[House Hearing, 112 Congress]
[From the U.S. Government Publishing Office]

QUALITY SCIENCE FOR QUALITY AIR

=======================================================================

HEARING

BEFORE THE

SUBCOMMITTEE ON ENERGY AND
ENVIRONMENT

COMMITTEE ON SCIENCE, SPACE, AND TECHNOLOGY
HOUSE OF REPRESENTATIVES

ONE HUNDRED TWELFTH CONGRESS

FIRST SESSION

__________

TUESDAY, OCTOBER 4, 2011

__________

Serial No. 112-41

__________

Printed for the use of the Committee on Science, Space, and Technology

Available via the World Wide Web: http://science.house.gov

U.S. GOVERNMENT PRINTING OFFICE
70-587 WASHINGTON : 2011
-----------------------------------------------------------------------
For sale by the Superintendent of Documents, U.S. Government Printing Office,
http://bookstore.gpo.gov. For more information, contact the GPO Customer Contact Center, U.S. Government Printing Office. Phone 202�09512�091800, or 866�09512�091800 (toll-free). E-mail, gpo@custhelp.com.

COMMITTEE ON SCIENCE, SPACE, AND TECHNOLOGY

HON. RALPH M. HALL, Texas, Chair
F. JAMES SENSENBRENNER, JR., EDDIE BERNICE JOHNSON, Texas
Wisconsin JERRY F. COSTELLO, Illinois
LAMAR S. SMITH, Texas LYNN C. WOOLSEY, California
DANA ROHRABACHER, California ZOE LOFGREN, California
ROSCOE G. BARTLETT, Maryland BRAD MILLER, North Carolina
FRANK D. LUCAS, Oklahoma DANIEL LIPINSKI, Illinois
JUDY BIGGERT, Illinois GABRIELLE GIFFORDS, Arizona
W. TODD AKIN, Missouri DONNA F. EDWARDS, Maryland
RANDY NEUGEBAUER, Texas MARCIA L. FUDGE, Ohio
MICHAEL T. McCAUL, Texas BEN R. LUJAN, New Mexico
PAUL C. BROUN, Georgia PAUL D. TONKO, New York
SANDY ADAMS, Florida JERRY McNERNEY, California
BENJAMIN QUAYLE, Arizona JOHN P. SARBANES, Maryland
CHARLES J. ``CHUCK'' FLEISCHMANN, TERRI A. SEWELL, Alabama
Tennessee FREDERICA S. WILSON, Florida
E. SCOTT RIGELL, Virginia HANSEN CLARKE, Michigan
STEVEN M. PALAZZO, Mississippi VACANCY
MO BROOKS, Alabama
ANDY HARRIS, Maryland
RANDY HULTGREN, Illinois
CHIP CRAVAACK, Minnesota
LARRY BUCSHON, Indiana
DAN BENISHEK, Michigan
VACANCY
------

Subcommittee on Energy and Environment

HON. ANDY HARRIS, Maryland, Chair
DANA ROHRABACHER, California BRAD MILLER, North Carolina
ROSCOE G. BARTLETT, Maryland LYNN C. WOOLSEY, California
FRANK D. LUCAS, Oklahoma BEN R. LUJAN, New Mexico
JUDY BIGGERT, Illinois PAUL D. TONKO, New York
W. TODD AKIN, Missouri ZOE LOFGREN, California
RANDY NEUGEBAUER, Texas JERRY McNERNEY, California
PAUL C. BROUN, Georgia
CHARLES J. ``CHUCK'' FLEISCHMANN,
Tennessee
RALPH M. HALL, Texas EDDIE BERNICE JOHNSON, Texas

C O N T E N T S

Tuesday, October 4, 2011

Page
Witness List..................................................... 2

Hearing Charter.................................................. 3

Opening Statements

Statement by Representative Andy Harris, Chairman, Subcommittee
on Energy and Environment, Committee on Science, Space, and
Technology, U.S. House of Representatives...................... 6
Written Statement............................................ 7

Statement by Representative Brad Miller, Ranking Minority Member,
Subcommittee on Energy and Environment, Committee on Science,
Space, and Technology, U.S. House of Representatives........... 8
Written Statement............................................ 10

Witnesses:

Dr. Roger O. McClellan, Advisor, Toxicology and Human Health Risk
Analysis
Oral Statement............................................... 12
Written Statement............................................ 14

Dr. George Thurston, Professor, New York University School of
Medicine
Oral Statement............................................... 19
Written Statement............................................ 20

Dr. Michael Honeycutt, Chief Toxicologist, Texas Commission on
Environmental Quality
Oral Statement............................................... 26
Written Statement............................................ 27

Dr. Robert F. Phalen, Professor of Medicine, and Co-Director, Air
Pollution Health Effects Laboratory, University of California,
Irvine
Oral Statement............................................... 36
Written Statement............................................ 38

Dr. Anne E. Smith, Senior Vice President, Nera Economic
Consulting
Oral Statement............................................... 39
Written Statement............................................ 40

Mr. J. Edward Cichanowicz, Consultant
Oral Statement............................................... 51
Written Statement............................................ 53

Discussion 61

Appendix 1: Answers to Post-Hearing Questions

Dr. Roger O. McClellan, Advisor, Toxicology and Human Health Risk
Analysis....................................................... 82

Dr. George Thurston, Professor, New York University School of
Medicine....................................................... 92

Dr. Michael Honeycutt, Chief Toxicologist, Texas Commission on
Environmental Quality.......................................... 96

Dr. Robert F. Phalen, Professor of Medicine, and Co-Director, Air
Pollution Health Effects Laboratory, University of California,
Irvine......................................................... 110

Dr. Anne E. Smith, Senior Vice President, Nera Economic
Consulting..................................................... 112

Mr. J. Edward Cichanowicz, Consultant............................ 118

Appendix 2: Additional Material for the Record

Reprint of article by Dr. Roger O. McClellan, Advisor, Toxicology
and Human Health Risk Analysis: "Role of Science and Judgment
in Setting National Ambient Air Quality Standards: How Low Is
Low Enough?", Air Quality and Atmospheric Health (published
online 01 June 2011)........................................... 122

Letter from Dr. Roger O. McClellan, DVM, MMS, Dsc (Honorary) to
Honorable Lisa P. Jackson, Administrator, U.S. Environmental
Protection Agency, May 6, 2011................................. 138

Memo from Honorable Cass R. Sunstein, Administrator, Office of
Management and Budget, Office of Information and Regulatory
Affairs, September 2, 2011..................................... 143

QUALITY SCIENCE FOR QUALITY AIR

----------

TUESDAY, OCTOBER 4, 2011

House of Representatives,
Subcommittee on Energy and Environment,
Committee on Science, Space, and Technology,
Washington, DC.

The Subcommittee met, pursuant to call, at 10:04 a.m., in
Room 2318 of the Rayburn House Office Building, Hon. Andy
Harris [Chairman of the Subcommittee] presiding.

hearing charter

COMMITTEE ON SCIENCE, SPACE, AND TECHNOLOGY

SUBCOMMITTEE ON ENERGY AND ENVIRONMENT

U.S. HOUSE OF REPRESENTATIVES

Quality Science for Quality Air

tuesday, october 4, 2011
10:00 a.m.-12:00 p.m.
2318 rayburn house office building

Purpose

On Tuesday, October 4, 2011, the Subcommittee on Energy and
Environment of the Committee on Science, Space, and Technology held a
hearing to examine the Environmental Protection Agency's (EPA) process
for setting standards under the Clean Air Act including (1) the role of
scientific advice from the Clean Air Scientific Advisory Committee
(CASAC) and similar bodies, (2) the economic underpinnings of EPA's
Regulatory Impact Analyses, and (3) the assumptions, models, and data
used in projecting compliance, technological standards necessary to
achieve compliance, and environmental benefits associated with proposed
and finalized rules.

Witnesses

Dr. Roger O. McClellan, Advisor, Toxicology and Human
Health Risk Analysis

Dr. George Thurston, Professor, New York University
School of Medicine

Dr. Michael Honeycutt, Chief Toxicologist, Texas
Commission on Environmental Quality

Dr. Robert F. Phalen, Professor of Medicine and Co-
Director, Air Pollution Health Effects Laboratory, University of
California, Irvine

Dr. Anne E. Smith, Senior Vice President, NERA Economic
Consulting

Mr. J. Edward Cichanowicz, Consultant.

Background

Originally passed in 1963, the Clean Air Act underwent significant
amendments in 1970, 1977, and 1990. The CAA provided the EPA the
statutory authority to regulate air pollution to address public health
and welfare concerns. Under the CAA statutory framework, the Agency is
required to set goals of reducing emissions from both stationary and
mobile sources.

National Ambient Air Quality Standards

The foundation of the CAA is based primarily on the concept of
nationwide air quality goals and the development of individual state
plans to meet those goals. EPA has identified six ``criteria
pollutants'' that are most prevalent and necessary to the protection of
public health and welfare for National Ambient Air Quality Standards
(NAAQS): sulfur dioxide (SO2), particulate matter (PM), \1\ nitrogen
oxides (NOx), carbon monoxide (CO), ozone (O3), and lead (Pb). For each
of these pollutants, EPA established a ``primary'' standard at a level
designed to protect the public health within an ``adequate margin of
safety.'' In addition, the statute allows EPA to set a secondary NAAQS
to protect public welfare. At this point, EPA has not set secondary
standards at different levels than the primary standards.
---------------------------------------------------------------------------
\1\ For the first time, during the 1997 revision of the PM NAAQS,
EPA established separate standards for fine particulate matter (smaller
than 2.5 micrometers or PM2.5) and coarse particulate matter (smaller
than 10 micrometers or PM10).
---------------------------------------------------------------------------
The standards themselves are not directly enforceable. Rather,
NAAQS establish ceilings for concentrations of criteria pollutants in
ambient air. States are required to develop their own State
Implementation Plans (SIPs) which outline the measures the State will
take to meet the reduction required by the standard (attain) or stay in
compliance with the standard (maintain). For example, a SIP may include
emission limits for power plants, refineries and manufacturing
facilities within the state, or fuel specifications for emission
reductions from mobile sources. SIPs must be approved by EPA. If EPA
determines that a SIP will not be able to attain or maintain the NAAQS
concentrations, EPA can require States to abide by a Federal
Implementation Plan (FIP) until such time that the State develops an
approvable SIP. Further, if a State fails to submit a SIP, fails to
submit an adequate SIP, or fails to implement a SIP, certain sanctions
may be imposed; for example, the State may be banned from receiving
Federal highway grants.
Under the CAA, each NAAQS must go through a review every five years
in order to ensure the standards were protecting public health
according to the most recent scientific findings. After a scientific
assessment and receipt of expert advice, the Administrator uses his or
her own judgment to determine whether or not and to what extent an
NAAQS is to be revised. Several Supreme Court cases \2\ limited the
ability of EPA to take cost into consideration when setting the NAAQS.
However, EPA still prepares a Regulatory Impact Assessment (RIA) that
details the Agency's expected costs and benefits.
---------------------------------------------------------------------------
\2\ Lead Industries Assn., Inc. v. EPA, 647 F.2d 1130, 1148 (CADC
1980) and Whitman v. American Trucking (February 2001).

---------------------------------------------------------------------------
Clean Air Scientific Advisory Committee

NAAQS reviews also include a scientific assessment phase in which
EPA assesses the scientific and technical data and provides
opportunities for public and expert review of relevant staff documents.
EPA then provides these documents to the Clean Air Scientific Advisory
Committee (CASAC) for review and feedback. CASAC typically provides the
Administrator of the EPA with a recommended concentration range for a
particular NAAQS that it believes the scientific literature justifies.
According to the EPA, CASAC ``provides independent advice to the
EPA Administrator on the technical bases for EPA's national ambient air
quality standards. Established in 1977 under the Clean Air Act (CAA)
Amendments of 1977 (see 42 U.S.C. Sec. 7409(d)(2)), CASAC also
addresses research related to air quality, sources of air pollution,
and the strategies to attain and maintain air quality standards and to
prevent significant deterioration of air quality.'' \3\
---------------------------------------------------------------------------
\3\ http://yosemite.epa.gov/sab/sabpeople.nsf/WebCommittees/CASAC.
---------------------------------------------------------------------------
In providing this advice, CASAC comments on EPA staff documents and
responds to charge questions from EPA staff. CASAC is comprised of
seven permanent members that are supplemented by more than a dozen
additional scientists that are appointed to join them for individual
NAAQS reviews.
In recent months, several Members of Congress \4\ have raised
questions regarding the objectivity and independence of the CASAC in
providing this scientific advice to EPA. \5\
---------------------------------------------------------------------------
\4\ http://republicans.energycommerce.house.gov/Media/file/
Letters/112th/030811inhofe.pdf.
\5\ http://epw.senate.gov/public/
index.cfm?FuseAction=Files.View&FileStore-id=d55fa42f-7c41-
456e-893f-2963eb26e07e.

---------------------------------------------------------------------------
Hazardous Air Pollutants

The CAA distinguishes between two types of pollutants:
aforementioned criteria pollutants (e.g., NOx, SO2, PM, etc.) and
hazardous air pollutants (HAPs). HAPs theoretically pose similar public
health concerns as criteria pollutants but are much less ubiquitous;
therefore, a different standard setting regime was established. The
National Emission Standards for Hazardous Air Pollutants (NESHAPs) was
established to deal with these nonconventional pollutants. The 1990
amendments required HAPs regulations to consider cost and technological
feasibility. Further, the statute directed EPA to develop standards by
industrial source category (e.g., acid gases) rather than focus on
individual pollutants.

Maximum Achievable Control Technology

The mandating of NESHAPs by the 1990 CAAA set the course for the
rapid development of technology-based standards for all major and
industrial source categories that emit HAPs. These standards are known
as Maximum Achievable Control Technologies, or MACT. MACT standards are
to be based on the ``maximum degree of reductions and emissions deemed
achievable for the category or subcategory, the EPA administrator,
taking into consideration the cost of achieving the reduction, any non-
air-quality health and environmental impacts and energy requirements,
determines is achievable for new or existing sources.'' \6\
---------------------------------------------------------------------------
\6\ CAA 112(d)(2).

---------------------------------------------------------------------------
Scientific Inputs for Standard Setting Under the Clean Air Act

Throughout the development of both NAAQS and NESHAP standards, EPA
is required to provide scientific justification for the regulations.
The initial inputs include information regarding the effects of
pollutants on public health and welfare. EPA must provide information
that demonstrates that criteria pollutants or HAPs within the ambient
air at current concentrations constitute a threat to public health. The
health risk is estimated through a scientific assessment, and the
public and expert advice is provided to EPA. The Court has ruled that
EPA may not take cost into account when establishing NAAQS levels,
though Executive Orders have required that such costs must still be
analyzed through a regulatory impact assessment. For MACT, EPA must
take cost into account, and may not set a standard that protects the
public health to a level that has no risk of health effects. Finally,
EPA is required for the MACT to conduct a technological feasibility
analysis to determine if the technology to reduce emissions of
pollutants is available and cost effective. Again, although EPA is not
required to conduct a similar analysis in the case of NAAQS levels, the
Agency still does a technical assessment when developing the regulatory
impact assessment.
The results of these scientific inputs: health, risk, cost and
technology, provide the basis and necessary justification for EPA to
move forward with setting a standard or making an existing standard
more stringent. The Science, Space, and Technology Committee will
examine the process by which the quality of the scientific inputs
affects the overall justification for regulation, and the importance of
that process in ensuring that only appropriate and necessary rules are
promulgated.

Relevant Current Proposed and Finalized (but under review) Rules

The following regulations pertain to the aforementioned Clean Air
Act authorities:

National Emission Standards for Hazardous Air Pollutants
from Coal- and Oil-fired Electric Utility Steam Generating Units and
Standards of Performance for Electric Utility Steam Generating Units;

National Emission Standards for Hazardous Air Pollutants
for Area Sources: Industrial, Commercial, and Institutional Boilers;

National Emission Standards for Hazardous Air Pollutants
for Major Sources: Industrial, Commercial & Institutional Boilers and
Process Heaters;

Portland Cement Manufacturing NESHAP and NSPS;

Review of the Primary National Ambient Air Quality
Standard for Sulfur Dioxide;

Review of the Primary National Ambient Air Quality
Standard for Ozone;

Review of the Primary National Ambient Air Quality
Standard for Particulate Matter;

Review of the Primary National Ambient Air Quality
Standard for Lead;

Cross-State Air Pollution Rule.
Chairman Harris. The Subcommittee on Energy and Environment
will come to order.
Good morning. Welcome to today's hearing entitled ``Quality
Science for Quality Air.'' In front of you are packets
containing the written testimony, biographies and Truth in
Testimony disclosures for today's witness panel. I recognize
myself for five minutes for an opening statement.
I thank our witnesses for being here today to provide their
expertise on the process for incorporating quality science into
Clean Air Act standards. In the debate over EPA issues, it can
often seem like two ships passing in the night: one side
talking about jobs, the other discussing children's health.
This hearing is designed to provide context to this
conversation and to examine the science and technology
assumptions behind air quality standards.
It is important to note at the outset that overall air
quality in the United States is excellent. By any objective
metric, air quality and related human health has improved
dramatically, and the levels of every major air pollutant have
plummeted over the last three decades. Most of America meets
increasingly stringent EPA standards.
Despite these improvements, the unprecedented pace of EPA's
Clean Air Act agenda requires us to ask two basic questions:
are we using common sense in establishing environmental
standards, and how low is low enough?
Unfortunately, whether it is the Cross-State Air Pollution
Rule, National Ambient Air Quality Standards for ozone or fine
particulate matter, or the so-called utility MACT rules, these
questions are being ignored and EPA is moving ahead
promulgating major, job-destroying regulations on the basis of
shaky, and often secret, science.
Now, both as a physician and the Chairman of a Subcommittee
overseeing what is supposed to be science at EPA, I was alarmed
to hear Administrator Lisa Jackson explain two weeks ago that
particulate matter ``does not make you sick. It is directly
causal to dying sooner than you should'' and that ``if we could
reduce particulate matter to healthy levels, it would have the
same impact as finding a cure for cancer.''
Now two weeks ago, Assistant Administrator Gina McCarthy
had a hard time explaining how the Cross-State Air Pollution
Rule would avoid ``up to 34,000 deaths'' at a committee hearing
in this room. Given the imprecise justification for those
34,000 avoided deaths, the Administrator's claim of 572,000
avoided deaths, which is the number who die from cancer each
year, is patently ridiculous.
I would hope that, of all people, members of the
President's Cabinet would be responsible enough to ensure any
public health claims are grounded in science, not hyperbole,
and if our current air is such a threat to human health that it
is killing hundreds of thousands of people each year, I am very
interested to review the information that the Agency relies on
in establishing this relationship.
Accordingly, I have asked EPA to make the federally funded
data sets and associated science upon which these health claims
appear to be based publicly available. But because the EPA is
not transparent with the sources of their data, from what we
have seen so far, EPA seems to rely on making statistical hay
out of minor associations between pollutants and premature
mortality. This is not quality science; this is press-release
science in which public relations is considered more important
than an honest, transparent, scientific discussion of
environmental outcomes and human health.
One glaring example is the EPA's justification of these
major Clean Air Act regulations on the basis of double-counting
the health benefits of lower particulate matter levels. Without
these coincidental co-benefits, none of these rules would have
passed a simple cost-benefit analysis.
Just last week, EPA's Inspector General released a report
highlighting the agency's inability to follow basic peer review
and scientific integrity guidelines in developing its
endangerment finding on carbon dioxide. I am concerned that
similar problems plague EPA's Science Advisory Board and the
Clean Air Scientific Advisory Committee, or CASAC. We are not
seeking to denigrate the participating scientists, but there
are questions raised by the IG that need to be asked about the
independence of these bodies. In many cases, these panels
suffer from little turnover, financial conflicts, a lack of
balance and transparency, and, perhaps most importantly,
panelists that are peer reviewing their own work.
There are also a number of signs that this EPA is
underestimating the time and cost to install pollution control
technology that is required. For example, there is no power
plant in America that can meet the three requirements proposed
by EPA in the Utility MACT Rule.
I am pleased that the House of Representatives has begun
pushing back against this job-killing regulatory agenda through
legislation like the recently passed TRAIN Act, and I hope that
the recommendations of our panelists today will help guide our
oversight of EPA science going forward.
[The prepared statement of Mr. Harris follows:]

Prepared Statement of Subcommittee Chairman Andy Harris

The hearing will come to order. I thank our witnesses for being
here today to provide their expertise on the process for incorporating
quality science into Clean Air Act standards.
In the debate over EPA issues, it can often seem like two ships
passing in the night: one side talking about jobs, and the other
discussing children's health. This hearing is designed to provide
context to this conversation and to examine the science and technology
assumptions behind air quality standards.
It is important to note at the outset that overall air quality in
the United States is excellent. By any objective metric, air quality
and related human health has improved dramatically, and the levels of
every major air pollutant have plummeted over the last three decades.
Most of America meets increasingly-stringent EPA standards.
Despite these improvements, the unprecedented pace of EPA's Clean
Air Act agenda requires us to ask two basic questions: ``Are we using
common sense in establishing environmental standards?'' and ``How low
is low enough?''
Unfortunately, whether it is the Cross-State Air Pollution Rule,
National Ambient Air Quality Standards for ozone or fine particulate
matter, or the so-called utility MACT rules, these questions are being
ignored and EPA is moving ahead promulgating major, job-destroying
regulations on the basis of shaky (and often secret) science.
Both as a physician and the Chairman of a Subcommittee overseeing
what is supposed to be science at EPA, I was alarmed to hear
Administrator Lisa Jackson explain two weeks ago that particulate
matter ``does not make you sick. It is directly causal to dying sooner
than you should'' and that ``if we could reduce particulate matter to
healthy levels, it would have the same impact as finding a cure for
cancer.'' Two weeks ago, Assistant Administrator Gina McCarthy had a
hard time explaining how the Cross-State Air Pollution Rule would avoid
``up to 34,000 deaths.'' Given the imprecise justification for 34,000
avoided deaths, the Administrator's claim of 572,000 avoided deaths is
patently ridiculous.
I would hope that, of all people, members of the President's
cabinet would be responsible enough to ensure any public health claims
are grounded in science, not hyperbole, and if our current air is such
a threat to human health that it is killing hundreds of thousands of
people each year, I am very interested to review the information that
the Agency relies on in establishing this relationship. Accordingly, I
have asked EPA to make the federally-funded data sets and associated
science upon which these health claims appear to be based publicly
available.
Because the EPA is not transparent with the sources of their data,
from what we have seen so far, EPA seems to rely on making statistical
hay out of minor associations between pollutants and premature
mortality. This is not quality science; this is press release science
in which public relations is considered more important than an honest
and transparent discussion of environmental outcomes and human health.
One glaring example is EPA's justification of these major Clean Air Act
regulations on the basis of double-counting from the health benefits of
lower particulate matter levels. Without these coincidental co-
benefits, none of these rules would have passed a simple cost-benefit
analysis.
Just last week, EPA's Inspector General released a report
highlighting the Agency's inability to follow basic peer review and
scientific integrity guidelines in developing its endangerment finding
on carbon dioxide. I am concerned that similar problems plague EPA's
Science Advisory Board and the Clean Air Scientific Advisory Committee
or CASAC. We are not seeking to denigrate the participating scientists,
but there are questions raised by the IG that need to be asked about
the independence of these bodies. In many cases, these panels suffer
from little turnover, financial conflicts, a lack of balance and
transparency, and, perhaps most importantly, panelists that are peer
reviewing their own work.
There are also a number of signs that this EPA is underestimating
the time and cost to install pollution control technology that is
required. For example, there is no power plant in America that can meet
the three requirements proposed by EPA in the utility MACT rule.
I am pleased that the House of Representatives has begun pushing
back against this job-killing, regulatory agenda through legislation
like the recently-passed TRAIN Act, and I hope that the recommendations
of our panelists today will help guide our oversight of EPA science
going forward.

