paperwork Reduction Act, 44 U.S.C. 3501 et seq.
and has assigned OMB control numbers 2050-0073 (``New and Amended RCRA
Reporting and Recordkeeping Requirements for Boilers and Industrial
Furnaces Burning Hazardous Waste'') for the RCRA provisions and 2060-
0349 (``New and Amended Reporting and Recordkeeping Requirements for
National Emissions Standards for Hazardous Air Pollutants from
Hazardous Waste Combustors'') for the CAA provisions.
EPA is required under section 112(d) of the Clean Air Act to
regulate emissions of HAPs listed in section 112(b). The requested
information is needed as part of the overall compliance and enforcement
program. The ICR requires that affected sources retain records of
parameter and emissions monitoring data at facilities for a period of
five years, which is consistent with the General Provisions to 40 CFR
part 63 and the permit requirements under 40 CFR part 70. All sources
subject to this rule will be required to obtain operating permits
either through the State-approved permitting program or, if one does
not exist, in accordance with the provisions of 40 CFR part 71, when
promulgated. Section 3007(b) of RCRA and 40 CFR part 2, subpart B,
which defines EPA's general policy on the public disclosure of
information, contain provisions for confidentiality.
The public reporting burden for this collection of information for
the CAA provisions under OMB control number 2060-0349 is estimated to
average 297 hours per respondent per year for an estimated 229
respondents. The annual public reporting and record keeping burden for
collection of information is estimated to be 67,977 hours and a cost of
approximately $1.6 million. The total annualized capital costs and
total annualized operation and maintenance costs associated with these
requirements are $15,000 and nearly $1.6 million, respectively.
The estimates for RCRA provisions under OMB control number 2050-
0073 include an annual public reporting and record keeping burden
reduction for collection of information of 131,228 hours and a cost
burden reduction of $4.9 million. The reductions in total annualized
capital costs and total annualized operation and maintenance costs
associated with these requirements are $2.1 million and $2.8 million,
respectively. The negative cost represents the reduced burden on 25
facilities getting out of the hazardous waste combustor universe due to
the comparable fuels exemption. A further reduction in this RCRA
information collection requirement burden will occur after three years
when the combustors will start reporting under the CAA information
collection requirements.
Burden means the total time, effort, or financial resources
expended by persons to generate, maintain, retain, or disclose or
provide information to or for a Federal agency. This includes the time
needed to review instructions; develop, acquire, install, and utilize
technology and systems for the purposes of collecting, validating, and
verifying information, processing and maintaining information, and
disclosing and providing information; adjust the existing ways to
comply with any previously applicable instructions and requirements;
train personnel to be able to respond to a collection of information;
search data sources; complete and review the collection of information;
and transmit or otherwise disclose the information.
An Agency may not conduct or sponsor, and a person is not required
to respond to a collection of information unless it displays a
currently valid OMB control number. The OMB control numbers for EPA's
regulations are listed in 40 CFR part 9 and 48 CFR Chapter 15. EPA is
amending the table in 40 CFR part 9 of currently approved ICR control
numbers issued by OMB for various regulations to list the information
requirements contained in this final rule.

XIII. National Technology Transfer and Advancement Act of 1995 (Pub L.
104-113, Sec. 12(d) (15 U.S.C. 272 Note)

Was the National Technology Transfer and Advancement Act Considered?
The rulemaking involves technical standards. Therefore, EPA
conducted a search to identify potentially applicable voluntary
consensus standards (VCS). However, we identified no such standards,
and none were brought to our attention in the comments, that would
ensure consistency throughout the regulated community. Our response-to-
comments document discusses this determination. Therefore, we have
decided to use the Air Methods contained in part 60, appendix A.

[[Page 53026]]