Chairman Harris. The chair now recognizes Mr. Miller for an
opening statement.
Mr. Miller. Thank you, Mr. Chairman. I appreciate your
holding this hearing. The EPA does have before us the need,
often court imposed, to issue new and updated regulations, and
we should review what those regulations are based on and what
we really get out of them, but the Clean Air Act's history
speaks for itself. It is not just the EPA's own estimates.
Those estimates have been broadly supported by public health
experts, neutral experts, those who truly do not have a
financial ax to grind, who are not employed by the industry,
either directly or as consultants or as experts or whatever
else. EPA's own estimates are that in the first 20 years of the
Clean Air Act, the Clean Air Act has prevented more than
200,000 premature deaths and almost 700,000 cases of chronic
bronchitis, and the benefits continue to grow and grow. The
Chairman said correctly the air quality is much better in the
United States. That is because we have had the Clean Air Act in
effect for 40 years. It has not been because of the benevolence
of industry. It has been because there has been an Act in place
that has been enforced by the Environmental Protection Agency,
in some instances, kicking and screaming, but still enforced,
enforced if not by the Agency at least by the courts requiring
the Agency to follow the law.
And the blame for these regulations for the effect on jobs
is not really something we can attribute to the industry
because there are many in the industry that support the
regulations and think that they need to be tougher than they
are, and many have advocated for that. But politicians have
learned that there is a political value in creating a villain,
creating a demon, and they have made the EPA their demon, so
there is no room between their image of the EPA as a rogue
agency that is enforcing the Clean Air Act, signed by President
Nixon, later strengthened by President Bush. So demonizing the
EPA by making specious claims that their regulations kill jobs
also without any particularly well-grounded basis while
ignoring all the benefits for public health and the economy is
another cynical effort to gain votes and get Americans to vote
against their own self-interest. Americans aren't buying it.
Polls show that Americans understand the importance of the EPA
and understand the importance of environmental protection,
support clean air and water, and do not believe that pollution
and damage to the public health is the price that we have to
pay for prosperity.
But in this Congress we have seen an unrelenting attack on
environmental protection. There have been 136 votes so far, all
to ratchet back environmental protections, the most extreme
example, the TRAIN Act, and nobody opposes transparency. I have
pushed--for four years as Chairman of the Science Committee's
Subcommittee on Oversight, I have pushed hard for transparency
and we should have that, but we also know that the arguments
about process are usually driven by folks who are not happy
with the result of the process, and if you don't like the
result, you almost never like the process.
The Inspector General did criticize, it is true, some of
the processes of the EPA but also said that the EPA's processes
that they did use complied with the statute, and they said that
there was no reason to think that the result would have been
any different if the processes had been different. I think we
should continuously improve the processes. Process does matter.
But the characterization of the Inspector General's report is
simply not correct.
So with significant progress over the last 40 years, I
think we should look at how to build on that, continuing to
improve the quality of our air, recognizing the importance of
that to a strong economy. Premature deaths cost our economy.
People having bronchitis and not being able to go to work cost
our economy. Stronger environmental protections also push
industry to develop new technologies and have in fact increased
jobs in various areas.
So we have proven time and again that worker productivity
improves better if there is clean air. Agricultural yield is
improved if there's clean air. There is a reduction in
mortality and illness and other economic and public health
benefits that far outweighs the cost of compliance. Of course,
we should consider the costs of compliance but we should also
include the effect on the public health as well.
Thank you, and I yield back.
[The prepared statement of Mr. Miller follows:]

Prepared Statement of Ranking Member Brad Miller

I want to thank Chairman Harris for holding a hearing to discuss
the science underpinning the 40-year-old landmark legislation, the
Clean Air Act. While I disagree with some of the opinions of my
colleagues and the witnesses, I understand the timing and motivations
behind this hearing. As we look forward to the EPA issuing new and
updated pollution regulations, it is worth reminding ourselves of what
they are based on and what we get out of them. In that regard, the
Clean Air Act's history of protecting public health speaks for itself.
In the four decades since it was signed, the Clean Air Act has
saved hundreds of thousands of lives. Even in its first 20 years--as
emissions reductions were just beginning--EPA figures show that the
Clean Air Act prevented more than 200,000 premature deaths and almost
700,000 cases of chronic bronchitis.
And these benefits to the public will continue to grow. The EPA
projects that by 2020, the Clean Air Act will prevent roughly 230,000
deaths, 200,000 cases of heart disease, and 2.4 million asthma flare-
ups every year. These will have a real economic effect by keeping
children and adults out of the hospital and saving the nation from 22.4
million missed school and work days per year, providing upwards of $2
trillion in economic benefits by 2020. These benefits would far exceed
the original costs by 30 to 1. That's not a bad investment by any
standard.
The Clean Air Act is hardly the economy-killer that so many claim.
Over the last 20 years, while emissions of the six principal air
pollutants were reduced by an additional 41 percent, the Nation's Gross
Domestic Product has increased by more than 64 percent. And we not only
got cleaner air, but entirely new technology sectors to boot. In fact,
GDP has risen by more than 200 percent since the Act was signed 40
years ago, and this is in spite of the doomsday prophesies of
widespread economic disruption and industrial collapse that some said
would result from environmental regulations. These claims have been
proven wrong time and again, and we should expect to look back and
regard the alarmism of today as no different.
To be fair, the blame for misguiding the public on the costs and
benefits of regulations cannot be laid solely at the feet of industry
that in the end exists to turn a profit. In fact, ``industry'' is
hardly uniform in its regard for environmental regulations, with many
industrial stakeholders clearly advocating in favor of the new
regulations. It depends on how they have invested. And even those on
the losing end of EPA's regulations are often complaining about the
process or timeline, knowing that the eventual regulation will be the
same.
Unfortunately, there is a more troubling force at play here as
politicians have discovered the distinct political value of vilifying
the EPA. To them, there is no middle ground, and no room for
negotiation or compromise with this ``rogue agency.'' In politicizing
the issue--sometimes far beyond the comfort level of the industries
they profess to champion--there also seems to be no limit to the
hysterics. In a recent hearing on the EPA's transport ruIe before this
Committee, a witness actuaIly stated that the rule will ``jeopardize
the lives of our most medically fragile citizens.''
Demonizing environmental safeguards and the EPA by making specious
claims that regulations kill jobs--and even people--while completely
ignoring the proven positive effects they have on public health and the
economy, is another cynical ploy to get Americans to vote against their
own self-interest.
Thankfully, poll after poll shows that the public believes that EPA
should protect their right to clean air and water more than they
believe that pollution is the price they must pay for economic
security.
In this Congress, it seems we have seen every assault possible on
environmental protections such as the Clean Air Act, taking 136 anti-
environmental votes in the House thus far. Before the recess, the House
passed the so-called TRAlN Act, a piece of legislation that would
derail efforts to curb emissions of dangerous pollutants such as soot,
mercury, dioxins and acid gases. Of course, nobody would argue against
having ``transparency'' in our regulatory processes, but that is not
really what these bills are about. They are about politics first, and
buying time for polluters who must otherwise be dragged kicking and
screaming into environmental compliance, while more forward-looking
firms are deprived of making a return on their investments in cleaner
technology.
Efforts such as the TRAIN Act and the paralysis-by-analysis it
would impose are themselves anything but transparent or comprehensive.
It adds another layer of bureaucracy, essentially for the purpose of
weighing industry' s cost of compliance, without considering the
benefits for public health and the creation of new jobs. It also
removes any provision to ensure that such safeguards will ever take
effect, delaying them indefinitely. It is designed to ignore the
overwhelming evidence that saving Americans' lives is far cheaper than
saving polluters' dollars.
No regulatory process is ever going to make everyone happy because
someone must always change, but we should certainly look for ways to
make the processes more efficient and transparent. Instead of making
doomsday claims that never hold up and scaring the American public into
forgoing their own rights to a cleaner environment, we need to trust in
the EPA's reliable, established scientific processes for characterizing
the effects of emissions on public health and evaluating the costs and
benefits of new technologies. We should acknowledge that we did not get
40 years of dramatic pollution reductions with strong economic growth
because EPA's scientific processes are not tried and true. Put simply,
they work.
And speaking of process, I am disappointed that we do not have
before us an EPA witness nor one from the Science Advisory Board. If
we're going to talk about them, they ought to be here to defend
themselves. That's just fundamental fairness.
Quoting Republican President Nixon who first signed into law the
1970 Clean Air Act, ``I think that 1970 will be known as the year of
the beginning, in which we really began to move on the problems of
clean air and clean water and open spaces for the future generations of
America.''
Although significant progress has been made in the past 40 years,
it is our job now to build upon this legacy and ensure that we continue
to improve our environmental quality while fostering a strong economy.
This is not science fiction; it is our history. In the U.S., a healthy
environment and strong economy are not mutually exclusive. Stricter
pollutions limits force us to push the envelope of scientific
innovation and create new technologies. And, as it has been proven many
times over, improved worker productivity, increased agricultural yield,
reduction in mortality and illness, and other economic and public
health benefits far outweigh the costs of compliance.
Thank you, and I yield back.

Chairman Harris. Thank you very much, Mr. Miller.
If there are Members who wish to submit additional opening
statements, your statements will be added to the record at this
point.
At this time I would like to introduce our witness panel.
Our first witness, Dr. Roger McClellan, is an advisor to public
and private organizations on issues concerned with air quality,
utilizing his extensive experience in comparative medicine,
toxicology, aerosol science and risk analysis. He was also
formerly the Chair of EPA's Clean Air Science Advisory
Committee.
Our next witness will be Dr. George Thurston, who is a
Professor at the NYU School of Medicine. Dr. Thurston conducts
research into the human health effects of air pollution. He has
been published widely in the scientific literature and was a
consultant to EPA's Clean Air Science Advisory Committee's
Panel on Nitrogen Oxides.
Next, we have Dr. Michael Honeycutt, Chief Toxicologist at
the Texas Commission on Environmental Quality. Dr. Honeycutt is
the Director of the Toxicology Division there and an Adjunct
Professor at Texas A&M University and has published numerous
articles in the peer-reviewed literature.
Dr. Robert Phalen, Professor of Medicine and Co-Director of
the Air Pollution Health Effects Laboratory, University of
California, Irvine, will be the next witness. Dr. Phalen has
over 30 years of experience with inhalation studies of
hazardous materials. He is also a member of EPA's Clean Air
Science Advisory Committee Panel on Particulate Matter.
Next, we have Dr. Anne Smith, Senior Vice President at NERA
Economic Consulting. Dr. Smith is an economist and decision
analyst specializing in energy and environmental markets and
compliance planning. Before joining NERA, Dr. Smith headed the
Climate and Sustainability Group at Charles River Associates.
Finally, we have Mr. J. Edward Cichanowicz, an experienced
engineering consultant. He provides consulting services for
utility industry clients in developing and implementing
environmental control strategies to meet the mandates of the
EPA and state and local regulatory agencies. Mr. Cichanowicz
specializes in the technical feasibility, cost and risk of
adopting both mature and evolving technologies.
Now, as our witnesses should know, spoken testimony is
limited to five minutes each after which the Members of the
Committee will have five minutes each to ask questions, and we
do have a copy of your written testimony.
I now recognize our first witness, Dr. Roger McClellan, an
advisor on toxicology and human health risk analysis. Dr.
McClellan.

STATEMENT OF DR. ROGER O. MCCLELLAN,

ADVISOR, TOXICOLOGY AND HUMAN HEALTH RISK ANALYSIS

Dr. McClellan. Good morning, distinguished Members of the
Subcommittee. Thank you for the invitation to present my views
and the role of science in informing policy judgments on the
setting of National Ambient Air Quality Standards. I ask that
my written testimony be entered in the record as though read in
its entirety. Let me summarize it.
What I hope my comments will convey is how low is low
enough requires context. It is not an abstract question. For
more than five decades, I have been contributing to the
development of the science needed to address important societal
issues concerned with air quality. I am proud to have served on
many EPA advisory committees under the Administrations of both
parties. I have served on numerous Clean Air Scientific
Advisory Committee panels and chaired CASAC from 1988 to 1992.
I served on the panels that advised on the 1997 ozone standard,
the 2006 particulate matter standard. I did not serve on the
ozone panel that advised on the 2008 ozone standard. However, I
did offer comments to the Administrator and CASAC.
In March 2008, then-Administrator Johnson revised the ozone
standard using the policy judgment exclusively delegated to the
EPA Administrator in the Clean Air Act. He retained ozone as an
indicator, the averaging time of eight hours, the statistical
form, the standard attained when the fourth highest eight-hour
average is less than the numerical level of the standard, and
he reduced the concentration allowed from 84 to 75 parts per
billion. In announcing his decision, he noted that he depended
on the science to inform his decision and noted that the advice
provided by the Clean Air Scientific Advisory Committee was in
fact a blend of science and the committee's own personal policy
preferences.
In January 2010, EPA Administrator Jackson formally
announced a fast-track reconsideration proposal to set the
standard in the range of 70 to 60 PPB. This in my opinion was
the beginning of a serious misadventure and waste of resources.
In my opinion, her decision was discretionary, arbitrary,
capricious and without precedent. With the decision, she was
expressing a personal opinion. If I had been in office 22
months earlier, I would have made a different policy choice.
She wrapped herself in the cloak of science saying I will
follow the advice of CASAC. In taking this course, she
abdicated the specific and exclusive authority delegated to the
EPA Administrator by the Clean Air Act.
In January 2011, she asked CASAC to elaborate on the basis
of their recommended range. CASAC Chair Jonathan Samet wisely
noted the decision as to an adequate margin of safety to
protect public health was inherently a blend of science and
policy.
After repeatedly missing self-imposed deadlines,
Administrator Jackson sent a proposed final rule to OMB in mid-
summer. On September 2, 2011, Administrator Cass Sunstein of
the OMB Office of Information and Regulatory Affairs advised
Administrator Jackson her proposed rule was not mandatory, said
it was discretionary, it would produce needless uncertainty and
was being offered even while the next five-year review cycle
was proceeding in an orderly fashion to conclude in March 2013,
and he noted that her rule was not based on the latest science
as called for by the Clean Air Act, and most importantly, the
President had indicated he has made it clear he does not
support finalizing the rule at this time.
I applaud the common sense decision of Administrator
Sunstein and President Obama. It is unfortunate that the
President and his appointees did not have a discussion in 2009
that could have avoided this serious misadventure.
Looking to the current review, I urge EPA and CASAC to
carefully heed the thoughtful advice of Supreme Court Justice
Stephen Breyer, who offered the view that a comparative health
view was appropriate in setting the standards. I urge the
Administrator to recognize that even when all U.S. manmade
precursor emissions are eliminated, ambient ozone
concentrations expressed as the maximum eight-hour average are
just below 60 PPB. These background levels should be considered
in making policy judgments on setting the standard.
I think the Administrator and CASAC should be cognizant of
the very weak signal for air pollution impacting public health
currently. We have made great progress. And keep in mind, the
substantial adverse health signal associated with socioeconomic
status. Employment and jobs do matter.
Frequently we hear about the relative risks associated with
air pollution expressed as a few percent over a baseline. My
view is, we ought to be concerned more with the science of the
baseline. What can we do to drive down those baseline risks?
Future research should focus on that, and keep in mind, a big
number in terms of relative risk, the relative risk of
unemployment, lower socioeconomic status. That is not a couple
of percent, that is 100 percent looking at the lowest quartile
versus the upper quartile.
Thank you for the opportunity to testify, and I look
forward to addressing your questions later, including the
process used by CASAC.
[The prepared statement of Dr. McClellan follows:]

Prepared Statement of Dr. Roger O. McClellan, Advisor, Toxicology and
Human Health Risk Analysis

Major Points of Testimony of Roger O. McClellan--October 4, 2011

Clean Air Act is primary National Statute governing air
quality issues in the U.S.A. The CAA requires the Administrator of the
U.S. EPA to establish primary (health-based) and secondary (welfare-
based) National Ambient Air Quality Standards for six criteria
pollutants with science-based criteria to be reviewed every five years.

Primary NAAQS are to be established by the EPA
Administrator based on the ``latest scientific knowledge'' at levels
``requisite to protect public health'' while ``allowing an adequate
margin of safety'' without considering the cost of implementing the
standard.

In March 2008, then-Administrator Stephen Johnson revised
the Ozone NAAQS as required by the CAA using the scientific record
based largely on papers published in 2005 and earlier to inform his
policy judgments. He retained (a) ozone as the indicator for
photochemical oxidants, (b) the averaging time of eight hours, (c) the
statistical form (the standard is attained when the fourth highest
eight-hour average value over a three-year period does not exceed the
numerical level of the standard, and (d) reduced the level from 84 ppb
to 75 ppb. In announcing his decision he noted that the Clean Air
Scientific Advisory Committee had recommended the standard be set in
the range of 60 to 70 ppb, advice based on a blend of science and their
policy judgment.

In January 2010, Administrator Lisa Jackson announced
that she was going to ``reconsider'' Administrator Johnson's policy
decision and set the standard in the range of 60 to 70 ppb. She based
this discretionary, arbitrary and capricious action on (a) her personal
opinion that if she had been in office 22 months earlier she would have
made a different policy choice, and (b) wrapped herself in a ``cloak of
science'' saying I will follow the advice of CASAC. With this proposal
she abdicated the specific and exclusive authority delegated to the EPA
Administrator to make the policy judgments inherent in setting the
NAAQS.

On September 2, 2011, Administrator Cass Sunstein of the
Office of Information and Regulatory Affairs/OMB advised Administrator
Jackson that her proposed final rule was: (a) not mandatory, produced
needless uncertainty, and that her Agency was already proceeding with
five-year review cycle set to conclude in March 2013, (b) that her
proposed final rule was not based on the latest science, and (c) the
President had instructed Mr. Sunstein to return the rule to her--``He
has made it clear that he does not support finalizing the rule at this
time.''

I applaud the actions of Administrator Sunstein and the
President. My only regret is they did not have this ``common sense''
discussion with Administrator Jackson in early 2009. It would have
avoided the misuse of the substantial EPA resources spent on this
misadventure during 2009-2011.

Building on recent experience in revising the NAAQS for
Ozone and PM2.5, I will comment on the NAAQS setting process and the
role of CASAC.

I will emphasize that the language of the CAA and the
efforts of narrowly focused advocacy groups may not be promoting, but
rather damaging, public health.

I urge the Congress to refocus the Nation's effort on
public health revising the Clean Air Act, to allow consideration of
costs in setting NAAQS, as part of an omnibus legislative package--
``Promoting Public Health'' that recognizes a healthy economy with
people employed is the cornerstone of a healthy population.

Good morning, Mr. Chairman and Members of the Subcommittee. Thank
you for the invitation to present my views on the role of science in
informing policy judgments on the setting of National Ambient Air
Quality Standards.
Since 1999, I have served as an advisor to public and private
organizations on issues related to air quality in the ambient
environment and workplace, drawing on more than 50 years of experience
in comparative medicine, toxicology, aerosol science, and risk
analysis. Prior to 1999, I provided scientific leadership for two
organizations--the Chemical Industry Institute of Toxicology (1988-
1999) in Research Triangle Park, NC, and the Lovelace Inhalation
Toxicology Research Institute (1966-1988) in Albuquerque, NM. Both
organizations, under my leadership, earned an international reputation
for developing scientific information undergirding occupational and
environmental health standards.
The testimony I offer today also draws on my experience serving on
numerous scientific advisory committees. This has included service on
many EPA Scientific Advisory Committees from the origin of the Agency,
including the Clean Air Scientific Advisory Committee (CASAC), which I
chaired from 1988 to 1992, and on CASAC Panels that have considered all
the criteria pollutants at various times. I served on the CASAC Panel
that advised on the 2006 revision of the Particulate Matter MAAQS. I
served on the CASAC Ozone Panel that reviewed the basis for the NAAQS
promulgated in 1997. I did not serve on the most recent CASAC Ozone
Panel. However, I closely followed the current NAAQS Ozone review
process from its inception in September 2000 to present. The testimony
I offer today reflects my own views on that review process and the
science used to inform the policy judgments made in revising the NAAQS
for Ozone. Attachment 2 is a reprint of a recent paper I authored
entitled ``Role of Science and Judgment in Setting National Ambient Air
Quality Standards: How low is low enough?'', Air Quality and
Atmospheric Health (published online: 01 June 2011).
EPA Administrator Johnson's March 2008 Decision

This morning I would like to comment on the role of science and
judgment in the ``Final Rule for the National Ambient Air Quality
Standard for Ozone'' announced on March 12, 2008, by EPA Administrator
Stephen Johnson. That Final Rule revises the 1997 Standard and
concludes a process begun in September 2000. Throughout the review
process, there was debate over the numerical level of a revised
standard. In my view, much of the debate was premature and focused on
the outcome desired by various parties--a lowering of the ozone
standard--even before the review of the science was complete. That
resulted in a blurring of the boundary between the role of science and
judgment in the setting of the standard.
As I will discuss later, Administrator Lisa Jackson took advantage
of the CASAC's blended science and policy advice to initiate in January
2010 reconsideration of the March 2008 decision of then-Administrator
Johnson.
As required by a Court Decree, the EPA published a Proposed Rule on
July 11, 2007, and requested public comments on anticipated action in
issuing a Final Rule for the ozone standard. Release of the Proposed
Rule intensified the debate over the numerical level of the standard
and continued to blur the distinction between science and judgment in
the setting of the standard. Numerous comments were submitted to the
official ozone docket. I submitted my personal comments to the ozone
docket and also joined with nine of my scientific colleagues in
submitting a document--``Critical Considerations in Evaluating
Scientific Evidence of Health Effects of Ambient Ozone'' to the Docket.
The debate over the numerical level of the standard continues even
today as evidenced by this Hearing.
Much of the debate failed to acknowledge that the setting of the
standard involves policy judgments informed by science. The debate has
included repeated reference to the Clean Air Scientific Advisory
Committee (CASAC) Ozone Panel recommendation that the primary standard
be set within a specific narrow numerical range, i.e., 0.060-0.070 ppm.
In my opinion, the CASAC Ozone Panel moved from the science arena into
the policy arena in advocating an upper bright line value of 0.070 ppm
for the primary standard. That value represents the personal judgment
of the Ozone Panel Members, not just their interpretation of the
science. It is my opinion, the CASAC Ozone Panel never adequately
communicated the extent to which the recommendations they communicated
to the Administrator represented both their interpretation of the
science and their personal policy judgments on the numerical level of
the standard.
The EPA Administrator, under the authority of the Clean Air Act,
has the exclusive responsibility and authority for making policy
judgments, informed by science, in setting the ozone standard. Supreme
Court Justice Stephen Breyer, in the landmark case Whitman v. American
Trucking Association (531 U.S. 457, 2001), offered ``common sense''
guidance for setting the standards for criteria pollutants such as
ozone (Attachment 3). Justice Breyer expressed the opinion that while
the Administrator cannot consider cost in setting air quality standards
for the criteria pollutants, the EPA Administrator need not set
standards at zero risk. He advised the Administrator to use judgment in
a ``comparative health'' context when ``deciding what risks are
acceptable in the world in which we live.''
In short, Justice Breyer recognized that everyday life carries with
it a variety of risks. Justice Breyer's opinion provides ``common
sense'' guidance for deciding how low is low enough in setting air
quality standards--the numerical level of the standard and the
associated acceptable risk level, even if not specifically articulated,
are policy judgments that should be informed by science. In my opinion,
the Administrator could have made a policy judgment, informed by
science, with selection of a numerical value for the ozone primary
standard as high as the 1997 primary standard of 0.08 ppm. His
selection of a lower value was consistent with the original advice of
his own staff--0.075 ppm up to a level slightly below the current
standard.
In my own comments to the Ozone Docket, I reviewed the science
available on the health effects of ozone. In my comments, I noted the
substantial uncertainty and variability in the findings of an increase
in common health effects with ozone exposure in the range of the
current standard and below. These scientific uncertainties were also
detailed in the comments I and nine of my colleagues submitted to the
Docket. Both sets of comments also emphasized that the selection of any
specific numerical standard is a policy judgment informed by science.
The CASAC Ozone Panel, in proposing a bright line upper limit of
0.070 ppm, offered their collective judgment on, in the words of
Justice Breyer--``what risks are acceptable in the world in which we
live.'' The CASAC was advancing their collective policy choice; it
should not be postured as being exclusively science based. Science
alone can never provide a basis for deciding how low is low enough;
policy judgments are always required in deciding ``what risks are
acceptable.'' Any specific numerical value for the Standard has an
associated implied ``acceptable risk value,'' even if the level of
acceptable risk has not been explicitly stated.
The CASAC Ozone Panel's letter to the Administrator dated April 7,
2008, commenting on the Final Rule, continues to suggest that somehow
science and scientists alone can establish the appropriate numerical
level of the NAAQS for ozone. In that letter, the CASAC Ozone Panel
again failed to clarify the distinction between their interpretations
of the science and their policy judgment in offering an opinion on the
numerical level of the ozone standard. The Panel should have clearly
acknowledged that the numerical level they have advocated reflects
their personal policy preferences. Likewise, in arguing for ``further
lowering the national ambient ozone standards,'' the Panel fails to
acknowledge that this is a collective wish that goes well beyond
considering just the available scientific information. How low is low
enough for the ozone standard is ultimately a policy judgment informed
by scientific information and analysis. The Clean Air Act clearly
specifies that the EPA Administrator has the exclusive authority and
responsibility for using judgment in the setting of the standard.
Without question, the Administrator, in setting the standard,
should consider scientific advice received from many parties, including
the special advice provided by the Clean Air Scientific Advisory
Committee. However, it is clear that the Clean Air Act calls for an
Advisory Committee and not a Clean Air Standard Setting Committee. This
places a special responsibility on the Committee to distinguish between
their scientific advice and their personal policy judgments as to the
numerical level of the Standard.
It is noteworthy that the March 2008 Final Rule states--``the
Administrator observes that he reaches a different policy judgment than
the CASAC Panel based on apparently placing different weight in two
areas:--'' The Final Rule goes on to detail these differences. The Rule
goes on to state--``and fully considering the scientific and policy
views of CASAC, the Administrator has decided to revise the level of
the primary eight-hour O3 standard to 0.075 ppm.'' Without question,
the Final Rule clearly acknowledges that the CASAC Ozone Panel offered
both their scientific and policy views. It is unfortunate that the
CASAC Ozone Panel did not make this important distinction in its
communications to the Administrator in their public statements on the
Final Rule.