As noted in the proposed rule, the National Technology Transfer and
Advancement Act of 1995 (NTTAA) directs EPA to use voluntary consensus
standards in its regulatory activities unless to do so would be
inconsistent with applicable law or otherwise impractical. Voluntary
consensus standards are technical standards (e.g., materials
specifications, test methods, sampling procedures, and business
practices) that are developed or adopted by voluntary consensus
standards bodies. The NTTAA directs EPA to provide Congress, through
OMB, explanations when the Agency decides not to use available and
applicable voluntary consensus standards.
In the proposal, we discussed the manual emission test methods that
would be required for emission tests and calibration of continuous
emission monitors and relied heavily on the BIF methods in 40 CFR part
266, appendix IX. On December 30, 1997, we published a NODA which in
part questioned whether the task of determining the appropriate manual
method tests to be used for compliance should be simplified. The stack
sampling and analysis methods for hazardous waste combustors are under
the current BIF and incinerator rules for compliance tests (with a few
exceptions) that are located in SW-846. For compliance with the New
Source Performance Standard and other air rules, methods are located in
40 CFR part 60, appendix A. Potentially, you could be required to
perform two identical tests, one for compliance with MACT or RCRA and
one for compliance with other air rules, using identical test methods
simply because one method is an ``SW-846'' method and the other an
``air method.'' Further, the NODA stated that stack test methods
hazardous waste combustors use for compliance should be found in one
place to facilitate compliance. Therefore, we stated our intention to
reference 40 CFR part 60, appendix A (Except for dioxin/furans, where
we stated method 0023A of SW-846.), when it requires a specific stack-
sampling test method.
Since the time of the proposal, we instituted the ``Performance-
Based Measurement System.'' This system identifies performance related
criteria that can be used to evaluate alternative methods. Methods
determined to contain criteria or are a ``Methods-Based Parameters''
method are required, and are the only methods that can be used for
regulatory tests.
Commenters generally supported use of the Air Methods contained in
part 60, appendix A, or their ``SW-846'' equivalent. Furthermore,
because these methods were used to establish the final standards
contained in today's rulemaking, application of non approved methods
would result in unreliable and inconsistent measurements. Therefore,
today's rule will require the use of the Air Methods contained in part
60, appendix A. Section 63.7 describes procedures for the use of
alternative test methods for MACT sources. This procedure involves
using Method 301 of part 63, appendix A, to validate an alternate test
method and submitting the data to us. We then decide if the proposed
method is acceptable. Absent this approval under Sec. 63.7 procedures,
alternate methods cannot be used.
Today's rule, by requiring the use of only part 60, appendix A
methods (method 0023A of SW-846 for dioxin/furans) for compliance
determinations and particulate matter continuous emission monitor
correlations, would maintain national consistency with the selection of
specific manual stack sampling methods. We have determined that this
approach would facilitate ease of implementation with today's ``self
implementing'' MACT rule. Again, alternate methods may be approved by
the Administrator via the provisions of Sec. 63.7(f) and part Sec. 63,
appendix A, Method 301, Field Validation or Pollutant Measurement
Methods from Various Waste Media.

XIV. Executive Order 13084: Consultation and Coordination With Indian
Tribal Governments (63 FR 27655)

Were Tribal Government Issues Considered?
The requirements of section 3(b) of Executive Order 13084 do not
apply to this rule. They apply to rules that are not required by
statute, that significantly or uniquely affect the communities of
Indian tribal governments, and that impose substantial direct
compliance costs on those communities. EPA cannot issue those rules
unless the Federal government provides the funds necessary to pay the
direct compliance costs incurred by the tribal governments, or EPA
consults with those governments and gives required information to OMB.
But today's rule does not significantly or uniquely affect the
communities of Indian tribal governments.
For many of the same reasons described in the Unfunded Mandates
Reform Act discussion (section VI.C above), the requirements of
Executive Order 13084 do not apply to today's rule. Promulgation of
today's rule is under the statutory authority of the CAA. Also, while
Executive Order 13084 does not provide a specific gauge for determining
whether a regulation ``significantly or uniquely affects'' an Indian
tribal government, today's rule does not impose substantial direct
compliance costs on tribal governments and their communities. Tribal
communities are not predominantly located near hazardous waste
combustion facilities, when compared with other communities throughout
the nation. Finally, tribal governments will not be required to assume
any permitting responsibilities associated with this final rule because
permitting authority is voluntary for nonfederal government entities.
Shortly after forming the regulatory workgroup for this rulemaking
in April 1994, we looked for ways to obtain the input of state, local,
and tribal governments into the rulemaking process. As a result,
representatives from four State environmental agencies agreed to
participate in the workgroup. These representatives were asked to
consider the impacts of this rule of the state, local, and tribal
level. These representatives served on the workgroup until Final Agency
Review in November 1998. As members of the workgroup, they participated
in workgroup meetings and conference calls resulting in the development
of rulemaking issues and their solutions. They also provided written
comments on our work products on several occasions, including the
proposal, the May 1997 NODA, and the Final Agency Review package.
In their comments on the proposal and subsequent notices of data
availability, these representatives raised concerns over the following
issues:

--Use of site-specific risk assessments under RCRA
--Continuous emissions monitors
--Manual sampling methods
--Compliance schedule
--Use of test data to establish operating limits
--Automatic waste feed cutoffs
--Performance testing schedule
--Recordkeeping requirements
--Permitting issues
--Assessment of potential costs and benefits
--Human health benefits
--Area sources
--Notification and reporting requirements
--Protectiveness of human health as required by RCRA
--Redundant requirements
--State authorization
--Public participation
--CAAA and RCRA coordination

[[Continued on page 53027]]

Notices For
2009
2008
2007
2006
2005
2004
2003
2002
2001
2000
1999
1998
1997
1996
1995
1994

[[pp. 52977-53026]] NESHAPS: Final Standards for Hazardous Air Pollutants for

Local Navigation