Administrator Jackson's Misadventure

During 2009 there were rumors that the President Obama/
Administrator Lisa Jackson Administration was going to ``fast track'' a
``reconsideration'' of the March 2008 Ozone NAAQS issued by then-
Administrator Stephen Johnson. Thus, it was not surprising when
Administrator Jackson on January 19, 2010, announced a proposed
``reconsideration'' Ozone NAAQS to be based on the record used to set
the standard in March 2008. This included the science used for the
March 2008 policy decisions, scientific papers which had been published
primarily in 2005 or earlier. By initiating the ``reconsideration''
action, Administrator Jackson was in essence saying--``if I had been in
office in March 2008 (nearly a year before being appointed and
confirmed), I would have made a different policy judgment call.'' In my
opinion, Administrator Jackson's action was totally discretionary,
arbitrary, capricious and without precedent. I know of many NAAQS that
have been revised by EPA Administrators in accordance with the Clean
Air Act and using EPA's now well-established formal rulemaking process.
I know of no NAAQS established by a previous Administrator that has
been ``reconsidered'' by a new Administrator based on the old and aging
record.
In announcing the ``reconsideration'' proposal (EPA, 2010)
Administrator Jackson put on the ``cloak of science'' and said that she
would set the ``reconsideration'' standard in the range of 60 to 70 ppb
following the advice of the CASAC Ozone Panel. In taking this course,
she ignored the documented record of previous Administrator Johnson who
noted that the advice of the CASAC Panel was a blend of science and
policy. In the fall of 2008, the EPA was already initiating action on
the next review of the Ozone NAAQS (Martin, 2008). In initiating the
next review, it was noted that the CASAC advice on the previous review
was ``a mixture of scientific and policy considerations.'' By
proceeding with the ``reconsideration'' proposal based exclusively on
the advice of the CASAC Panel, Administrator Jackson abdicated her
responsibilities under the Clean Air Act to use her judgment in the
setting of NAAQS.
The ``fast track'' reconsideration proposal turned out to be on a
slow track with the target date for release of the final rule
repeatedly revised. My suspicion was that Administrator Jackson and her
senior advisors were continually spinning the ``Ozone Science
Kaleidoscope'' in an attempt to have the science justify a specific
numerical level. Indeed, in January 2011, Administrator Jackson went
back to CASAC and asked for yet another opinion on the setting of the
ozone NAAQS. The CASAC Panel had a difficult time dealing with this
serious question for several reasons.
First, the CASAC members found it difficult to offer an opinion on
the old science since many of them were already involved in reviewing
the new science that would inform policy judgments on potential
revision of an Ozone NAAQS in March 2013. Second, the CASAC Panel
meetings were actually teleconferences. With about 20 ``official''
participants such teleconferences are much like a ``Tower of Babylon.''
The third issue was the challenge of separating the Panel members'
views of the science from their personal policy preferences. The CASAC
Chair, Dr. Jonathan Samet, wisely offered the following summary comment
to Administrator Jackson in his letter dated March 30, 2011. Dr. Samet
wisely noted that establishing a margin of safety was apparently a
blend of science and policy. I offered comments to Administrator
Jackson on Comments on EPA-CASAC-11-004 Clean Air Scientific Advisory
Committee (CASAC) Response to Charge Questions on the Reconsideration
of the 2008 Ozone National Ambient Air Quality Standards (Attachment
4).
Apparently Administrator Jackson and her senior advisors spun the
``Ozone Science Kaleidoscope'' without a firm endorsement of CASAC and
in mid-summer sent forward a final rule for review by OMB's Office of
Information and Regulatory Affairs. Administrator Jackson has testified
that she had proposed 70 ppb. I have seen no indication as to specifics
of a revised Secondary Standard. It is important to recognize that the
CASAC Ozone Panel (Henderson, 2008) in a letter dated April 7, 2008,
based on a meeting scheduled even before then Administrator Johnson had
issued a final rule protested both the Primary and Secondary Standard.
They also expressed their displeasure with the involvement of then-
President Bush and Susan Dudley, who then headed OMB's Office of
Information and Regulatory Affairs. Such involvement was not a surprise
to students of the history of the NAAQS process. President Clinton
conferred with the EPA Administrator Carol Browner on the Ozone and
Particulate Matter NAAQS revisions in 1997.
The misadventure of Administrator Jackson with the
``reconsideration'' Ozone NAAQS was brought to a close on September 2,
2011. The legal basis for the decision to abandon the
``reconsideration'' proposal is contained in a memo from Cass Sunstein,
Administrator of the Office of Information and Regulating Affairs
within OMB (Sunstein, 2011) (Attachment 5). In his memo, he notes the
proposed final rule was (a) not mandatory and produced needless
uncertainty and that his Agency was already proceeding with the next
review that should be concluded in March 2013, (b) her proposed final
rule was not based on the latest science, recall the record is largely
based on pre-2006 scientific publications, and (c) the President had
advised Mr. Sunstein to return the proposal to Administrator Jackson--
``He has made it clear that he does not support finalizing the rule at
this time.''
I applaud the actions of Mr. Sunstein and President Obama for
making a sound common sense decision. My only regret is that the key
parties had not conferred in early 2009 and never have launched this
misadventure that wasted valuable EPA resources and those of many other
interested parties. In this time of crisis, the scarce resources could
have been used better on other endeavors. The really good news is that
a potential precedent setting actions did not take place. It is hard to
imagine the uncertainty and chaos that would occur if every change in
Presidential Administration were to be accompanied by a new EPA
Administrator that would ``reconsider'' the policy judgments of the
previous EPA Administrator.

The Wrong Scientific and Policy Focus

Remarkable progress has been made in improving air quality in the
United States during the last four decades using the various regulatory
tools provided by the Clean Air Act including the establishment of
NAAQS. Clean air is automatically equated with better health. Every
lowering of a NAAQS for each of the criteria pollutants has been
justified on the basis of health benefits.
It has been argued by some that a linear relationship, without a
threshold, exists between ambient concentrations of criteria pollutants
and increased risk of morbidity and mortality over and above the
baseline morbidity or mortality rate. Some scientists have argued that
the absence of a threshold and a linear concentration-response
relationship extends to background concentrations. Using that logic,
which I do not necessarily agree with, it can be argued that health
benefits result from every reduction in concentration, even reductions
in background. With this flawed logic and a prohibition in considering
cost in setting NAAQS the answer to how low is low enough becomes zero.
That is hardly realistic and certainly does not meet the common sense
comparative health approach advanced by Supreme Court Justice Breyer.
In my view, the USA is reaching a point of diminishing returns in
setting the NAAQS at lower and lower concentrations with each review
and treating each reduction as a success story for public health. In
examining this viewpoint, it is important to remember that each NAAQS
is a federal goal. The achievement of the goals is by and large left to
the States through the development of State Implementation Plans and
their actual implementation and, finally, to actions on the part of
private firms and the public.
In my opinion, this approach is flawed in that it fails to
recognize any untoward consequences of setting lower standards and
attempting to attain them. I submit the untoward consequences may be
substantial. Let me illustrate by discussing health impacts using a
common metric-all-cause mortality. Major population studies have
suggested that a 1 mg/m3 increase in particulate matter--2.5 micron
size causes a 0.5% increase in mortality. You may not recognize the
0.5% value because it is usually expressed as 2.5% increase per 5 mg/m3
of PM2.5. In reality, 5 to 10 mg/m3 is the background level for PM2.5
in most areas in the U.S.A. Does it make sense to talk about a 5 mg/m3
change in PM2.5? In my opinion, No! Thus, I use a more realistic 1 mg/
m3 change.
Some population studies suggest a 0.24% change in mortality for a 5
ppb change in eight-hour ozone concentration. Again, you may not
recognize the value because it has frequently been presented as 3.6%
for a 75 ppb change in eight-hour ozone. This is hardly a realistic
presentation recognizing background levels for the eight-hour highest
ozone concentrations approaches 60 ppb, the level simulated by models
when all man-made ozone precursors are shut off. I view a 5 ppb shift
in ozone as being more realistic.
Let me now turn to a real risk factor--socioeconomic status (SES).
The ratio of the mortality rate for all-cause mortality for men in the
lowest quartile of SES over the top quartile was found to be 2.02 by
Steenland et al. (2004). In other words, a doubling of the mortality
rate by dropping from the top quartile to the bottom quartile. Put
another way, moving to the second quartile from the lowest quartile
reduced the ratio to 1.69 and a move from the second to the third
quartile reduced the ratio to 1.25. Socioeconomic status matters--
employment and jobs matter. If the U.S. wants to improve the health of
the Americans, we need to create employment--JOBS.
Setting aside the issue of socioeconomic status, does it make sense
to keep pursuing risk factors that only contribute marginally to our
overall burden of disease? I think the answer is No! Recognizing the
small estimated burden of disease attributed to air pollution, it would
appear to make more sense to pursue what are the major factors that
contribute to the baseline incidence of disease. For example, there is
appropriate increasing concern for rising asthma rates. However, when
it is recognized that air pollution decreased substantially while
asthma rates increased, it would appear that the focus on air pollution
and asthma is misdirected.

A Path Forward

I am increasingly concerned that our policy for advancing public
health is being driven by advocacy groups with narrow interests.
Perhaps it is time for all the advocacy groups to step back and ask
what can be done to further improve the health of all Americans. A
starting point is to recognize that the steady progress made in
improving the health of Americans over the last half century has been
driven by a strong economy that provided jobs and improving income.
Perhaps the answer to the question of how low is low enough for each of
the NAAQS is low enough for now. I suggest it is appropriate for time
out on moving the goal posts.
I urge the Congress to refocus the nation's effort on public health
revising the Clean Air Act, to allow consideration of costs in setting
NAAQS, as part of an omnibus legislative package--``Promoting Public
Health'' that recognizes a healthy economy with people employed is the
cornerstone of a healthy population.

References

McClellan, R. O. Comments on National Ambient Air Quality Standards
for Ozone: Proposed Rule, Federal Register, Vol. 72, No. 132/Wednesday,
July 11, 2007, pp 37818-37919, Docket ID No. EPA-HQ-OAR-2005-0172,
submitted October 9, 2007.

Brauer, M., Frampton, M.W., Koutrakis, P., McClellan, R.O.,
McDonnell, W.F., Moolgavkar, S., North, D. W., Smith, A.E., Smith,
R.L., Utell, M.J. Critical Considerations in Evaluating Scientific
Evidence of Health Effects of Ambient Ozone. Comments on National
Ambient Air Quality Standards for Ozone: Proposed Rule, Federal
Register, Vol. 72, No. 132/Wednesday, July 11, 2007, pp 37818-37919,
Docket ID No. EPA-HQ-OAR-2005-0172, submitted October 9, 2007.

McClellan, R.O., ``Written Statement for Consideration by the Clean
Air Scientific Advisory Committee (CASAC) Ozone Review Panel'' at the
Panel's Teleconference Meeting on February 18, 2011 (Prepared on
February 7, 2011).

Chairman Harris. Thank you very much, Dr. McClellan.
I now recognize our second witness, Dr. George Thurston, a
Professor at New York University School of Medicine. Dr.
Thurston.

STATEMENT OF DR. GEORGE THURSTON,

PROFESSOR, NEW YORK UNIVERSITY SCHOOL OF MEDICINE

Dr. Thurston. Good morning.
The evidence is clear, as Mr. Miller was saying, the
adverse health consequences of breathing polluted air are well
documented in the published medical and scientific literature,
even at levels we experience today. This human evidence
includes impacts revealed by epidemiologic studies, natural
experiments and controlled chamber exposure experiments, also
including consistent associations between air pollution
exposure and increases in risk of adverse health impacts across
a wide range of human health outcomes including illness and
death.
These effects of air pollution include decreased lung
function, the ability of a person to breathe air in and out
freely, more frequent asthma symptoms, increased numbers of
asthma and heart attacks, more frequent emergency department
visits, additional hospital admissions, and increased numbers
of premature deaths, which, by the way, all have a huge
economic impact.
One of the air pollutants most carefully studied is fine
particulate matter, or PM2.5, which are particles that are so
small that they can become lodged deep in the lung when they
are breathed in and they can cause a variety of health
problems. For example, two of the larger cohort studies of air
pollution and death, the Harvard Six Cities study and the
American Cancer Society study, both of which I was involved in,
have demonstrated greater risk of premature death in more
polluted cities as compared to studies with cleaner air as
demonstrated in figure one of my testimony that I submitted.
Another major air pollution health threat is ozone, as Mr.
McClellan was discussing. Ozone is a highly irritant gas that
is formed in our atmosphere in the presence of sunlight and
other precursor air pollutants such as nitrogen oxides and
hydrocarbons that come from trucks and buses and cars and power
plants and various pollution sources. In my own research, I
found that ozone air pollution is associated with increased
numbers of respiratory hospital admissions and death. But these
effects of ozone are only the tip of the iceberg of adverse
health effects associated with this pollutant. Ozone is
especially a problem for children with asthma. My own asthma
camp study results have shown that children have more asthma
attacks on higher ozone air pollution days in the summer. The
air pollution health effects associated indicated by
epidemiological studies are supported by a large body of data
from controlled human exposure studies giving consistent
results and demonstrating pathways by which air pollution can
damage the human body when it is breathed.
It has also been well shown that reducing pollution in the
air can result in health benefits for the public. For example,
Arden Pope conducted a compelling study of the period during
the winter of 1986 and 1987 when the Geneva steel mill in Utah
Valley shut down during a strike. The PM levels dropped
dramatically and the hospital admissions in the valley showed
the same pattern as the air pollution, decreasing dramatically
during the strike as shown again in my written testimony. When
pollution levels diminish, the health of the general public
improves.
A recent follow-up analysis of a Harvard Six Cities study
has also shown that mortality is decreased as PM2.5 pollution
decreases over time, as shown in figure three of my written
testimony. Thus, recent research shows lowering of air
pollution levels in the air is an effective way to improve
public health.
As to the role of CASAC, the setting of the National
Ambient Air Quality Standards, or NAAQS, is an essential
mandate of the Clean Air Act. The Clean Air Science Advisory
Committee was authorized by Congress to provide the EPA
Administrator with independent advice on the setting of the
NAAQS. This process has worked well and has led to the
application of sound science to the setting of the U.S. NAAQS
standards to protect the health of the American people.
In conclusion, the science is sound and the results are
clear. Cleaning our air reduces air pollution and health
impacts, lowers our health care costs and saves lives. With the
independent advice of CASAC, the EPA's regulatory control of
ambient air pollution has led to reductions in both air
pollutant exposures and health risks to the American people.
This has caused the public to enjoy associated health benefits
including decreased asthma attacks, fewer hospital admissions,
fewer heart attacks, and increased length and quality of life.
Thank you for this opportunity to testify on this important
issue.
[The prepared statement of Dr. Thurston follows:]

Prepared Statement of Dr. George Thurston,
Professor, New York University School of Medicine

I am George Thurston, a tenured Professor of Environmental Medicine
at the New York University (NYU) School of Medicine. My scientific
research involves investigations of the human health effects of air
pollution. I have served on the U.S. EPA's Clean Air Science Advisory
Committee (CASAC) Panel, and have been a contributing author to EPA's
Integrated Science Assessment (ISA) documents.
I am also a member of the National Institute of Environmental
Health Sciences' (NIEHS) Center at the NYU Institute of Environmental
Medicine. One goal of this Center is to provide an impartial scientific
resource on environmental health issues to the public and decision
makers, and that is my purpose in speaking to you at this hearing.
The adverse health consequences of breathing polluted air are well
documented in the published medical and scientific literature. During
the past decades, medical research examining air pollution and public
health has shown that air pollution causes a host of serious adverse
human health effects. The human evidence includes impacts revealed by
epidemiologic studies, natural experiments and controlled chamber
exposures, all showing consistent associations between air pollution
and increases in adverse health impacts across a wide range of human
health outcomes, including illness and death.

Epidemiological Evidence of Air Pollution Effects on Health

Observational epidemiology studies provide the most compelling and
consistent evidence of the adverse effects of air pollution.
``Epidemiology'' is literally ``the study of epidemics,'' but includes
all statistical investigations of human health and potentially causal
factors of good or ill health. In the case of air pollution, such
studies follow people as they undergo varying real-life exposures to
pollution over time, or from one place to another, and then
statistically intercompare the health impacts that occur in these
populations when higher (versus lower) exposures to pollution are
experienced. In such studies, risks are often reported in terms of a
Relative Risk (RR) of illness, wherein a RR=1.0 is an indication of no
change in risk after exposure, while a RR>1.0 indicates an increase in
health problems after pollution exposure, and that such exposure is
damaging to health.
These epidemiological investigations are of two types: (1)
population-based studies, in which an entire city's population might be
considered in the analysis; and (2) cohort studies, in which selected
individuals, such as a group of asthmatics, are considered. Both of
these types of epidemiologic studies have confirmed associations
between air pollution exposures and increasing numbers of adverse
impacts, including:

decreased lung function (a measure of our ability to
breathe freely);

more frequent asthma symptoms;

increased numbers of asthma and heart attacks;

more frequent emergency department visits;

additional hospital admissions; and

increased numbers of premature deaths.

The fact that the effects of air pollution have been shown so
consistently for so many health endpoints, and in so many places,
indicates these associations to be causal.

Particulate Matter Air Pollution

One of the air pollutants most carefully studied is particulate
matter (PM). Fine particles (PM2.5), such as those that result from
power plants and diesel trucks, defeat the defensive mechanisms of the
lung, and can become lodged deep in the lung where they can cause a
variety of health problems. New evidence indicates that short-term
exposures to air pollution cause both respiratory and cardiac effects,
including more heart attacks. In addition, my own research indicates
that long-term exposure to fine particles increases premature
mortality, and such exposures in the general population have been
estimated to take years from the life expectancy of people living in
our most polluted cities, relative to those living in cleaner cities
(e.g., see Brunekreef, 1997).
PM2.5 air pollution may be emitted directly from tailpipes and
smokestacks (known as ``primary'' particulate matter), but much PM2.5
that we breathe comes from the conversion of gaseous pollution
emissions, such as sulfur dioxide, in the atmosphere to form
``secondary'' PM2.5.
The hazards of PM air pollution have become particularly clear in
the past two decades of research. Two of the largest studies on air
pollution and death, the Harvard Six Cities Study, published in 1993,
followed by the American Cancer Society (ACS) Study report in 1995,
have demonstrated greater risk of premature death in higher PM cities
compared to cities with cleaner air. The Harvard Six Cities study
monitored air pollution and tracked mortality in six U.S. cities and
discovered a 25 percent increased risk of death in the most polluted
city (Dockery et al., 1993). Similarly, the ACS study examined half a
million people in over 150 metropolitan areas throughout the United
States and found a 17 percent increase in risk of mortality between the
city with the least PM and the city with the highest levels of this
pollution (Pope et al., 1995). The results of these two landmark
studies were challenged by industry, resulting in an independent
reanalysis by the Health Effects Institute (HEI)-funded by industry and
EPA. HEI found the results to be robust, and confirmed the
relationships documented by the original investigators (Krewski et al.,
2002).
More recent follow-up analyses of the Harvard and ACS studies have
now considered longer records of time, and have confirmed and extended
the conclusions from these two major studies. Indeed, a recent National
Institute of Environmental Health Sciences (NIEHS)-funded extension of
the ACS study, of which I was Principal Investigator, strengthens the
original conclusions of the ACS study and, importantly, now links
increased risk of lung cancer to long-term exposure to particulate
matter (Pope et al., 2002), as shown in Figure 1.

Ozone Air Pollution

Another major air pollution health threat, ozone (O3), is a highly
irritant gas that is formed in our atmosphere in the presence of
sunlight from other ``precursor'' air pollutants, including nitrogen
oxides and hydrocarbons. These precursor pollutants are emitted by
pollution sources including automobiles, electric power plants, and
industry.
In my own research, I have found that ozone air pollution is
associated with increased numbers of respiratory hospital admissions in
U.S. and Canadian cities. But these effects of ozone are only the ``tip
of the iceberg'' of adverse effects associated with this pollutant, and
they are best viewed as indicators of the much broader spectrum of
adverse health effects being experienced by the public today as a
result of air pollution exposures, such as more restricted activity
days and doctors' visits.
Airway inflammation induced by ozone is especially a problem for
children and adults with asthma, as it makes them more susceptible to
having asthma attacks. My own asthma camp results have shown that
children have more asthma attacks on high ozone days in the summer
(Thurston et al., 1997). In addition, recent controlled human studies
have indicated that prior exposure to ozone enhances the reactivity of
asthmatics to aeroallergens, such as pollens, which can trigger asthma
attacks. In addition, the increased inflammation and diminished immune
system ozone effects in the lung can make the elderly more susceptible
to pneumonia, a major cause of illness and death in this age group.

Controlled Exposure Studies

The air pollution--health effects associations indicated by
epidemiologic studies are supported by a large body of data from
controlled exposure studies giving consistent and/or supportive
results, and demonstrating pathways by which ozone can damage the human
body when it is breathed. For example, clinical studies have
demonstrated ozone-related decreases in lung function, increased
frequencies of respiratory symptoms, heightened airway hyper-
responsiveness, and cellular and biochemical evidence of lung
inflammation in healthy exercising adults exposed to ozone. Similarly,
animal exposures to combustion-related PM2.5 have been shown to have
significant adverse effects on the lung, including diminished
respiratory defense mechanisms, opening the lung to illness from other
causes.

The Benefits of Cleaner Air

Most published studies evaluate whether rising air pollution levels
worsen health, but it has also been shown that reducing pollution in
the air can result in health benefits to the public. For example, Pope
(1989) conducted a compelling study clearly showing that, when
pollution levels diminish, the health of the general public improves.
He investigated a period during the winter of 1986-87 when the Geneva
Steel mill in the Utah Valley shut down during a strike. The PM levels
dropped dramatically in that strike-year winter, as opposed to the
winters preceding and following when the steel mill was in operation.
As shown in Figure 2 below, hospital admissions in the valley showed
the same pattern as the PM air pollution, decreasing dramatically
during the strike. As a control, Pope also examined the pollution and
hospital admissions records in nearby Cache Valley, where the mill's
pollution was not a factor, and no such drop in respiratory admissions
was seen, showing that the drop in admissions in the Utah Valley was
not due to some cause other than the reduction in the air pollution
levels.

A more recent study considers a broadly relevant case showing the
benefits of cleaner air. During the Atlanta Summer Olympics of 1996,
traffic-related ozone and PM declined significantly as a result of the
alternative mass transportation strategy implemented to reduce road
traffic during the Games (Friedman et al., 2001). These improvements
were correlated with changes in the rate of children's hospital
admissions. Compared to a baseline period, traffic-related ozone and PM
levels declined by 28% and 16%, respectively. Concentrations of both PM
and ozone also rose noticeably after the end of the Olympics. The study
showed a significant reduction in asthma events associated with these
pollution improvements. This study indicates that improvements in acute
air pollution can provide immediate public health benefits.
Furthermore, a recent follow-up analysis of the Harvard Six Cities
Study discussed earlier (Dockery et al., 1993) has shown that mortality
was decreased by lowering pollution (Laden et al., 2006). An extended
analysis of the Harvard Six Cities Study (to include follow-up through
1990) has now shown that reductions in long-term ambient PM pollution
results in concomitant reductions in the health risks associated with
PM. As shown in Figure 3, large reductions in PM at Harvard study
cities have resulted in likewise large reductions in the relative risk
(RR) of mortality in those cities: Steubenville, OH (S), Harriman, TN
(H), St. Louis, MO (L), and Watertown, MA (W). The authors found that,
for each decrease of 1 mg/m3 in averge PM2.5, the overall death rate
from causes such as cardiovascular disease, respiratory illness and
lung cancer decreased by some 3 percent, while also extending the lives
of study subjects. Thus, although we still carry very large health
risks in the United States from our present levels of air pollution,
amounting to tens of thousands of premature deaths per year, recent
research shows that the lowering of air pollution levels in the air is
an effective way to improve public health.

The evidence is clear and consistent: air pollution is adversely
affecting the health and lives of Americans across our Nation. There is
a coherence between the epidemiologic study associations and
experimental study results, validating that there is indeed a cause-
effect relationship between air pollution and adverse human health
effects. The importance of these health effects relationships is made
all the more imperative by the fact that virtually every American is
directly impacted by this pollution. Cleaning the air causes
improvements in public health, saving lives and improving the quality
of life of all Americans.

The Role of CASAC

The setting of the National Ambient Air Quality Standards (NAAQS)
is an essential mandate of the Clean Air Act. These NAAQS apply to
``criteria'' air pollutants that, in the words of the legislation,
``endanger public health or welfare.'' At present, six pollutants are
designated as criteria pollutants. The Clean Air Science Advisory
Committee (CASAC) was authorized by the Congress in 1977 to aid the EPA
Administrator in the setting of these NAAQS. Its members are derived
largely from academia and from private sector research institutes, and
are appointed by the EPA Administrator. CASAC's role is primarily to
review the agency's work in the process of setting each NAAQS, which
are each reviewed on an every-five-year schedule, and then to provide
the EPA Administrator with independent advice on the interpretation of
agency documents for the setting of the NAAQS. Most notably, CASAC
reviews the agency's Integrated Science Assessment (ISA) and the Risk
and Exposure Assessment (REA) that summarize the science and the policy
analyses, respectively. The assessment of these pollutants individually
during the CAAA approach, potentially missing effects of synergisms
among the various air pollutants, combined with the fact that not all
potential health impacts can be quantified in the REA process, has
likely caused the EPA to underestimate the total benefits of pollution
reductions when viewed on an individual pollutant-by-pollutant basis.
Despite these facts, the EPA's REA and RIA (Regulatory Impact Analysis)
analyses have consistently indicated that the valuation of the health
benefits of cleaner air, such as fewer hospital admissions and deaths,
far outweigh the costs of applying emission controls to reduce air
pollution.
Based upon the EPA ISAs, REAs, and CASAC independent expert advice
on the setting of the NAAQS (usually provided as a range of possible
standards), the EPA Administrator proposes an updated standard for each
pollutant that can be the same, more stringent, or less stringent than
the existing NAAQS standard for a criteria air pollutant. This process
has generally worked well in the past, and has led to the application
of sound science to the setting of the U.S. NAAQS standards to protect
the health of the American public.

Conclusions

The science is sound and the results are clear: cleaning
our air reduces air pollution health impacts, lowers our health care
costs, and saves lives.

With the independent advice of CASAC, as stipulated by
the Congress, the EPA's regulatory control of ambient air pollutants
has led to reductions in both air pollutant exposures and health risks
to the American people. This has caused the public to enjoy associated
health benefits, including decreased asthma attacks, fewer hospital
admissions, fewer heart attacks, and increased length and quality of
life.

Thank you for the opportunity to testify on this important issue.

References

Brunekreef B. (1997) Air pollution and life expectancy: is there a
relation? Occup Environ Med. Nov; 54(11): 781-4.

Dockery DW, Pope CA 3rd, Xu X, Spengler JD, Ware JH, Fay ME, Ferris
BG Jr, Speizer FE. (1993) An association between air pollution and
mortality in six U.S. cities. N Engl J Med. Dec 9; 329(24):1753-9.

Krewski, D. et al. (2000) Reanalysis of the Harvard Six Cities
Study and the American Cancer Society Study of Particulate Air
Pollution and Mortality: Investigators' Report Part I: Replication and
Validation. Health Effects Institute, Cambridge, MA.

Laden F, Schwartz J, Speizer FE, Dockery DW. (2006) Reduction in
fine particulate air pollution and mortality: Extended follow-up of the
Harvard Six Cities study. Am J Respir Crit Care Med., Mar 15;
173(6):667-72.

Pope CA 3rd. (1989). Respiratory disease associated with community
air pollution and a steelmill, Utah Valley. Am J Public Health. 1989
May; 79(5):623-8.

Pope CA 3rd, Thun MJ, Namboodiri MM, Dockery DW, Evans JS, Speizer
FE, Heath CW Jr. Particulate air pollution as a predictor of mortality
in a prospective study of U.S. adults. Am JRespir Crit Care Med. 1995
Mar; 151(3 Pt 1): 669-74.

Pope CA III, Burnett RT, Thun MJ, Calle EE, Krewski D, Ito K, and
Thurston, GD. (2002). Lung Cancer, Cardiopulmonary Mortality and Long-
Term Exposure to Fine Particulate Air Pollution. Journal of the
American Medical Association (JAMA). March 6, 2002, Vol. 287, No. 9,
pp. 1132-1141.

Thurston, G. D., Lippmann, M., Scott, M.B., and; Fine, J.M.
Summertime haze air pollution and children with asthma. Am. J. Respir.
and Crit. Care Med. 155:654-660 (1997).

Chairman Harris. Thank you very much, Dr. Thurston.
I now recognize our next witnesses, Dr. Michael Honeycutt,
Chief Toxicologist for the Texas Commission on Environmental
Quality. Dr. Honeycutt.

STATEMENT OF DR. MICHAEL HONEYCUTT,

CHIEF TOXICOLOGIST, TEXAS COMMISSION ON

ENVIRONMENTAL QUALITY

Dr. Honeycutt. Good morning, Mr. Chairman and Members of
the Committee. I have submitted more detailed written comments
on the science behind the ozone and PM NAAQS and the Utility
MACT, but I will touch on the highlights right now.
Regarding ozone, I would first like to talk about the
concept of personal exposure. As regulators, we set health
protective standards for chemicals in outdoor air and we
measure those pollutants at outdoor monitors. However, it is a
well-established fact that we are not exposed to ozone
concentrations at concentrations measured at outdoor monitors.
This is because Americans spend on average 90 percent of our
time indoors, especially on hot days when ozone levels are at
their highest. Because ozone is formed in sunlight, ozone
concentrations are always much lower indoors than outdoors and
in fact are practically zero in air conditioned buildings.
Keeping that in mind, I want to talk briefly about the
ecological epidemiology studies that EPA is using as the
primary basis for the ozone NAAQS.
Ecological epidemiology studies are exploratory studies
designed to look for correlations. They are supposed to be
followed up by more rigorous epidemiology studies to see if the
correlations are real. These studies are not supposed to be
used quantitatively and they are certainly not rigorous enough
to set environmental policy. Statisticians can run data through
elegant models to try to find statistically significant
correlations but the output of those models is only as good as
the input, and any researcher worth their salt will tell you
that correlation is not causation.
With the ozone NAAQS, EPA has set environmental policy
based on these ecological epidemiology studies. Specifically,
the studies gathered death certificates for people who died
from non-accidental causes including cancer, liver disease or
any other disease. The assumption is that breathing ozone made
them die earlier than they would have otherwise. The
researchers correlated the outdoor ozone levels, usually from
the highest monitor in the city, with various time periods just
before the time of death for these thousands of people. Using
the highest monitor or even an average value in the city is not
scientific since ozone levels can vary tremendously across a
city. These studies did not look at whether the people who died
were actually outdoors for eight hours just prior to their
death to actually breathe that ozone. Even if we assume that
these ill people spent eight hours outdoors just before they
died, were they near the monitor the EPA assumed they were?
Were they exposed to other pollutants during the day? Did they
take their medications that day? There are a whole host of
common sense questions that go unanswered in these studies.
Simply put, these studies cannot tell us if ozone caused these
deaths or if these people died prematurely, much less tell us
what level of ozone caused their premature death.
EPA also used clinical studies conducted by Professor
William Adams, formerly from USC, in setting the ozone NAAQS.
EPA reanalyzed his data inappropriately and called the mild
effects he observed adverse.
Turning to the Utility MACT, EPA themselves determined that
the rule will not have an effect on mercury levels in fish in
America's watersheds. EPA continues to overstate the health
risk of lower IQ and heart disease from mercury while ignoring
the very well-demonstrated health benefits of eating seafood.
EPA used a study known as the Faroe Island study to set their
safe level for mercury where the mothers ate whale meat and
blubber contaminated with PCBs in addition to eating the fish.
The Faroe Island infants ingested 600 times EPA safe dose of
PCBs. I will say it again: 600 times EPA safe level. The
effects EPA attributed to mercury can more justifiably be
attributed to the PCBs. A similar study in the Seychelles
Islands that did not include PCB exposures was essentially
negative. EPA ignores the fact that Japanese eat 10 times more
fish than Americans do and have higher levels of mercury in
their blood but they have lower rates of coronary heart disease
and they have high scores on their IQ tests. Methyl mercury is
a toxic chemical but scientific data overwhelmingly do not
support EPA's position on the risk of mercury. In fact, EPA may
have the most conservative safe level for mercury in the world.
The FDA, the ATSDR, the World Health Organization and Canada
all have set a higher safe level for mercury, and EPA still
uses decade-old data whenever they say that six percent of the
women in the United States have unsafe levels of mercury in
their blood. Newer data time and time again clearly shows this
isn't the case.
Thank you for the opportunity to testify.
[The prepared statement of Mr. Honeycutt follows:]

Prepared Statement of Dr. Michael Honeycutt,
Chief Toxicologist, Texas Commission on Environmental Quality

Main Conclusions

On behalf of the Texas Commission on Environmental Quality (TCEQ),
I disagree with the United States Environmental Protection Agency's
(EPA) proposed range of values for the eight-hour ozone and PM
standards because of uncertainties relating ambient concentrations to
personal exposures and limitations of the epidemiological and clinical
studies used as the basis of the revisions. The TCEQ strongly
recommends that EPA use robust scientific data as the basis for the
ozone and PM standards and Utility MACT, and more meaningful
consideration of risk management issues in its final policy decisions.
The roles of uncertainty and bias in EPA's assessments have been
severely downplayed and should be reexamined. This is particularly true
in EPA's analysis of personal exposure. For ozone, EPA relies on
studies that estimate personal exposure (the amount of ozone a person
actually breathes) by using ambient monitoring data, which
oversimplifies personal exposure by assuming that ambient monitoring
data accurately reflects personal exposure. Further, EPA doesn't
acknowledge or account for this potential overestimate in their
standard calculations. Also, it is essential that EPA clearly discuss
the uncertainties associated with adverse health effects reported in
both ecological epidemiology and clinical studies. These uncertainties
should also be clearly communicated in publicly accessible documents in
consideration of new standards.
EPA should be more critical and conscientious in its selection of
studies they use to calculate proposed numerical standards.
Specifically, EPA should consider ecological epidemiology studies in a
more broad, supportive context, rather than as the primary basis for
calculating air quality standards. Ecological epidemiology studies are
not scientifically rigorous enough to draw conclusions about the cause
of health effects identified in the studies for ozone or any other
pollutant and are not suitable for policy decisions. As with all
observational studies, the results may provide valid areas for further
inquiry and be informative, but should not be considered conclusory.
EPA's criteria for the selection of key studies should emphasize not
only statistical significance but also biological significance of the
observed adverse health effects Furthermore, EPA should focus on the
entire weight of evidence of more robust epidemiology and toxicology
studies for the basis of its policy decisions.
Finally, EPA should avoid unnecessary regulation that will not
improve human health. EPA's own analysis demonstrates that the Utility
MACT will not have an effect on mercury levels in fish in U.S.
watersheds. EPA's claims of mercury causing lower IQ and heart disease
scares the public into avoiding seafood. EPA ignores the fact that
Japanese eat 10 times more fish than Americans do and have higher
levels of mercury in their blood, but have lower rates of coronary
heart disease and high scores on their IQ tests. To claim that a policy
decision is ``based on the science'' without putting those decisions in
appropriate context with real world implications is not just a misuse
of science but causes harm to the public. It is a disservice to her
citizens when government exaggerates, misstates or misleads the public
about the ``real'' risk of environmental effects.

Ozone NAAQS

Ecological Epidemiology Studies
EPA used ecological epidemiology studies, also known as time-series
analyses, as the primary basis of the most recent proposed ozone
standard. Ecological epidemiology studies are observational studies
designed to look for correlations. To accomplish this they examine the
relationships between exposure and disease at a population level rather
than on an individual level. These types of studies are intended to be
followed up by more rigorous epidemiology studies to determine if the
correlations are real. While ecological epidemiology studies are useful
in evaluating potential associations between health effects and ambient
exposures to environmental pollutants, they are severely limited due to
their study design. Policy conclusions should not be based on
ecological epidemiology studies for the following reasons:

1. Ecological epidemiology studies are not designed to
determine if ozone caused the health effects evaluated. The assumption
that ozone caused all evaluated health effects, including aggravation
of asthma and premature mortality, in ecological epidemiology studies
is not well-grounded in science. Ecological epidemiology studies do not
collect data on when, how long, and how much exposure occurred; if
exposure occurred before the health effects; or if it makes biological
sense that the chemical could cause the effect. In other words, the
study designs are incomplete. Scientists agree that the incomplete
study design does not provide enough information to determine the
actual cause of studied effects. Ecological epidemiology studies are
not supposed to be used quantitatively and they certainly are not
rigorous enough to set environmental policy.

2. Lack of personal exposure data severely limits the
utility of ecological epidemiology studies. The issue of limited or
entire absence of personal exposure data is significant. Personal
exposure is a measurement of the amount of an air pollutant that a
person actually breathes. In the case of air pollutants like ozone,
ecological epidemiology studies rely on ambient monitoring data as a
surrogate for personal exposure for percentages of people with a health
issue in an area (i.e., census tract, county, or state). However, it is
very unlikely that people would ever be exposed to those pollutants at
concentrations measured at outdoor monitors for very long. This is
partly because the average American spends 90% of his/her time indoors,
especially during the heat of the summer when ozone concentrations tend
to be at their highest. Ozone concentrations in most buildings are
characteristically low, due to the reactive nature of ozone, the
tendency of ozone to deposit on surfaces, and the ventilation systems
inside buildings (McClellan et al., 2009). Other additional factors
such as time spent outdoors, outdoor activity level, and weather
(especially temperature and relative humidity) can dramatically change
the potential for ozone exposure and the resultant estimate of risk.
Therefore, ambient ozone concentrations alone do not adequately
characterize, and easily overestimate, personal exposures (Sarnat et
al., 2006). This position is shared by the National Academies of
Science (NAS 2008) and the Clean Air Science Advisory Committee (CASAC
2006). That ecological epidemiology studies continue to derive
inconsistent and vastly differing conclusions about the adverse effects
of ozone is perhaps evidence of this fact.

3. Ecological epidemiology studies frequently do not take
into account the heterogeneity of regional air pollution and
oversimplify their exposure analysis by relating health effects to only
ozone. In most ecological epidemiology studies, \1\ exposure is
estimated to be either some statistical representation (e.g., average
or weighted average) of several air monitors or concentrations at the
monitor with the highest readings. This assumption oversimplifies
outdoor exposure because concentrations vary across a given area. \2\
Moreover, few studies fully account for simultaneous exposure to
multiple other pollutants, such as particulate matter, nitrogen
dioxide, and sulfur dioxide. The ratios of these pollutants can vary
tremendously from region to region, making it difficult to determine
which effects are related to which pollutants. This complication blurs
the association between health effects and ozone exposure, as
documented in recent studies. Furthermore, it has been documented in
studies that the association between ozone and health effects is
confounded by temperature and relative humidity (which alone can cause
physical stress), and population characteristics, such as age, health
status, socioeconomic status, and exercise.
---------------------------------------------------------------------------
\1\ Many ecological epidemiology studies do not look at the area
immediately around a monitor but rather at a conglomeration of several
cities. For example, Bell and Dominici (2008) looked at communities,
which they defined as a county or contiguous counties.
\2\ EPA acknowledges the variance in ozone concentrations across a
region within its state implementation planning (SIP) process by its
requirements to have multiple monitors within a populated region and
its requirement to further analyze unmonitored areas during the
planning process.

4. Ecological epidemiology studies have considerable
uncertainty in their identification of health effects. To determine
prevalence of a health issue, epidemiologists frequently use readily
available information, including hospital admissions records and death
certificates, or participant surveys. In some of the ecological
epidemiology studies EPA used for the proposed ozone standard, death
certificates for thousands of people who died at a hospital from any
non-accidental cause were compared to outdoor ozone levels from up to
three days before the person died. Because of the broad selection
criteria, it is highly likely that many of these people died due to
non-respiratory health issues unrelated to ozone exposure. This problem
is compounded when paired with the lack of personal exposure data,
making it impossible to know if decedents were actually well enough to
be outdoors in the days preceding their deaths. In this case, patient
history records from physicians would be more reliable than hospital
admission records or death certificates for determining the presence
and severity of any health effects potentially caused or aggravated by
ozone exposure. EPA could better serve the public trust to recognize
the limitations on the information and data used and to fully consider
---------------------------------------------------------------------------
these limitations when making policy decisions.

5. Additional statistical analysis (time-series and
multi-city time-series studies) further complicates the interpretation
of ecological epidemiology studies. The shortcomings of ecological
epidemiology studies are compounded when researchers perform time-
series studies, which try to correlate health effects collected from
epidemiology studies and ambient ozone concentrations measured during
the hours and days leading up to their hospital visit or death. Some
studies compare even broader sets of data from multiple cities averaged
over multiple years. In addition to the issues regarding uncertainty in
the original ecological epidemiology study discussed previously, this
additional analysis fails to take into account:

The high degree of variability between cities, seasons,
and years;

The effect of other pollutants that contribute or cause
the same effects;

The inconsistent ambient air sample collection period
between cities;

Socioeconomic factors such as age, access to health
care, etc., and

Mortality differences among cities.

In addition, further analysis of time-series data indicates the
studies are highly influenced by the type of statistical model used
(often, the model showing the most health effects) and publication bias
(studies showing effects are more likely to get published than those
showing no effects). Due to the substantial uncertainty in these
studies, policy decisions should not be based on these studies and EPA
should revise its study selection criteria to use studies of higher
scientific quality.

. 6. Results of ecological epidemiology studies are
inconsistent and it remains unclear if ozone is truly related to
increased health effects. Ecological epidemiology studies have provided
vastly different conclusions regarding the effects of ozone on the
population, with studies showing significant adverse effects, no
effects, or even protective effects of ozone. In particular, one
reanalysis of an ecological epidemiology study frequently cited by EPA
identified only six out of 95 cities evaluated with a significant
correlation of mortality and ozone (Smith et al., 2009). Furthermore,
although it has been repeatedly hypothesized that ozone is a potent
inducer of asthma attacks, Texas Inpatient Hospital Discharge data on
the numbers of hospital visits for asthma between 1999 and 2001
actually showed that people were more likely to visit the hospital for
asthma during winter when ozone is at its lowest than they were in the
summer when ozone concentrations are high. In a relative risk sense,
cold weather and pet dander are more potent inducers of asthma hospital
visits than ozone. Furthermore, results from a four-year (2000-2003)
air quality study conducted by Texas A&M University and Driscoll
Children's Hospital indicate hospital admissions to be weakly
correlated with ambient daily maximum ozone levels.

Clinical Studies

Clinical studies expose humans to a known concentration of ozone
for a known period of time and monitor their health. Although these
studies do not have the significant limitations of ecological
epidemiology studies, there is confusion among scientists and
regulators about whether subtle clinical changes documented in the
studies represent adverse effects. In its ozone reassessment, EPA
failed to consider key recommendations regarding this issue and
conducted a reanalysis of clinical data that was not scientifically
appropriate.

1. EPA should rely on biological, not just statistical,
significance in identifying an adverse health effect in clinical
studies. Ambiguity exists in defining what constitutes an adverse
effect on exposure to air pollution. Clinical studies evaluating health
effects due to ozone exposure have mainly focused on decreases in lung
function as measured by forced expiratory volume in one second (FEV1)
\3\ and other similar measures. Daily normal activities, exercise, and
diurnal variations can themselves cause changes in the FEV1. Within a
single day, FEV1 in normal subjects can vary by over 5% (Pellegrino et
al., 2005) and be as high as 17.6% (Medarov et al., 2008). Therefore,
controlled exposure studies must properly account for normal changes by
including filtered air (FA) exposures and a range of concentrations and
exposure durations. The American Thoracic Society (ATS) recommends a
comprehensive description of ``adverse'' effects by combining the loss
of lung function in conjunction with respiratory symptoms, such as
cough and discomfort while breathing (ATS 2000). Further, OEHHA, the
TCEQ, and jointly the ATS and the European Respiratory Society (ERS)
consider decrements in FEV1 of : % as ``mild,'' not ``adverse.''
However, in its reevaluation of the Adams (2006) study, EPA identified
FEV1 decrements of only 2.8% to be adverse effects. According to the
sources listed previously and Adams himself, the decrements in the
Adams (2006) study at 0.06 ppm are not of biological significance, even
though they may be of statistical significance. Therefore, it is also
prudent that the EPA justify the importance of key study results to
indicate not just statistical significance, but also biological
significance before labeling the result as an adverse effect.
---------------------------------------------------------------------------
\3\ FEV1 is a measure of the forced expiratory volume during the
first second of an active exhalation. This measurement is used to
assess lung function and is often used in epidemiological or controlled
clinical studies. The measurement is accomplished by having a subject
inhale deeply and then exhale quickly. A significant reduction in FEV1
may be indicative of impaired ventilation.

2. EPA's reanalysis of Adams (2006) data is not
scientifically appropriate and should not be included as part of the
final ozone policy decision. In addition to the issue of whether or not
the decrease in FEV1 was adverse, the EPA also conducted a highly
contentious statistical reanalysis of the Adams (2006) data to show
statistical significance in the absence of the effect (Brown 2007,
Brown 2008). Dr. Adams himself disagreed with the EPA's reanalysis and
statistical reinterpretation of his study during a teleconference on
March 5, 2007, and in written comments to the EPA during the 2007
comment period. EPA's reanalysis was also criticized by other
statisticians and scientists, as stated in comments submitted to EPA by
Drs. RL Smith and JE Goodman. The TCEQ concurs with Dr. Adams' peer-
---------------------------------------------------------------------------
reviewed results.

3. EPA should consider more recent studies as part of the
ozone weight of evidence. Recent clinical studies (Kim et al., 2011;
Schelegle et al., 2009) of ozone exposure at concentrations lower than
0.08 ppm have further confirmed the Adams (2006) results, showing no
adverse effects at 0.06 ppm. When compared to filtered air, Schelegle
et al. (2009) reported statistically significant mean percent change in
FEV1 at 0.07 ppm (5.34%) and Kim et al. (2001) reported statistically
significant mean percent change in FEV1 at 0.06 ppm (1.71%), these are
not only within in the range of intra-individual variability but are
also substantially less than the 20% decrease identified as adverse.

4. EPA needs to emphasize the importance of having
realistic controls for clinical studies. Many of the clinical studies
use filtered air (no ozone) for the control groups (Schelegle et al.,
2009; Kim et al., 2011), which creates an unrealistic scenario as the
natural background ozone concentration in the atmosphere is around 0.04
ppm (Last et al., 2010). In its analysis of the clinical studies, EPA
has not adjusted for this background factor and has not provided any
justification for not doing so. Not adjusting for background can result
in overestimating the severity of the observed effects as ``adverse
effects,'' when in fact the effects were ``not adverse.'' Based on the
clinical studies, it can be inferred that the weight of evidence at the
lower range of exposure levels (i.e, 0.06-0.07 ppm) is weak and
inconclusive. Thus, I can conclude that the clinical studies used to
justify the lower end of the proposed range do not support lowering the
ozone standard below the present NAAQS of 0.075 ppm (Adams 2002 and
2006; Schelegle et al., 2009; Kim et al., 2011). Further, these studies
are conservative since they do not consider personal exposure.

Differing Roles of Policy and Science

EPA's recent attempts at using science to justify policy decisions
are particularly troubling. In its reconsideration of the ozone
standard, EPA attempts to establish a health basis as the need for a
new, reduced standard. However, the assumption that the reduced
standard would prevent up to 12,000 deaths is based on dubious studies
and the use of such an analysis signals an unfortunate shift in the
roles of scientists and risk managers.

1. The basis of the theoretical number of lives saved is
meaningless and unrealistic. EPA relied on studies that took mortality
data from ecological epidemiology studies to calculate the number of
theoretical deaths that would be avoided with a lower standard. Not
only do these studies suffer from the severe limitations described
above, but theoretical lives saved estimates are also meaningless from
a scientific and practical standpoint. It is not possible to verify
either the current number of deaths due to ozone exposure or the future
change in deaths if the standard is lowered because there is still no
conclusive evidence that ozone causes mortality at ambient
concentrations. \4\ There is no guarantee of increased life expectancy
or degree of confidence in this estimation, since some degree of risk
is present in all aspects of daily life.
---------------------------------------------------------------------------
\4\ In fact, EPA has provided no data to illustrate lives saved
under previous standards. All estimates of lives saved are projections,
not factual.

2. EPA misuses scientific studies to justify policy
decisions. Scientific studies should be just one aspect of responsible
policymaking. Rigorous scientific studies focus on expanding the
knowledge of how a chemical interacts with the body at different tested
doses. However, even the most extensive studies are not able to define
an acceptably safe level of a chemical. In the specific case of ozone,
scientific studies have still been unable to clearly identify human
risk at current ambient levels and have certainly not shown if 0.065
ppm ozone is substantially more protective than 0.08 ppm. Determining
what level of risk is acceptable is and should remain a decision for
---------------------------------------------------------------------------
risk managers, not scientists.

Responsible risk managers and policymakers consider science as one
of many aspects to be considered in setting policy. Science cannot
determine practical issues, such as the feasibility of implementation
and to what extent society would accept the trade-offs associated with
the standard. For example, an overly restrictive health-based standard
might be more detrimental to public health if it forces an industry out
of business due to the cost of compliance and its employees are unable
to find work to support their families. Studies have consistently
indicated that poverty is a much better predictor for premature
mortality than exposure to environmental pollutants. Public officials
with a broader perspective of potential policy implications are better
equipped to evaluate these important aspects.

PM NAAQS

The Proposed PM Standard
EPA has proposed a new particulate matter (PM) standard that is
twice as stringent as the current standard. Attainability of the
proposed standard, especially in rural and agricultural areas, is
impractical and even EPA staff acknowledges that the available
scientific evidence supports the effectiveness of the current standard
in protecting public health. There is no scientific basis supporting a
reduction in the current standard, let alone a two-fold reduction.

1. EPA based the proposed PM standard on an ecological
epidemiology study. EPA used a study by Zanobetti and Schwartz (2009),
which is an ecological epidemiology study, as a basis for the proposed
PM standard. This ecological epidemiology study concludes that exposure
to coarse PM is ``suggestive'' of a causal relationship with adverse
effects. As stated above, ecological epidemiology studies are
incomplete studies plagued with limitations and should not be used as
the basis for policy conclusions.

2. EPA assumes all PM composition is equal. Not all PM is
created equally; however, EPA makes the assumption that it is. Coarse
PM is produced by surface abrasion or suspension of biological material
and fragments of living things. Because of this, PM in urban and
industrial areas is likely to be vastly different from PM in rural and
agricultural areas. Urban and industrial PM is expected to be enriched
with pollutants; pollutants that are inherently more toxic than the
dust predominantly found in agricultural operations and arid rural
areas. EPA didn't take this scientific fact into consideration when
they developed their proposed PM standard. When they assume all PM
composition is the same, they ignore the fact that agricultural and
rural areas will likely exceed the standard due to natural occurrences
rather than man-made sources.

3. PM composition varies greatly by geographic regions.
The PM data EPA used in their assessment for the proposed PM standard
were not uniformly distributed across the United States or even within
counties. Therefore, potential differences in PM composition may be
reflected in the EPA estimates. Geographic variability is also strongly
influenced by region-specific sources, meteorology (e.g., wind speed
and direction), and topographical conditions (e.g., trees, mountains).
When PM composition differs geographically, the conclusions drawn may
not apply equally to all parts of a geographic region.

Utility MACT

Mercury and the Utility MACT

EPA has proposed a National Emission Standards for Hazardous Air
Pollutants (NESHAP) rule for coal- and oil-fired electric utility steam
generating units (EGU). This proposed NESHAP rule (the Utility MACT)
would establish maximum achievable control technology (MACT) emission
limits for certain hazardous air pollutants (HAP), including mercury.
In EPA's analysis for mercury, they state ``if U.S. EGU impacts to
watersheds included in the risk assessment were zeroed out, for a
significant majority of those watersheds, total exposure would still
exceed (and in most cases, significantly exceed) the RfD [Reference
Dose].'' In EPA's own words they are admitting control of U.S. EGU
mercury emissions will not have an effect on mercury levels in fish in
U.S. watersheds; however, they still insist on the necessity to require
these controls. Concurrent with the Utility MACT, the EPA's National-
Scale Mercury Risk Assessment Supporting the Appropriate and Necessary
Finding for EGUs (mercury risk assessment) was released for public
comment and review by EPA's Science Advisory Board Mercury Review
Panel. Currently, the Mercury Review Panel's support for the mercury
risk assessment is contingent upon development of a revised document
that addresses numerous issues. The Panel's comments to EPA on the
mercury risk assessment were finalized in September 2011, illustrating
the limited time allowed for review and revisions of such an important
document whose purpose was to determine whether a public health hazard
is associated with U.S. EGU emissions. One could easily conclude that
the Panel's input was merely a formality and was not intended to be
seriously considered, much like EPA treats input from the States.

The EPA 2000 appropriate and necessary finding estimates
were inaccurate. \5\ The risk analysis estimates of hazard quotients
due to U.S. EGU-attributable emissions of mercury have already
decreased significantly between the 2005 and 2016 scenarios, mainly due
to PM controls. In fact, 2010 levels of mercury emissions are already
at levels predicted for 2016. In addition, the 2000 appropriate and
necessary finding was based on estimates that U.S. utility mercury
emissions would increase from 46 tons in 1990 to approximately 60 tons
in 2010. In reality, emissions were reduced to 29 tons in 2010.
---------------------------------------------------------------------------
\5\ In December 2000, EPA issued a ``regulatory determination''
under the Clean Air Act (CAA) that it is ``apppropriate and necessary''
to regulate mercury emissions from coal-based power plants and nickel
emissions from oil-based power plants. This regulatory determination
listed coal- and oil-based EGUs as a source category under section
112(c) of the CAA, the first step to setting MACT standards. On January
30, 2004, EPA proposed to remove EGUs from the 112 list based on a
finding that it was neither appropriate nor necessary to regulate EGUs
under this section of the CAA. On March 29, 2005, EPA issued a final
revision of the appropriate and necessary finding for coal- and oil-
fired EGUs and removed such units from the 112 list. The removal of
EGUs from the 112 list was challeneged in court. On February 8, 2008,
the court determined that EPA violated the CAA by removing EGUs from
the 112 list. As a result, EGUs remain a CAA section 112(C) listed
source category according to EPA. The basis of the court ruling was
that EPA did not follow the requirements of 112(C)(9) in removing EGUs
from the 112 list. As such, the court did not reach a determination on
the merits of the case.

U.S. EGU mercury emissions are insignificant compared to
other sources. The Utility MACT preamble states that on average, U.S.
EGUs are estimated to contribute only 2% to total mercury deposition in
the U.S. Therefore, any health benefits related to mercury reductions
would pose an insignificant change in the overall risk from mercury
from all sources. Only in combinations of the worst-case watersheds
with fish consumption rates (e.g., 95th and 99th percentile fish
consumption rates paired with the 95th and 99th percentile watersheds)
did estimates of U.S. EGU-attributable hazard quotients (HQs) exceed
1.5 (EPA considered an HQ>1.5 to represent a potential public health
hazard). U.S. EGUs contributed insignificantly to the total risks posed
by other sources of mercury; thus, regardless of this regulation, risk
---------------------------------------------------------------------------
from mercury deposition will remain from sources other than U.S. EGUs.

Mercury is a global pollutant. It travels beyond boundaries of
states and continents. EPA modeling estimates that, on average, 83% of
the mercury deposited in the U.S. originates from international
sources, excluding Canada; the remaining 17% comes from U.S. and
Canadian sources. As such, control strategies related to EGUs may not
affect change in fish tissue concentrations of mercury. According to
EPA (2007), ``The mix of long-distance and local sources makes it
difficult in some water bodies to achieve water quality standards for
mercury.''

EPA uses a worst-case scenario for risk and does not
characterize risk for realistic U.S. populations. EPA should have
characterized risk for the more realistic general recreational angler
population to provide perspective and information to that population.
Instead, the EPA's mercury assessment is essentially a worst-case
scenario that focuses on subsistence fishing populations and may
overestimate risk for the majority of the U.S. population. EPA's own
Science Advisory Board Mercury Review Panel states ``There is scant
evidence documenting the prevalence or extent of subsistence fishing in
the United States.''

EPA states that about seven percent of women of child-
bearing age are exposed to mercury at a level capable of causing
adverse effects in the developing fetus. Several well-conducted studies
examining effects of mercury on children have been conducted, including
the Seychelles Child Development Study (Seychelles) and the Faroe
Island Study (Faroe). A blood mercury No Effect Level (NEL) of 85 parts
per billion (ppb) was observed in the Seychelles study. Interestingly,
this study also observed positive improvements on IQ as mercury levels
increased; a phenomenon likely due to nutrients such as omega-3 fatty
acids and selenium from high fish consumption. A blood mercury NEL of
58 ppb was observed in the Faroe study; however, these residents also
consumed large quantities of whale meat and blubber that contained
unsafe (according to EPA) levels of polychlorinated biphenyls (PCBs).
Since neither study found effects below 58 ppb blood mercury levels,
one would only expect to find health effects in children whose mothers
had mercury levels higher than 58 ppb in their blood. EPA's safe level
(the RfD) is set to prevent blood mercury levels exceeding 5.8 ppb, 10
times lower than the NEL of 58 ppb from the Faroe study.

Data from the Centers for Disease Control's (CDC's) National Health
and Nutrition Examination Survey (NHANES), 2003-2008, show the mean
blood mercury level for pregnant women is 0.69 ppb (well below EPA's
safe blood mercury level) (Jones et al., 2010). Although some
individuals have blood mercury levels greater than EPA's safe blood
mercury level, none have blood mercury levels above the Faroe study NEL
of 58 ppb, and therefore adverse health effects would not be expected
in their children. A 2005 study conducted by Texas Department of State
Health Services (DSHS 2005) determined that even when subsistence
fishers are eating fish from Caddo Lake with elevated mercury, women of
child-bearing years did not have blood mercury levels greater than the
EPA's safe blood mercury level.
On comparing U.S. blood mercury levels to other countries, both the
United Kingdom (UK) and Japan median blood mercury levels are higher.
Using EPA's RfD to describe Japan's data, 66% of Japanese women are
exposed to levels above EPA's safe blood level. From this the claim
could be made (falsely) that 66% of Japanese children are born at risk
for adverse effects. On the contrary, the Japanese population consumes
10 times more fish than the U.S. population but only shows positive
outcomes; they have lower rates of coronary heart disease and high IQ
scores. EPA is causing unnecessary alarm in the public with their
assertions that 7% of women of child-bearing age are exposed to mercury
at a level capable of causing adverse effects in the developing fetus
when the evidence clearly shows this statement to be false and
misleading.

EPA uses an RfD that is more conservative than most
other Agencies (U.S. and World). The Agency for Toxic Substances and
Disease Registry (ATSDR) and the U.S. Food and Drug Administration
(FDA) both have established safe levels three-fold higher than EPA's
conservative RfD. The World Health Organization (WHO) recommends a
level that is two times higher than EPA's RfD; Health Canada uses a
value similar to the WHO recommended value. The TCEQ agrees with ATSDR
and FDA that it is more appropriate to use a study that reflects U.S.
fish consumption (e.g., saltwater fish such as tuna) rather than a
study based entirely on consumption of saltwater fish and mammals
(e.g., whale).

References

Adams, WC. 2002. Comparison of chamber and face-mask 6.6-hour
exposures to ozone on pulmonary function and symptoms responses. Inhal.
Toxicol. 14(7): 745-764.

Adams, WC. 2003. Comparison of chamber and face mask 6.6-hour
exposure to 0.08 ppm ozone via square-wave and triangular profiles on
pulmonary responses. Inhal. Toxicol. 15: 265-281.

Adams, WC. 2006. Comparison of chamber 6.6-h exposures to 0.04-0.08
ppm ozone via square-wave and triangular profiles on pulmonary
responses. Inhal. Toxicol. 18(2): 127-136.

Adams, WC. 2007. Public Comment to CASAC Ozone Review Panel
Teleconference. University of California, Davis, CA, 2p., March 5.

Agency for Toxic Substances and Disease Registry, Minimum Risk
Level for Mercury, www.atsdr.cdc.gov/toxprofiles/tp46-a.pdf.

American Thoracic Society (ATS). 2000. What constitutes an adverse
health effect of air pollution? Am. J. Respir. Crit. Care Med. 161:
665-673.

American Thoracic Society and European Respiratory Society. 2008.
Outcomes for COPD pharmacological trials: from lung function to
biomarkers. Available online at http://erj.ersjournals.com/content/31/
2/416.full#sec-149.

Bell, ML; Dominici, F. 2008. Effect modification by community
characteristics on the short-term effects of ozone exposure and
mortality in 98 U.S. communities. Am J Epidemiol. 167(8): 986.997.

Brown, JS. [U.S. EPA]. 2007. Memo to Ozone NAAQS Review Docket
(OAR-2005-0172) re: The effects of ozone on lung function at 0.06 ppm
in healthy adults. 8p., June 14.

Brown, JS; Bateson, TF; McDonnell, WF. 2008. Effects of exposure to
0.06 ppm ozone on FEV1 in humans: A secondary analysis of existing
data. Environ. Health Perspect. 116(8): 1023-1026.

California Environmental Protection Agency (California EPA). 2008.
Technical support document for the derivation of noncancer reference
exposure levels, Oakland, California.

Daniels, JL; Longnecker, MP; Rowland, AS; and Golding, J. (2004).
Fish intake during pregnancy and early cognitive development of
offspring. Epidemiology. 15(4): 394-402.

Davidson, P; Myers, G; Cox, C; et al. (1995) Longitudinal
neurodevelopmental study of Seychellois children following in utero
exposure to methylmercury from maternal fish ingestion: outcomes at 19
and 29 months. NeuroToxicology 16:677-688.

Davidson, PW; Myers, GJ; Cox, C; et al. (1998) Effects of prenatal
and postnatal methylmercury exposure from fish consumption on
neurodevelopment: outcomes at 66 months of age in the Seychelles child
development study. JAMA 280:701-707.

Dourson, ML; Wullenweber, AE; Poirier, KA. (2001) Uncertainties in
the Reference Dose for Methylmercury, NeuroToxicology 22 (5) (2001) pp.
677-689.

EPA. 1997. Mercury Study Report to Congress: Volume VII:
Characterization of Human Health and Wildlife Risks from Mercury
Exposure in the United States. EPA-452/R-97-009. Available online at
http://www.epa.gov/ttn/oarpg/t3/reports/volume7.pdf.

EPA. 2007. Memorandum on Listing Waters Impaired by Atmospheric
Mercury Under Clean Water Act Section 303(d): Voluntary Subcategory 5m.
March 8. Available online at http://water.epa.gov/lawsregs/
lawsguidance/cwa/tmdl/mercury5mfactsheet.cfm.

EPA Notice of Regulatory Finding, 65 Fed. Reg. 79,825, 79,830 (Dec.
20, 2000). Available online at: http://frwebgate.access.gpo.gov/cgi-
bin/getdoc.cgi?dbname=2000-register&docid=00-32395-
filed.pdf.

Goodman, JE; Prueitt, RL. 2010. Comments on U.S. EPA's Health
Effects Draft Materials for the Ozone Integrated Science Assessment
(ISA). Gradient, Report to American Petroleum Institute, 49 p.,
September 9.

Grandjean, P; Weihe, P; White, R; et al. (1997) Cognitive deficit
in 7-year-old children with prenatal exposure to methylmercury.
Neurotoxicol Teratol 20:1-12.

Henderson, R. 2006. Letter from CASAC Chairman Rogene Henderson to
EPA Administrator Stephen Johnson, June 5, 2006, EPA-CASAC-06-007.

Jones , L, Parker, JD, and Mendola, P. 2010. Blood lead and mercury
levels in pregnant women in the United States, 2003-2008. NCHS data
brief no 52. Hyattsville, MD: National Center for Health Statistics.
Available online at: http://www.cdc.gov/nchs/data/databriefs/db52.pdf.

Kim, CS; Alexis, NE; Rappold, AG; Kehrl, H; Hazucha, MJ; Lay, JC;
Schmitt, MT; Case, M; Devlin, RB; Peden, DB; Diaz-Sanchez, D. 2011.
Lung function and inflammatory responses in healthy young adults
exposed to 0.06 ppm ozone for 6.6 hours. Am. J. Respir. Crit. Care Med.
183: 1215-1221.

Kjellstrom, T; Kennedy, P; Wallis, S; et al. (1986) Physical and
mental development of children with prenatal exposure to mercury from
fish. Stage 1: Preliminary test at age 4. Natl Swed Environ Protec Bd,
Rpt 3080 (Solna, Sweden).

Kjellstrom, T; Kennedy, P; Wallis, S; et al. (1989) Physical and
mental development of children with prenatal exposure to mercury from
fish. Stage 2: Interviews and psychological tests at age 6. Natl Swed
Environ Prot Bd, Rpt 3642 (Solna, Sweden).

Last, JA; Hyde, DM, and Schelegle, E.S. 2010. Ozone and Ozone
Toxicity. Comprehensive Toxicology (Second Edition) 8: 261-275.

McClellan, RO; Frampton, MW; Koutrakis, P; McDonnell, WF;
Moolgavkar S, North, DW; Smith, AE, Smith, RL; Utell, MJ. 2009.
Critical considerations in evaluating scientific evidence of health
effects of ambient ozone: a conference report. Inhal. Toxicol. Sept 21
Suppl. 2: 1-36.

Myers, GJ; Marsh, DO; Cox, C; et al. (1995a) A pilot
neurodevelopmental study of Seychellois children following in utero
exposure to methylmercury from a maternal fish diet. Neurotoxicology
16(4):629-638.

Myers, GJ; Marsh, DO; Davidson, PW. (1995b) Main neurodevelopmental
study of Seychellois children following in utero exposure to
methylmercury from a maternal fish diet: outcome at six months.
Neurotoxicology 16(4):653-664.

Myers, GJ; Davidson, PW; Cox, C; et al. (1995c) Neurodevelopmetnal
outcomes of Seychellois children sixty-six months after in utero
exposure to methylmercury from a maternal fish diet: pilot study.
Neurotoxicology 16(4):639-652.

Myers, GJ; Davidson, PW; Shamlaye, CF; et al. (1997) Effects of
prenatal methylmercury exposure from a high fish diet on developmental
milestones in the Seychelles Child Development Study. Neurotoxicology
18(3):819-830.

National Academy of Science, Toxicological Effects of Methylmercury
(2000).

National Academy Sciences (NAS) 2008. Committee on Estimating
Mortality Risk Reduction Benefits from Decreasing Tropospheric Ozone
Exposure, National Research Council Copyright National Academy of
Sciences. Available online at http://www.nap.edu/
catalog.php?record-id=12198#description.

Pellegrino, R; Viegi, G; Brusasco, V; Crapo, RO; Burgos, F;
Casaburi, R; Coates, A; van der Grinten, CP; Gustafsson, P; Hankinson,
J; Jensen, R; Johnson, DC; MacIntyre, N; McKay, R; Miller, MR; Navajas,
D; Pedersen, OF; Wanger, J. 2005. Interpretative strategies for lung
function tests. Eur. Respir. J. 26(5): 948-968.

Sarnat, SE; Coull, BA; Schwartz, J; Gold, DR; Suh, HH. 2006.
Factors affecting the association between ambient concentrations and
personal exposures to particles and gases. Environ. Health Perspect.
114: 649-654.

Schelegle, ES; Morales, CA; Walby, WF; Marion, S; Allen, RP. 2009.
6.6-Hour inhalation of ozone concentrations from 60 to 87 parts per
billion in healthy humans. Am. J. Respir. Crit. Care Med. 180(3): 265-
272.

Smith, RL; Xu, B; Switzer, P. 2009. Reassessing the relationship
between ozone and short-term mortality in U.S. urban communities.
Inhal. Toxicol. 21(Suppl. 2): 37-61.

Texas A&M University and Driscoll Children's Hospital Report--
Hamilton, M. J. ; Banks, J. M; Ricard, J. R. No Date. Air Quality,
Ozone Level, Respiratory Response and Acute Pediatric Hospital
Admissions.

Texas Commission on Environmental Quality (TCEQ). 2011. Proposed
Guidelines to Develop Inhalation and Oral Cancer and Non-Cancer
Toxicity Factors. Chief Engineer's Office. Available online at http://
www.tera.org/peer/tceqesl/.

Texas Department of State Health Services (DSHS). Health
Consultation: Mercury Exposure Investigation Caddo Lake Area--Harrison
County Texas. Agency for Toxic Substances and Disease Registry (2005).
Available online at: http://www.atsdr.cdc.gov/HAC/pha/CaddoLake/
CaddoLakeHC.pdf.

Weil, M, Bressler, J, Parsons, P, Bolla, K, Glass, T, and Schwartz,
B 2005. Blood Mercury Levels and Neurobehavioral Function. JAMA.
2005;293(15):1875-1882.

Zanobetti A, Schwartz J (2009). The effect of fine and coarse
particulate air pollution on mortality: A national analysis. Environ
Health Perspect., 117: 1-40.

Chairman Harris. Thank you very much, Dr. Honeycutt.
I now recognize Dr. Robert Phalen, Professor of Medicine,
and Co-Director of the Air Pollution Health Effects Laboratory,
for five minutes to present his testimony. Dr. Phalen.

STATEMENT OF DR. ROBERT F. PHALEN,

PROFESSOR OF MEDICINE, AND CO-DIRECTOR,

AIR POLLUTION HEALTH EFFECTS LABORATORY,

UNIVERSITY OF CALIFORNIA, IRVINE

Dr. Phalen. Chairman Harris, Mr. Miller, thank you for this
opportunity.
My participation in the Clean Air Scientific Advisory
Committee on Particulate Matter was interesting and
stimulating. Everyone involved--the members of CASAC, the EPA
staff and public presenters--was qualified, efficient and
dedicated. But we are here to find, and I think all of us
agree, to find possible ways to improve the CASAC process of
incorporating good science. Therefore, I will present seven
points for you to consider that are my observations as a result
of my participation in the CASAC Particulate Matter
subcommittee.
One, the EPA mandate is too restrictive. It does not allow
the full competence of the EPA to be used in protecting public
health. They are required to evaluate pollutants one by one,
and pushing one pollutant way down can raise the risks from
other factors, including unemployment. The EPA also has to err
on the side of increased safety, and if you overdo this, you
actually increase the overall risk to public health.
Two, linear incrementalism. This is where CASAC is only
allowed to comment on questions that are supplied by EPA, and
each of these comments seems to lead step by step to an
inevitable conclusion. Therefore, the committee can't really
express their concerns and their doubts along the way.
Three, the definition of particulate matter just by
weighing it, and not taking the composition into account, to me
is very poor science mainly because some areas of the country
have more toxic particulate matter than others. And by setting
one standard nationwide that adequately protects maybe two or
three cities, you are punishing major industries, including
agriculture, and you are punishing areas where the particulate
matter has a different, less toxic composition.
Four, the current risk assessment process is very seriously
flawed, as it is based on ``individual pollutants.'' In 2009,
the National Academy of Sciences at EPA's request looked at
risk assessment and said in reality the risks are not the just
risks of the pollutant, the risks are the risks of the
``decision'' about that pollutant. For example, if you lower a
particle standard, it is going to affect agriculture, it is
going to affect the price of gas, and it is going to affect
small businesses in particular. The National Academy said the
risk assessment has to be on the decision itself because that
is what the public lives with. So the decision itself and all
of its important health consequences have to be taken into
account in the risk assessment.
Five, the public commenters that make presentations to
CASAC were eloquent and well reasoned in their presentations,
but CASAC did not even discuss them or weigh their comments and
that was a shame. So the feedback from industry, interested
scientists, and the American Lung Association, for example,
were not effectively taken into account by CASAC.
Six, the subcommittee that I was on did not adequately
inform the Administrator on the pitfalls, the scientific
limitations, and even the adverse health consequences that
would flow from a more stringent regulation. Not understanding
the feasibility of compliance, the economic hardships and
unintended adverse consequences places the Administrator in an
embarrassing position of possibly having to issue a standard
that might do more harm than good to public health.
Number seven, the final, and I think this is the most
important, the public is not going to be adequately informed
about the adverse effects (i.e., unintended consequences)
associated with new standards. They (the public) are informed
about the benefits but not the adverse side of the coin.
``Informed consent'' is a fundamental ethical principle that
has to be applied when you make decisions that will affect
people's lives. Informed consent must include, and elucidate,
the adverse consequences that flow from a standard or decision.
CASAC was not allowed to discuss any of the adverse
consequences associated with setting new standards. But the
public must live with all the consequences of the new standard.
In sum, the current process is very elegant, it is very
highly evolved, and it is very efficient for getting scientists
involved. But in my opinion, it is seriously flawed, it is
narrowly focused, and it is even ethically questionable.
It is important to reiterate that no one is really to blame
because all of the people that were involved, in my opinion,
performed their tasks enthusiastically and competently. Thank
you.
[The prepared statement of Dr. Phalen follows:]

Prepared Statement of Dr. Robert F. Phalen,
Professor of Medicine, and Co-Director,
Air Pollution Health Effects Laboratory,
University of California, Irvine

My participation in the CASAC subcommittee on Particulate Matter
(CASAC-PM) was stimulating and enlightening. Everyone involved,
committee members, EPA staff, and public presenters, were well-
qualified, efficient, and dedicated. But we are here, in part, to
explore possible improvements in the process, so I will summarize some
of my personal observations to that end. Many of the problems arose
from the outdated mandate that the U.S. EPA had to follow.

1. The mandate is too restrictive, and does not allow the
full competence of the EPA to be used in protecting public health.
Evaluating air pollutants one by one can lead to air standards that do
not make sense given the complexities of air chemistry (e.g.,
suppression of one pollutant can cause the mixture to have increased
toxicity). The mandate to err on the side of increased safety can also
be a disservice to public health. And the policy to set nationwide
standards can place unreasonable burdens on some industries and some
regions of the U.S.

2. Linear incrementalism, in which CASAC only comments on
each step in a long process, can lead to conclusions that do not pass a
``common-sense criterion.'' The questions posed to CASAC-PM appeared to
be restrictive, carefully crafted, and led to inevitable conclusions.

3. Defining particulate matter by aerodynamic mass
fractions, with composition not taken into account, is poor science in
my opinion, and it punishes some regions and industries. Furthermore,
it does not apply to ultrafine particles (the count can be quite large
without having any appreciable mass).

4. The current risk assessment process is seriously
flawed. It is based on individual mass fractions and can lead to
regulations that do not serve public health. The 2009 National Academy
of Sciences Report (National Research Council, ``Science and Decisions:
Advancing Risk Assessment,'' The National Academy Press, Washington,
DC, 2009) advises that the ``decision'' to set a standard, not the
``pollutant,'' is what must undergo risk assessment. The public must
live with all of the relevant consequences of an air standard, not just
selected effects of the substance under consideration (the general
economy, jobs, and costs of goods and services have dominant impacts on
public health, but they are not even considered by CASAC).

5. The public comments were not weighed and discussed by
CASAC-PM in spite of the fact that most were well-reasoned and
relevant. If the agenda included time for discussion of public comments
and formal acceptance or rejection of their recommendations, the
process might be improved.

6. The CASAC-PM subcommittee did not adequately inform
the EPA Administrator on the pitfalls, scientific limitations, and even
the range of adverse health consequences associated with the
recommended PM standards. Not understanding the feasibility, economic
hardships, and unintended adverse health consequences can place the
Administrator in the embarassing position of issuing a standard that
may harm public health.

7. The public will not be adequately informed about the
adverse effects associated with new standards. ``Informed consent'' is
a fundamental ethical principle that should be applied to mandates,
including air standards. Informed consent must include, and elucidate,
the adverse consequences that flow from a decision. CASAC-PM was not
allowed to adequately discuss the adverse consequences associated with
air standards.

In sum, the current process, although elegant and efficient, in my
opinion is flawed, narrow, and possibly ethically questionable.
It is important to reiterate that all of the people involved
performed their tasks enthusiastically and competently. Thank you for
this opportunity to provide what I hope are constructive comments.

Chairman Harris. Thank you very much, Dr. Phalen.
I now recognize Dr. Anne Smith, Senior Vice President at
NERA Economic Consulting, for five minutes. Dr. Smith.

STATEMENT OF DR. ANNE E. SMITH,

SENIOR VICE PRESIDENT, NERA ECONOMIC CONSULTING

Dr. Smith. Mr. Chairman, Members of the Committee, thank
you for inviting me. I am Anne Smith. My statements today
reflect my own opinions and not those of my company, NERA
Economic Consulting.
EPA is relying on benefit estimates, the so-called co-
benefits from ambient PM2.5 reductions to claim that the rules
that are not intended to address PM2.5 at all have benefits
larger than their costs. For instance, for the ozone
reconsideration, up to 91 percent of the estimated benefits
were not from reductions in ozone risk but from EPA's
predictions of coincidental PM2.5 reductions under that rule.
Ozone benefits alone always fell short of their costs by tens
of billions of dollars per year.
For the Ozone Utility MACT Rule, EPA claims this air toxics
rule will save up to 17,000 lives per year and many other
respiratory and heart ailments but all of those purported
benefits are due to PM2.5, not to air toxics. Over 99.99
percent of the Utility MACT Rule's estimated benefits are due
to PM2.5 co-benefits and not the air toxics that are its
purpose.
EPA has developed a habit since 1996 of relying on PM2.5
co-benefits to create a benefit-cost case for its non-PM
rulemakings. PM2.5 co-benefits were the driving factor in all
but two of the 24 non-PM air rules which EPA quantified any
benefits for at all as you can see listed here in this figure
with the Xs in the first column, and EPA's co-benefits habit
has really taken over since 2009. Claims of PM2.5 mortality co-
benefits now account for over 99.9 percent of the benefits in
all of the RIAs since 2009 except the ozone reconsideration, as
you can see in the last column of the table.
Where did all these PM co-benefits come from? How could
they be increasing in importance even though ambient 2.5 is
declining to the levels that EPA deems safe? This is how. In
2009, EPA changed the assumption that in one fell swoop nearly
quadrupled its estimates of the number of U.S. deaths due to
PM2.5. EPA decided to calculate risks from PM2.5 exposures that
occur far below the level it deems safe under the PM2.5 NAAQS.
Prior to this change, EPA was assuming that less than four
percent of all current U.S. deaths were due to PM2.5, and after
this change it is assuming that 13 percent of all current
deaths in the United States are due to PM2.5. All of these
newly calculated PM2.5 risks come from the most non-credible
source of risk calculation. By assuming that a unit of exposure
to PM2.5 at concentrations well below any epidemiological
studies will increase your risk of death by just as much as a
unit of exposure at the much higher PM2.5 levels that
statisticians have observed correlations for. This new
assumption is scientifically dubious but its lack of realism
shows in how it has driven the underlying estimate of PM-
related deaths in the United States to implausible levels.
EPA's new PM risk estimates now imply that 25 percent of all
deaths nationwide were due to PM2.5 as recently as 1980, 25
percent. EPA projects PM2.5 will be almost entirely in the safe
range, that is, in attainment with the PM NAAQS, after the
Cross-State Air Pollution Rule is implemented in 2014. But EPA
is also assuming there will still be about 250,000 deaths per
year due to PM2.5 after that and it keeps tapping into this
implausible estimate of a reservoir of PM benefits to come up
with the very large co-benefits estimates for its new rules,
and the Utility MACT is a shining example. The 17,000 lives
that EPA claims it would save in 2016 are taken from that
highly dubious pool. Effectively, all of those 17,000 deaths
are from exposures to PM that according to the EPA ambient
standard are safe. About 4,000 of those 17,000 deaths, if
believed to really exist, will be prevented anyway if EPA
tightens its PM2.5 standard down to 11 micrograms per cubic
meter as it appears to be poised to do this year. The Utility
MACT should not be credited with those benefits. They belong on
the PM NAAQS benefits ledger. But the remaining 13,000 of those
deaths should not be counted at all because they are from
exposures that are already below 11 micrograms per cubic meter,
which EPA does not consider to be risks that are believable
enough to set a PM NAAQS standard that would stop them. So in
sum, EPA's use of highly dubious co-benefits gives it a shield
to justify a complex web of rules.
The fact that EPA does not just use its streamlined
regulatory option, which is to set a PM2.5 NAAQS, hints at the
degree to which it realizes that those co-benefits calculations
do not reflect true public health risk but also it is just bad
policy. A complex web of non-PM rules cannot possibly be a
cost-effective path to addressing a nation's clean air needs,
and no one disputes that these rules do have significant costs
that need to be streamlined to the maximum extent possible.
Thank you very much.
[The prepared statement of Dr. Smith follows:]

Prepared Statement of Dr. Anne E. Smith,
Senior Vice President, Nera Economic Consulting
Mr. Chairman and Members of the Committee, thank you for your
invitation to participate in today's hearing. I am Anne E. Smith, and I
am a Senior Vice President of NERA Economic Consulting. I am a
specialist in environmental risk assessment and integrated assessment
to support environmental policy decisions, which was a core element of
my Ph.D. thesis at Stanford University in economics and decision
sciences. I have performed work in the area of air quality cost and
benefits analysis and risk assessment over the past 30 years, including
as an economist in the USEPA's Office of Policy, Planning, and
Evaluation, as a consultant to the USEPA Air Office, and in many
consulting engagements since then for government and private sector
clients globally. I have also served as a member of several committees
of the National Academy of Sciences focusing on risk assessment and
risk-based decision making. I have been deeply involved in assessment
of the evidence on risks from ambient fine particulate matter (PM2.5)
since EPA first turned to the task of identifying an appropriate
National Ambient Air Quality Standard (NAAQS) for PM2.5 over 15 years
ago. I have also analyzed costs, risks and benefits of many other key
U.S. air policies, including ozone, regional haze, mercury and other
air toxics, NO2, SO2, and greenhouse gases. I thank you for the
opportunity to share my perspective today on the economic underpinnings
of EPA's policy analyses for setting air quality standards. My written
and oral testimonies reflect my own opinions, and do not represent any
position of my company, NERA Economic Consulting.
The Chairman has asked me to describe my work analyzing major Clean
Air Act regulations including National Ambient Air Quality Standards
(NAAQS) and National Emission Standards for Hazardous Air Pollutants
(NESHAPs), and to discuss any trends I have identified in EPA's
analyses of such regulations. Although I have worked on these issues
for over 30 years, I would like to focus my testimony today on analyses
and research that I have done during 2011. In the past several months,
I have reviewed and commented on the costs and benefits in EPA's
Regulatory Impact Analysis (RIA) for the reconsideration of the ozone
NAAQS. I have also prepared technical comments on the RIA for the
proposed NESHAP for electric generating units, which was proposed in
May 2011. That rule is commonly called the ``Utility MACT'' rule
because it would impose maximum achievable control technology (MACT)
standards on several categories of air toxics emitted by electricity
generators. I am presently in the process of reviewing the entire body
of RIAs that EPA has produced for air quality regulations, to trace the
history of some troubling patterns that I found in EPA's ozone and
utility MACT RIAs.
My key findings, which I will explain in more detail below, are:

EPA is relying to an extreme degree on coincidental ``co-
benefits'' from PM2.5 reductions to create the impression of benefit-
cost justification for many air regulations that are not intended to
address PM2.5.

In 2009, EPA vastly increased the levels of mortality
risks that it attributes to PM2.5 simply by starting to assign risks to
levels of PM2.5 down to zero exposure, thus ``creating'' risks from
ambient exposures that are well within the safe range established by
the PM2.5 NAAQS.

This single change nearly quadrupled the pool of
purported U.S. deaths due to PM2.5 that RIAs can now count as ``saved''
by minor incremental reductions in already-low ambient PM2.5 levels
projected under new rules.

This additional pool of PM2.5-related mortality
consists of the most non-credible sort of risk estimate, as it is
derived from an assumption that a unit of exposure at PM2.5 levels well
below any observed in the epidemiological studies poses just as much
risk as a unit of exposure at the higher PM2.5 levels where
associations have been detected.

With this change, EPA is now assuming that 13% to 22%
of all deaths in the Eastern U.S. were due to PM2.5 in 2005, and that
25% of all deaths nationwide were due to PM2.5 as recently as 1980.

The decision to inflate the PM2.5 risk estimates by
presuming risks continue down to zero has its greatest impact on co-
benefits estimates because--for rules that do not address PM2.5
directly--a much greater share of their incremental reduction of PM2.5
will occur in areas that are already in attainment with the PM2.5 NAAQS
(and thus that have PM2.5 levels that EPA has deemed safe). Yet, EPA
now attributes about 200,000 more PM2.5-related deaths per year to
exposures in those areas.

If it were viewed as credible that such large effects
exist below the level of the PM2.5 NAAQS, the appropriate policy remedy
would be to tighten the PM2.5 standard, and not to regulate something
else altogether in order to obtain those benefits through
``coincidence.''

Co-benefits from a pollutant that EPA already can and
does regulate should not be allowed to serve as the predominant benefit
in RIA's for rules that target a different public health concern.

Otherwise, RIAs will only help drive our nation towards
regulatory complexity by creating the false appearance of a benefit-
cost justification for regulations that are very costly compared to
their own benefits.

EPA is relying to an extreme degree on coincidental "co-benefits" from
PM2.5 reductions to justify air regulations that are not
intended to address PM2.5.

As EPA releases each of its proposed and final air quality rules,
it typically emphasizes that the rule will generate health benefits
that exceed its costs. However, close inspection of the associated RIAs
reveals that a majority of those benefits--sometimes all of them--are
not from reductions in the pollutant(s) being targeted by the new
regulation, especially in the case of air regulations that are
targeting clean air objectives other than PM2.5. For many of those, the
bulk of the benefits estimates in their RIAs are attributable to
reductions in already-low concentrations of ambient PM2.5 that EPA has
predicted will occur coincidentally as a result of regulation of those
non-PM pollutant(s).
For example:

In the Ozone Reconsideration RIA, up to 91% of EPA's
benefits estimate for its preferred standard was due to EPA's
predictions of coincidental PM2.5 reductions rather than to reductions
in ozone risks that were the target of the rule. \1\ Not a single one
of EPA's benefits estimates in that RIA exceeded its costs unless
PM2.5-mortality co-benefits were added in. By EPA's own calculations,
all of the alternative ozone standards had ozone benefits that fell
short of their costs by billions of dollars per year. \2\
---------------------------------------------------------------------------
\1\ The preferred standard that EPA had forwarded to OMB for the
Ozone Reconsideration was 70 ppb.
\2\ A copy of my full review of the ozone RIA is available at
http://www.nera.com/67-7390.htm. Parts of it are excerpted
in the Appendix of this testimony.

EPA has widely claimed that the Utility MACT rule, which
targets air toxics, will save up to 17,000 lives per year, 11,000 heart
attacks, and numerous other respiratory and cardiovascular ailments.
But all of those purported health benefits are due to EPA's predictions
of coincidental reductions of PM2.5--which is not an air toxic. Of all
the air toxics targeted by this rule, EPA has estimated benefits for
only one--mercury--and EPA's highest estimate of those mercury benefits
is only $6 million per year, compared to EPA's estimate of $10.9
billion in costs per year. In the Utility MACT's RIA, over 99.99% of
the benefits that EPA has attributed to the rule are due to PM2.5 co-
benefits rather than to the air toxics that are its purpose. \3\
---------------------------------------------------------------------------
\3\ A copy of my full review of the Utility MACT RIA is available
at http://www.nera.com/67-7412.htm. Parts of it are
excerpted in the Appendix of this testimony.

In my ongoing review of all air regulation RIAs, I have identified
28 RIAs released since 1996 that were for rules not targeting PM2.5-
related health risks. These are listed in Table 1 in chronological
order.

Table 1 shows that in 22 of those 28 RIAs, I found that a majority
of the total benefits were due to PM2.5 mortality co-benefits. In fact,
PM2.5 co-benefits were the only benefits, or accounted for more than
99.9% of the quantified benefits in 13 of those 22. Of the remaining
six, four did not quantify health benefits at all (yet most of those
discussed PM2.5 co-benefits qualitatively as well as direct benefits of
the rule's targeted pollutants). This leaves just two of the 28 RIAs
that were not specifically targeting ambient PM2.5 yet did not find
that most or all of the quantified benefits were actually co-benefits
due to PM2.5. Overall:

PM2.5 health-related co-benefits have been relied on to
create the benefit-cost case for regulations that were actually
intended to address mercury, a host of other air toxics, ozone,
regional haze, lead, NO2, and SO2.

The trend towards almost complete reliance on PM2.5-
related health co-benefits has grown over time.

In 2009, EPA changed its RIA calculations to vastly increase the levels
of PM2.5 co-benefits that it can attribute to non-PM2.5 rules.

As noted above, EPA has been increasingly relying on PM2.5 co-
benefits to produce a benefit-cost case for a host of non-PM2.5 rules.
However, in my review of RIAs, I also realized that EPA made a move in
2009 that greatly increased those co-benefits estimates --and did so in
a way that I consider to have no scientific credibility. The co-
benefits that EPA estimates for rules that are not targeting ambient
PM2.5 are calculated from very small changes in PM2.5 concentrations
that are already well below the safe level established by the PM2.5
NAAQS. This is because those co-benefits are supposed to be computed
only for incremental improvements beyond existing regulations, such as
the existing PM2.5 NAAQS. The PM2.5 NAAQS imposes a maximum annual
average ambient concentration of 15 mg/m3, which the EPA Administrator
deemed to protect the public health with an adequate margin of safety
in 2006. That NAAQS is under review now, and EPA staff (with CASAC's
concurrence) has stated that the lowest level that it may be revised to
is 11 mg/m3. \4\ Nevertheless, in 2009, EPA suddenly started to
calculate PM2.5 risks in its RIAs down to the lowest level its air
quality models predict, which can be as low as 4 or 5 mg/m3. This
results in risks being attributed to exposures that are far below the
level of PM2.5 deemed safe. As I will show, those increased risk
estimates are very large. EPA is using those greatly inflated risk
estimates to justify a wide range of regulations other than PM2.5, even
though it is not prepared to argue that those risks are credible enough
to justify action in the form of an even-tighter PM2.5 NAAQS.
---------------------------------------------------------------------------
\4\ EPA, Policy Assessment for the Review of the Particulate
Matter National Ambient Air Quality Standards. EPA-452/R-11-003. Office
of Air Quality Planning and Standards, Research Triangle Park, N.C.,
April 2011, p. 2-106. (Available at: http://www.epa.gov/ttnnaaqs/
standards/pm/data/20110419pmpafinal.pdf.)

Risk Estimates Have Been Nearly Quadrupled. This decision by EPA to
calculate risks down to the lowest level that its models project,
rather than just to the lowest measured level (LML) in the
epidemiological study that serves as the basis for its risk
relationship greatly increased EPA's estimates of PM2.5 co-benefits in
its RIAs. This large inflationary effect can be observed just by
comparing EPA's baseline 2005 risk estimates in its 2010 PM2.5
Quantitative Health Risk Assessment for PM2.5--which does not
extrapolate below the LML--to those in its post-2009 RIAs which do
extrapolate below the LML. The former is being used the current review
of the PM2.5 NAAQS mentioned above, and in it, EPA estimates 88,000
deaths were due to PM2.5 in 2005 based on an epidemiological study by
Laden et al. \5\ In its concurrent RIAs, however, EPA estimates fully
320,000 deaths due to PM2.5 for the same year, the same estimated air
quality, and using the same Laden et al. study. \6\ The former is 4% of
total annual U.S. deaths of 2.4 million and the latter is 13% of 2.4
million annual U.S. deaths. Notably, EPA is now using both of these
contradictory estimates of baseline PM2.5-related deaths simultaneously
in different regulatory proceedings--EPA is using the smaller number of
baseline deaths in its CASAC-reviewed risk analyses for the PM2.5 NAAQS
review, and it is using the larger number of baseline deaths in its
RIAs that are generating the large co-benefits for non-PM2.5
regulations, such as for air toxics regulations and for non-PM NAAQS,
such as ozone.
---------------------------------------------------------------------------
\5\ EPA, Quantitative Health Risk Assessment for Particulate
Matter. EPA-452/R-10-005. Office of Air Quality Planning and Standards,
Research Triangle Park, N.C., June 2010, p. G-2. (Available at: http://
www.epa.gov/ttn/naaqs/standards/pm/data/
PM-RA-FINAL-June-2010.pdf.)
\6\ EPA, Regulatory Impact Analysis (RIA) for the Final Transport
Rule, Docket ID No. EPA-HQ-OAR*2009-0491, p. 3. (Available at http://
www.epa.gov/airtransport/pdfs/FinalRIA.pdf.)
---------------------------------------------------------------------------
Thus, with this single change in its RIA calculations, EPA has
caused the estimate of total PM2.5-related deaths to nearly quadruple,
from 88,000 to 320,000. In effect, in 2009, EPA quietly ``created'' an
additional reservoir of 232,000 PM2.5-related deaths that it could
continue to tap into in its future RIAs as co-benefits for the many
non-PM clean air regulations that it will be proposing and promulgating
in the future. The RIAs for the proposed Utility MACT and the Ozone
Reconsideration are recent RIAs that benefited from the dramatic
inflation of EPA's estimates of total PM2.5 risks, as I will show next.

Inflated Co-Benefits Estimates Are Being Calculated for Small
Changes in Exposure to PM2.5 that EPA Deems Safe. The Cross-State Air
Pollution Rule (CSAPR) that was promulgated in July 2011 is intended to
help bring the nation into compliance with the present PM2.5 NAAQS. The
RIA for the CSAPR reports that in 2014 it will save up to 34,000 lives
that would otherwise end prematurely due to PM2.5 exposures, as
compared to premature deaths in a baseline that did not even include
CAIR. \7\ One can think of this as a reduction from the 320,000
underlying deaths associated with 2005 levels of PM2.5. Even if we
assume that control measures between 2005 and 2014 additional to those
of CSAPR would double the estimated lives saved that EPA attributes to
CSAPR alone, EPA is estimating that there still will remain some
250,000 deaths due to PM2.5 even after CSAPR has been implemented in
2014. It is from this remaining reservoir of ``premature deaths''
(still nearly 10% of all U.S. deaths per year!) that EPA finds the
17,000 lives that it purports would be ``saved'' as a co-benefit of the
Utility MACT, when it comes into effect in 2015 and mandates reductions
of acid gases. \8\ When placed in the context of such a huge pool of
lives that still ``could be saved'' if PM2.5 were to be 100%
eradicated, it becomes apparent that the 17,000 lives of ``co-
benefits'' is a small percentage change that reflects the small
difference in PM2.5 exposures offered by the Utility MACT. The RIA for
the Utility MACT confirms that it provides not only a small percentage
risk reduction, but that its comes from very low exposures, as Figure
1, copied from Figure 6-15 of the Utility MACT RIA, shows.
---------------------------------------------------------------------------
\7\ EPA, Regulatory Impact Analysis (RIA) for the Final Transport
Rule, Docket ID No. EPA-HQ-OAR-2009-0491, p. 1. (Available at http://
www.epa.gov/airtransport/pdfs/FinalRIA.pdf.)
\8\ The benefits of the Utility MACT rule are calculated after
having modeled full implementation of the proposed Clean Air Transport
Rule, which was the proposed version of the final CSAPR. It is thus
fairly similar to an analysis of benefits after accounting for the
reductions expected from CSAPR.

Figure 1 shows that effectively all of the Utility MACT's purported
PM2.5 co-benefits are due to reductions in exposures to PM2.5 that are
already below the annual NAAQS standard of 15 mg/m3. This fact can be
inferred from the figure in the following way. The blue S-shaped curve
in Figure 1 indicates on the vertical axis the percent of the RIA's
PM2.5 co-benefits estimate that is attributable to baseline PM2.5
exposures at or below the PM2.5 concentration on the horizontal axis.
This is known as a ``cumulative distribution.'' The point on the
horizontal axis where the S-shaped curve just reaches 100% indicates
the level of baseline PM2.5 at or below which all (i.e., ``100%'') of
the estimated PM2.5 co-benefits occur. I have added a vertical dotted
red line to Figure 1 at the level of the current annual NAAQS (i.e., at
15 mg/m3 on the horizontal axis). As one can see, the vertical reading
on the blue S-shaped curve is about 100% at 15 mg/m3, which means that
about 100% of EPA's estimated PM2.5 co-benefits from the Utility MACT
would be based on reductions in annual average PM2.5 exposures that are
already below the health-protective level of the current standard. Not
only are most of the benefits occurring at very low PM2.5 exposures to
start with, but RIA also tells us that they are due to very small
exposure changes. The changes in exposure are only 0.7 mg/m3 on
average, and do not exceed 1.49 mg/m3 in any location. \10\
---------------------------------------------------------------------------
\10\ See Utility MACT RIA, p. 4-5, (at http://www.epa.gov/
ttnecas1/regdata/RIAs/ToxicsRuleRIA.pdf.)
---------------------------------------------------------------------------
EPA is presently considering whether to tighten the PM2.5 NAAQS,
with a Proposed Rule expected later in 2011. EPA is considering a range
of possible alternative annual standards that extends as low as 11 mg/
m3. If EPA revises the NAAQS to the lowest of those levels, the RIA's
figure also tells us that 20% of the co-benefits being attributed to
the Utility MACT (i.e., those that occur in locations where pre-rule
PM2.5 is above 11 mg/m3) are going to occur anyway, as a result of
NAAQS attainment. \11\ They therefore are inappropriate to count as co-
benefits of the Proposed Rule for air toxics--they should be counted as
the direct benefits of the new PM2.5 standard. Moreover, the remaining
80% of the Utility MACT's PM2.5 co-benefits are for reductions in PM2.5
exposures that will still be deemed safe by EPA.
---------------------------------------------------------------------------
\11\ 11 Some might argue that these PM2.5 benefits will appear
sooner because the Proposed Utility MACT Rule will be fully implemented
by 2016, while full implementation of a tightened PM2.5 NAAQS will be
several years later. However, that difference is only temporary, and
many have argued that the accelerated time frame for implementation of
the Utility MACT rule will be far more disruptive than EPA's cost
analysis indicates due to its exceedingly rapid implementation. Thus,
making a point that these could be considered valid temporary co-
benefits for the years 2016 through perhaps 2020 only raises the
question of whether that accelerated time frame is reasonable and
justifiable.
---------------------------------------------------------------------------
The Additional PM2.5 Benefits Estimates Are Not Scientifically
Credible. The significant inflation in PM2.5 health benefits that EPA
has introduced into its RIA calculations since 2009 is accomplished by
adding in benefits of the least credible sort because most of that
increase is due to benefits estimates below--often far below--the
levels of PM2.5 that have been observed in the scientific studies that
form the basis of the PM2.5 health effects literature. Thus, overnight
in 2009, in the course of preparing RIAs that are not subject to public
peer review, EPA dramatically escalated its estimates of benefits for
all of its RIAs. This had the most profound impact on its estimates of
benefits in the vast swath of the U.S. that has PM2.5 concentrations
below 10mg/m3: small changes in modeled PM2.5 in these areas used to
contribute nothing to the total estimated benefits of a regulation, but
they now contribute as much as 70% of the co-benefits estimates (as can
be seen in the case of the co-benefits in the Utility MACT RIA from
Figure 1). EPA accomplished this enormous benefits inflation without
changing the epidemiological studies it relies on, but by altering a
much more obscure assumption in its risk analysis calculations, the use
of the ``LML.''
One associated and interesting effect of this benefits inflation,
however, is the degree to which it makes the total number of deaths
attributed to PM2.5 implausible. EPA's presumption that fully 320,000
deaths in the U.S. were ``due to PM2.5'' in 2005 represents over 13% of
all deaths in the U.S. on average. And behind that average is the
presumption that in large expanses of the Eastern U.S., between 16% and
22% of all deaths in 2005 were ``due to PM2.5.'' By extension (although
EPA has not reported this calculation), EPA's estimates imply that
about 25% of all deaths nationwide were due to PM2.5 as recently as
1980. \12\ These fundamental assumptions that underpin EPA's co-
benefits calculations stretch the bounds of credibility, and thus
undercut the credibility of all the co-benefits estimates themselves.
---------------------------------------------------------------------------
\12\ See pp. 14-16 of my technical comments on the Utility MACT
(at: http://www.nera.com/67-7412.htm).
---------------------------------------------------------------------------
The simple reason why these new baseline risks are so large--
implausibly large in my view--is that EPA assumes in its risk analysis
calculations that there is no tapering off of relative risk as PM2.5
exposure approaches zero. For years there has been a debate about
whether the concentration-response relationship can truly be linear
down to zero, but this debate has been focused on questions of
statistical power and on basic principles of toxicology. The
implication of the linear-to-zero/no-threshold assumption has never
been debated in terms of its implication that an implausible proportion
of total deaths in the U.S. would be due to PM2.5--but perhaps now it
should be debated that way too.

The decision to inflate the PM2.5 risks by presuming risks continue
down to zero has its greatest impact on co-benefits estimates.

The vast increase in total deaths that EPA now attributes to PM2.5
exposures (i.e., the increase from 88,000 to 320,000 for the year 2005)
has a greater inflationary effect on estimates of co-benefits from
rules that do not address PM2.5 risks directly than it does on the
direct benefits of rules to steer ambient PM2.5 into attainment of its
NAAQS. In rules not targeting ambient PM2.5 directly, the changes in
PM2.5 are only coincidental and presumably incremental to attainment of
the PM2.5 NAAQS. Such changes are most likely to occur in areas that
are either already in attainment or will be pushed into attainment by
rules implementing the PM2.5 NAAQS. In fact, coincidental and
incremental reductions of PM2.5 that could qualify as co-benefits from
a non-PM rule must occur in locations that are already in attainment
with the PM2.5 NAAQS, or else those benefits are being double-counted,
because they have already or soon will be counted as the direct
benefits of the PM2.5 NAAQS itself. Hence, by inflating its PM2.5
benefits estimates with additional risk estimates of the least credible
form, EPA has enhanced its ability to justify non-PM2.5 regulations
through PM2.5 co-benefits. The practice of basing the benefit-cost case
for new rules almost solely on co-benefits rather than on direct
benefits is troubling to start with, but this recent change in EPA's
RIA benefits estimation methods now causes the bulk of the co-benefits
that it estimates to be quite suspect from a scientific basis.

If it were viewed as credible that such large effects exist below the
level of the PM2.5 NAAQS, the appropriate policy remedy would
be to tighten the PM2.5 standard, and not to regulate something
else altogether in order to obtain those benefits through
``coincidence.''

There remain many reasons to continue to have doubts about the
causality in the presumed relationship between ambient PM2.5 and
mortality. These calculations continue to rely solely on statistical
associations with little to no clinical evidence to support the causal
interpretation of these correlations. Despite many efforts to provide
statistical controls, the ability to tease out other explanations based
on phenomena that are correlated with variations in ambient PM2.5
levels remains elusive. Alternative explanatory factors may include
traffic, noise, and even socioeconomic conditions that have not been
possible to characterize fully with statistically useful data. Tighter
controls on PM2.5 may therefore not produce the benefits that EPA
calculates even for reductions from levels of PM2.5 that are in the
ranges of concentrations that have been measured in the epidemiological
studies. But to also assume that the presumed causal relationship
remains in effect with equivalent potency down to essentially zero
concentration levels is simply inappropriate scientifically. \13\
---------------------------------------------------------------------------
\13\ 13 For a more complete discussion of these points, see my
technical comments on the Utility MACT RIA, pp. 19-20, pp. 35-36, and
Appendix C, available at http://www.nera.com/67-7412.htm.
---------------------------------------------------------------------------
EPA and CASAC have not shown any willingness to argue for setting a
PM2.5 standard at those very low levels that have not yet been studied,
even though there is a complete and thoroughly effective mechanism in
the Clean Air Act that gives the Administrator the ability to protect
the public health from such exposures if they really do pose risks as
large as EPA assumes in its RIAs. EPA therefore should not continue its
practice of reporting that regulations that do not address ambient
PM2.5 will have benefits that exceed their costs based on estimates of
PM2.5 risks that EPA is not prepared to directly reduce through the
PM2.5 NAAQS.

Co-benefits from a pollutant that EPA already regulates should not be
allowed to serve as the predominant benefit in an RIA for a
rule that targets a different public health concern.

EPA's use of co-benefits in its RIAs scares the public into
believing that people would be dying in droves were it not for
implementation of new rules on pollutants for which EPA has not
actually identified any current public health risk. It gives EPA a
shield to justify building a complex web of many different rules, when
EPA could provide almost all of those purported health-protective
benefits with just a single rule: the PM2.5 NAAQS. That EPA does not
take this simple, streamlined approach hints at the degree to which it
realizes that its co-benefits calculations do not reflect true public
health risks. But also, it is just bad policy to promote the goal of
further PM2.5 risk reductions by way of expanding MACT rules for
mercury, acid gases, metallic air toxics and by way of striving to
attain tighter NAAQS for ozone, lead, SO2 and NOx. This cannot possibly
result in a cost-effective path to addressing a nation's clean air
needs.

Appendix

More Details From My Technical Comments on the Utility MACT and Ozone
RIAs

I have described and discussed the key trends of concern that I
have observed in my review of many RIAs, but I also would like to also
provide the summaries of the specific issues that I found in the two
RIAs for which I have written full technical comments. I believe that a
recap of my summaries for those two individual RIAs may help illustrate
the depth of the problems that are created by EPA's reliance on PM2.5
co-benefits as the central feature of its benefits analyses for clean
air rules that are not purposefully reducing PM2.5-related health
risks.

Summary of Key Findings From My Review of the Proposed Utility MACT
Rule

This section is excerpted from my Technical Comments on the
Regulatory Impact Analysis Supporting EPA's Proposed Rule for Utility
MACT and Revised NSPS (76 FR 24976) which was entered into the Utility
MACT docket as part of comments submitted by the Utility Air Regulatory
Group (UARG). The full comments can be downloaded from http://
www.nera.com/67-7412.htm.

Although EPA reports that the Proposed Rule will produce
annual benefits ranging from $53 billion to $140 billion, these
benefits have nothing to do with air toxics at all.

EPA's estimates of the direct benefits due to reduction
of the air toxics that are the specific purpose of this rulemaking
range from only $0.0005 billion to $0.006 billion per year \14\--less
than .01% of EPA's total benefits estimate--and this is due to
reduction of just one of the HAPs, mercury (Hg). EPA concluded it had
no basis for estimating benefits from reduction of any of the other EGU
HAPs.
---------------------------------------------------------------------------
\14\ Stated in a more readable format, the range of benefits
estimated for the air toxics is $500,000 per year to $6 million per
year. The Utility MACT RIA's summary Table 1-3 incorrectly states the
lower bound, and I am reporting the values from RIA Chapter 5 (Table 5-
7), and in the Proposed Rule (at 24979).

Effectively all of the $53 billion to $140 billion of
estimated benefits is due to ``co-benefits'' from coincidental
reductions of fine particulate matter (PM2.5), a pollutant that is
separately and independently regulated under the Clean Air Act (CAA) as
---------------------------------------------------------------------------
a criteria pollutant.

The PM2.5 co-benefits lack credibility because almost all
of that dollar value comes from exposures that are so low that EPA
deems them safe and is expected to continue to deem them safe after
completing its review of the current PM2.5 health standard this year.
Further, the reductions in exposure levels are very small, averaging
only 0.7 mg/m3 in annual average concentrations. \15\
---------------------------------------------------------------------------
\15\ 15 Utility MACT RIA, p. 4-5. To put this in context, the
annual average standard (i.e., the level protective of public health
with an adequate margin of safety) is 15 mg/m3, about 20 times larger.
Even the maximum decrease in PM2.5 projected under the Proposed Utility
MACT Rule is only 1.49 mg/m3 (ibid.).

The PM2.5 co-benefits also lack credibility because of a
long list of well-documented technical problems with the way EPA
chooses to calculate actual health risks from statistical associations
that have not been reliably shown to reflect causal relationships.
These causality questions are particularly pronounced with respect to
individual PM2.5 constituents such as sulfate, which is almost the only
---------------------------------------------------------------------------
constituent accounting for the Proposed Rule's co-benefits.

Prima facie evidence of the non-credibility of EPA's co-
benefits estimates exists in EPA's baseline estimates of risk in this
RIA: deaths that were ``due to'' ambient PM2.5 exposures exceeded 20%
in areas of the U.S. in 2005. These co-benefits assumptions also imply
that over 40% of deaths were due to PM2.5 in parts of the U.S. during
the period 1979-1983 when PM2.5 concentrations were approximately
double those for 2005. These surprisingly high assumptions about
baseline risk, which in my opinion stretch the bounds of plausibility,
are the result of a single assumption change in 2009 in EPA's RIAs to
extrapolate risks below the ambient PM2.5 levels that have been
studied, to as low as background (i.e., nearly zero).

RIAs are not subject to peer review by EPA's Clean Air
Scientific Advisory Committee (CASAC) or to a public comment period.

EPA has not made this assumption change in any of the
risk analyses supporting its current review of the PM2.5 health
standard, which are subject to CASAC review.

The PM2.5 co-benefits estimates are virtually all tied to
attainment of the Proposed Rule's MACT for acid gases, which is the one
MACT category in this Proposed Rule for which EPA has not offered any
evidence of health risk.

Given that almost all of the co-benefits are solely
attributable to the acid gas MACT portion of the Proposed Rule, there
is no cost-benefit case for the remainder of the HAPs control
requirements in the rule, whether their estimated co-benefits are
included or not.

In light of the above points, which are further elaborated in the
rest of my comments, I conclude that the lower bound of the PM2.5 co-
benefits should be zero, and that EPA's upper bound PM2.5 co-benefits
estimate is just not credible. EPA has not even quantified any benefits
for the HAPs themselves, other than a tiny benefit from Hg reduction.
More importantly, I conclude that EPA's argument that there is a
strong cost-benefit justification for the Proposed Rule is
inappropriate because it is based solely on a preponderance of co-
benefits from a pollutant that is already regulated, and not an air
toxic. Moreover, the estimate is almost entirely derived from changes
in very low concentrations that EPA has deemed adequately protect the
public health. In the meantime, EPA has not been able to quantify, or
even clearly identify, any meaningful amount of direct benefits from
the reductions in air toxics that this rule mandates. The maximum ratio
of direct benefits to costs for all three MACT groupings is 0.0006-to-
1, with a net loss of about $10.9 billion per year. Each individual
MACT grouping appears to impose a net benefit-cost loss on the basis of
its direct benefits only, and two of those groupings appear to impose
net losses even if their share of the upper bound estimates of co-
benefits is included in the net benefit calculation.

Summary of Key Findings From My Review of the Ozone Reconsideration RIA

This section contains excerpts from the beginning and end of my
report, ``Summary and Critique of the Benefits Estimates in the RIA for
the Ozone NAAQS Reconsideration,'' which was prepared for the American
Petroleum Institute. The full report can be downloaded from http://
www.nera.com/67-7390.htm.
The excerpts below have been modified to fit the current document's
figure numbering.
EPA's statements on health benefits from lowering the Ozone NAAQS
grossly misrepresent what EPA is actually estimating as the potential
benefits of reducing public exposures to ozone. If based on ozone
benefits alone, not one of EPA's estimates of the benefits of reducing
ozone to a tighter alternative ozone standard is as large as the costs
of attaining that respective ozone standard--all cost more than the
ozone benefits they might provide.
EPA's estimates of ozone benefits are less than their costs despite
the fact that EPA has now escalated those benefits by always including
benefits due to ozone-related mortality. EPA's science advisors (CASAC)
found no ``causal'' link established between ozone and mortality during
their deliberations, but EPA now presumes, as part of the
reconsideration, a causal link between ozone and mortality risk.
Despite this change that is unsupported by CASAC, EPA's net benefits
estimates for ozone standards tighter than 0.075 ppm are all still
deeply negative.
The only way EPA finds benefits greater than costs for a tighter
ozone standard is to add in health gains from concomitant reductions in
PM2.5 that may occur while reducing ozone precursors--``co-benefits''
that have nothing to do with ozone exposures. Thus, EPA's claim that
tightening the Ozone NAAQS has greater benefits than costs has nothing
to do with reducing risks from ozone. EPA also has inflated the
magnitude of these co-benefits as part of the reconsideration through
several specious assumption changes. The Agency's inflated co-benefits
assumptions during this reconsideration represent a change compared to
those assumed in the original Ozone NAAQS review ending in 2008. Even
with both ozone mortality benefits and PM2.5 mortality co-benefits, a
large fraction of EPA's net benefits estimates are negative.
Figure 2 illustrates the Supplemental RIA's estimates of the net
benefits of each of the alternative ozone standards (relative to the
standard of 0.084 ppm) when no PM2.5 co-benefits are included. Even
using the highest estimate of ozone mortality benefit in the RIA
combined with the lowest EPA cost estimate, the estimated net benefits
of the 0.075 ppm standard are about ^$4.5 billion relative to the 0.084
ppm standard while the yet-tighter alternative standards (i.e., 0.070
through 0.055 ppm) have estimated net benefits ranging from ^$8.8
billion to ^$12.7 billion. If one treats the ozone-mortality
association as non-causal, EPA estimates that the current ozone
standard of 0.075 ppm would have net benefits of ^$7.5 billion and the
yet-tighter alternative standards of 0.070 through 0.055 ppm would
range from ^$18.8 billion to ^$76.7 billion. In fact, if there is no
causal relationship between ozone and mortality risk, the net benefits
estimates for standards tighter than 0.075 ppm remain negative even
with the inclusion of the highest of EPA's PM2.5 mortality and
morbidity co-benefits and using the low end of its cost range.

Chairman Harris. Thank you, Dr. Smith.
And I now recognize our final witness, Mr. J. Edward
Cichanowicz, for five minutes to present his testimony. Mr.
Cichanowicz.

STATEMENT OF MR. J. EDWARD CICHANOWICZ, CONSULTANT

Mr. Cichanowicz. Thank you. Good morning, Mr. Chairman,
Members of the Committee. This morning I would like to
summarize my opinion on the time that it takes to design and
install environmental controls, and then I would also like to
present a few graphics that convey the size of this equipment
and how it fits into a power plant.
I will discuss three types of equipment: flue gas
desulfurization scrubbers that remove sulfur dioxide, catalytic
converters that remove nitrogen oxides, and a fabric filter, or
baghouse, that removes particulate matter. Experience with this
equipment in the last 10 years demonstrates that the timeline
for most projects starting all the way from conceptual design
including permitting and picking the right contractor and
through commercial operation is more than 40 months for
scrubbers and catalytic converters. Fabric filters will require
less time.
The mandates of the Cross-State Air Pollution Rule and the
Utility MACT are based on EPA's assumption of a much shorter
time frame, less than 30 months. Notably, others within EPA
agree with the estimates of near 40 months. Specifically, in a
rulemaking addressing the retrofit of best available control
technology to reduce NOx at a power station in New Mexico, EPA
staff conducted their own survey and concluded that 37 months
on average was required to retrofit catalytic converters. In
summary, the mandates for the Cross-State Air Pollution Rule
and the Utility MACT presume an installation time that is not
supported by recent experience.
If I could ask you to turn your attention to the graphics,
I would like to show first a power station. This is a satellite
image from a power station that is in Georgia. The reason why
it is important is that they have already retrofit or are in
the process of retrofitting the controls that we have just
talked about. They have done it in response to a state mandate,
and what we are going to do is look at the layout of the
equipment as originally designed, and then we are going to walk
around the unit and see what kinds of things have been put in
and where they have been put. First, I will call your attention
to the original equipment as designed, and that is denoted by
the blue callouts. On the top, you see the boiler house and
below it the initial original particulate collectors. The
boiler house is indicated by the red rectangle and it contains,
as you would guess, the boiler and the steam turbine and a lot
of the fuel handling equipment. Right below it is the original
particulate control device, and you will see it has its own
stack. The first thing we will do is look and see what happens
when we install a catalytic converter. This station is so big
that it actually has two parallel flow paths so it is kind of
like having two environmental control systems on one site, and
what you see in the green is one of the two catalytic reactors
being retrofit. At the time of the image, that construction was
almost complete on one and was starting on the second. It is
important to know that these catalytic converters are basically
hung off the back of the boiler house so they are a couple
hundred feet in the air. They are quite large devices.
As you continue to walk around the station, we see first
one of the fabric filters that has been retrofit. Now, the good
news for this site is that there is actually room to put one.
The bad news is, it is not real close to the original equipment
so you can see the fair amount of ductwork that had to be
configured to tie it into the existing equipment.
As we continue to walk around, we see the location for the
scrubber. It is being constructed within the green circle, and
again, there is space for it, it just isn't very close to the
balance of plant equipment and you need the additional
ductwork. Also please note, there is an additional stack that
is being constructed. For the type of scrubbers that most
people prefer, they essentially saturate the gas with water,
and what that means is that when the gas goes up the stack, it
basically rains inside the stack and so you kind of have to
build a new stack to tolerate those set of conditions. And then
finally, the second of the two fabric filters is shoehorned in,
and again, there is space for it, it just has to be configured
within the site.
The next graphic or the only other graphic I will show is
an absorber tower for a scrubber. First, I will call your
attention to the red circle in the lower right. That is the
standard reference person, and it gives you an idea of how
large the equipment is. The gas enters on the lower right,
proceeds vertically upward and then exits through the top to
the left. The reason why these devices are so big is that the
speed of the gas has to be slowed down to about anywhere from
four to seven miles per hour. We have to mix other chemicals
with it and we also have to make sure there is enough residence
time for the reactions to take place. If you will note on the
top of the tower, there is a series of nozzles that spray an
alkaline material.
Let me give you an idea of how much water is there. If you
were to, say, stumble into a monsoon on the Pacific Rim, you
might have a rainfall of 15 to 20 inches of rain per hour.
Within that scrubber, it is generated 80 to 100 inches of rain
per hour. So the point is, it is a very powerful spray and it
sets up an engineered set of really perfect conditions that are
intended to annihilate almost anything coming out in the gas,
and for many cases, they do so.
So in summary, the reason why these devices take the time
that they do is that they are complex, engineered systems.
Their design needs to be thought out and they need to be
constructed with precision.
Thank you.
[The prepared statement of Mr. Cichanowicz follows:]

Prepared Statement of Mr. J. Edward Cichanowicz, Consultant

Summary

The U.S. Environmental Protection Agency (EPA) states that
planning, designing, and constructing state-of-the-art emission control
systems can be accomplished in less than 30 months. EPA made such
claims when publishing its Cross State Air Pollution Rule (CSAPR) this
summer and when proposing National Emissions Standards for Hazardous
Air Pollutants for electric generating units (the Utility MACT
proposal) last spring.
Contrary to EPA's claims, however, the ``start to finish'' times
for this equipment--flue gas desulfurization (FGD) ``scrubbers'' to
reduce sulfur dioxide (SO2); selective catalytic reduction (SCR)
catalytic reactors to reduce nitrogen oxides (NOx); and fabric filters
to remove particulate matter--will be much more than 30 months. The
most recent experience shows retrofits of FGD and SCR systems will
typically take between 40 and 50 months.
There are two key reasons why it takes this long to plan, permit,
fabricate, and install these control systems. First, the equipment is
large and must be configured to fit into often-crowded plant sites.
Second, these are not off-the-shelf designs; each must be custom-
tailored to the site and coal. The problems are compounded when several
control technologies are retrofit simultaneously at one plant site. For
example, in order to comply with the CSAPR and the Utility MACT rule as
proposed, owners may need to install FGD scrubbers and SCR catalytic
converters at their plant sites at about the same time they are
installing fabric filters to reduce mercury and other pollutants
targeted by the proposed Utility MACT rule.
Few ``short cuts'' are available to significantly reduce this
schedule by more than a few months. I do not concur with EPA's
statement in the CSAPR rulemaking that years can be carved off the
``start-to-finish'' times of FGD and SCR systems by fast tracking
design and procurement processes. And there is a downside to fast
tracking these kinds of projects: it could compromise the quality of
design or construction of the equipment, forcing plant operators for
decades to use control systems that are ill-suited or otherwise not
optimal for their sites.
Similarly, EPA has no basis to predicate the feasibility of its
very tight Utility MACT compliance deadlines on principally one
methodology: injecting specially prepared powders or sorbents into
power plant gas streams to remove mercury and other pollutants such as
hydrogen chloride. These sorbents will work in some instances, but not
across the board as envisioned by EPA to achieve the targeted
reductions. And where the use of sorbents does not achieve the very low
emission limits proposed by EPA, owners will have to retrofit fabric
filters to meet the proposed Utility MACT rule. EPA predicts as much as
half of the generating inventory in the U.S. will have to do so. Under
the best of circumstances, it would be difficult (if not impossible) to
retrofit fabric filter controls at so many sites in the short MACT
compliance timeframe. The challenge becomes even greater, though, if
owners must install fabric filters at roughly the same time they are
installing FGD and SCR systems to comply with the CSAPR.

Introduction

Chairman Hall, Ranking Member Johnson, and Members of the
Subcommittee, thank you for the opportunity to speak with you today. I
will provide an overview of the factors that affect the retrofit of
environmental control technologies to coal-fired power stations for the
purpose of meeting the mandates of the Cross State Air Pollution Rule
(CSAPR) and the National Emissions Standards for Hazardous Air
Pollutants (i.e., Utility MACT), particularly factors that influence
the timing of installation.
The power industry in the last 10 years has successfully retrofit
state-of-the-art environmental controls to a large fraction of
generating units. Consequently, much of the power delivered into
today's markets is generated by units equipped with effective
environmental controls. The industry will continue to strive to meet
future environmental mandates. However, as I will describe, the type of
equipment that must be retrofit is exceedingly large in size, can be
very complex, and can require special engineering and preparation
tasks. To do this right simply takes time. Further, we have learned
from experience what happens when the design or fabrication of a
control technology is rushed, or is not optimized or properly designed
for a given site and fuel. The outcome is never good.
Based on my years of experience advising power generation equipment
owners in the retrofit of environmental control technology, I believe
that typically between 40 and 50 months will be required to retrofit
control options to meet the mandates of the CSAPR and the Utility MACT.
It may be possible to reduce a few months from the schedule by fast
tracking design and procurement, and using so-called ``lean''
construction methods, but in general it will not be possible to achieve
this outcome in less than 30 months. Further, a result of fast tracking
these duties could be a compromise in the quality of design or
construction of this equipment. Operators would be forced for decades
to use equipment that is not optimal for the site, orotherwise ill-
suited.

Description of Environmental Controls

The industry selects from a suite of environmental controls those
appropriate for a given task: removing sulfur dioxide (SO2), nitrogen
oxides (NOx), particulate matter, and trace species commonly referred
to as hazardous air pollutants (HAPS). HAPS include mercury and acid
gases such as hydrogen chloride. There are two distinguishing features
of environmental controls for power plant effluent gases--first, the
equipment is very large, and second, there is no one-size-fits-all
design. Most equipment represents a custom design tailored to the
characteristics of a particular site, and coal.
Exhibits 1-3 depict three of the key control technologies utilized.
Exhibit 1 depicts a flue gas desulfurization (FGD) process that removes
SO2. Exhibit 2 depicts a selective catalytic reduction (SCR) module to
reduce NOx; this module is basically a catalytic converter for a power
station. Exhibit 3 depicts a fabric filter or baghouse, which filters
out particulate matter. As will be discussed subsequently, the fabric
filter device is important not only to control particulate matter, but
also to contribute to limiting emissions of HAPS. Each of Exhibits 1-3
is meant to convey the large equipment size that is necessary to
process the volume of gaseous combustion products. The equipment must
be of large flow cross-section to reduce the velocity or speed of the
gas to very low levels, to allow mixing of chemical reagents, and to
provide time for reactions to take place to completion. For example,
the speed of the gas in the FGD absorber tower is typically 5-10 feet
per second--or about 3-7 mph--necessitating a large reactor.
The vessel size becomes problematic when it must be fit into an
existing crowded site. It is the challenge of fitting these vessels
into crowded sites, all of which differ in almost limitless ways, which
can require a protracted design and installation effort. Exhibit 4
presents a plant layout--in my opinion of intermediate difficulty--
containing this equipment, showing how environmental controls can be
arranged. Some sites can be open and offer less challenge, but a
notable number of units will face space limits.
In summary, the retrofit of environmental controls--all utilizing
large vessels or reaction chambers, some with chemical and byproduct
support plants--requires custom design, shop fabrication, and
installation that take a lot of time.

Timeline for Equipment Installation

The ``start-to-end'' time line to install these control systems
includes many steps above and beyond just the work to prepare a
detailed design and install the equipment. A discussion of the key
steps required and a summary of recent experience is insightful. A
detailed description of the key steps and recent relevant experience is
presented in a report submitted to EPA on October 1, 2010 (a copy of
which is attached to these comments), as part of comments to the
rulemaking process. I'll summarize both topics.

Ten Steps: Project Initiation to Completion

The complete scope of activities to retrofit environmental controls
can require as many as 10 separate steps, each of which will vary by
project. Several of these steps can be conducted in parallel, but most
require some sequence--at least some portion of one activity must be
completed before the next is started. The 10 steps, including the range
of time (in months) for execution, are:

Conceptual Design and Preparing a Specification. What you
want to build must be described in a way bidders can use to derive a
design (6-12 months).

Identification of Qualified Bidders. Potential
contractors are to be identified; this process is typically conducted
in parallel with the preceding (one month).

Solicitation and Review of Bids and Contractor Selection.
Onsite ``walkdowns'' are essential to acquaint bidders with the
project. Evaluating capabilities is key; cost alone is not the
determining factor in contractor selection (3-5 months).

Negotiating Contract Terms and Conditions. Acceptable
terms and conditions for labor and material, including escalation, are
negotiated in advance (1-5 months).

Securing Construction, Operating Permits. Permits--issued
by local regulatory agencies and public utilities commissions--are
required before construction can begin. Some preliminary design must be
completed to define equipment and estimate emissions. Opening a storage
site for byproduct material is most challenging (4 months-4 years).

Finalizing Design. Producing engineering drawings is key,
with detailed estimates of media emissions, to enable equipment
purchase and fabrication (15-45 months).

Mobilizing the Workforce. Identifying and securing the
services of the mobile, specialized workforce has been rate-limiting
for some projects (1-3 months).

Construction. Includes soil, foundation, and structural
preparatory work; fabricating, transporting, and erecting equipment.
This is the most protracted onsite activity (25-40 months).

Process Tie-In (1-3 months) and Process Start-Up (1-3
months) are the final steps.

Each of these steps is essential, although some can be
expeditiously conducted depending on the site. For example, where an
owner has negotiated a long-term strategic agreement with one supplier,
the steps of contractor selection, evaluation, and contract negotiation
for any project may take less time. However, getting the long-term
agreement in place at the start is a lengthy process.
Projected time lines that do not consider each of these steps may
not reflect the true ``start'' date, and will not be accurate.

Recent Experience and Lessons Learned

The power industry, working with the community of equipment
suppliers, has extensive experience retrofitting environmental control
technology to generating units. Most recently, a significant fraction
of the generating inventory was retrofit with FGD scrubbers and SCR
catalytic converters over the time period of 2008 through 2010.
Specifically, a total of 123 generating units were retrofit with FGD,
and a total of 40 units were retrofit with SCR catalytic converters
from 2008 through 2010.
During the past two years, I have been involved with or had the
opportunity to review retrofit projects for 22 FGD scrubbers and 14 SCR
catalytic converters. As I described in the previously referenced
October 1, 2010, report that was submitted to EPA, the time required to
execute each retrofit--from ``start-to-finish''--varied between units
and sites. For FGD retrofits, completing all duties for the least
complex projects--those that retrofit a single FGD process at a single
site--took from 40 to 64 months, with the average of projects being 48
months. The shortest of these schedules--40 months--was incurred for a
unit that applied the process design from a near-identical ``sister''
unit, and was able to construct several critical facilities in
parallel. The retrofit of multiple FGD equipment to more complex sites
can require more time.
For SCR catalytic converters, the complete scope of duties for the
least complex projects required from 28 to 46 months, with an
intermediate project taking 40 months. The shortest of these
schedules--28 months--was achieved as the subject unit was on the
``end'' of a row, providing improved access for cranes and other heavy
fabrication equipment. Similar to FGD, the retrofit of SCR equipment to
more complex sites with multiple units requires more time, up to 60
months.
Some within EPA appear to agree it will typically take more than 21
(or even 30) months to install SCR. In an unrelated rulemaking to
establish Best Available Retrofit Technology (BART) to limit NOx
emissions from the San Juan Generating Station (SJGS) in New Mexico,
EPA determined recently (on August 22, 2011), that on average it takes
37 months to retrofit an SCR system on an existing unit. And EPA
determined that it would be reasonable for the owners of SJGS to have
five years to undertake and complete the SCR retrofit at SJGS.
In summary, under the best conditions, an FGD scrubber will require
at least 40 months to retrofit, with most applications between 40 and
50 months. For SCR, under the best conditions an SCR catalytic reactor
will require 28 months, with most applications averaging 44 months.

Compliance Timing: ``Logjam'' of Events 2012 to 2015

The emissions reductions provisions of the CSAPR and the Utility
MACT require control technologies to be installed and operational at
almost the same time--January 1 of 2014 for the CSAPR, and January 1 of
2015 for the Utility MACT rule. Given the time required to prudently
design and install control equipment, it is not possible for operators
affected by these regulations to meet these deadlines. This becomes
clear with a further elaboration of the needs of each mandate.

2014 Mandates of the Cross State Air Pollutant Rule (CSAPR)

The CSAPR requires affected companies in the so-called ``Group 1''
states to achieve the mandated SO2 reductions by January 1 of 2014. The
amount of generating capacity, and the number of FGD scrubbers that
need to be installed to achieve this compliance, has been projected by
EPA as part of the Agency's analysis in support of the rule.
EPA's initial estimates of technology retrofit for the CSAPR, as
first published in 2010, projected that 85 units generating 25 GW of
capacity would retrofit FGD to comply with the 2014 mandate. In the
final proposal for the CSAPR in July of 2011, EPA revised downward the
estimates of FGD to 39 units generating 17.4 GW of capacity. The basis
of EPA's downward revision appears to be a consequence of altering the
modeling details and lowering the projected load growth. Based on the
typical FGD ``start-to-finish'' scope discussed of 40 to 50 months, any
owners that must comply would already have had to start--and in fact
should be more than one year into these efforts. Given the date of the
final release of the CSAPR--less than 60 days ago on July 7 of 2010--
the timing presumes owners started engineering well in advance of
finalizing EPA's rule.
EPA's rationale in proposing the 2014 date is not only that a 27-
month time line is typical for FGD, but also that owners can start work
without risk prior to the promulgation of a final regulation. This is
not the case. Historically, there have been instances where owners have
quickly and proactively responded to a pending rule, only to witness
the rule being changed or delayed. As a result, construction is
terminated, or acquired SO2 allowances cannot be utilized. The owner
must absorb any ``sunk'' costs for equipment or allowance purchase.

2015 Mandates of the Utility MACT Rule

Perhaps more challenging is the schedule presented by the National
Emissions Standards for Hazardous Air Pollutants (NESCHAPS)--the
Utility MACT mandate. Compliance strategies for this proposed rule--
scheduled to be finalized by the end of this year--are uncertain. The
control technologies discussed in this testimony so far--the FGD
scrubber, the SCR catalytic converter, and the fabric filter--can
contribute in ways both large and small to MACT compliance. Owners of
generating units are investigating how to best utilize these
technologies for MACT, recognizing the degree of control required for
both mercury and hydrogen chloride is at or beyond the capabilities of
these controls in most applications.
However, EPA is predicating success--timely compliance with the
MACT--based principally on one methodology. This method entails
injecting into the gas one or more specialty powder(s), referred to as
sorbents, to remove mercury and hydrogen chloride. One class of
sorbents, known as activated carbon, is intended to remove mercury from
combustion products. A second class of sorbents--actually a family of
materials derived from the mineral trona--is intended to remove acid
gases, such as hydrogen chloride. EPA believes any shortcomings in
sorbent performance can be compensated by retrofitting fabric filters
to 166 GW of capacity--more than half of the national inventory of
units in 2015.
Regarding mercury, experience with activated carbon in
demonstration tests suggests this sorbent will be successful on many
units. However, as noted in an August 2011 report addressing mercury
control technology (a copy of which is attached to this testimony) for
the proposed Utility MACT rule, there may be an equal population of
units that will not meet the targeted mercury limit. There may also be
units where the carbon sorbent induces operating problems, or increases
the emissions of particulate matter.
Regarding control of ``acid gases'' such as hydrogen chloride, the
uncertainty is far greater. EPA, to its credit, developed an extensive
database of emissions of HAPS species from power generation equipment.
Regrettably, certain elements of the database were either ignored or
not properly utilized. EPA's proposed hydrogen chloride limits presume
that sodium sorbents can be a sole means to comply--despite the fact
that of the 11 units in EPA's database using this approach, there are
only two units with data suggesting such success. EPA predicted in a
March 17, 2011, document that 56 GW of capacity would deploy this
sodium-based sorbent approach. It is hard to believe the design for so
many commercial systems can be successfully scaled, and equipment
installed, on such limited experience.
For both the mercury and hydrogen chloride MACT mandates, EPA's
``backstop'' approach is broad application of fabric filters to 166 GW
of capacity. Again, it is hard to believe that such capacity can be
retrofit with both sorbent injection systems and fabric filters, and
successfully operate as predicted, in slightly more than three years.
Furthermore, the proposed fabric filter retrofits are to be achieved at
the same time the technologies for CSAPR are being deployed. Such a
schedule would stretch supply sources in 2013 and, in my opinion, well
into 2014 as the FGD units are delayed. Although the task of installing
any single fabric filter collector may be less onerous than a FGD or a
SCR catalytic converter, many of the steps are still the same.

Review of Key Uncertainties

In summary, several key uncertainties behind the proposed mandates
in 2014 and 2015 should be considered:
Equipment Installation Timeline: EPA's assumed time line for
equipment installation--based on experience gathered from 2008 through
2010--is unrealistic. The FGD and SCR installations completed prior to
2010 were mandated five years prior to the compliance date. As EPA
hasnoted in CSAPR rulemaking documents, some large system owners
initiated work prior to 2005, but in response to incentives to acquire
SO2 allowances. Owners had a financial incentive to deploy technology
early--and not a disincentive of putting capital at risk, which is the
present case.
Capability of Sorbent Injection for Hg Control. The use of
activated carbon sorbent to remove mercury has been demonstrated to
meet the proposed MACT mercury limit for several categories of
generating units. However, an equal number of generating units could be
at risk to meet the proposed MACT limit using activated carbon
sorbents, unless a fabric filter is retrofit.
Capability of Sodium-Based Sorbents to Remove Hydrogen Chloride to
the MACT Limit. Sodium-derived sorbents have been used to remove acid
gases such as hydrogen chloride, but there is limited experience in
achieving the low levels mandated by the MACT. At this time there are
only two operating units with data suggesting this option can
potentially meet the proposed Utility MACT rule.
Capability to Broadly Retrofit Fabric Filters. EPA's analysis of
complying with the MACT is predicated on the ability to successfully
retrofit 166 GW of generating capacity with fabric filter controls by
January of 2015. As noted in an analysis that I co-authored and
submitted in July of 2011 as comments to MACT (a copy of which is
attached to this testimony), it is unlikely this amount of fabric
filter control technology can be retrofit by January of 2015.
Successfully retrofitting fabric filters to this capacity alone would
be a challenge, much less conducting this work contemporaneous with FGD
scrubber retrofit for the CSAPR.

Chairman Harris. Thank you very much, Mr. Cichanowicz.
And what we will do now is, we will start with five minutes
of questions, a five-minute round of questions, and I recognize
myself for the first five minutes.
Dr. Phalen, let me just ask you a question just to
elaborate a little bit about your testimony because what you
imply is that the decisions made by CASAC are made--or I should
say the advice is not made as to the entirety of the decisions
and the other potential ramifications of that decision. Rather,
it goes to a very specific question that is asked without
taking the entire context. Is that correct?
Dr. Phalen. Yes, sir, that is correct. And when you look at
such isolation of an issue, it causes trouble everywhere else.
Chairman Harris. Sure, so that when you have--for instance,
there is a study from Stony Brook University earlier this year
that found that people who experienced a period of unemployment
were more than 60 percent more likely to experience premature
mortality. So you have some co-factor there, and in this case,
everybody agrees, I think that if you lose your health
insurance, you are more likely to have premature mortality. I
mean, I think probably there is uniform agreement on that. If
you don't have a job, you are more likely not to have health
insurance. Therefore, you can connect something that reduces
the number of jobs to actually an adverse health effect. Is
that what you are talking about when you are saying that the
panel should be allowed to make statements regarding the
entirety of a decision?
Dr. Phalen. Affirmative, Mr. Chairman.
Chairman Harris. Okay. Thank you very much.
Dr. McClellan, you stated in your testimony that CASAC
specifically said that it was a science and policy judgment
when these standards are being suggested. Now, I am a little
puzzled by that. How are you making policy judgments without
science? I mean, is that what is happening at EPA, that in
fact, you are infusing policy into these decisions that are not
based on science or without a scientific background?
Dr. McClellan. In the CASAC review of the science in the
two recent cases, the PM2.5 and then the ozone, they chose to
set a bright line. They said in the case of the PM2.5 that the
Administrator should set the annual standard and higher than 14
micrograms per cubic meter. The Administrator renewed it at 15
and they stamped their feet and said you didn't follow our
advice. Anybody knows that 14 is one integer less than 15 but
to suggest that the science was so compelling that 14 was
acceptable, 15 was not, that is a blending of policy and
science. In the case of ozone, they said 60 to 70 PPb. They
tried to sort of justify it by saying well, we gave you a
range. I don't presume to say that they would have stamped
their feet if you had gone below 60 but they stamped their feet
when the Administrator set it at 75 PPb. They actually had
scheduled a meeting before the formal rule was announced, and
at that meeting protested and said no, we told you 70 PPb, you
set it at 75 PPb. Their advice was a blend of science and their
personal policy preferences. There is no scientific methodology
that can tell you this is where the standard should be set.
That represents a policy choice exclusively delegated to the
Administrator of EPA.
Chairman Harris. Thank you. Thank you very much.
Dr. Thurston. I would like to respond to that.
Chairman Harris. Well, I have a question for you, Dr.
Thurston, because I only have about a minute left. Dr.
Thurston, two questions. One is, one of your graphs kind of
presumes that there are linear effects and some of the science
assumes that there is a linear effect of increasing and
decreasing exposure but we know from carcinogens, for instance,
specifically that is likely not true, that at some low levels,
you know, an effect seeing a reducing carcinogens at a higher
level may not have the same effect on mortality as reducing the
amounts at a lower level. So would you agree that the science
about linearity of effect of air pollutants is not clear?
Dr. Thurston. Well, I would say the evidence we have is
consistent, or certainly not inconsistent with a linear effect,
and I think that, you know----
Chairman Harris. Right. Okay. Thanks. I just need a very
simple answer because I have one more question.
Dr. Thurston. But you mentioned figure one does show the
association going down to seven micrograms per cubic meter,
well below the standard.
Chairman Harris. I know, and it has a bunch of lines. I
understand that.
Dr. Thurston. But in response to Dr. Smith's comment----
Chairman Harris. I have to ask this question because this
puzzles me.
Dr. Thurston. Go look at figure one.
Chairman Harris. In the past 20 years, and I have observed
this being a physician even in the operating room, the
incidence of asthma has clearly increased. The incidence and
prevalence of asthma in the population has clearly increased in
the past several decades. Would you agree with that?
Dr. Thurston. Yes. It is a multifactoral disease.
Chairman Harris. Has air pollution clearly decreased
significantly in the past several decades?
Dr. Thurston. Yes, but it is not the only cause of asthma.
Chairman Harris. These are very complicated interactions;
aren't you afraid that we simplify them too much when we make
statements as were made by the Assistant Air Quality
Administrator here that hundreds of thousands of asthma
episodes are going to be avoided if we change these standards
when in fact there are hundreds of thousands more episodes
since the last time we changed these standards?
Dr. Thurston. I think it raises a good point, that there is
a whole pyramid of effects that are not measured by the system.
In other words, I can't answer that question because we don't
have good records of those hospital--you know, the asthma
exacerbations, and that is not considered by the process, so we
are really underestimating the benefits of clean air and the
process the EPA is using today, and that is a good example.
Chairman Harris. Thank you. And Dr. Thurston, by the same
logic, I would say we may be overestimating the benefits by the
very same example of the poor data we have.
Anyway, I recognize Mr. Miller for five minutes.
Mr. Miller. Thank you, Mr. Chairman. This hearing is about
process, but the process of the Committee I think also bears
some scrutiny. It is very strange for the EPA's processes to be
the subject of this Committee's hearing and for the EPA not to
be present. I think most Americans understand fundamental
fairness is that if you are going to be criticized in public,
that you should be there to defend yourself, but the EPA has
not been invited today, and in past hearings when they have
been criticized, they have not been called to testify.
There was a book about a famous singer of the last
generation. I will not say his name although all of you know
who I am talking about, because I think he denied this and his
family denied this, but according to the book, the singer as a
young man would go to bars, have too much to drink, get in
fights, but then after he became famous and as a middle-aged
man he would continue to go to bars and get in fights but the
fights would be different. He would have his bodyguards hold
the person he was in the fight with so he could punch them.
This hearing does seem like one of those fights. The EPA is
being punched and cannot punch back. Fundamental fairness does
require that they be here and they are not.
Also, the Chairman used the term ``financial conflicts'' to
describe those involved in EPA decision-making, and I think
that is something we should be concerned about. I think members
of this Committee and the American people are entitled to know
what financial interest anyone has who is involved in any
process of government. That is why I criticize the Committee's
change in rules at the beginning of this Congress to require
much less in financial disclosure; the Truth in Testimony forms
require all that a witness disclose is that they are being paid
to testify at the hearing without any other disclosure of what
their financial interests are, and we have had again and again
witnesses whose entire livelihood comes from the industries
whose interests are very much the subject of Committee
hearings, and all we see on the financial disclosure forms are
that they are not being paid to attend, they are simply public-
spirited citizens.
Just to pick on one of you, Mr. Cichanowicz, your
information available on the Internet about you does say that
you--and actually in the introduction it says that you do
primarily work for utility industry clients. It does appear
from the Internet that your clients have included Edison
Electric, Midwest Ozone Group, American Coalition for Clean
Coal Electricity, American Public Power Association, the
National Rural Electric Cooperative Association. Mr.
Cichanowicz, what percentage of your income comes from the
utility industry?
Mr. Cichanowicz. Probably 75 percent.
Mr. Miller. Okay. Do you not think the American people and
the members of this Committee ought to know that in evaluating
your testimony?
Mr. Cichanowicz. It is evident, isn't it? I mean, in my
resume----
Mr. Miller. Well, we had to ask about it. It was not
available in the regular course of things. It is something that
I have had to use my time to ask about today.
Dr. Thurston, it seems like you could barely get a word out
in response to Dr. McClellan's statements and the questioning.
Do you want to elaborate upon what you said earlier?
Dr. Thurston. Well, you know, I did want to just point out
that first of all, I don't remember any foot stamping in the
CASAC process. But the fact is that the reason why there was a
problem, this was the first time that the Administrator in many
years of CASAC, that I know of, that the Administrator did not
follow the advice of CASAC, and that is really what this whole
problem was about, and that is why, of course, Administrator
Jackson then went ahead and she said well, okay, the last
Administrator wouldn't follow CASAC, but I will go back to the
tradition of following CASAC, and that is really in a nutshell
what happened there.
Mr. Miller. Okay. I will yield back the little bit of time
I have left.
Chairman Harris. Thank you.
I recognize the gentleman from California, Mr. Rohrabacher,
for five minutes.
Mr. Rohrabacher. Thank you very much, Mr. Chairman.
Let me just note that the personal attack on our witness
was obnoxious and not reflective of higher standards that I
have had here and you have also had, to my colleague from North
Carolina. This type of personal attack is not acceptable. The
fact is, every witness we ever have has some contact with the
people that they are expert with.
Mr. Miller. Will the gentleman yield?
Mr. Rohrabacher. No, not yet. The fact is, is that people
who are expert tend to have worked in the industry in which
they are expert, and the fact is that when you confront
someone's arguments, you confront someone's arguments with
better arguments or you challenge what they are saying by
suggesting there is another explanation. That is totally within
the area of acceptability with these kind of hearings. Just
simply trying to dismiss someone on a personal level because of
the way he has earned his living is just absolutely
unacceptable to me. Anyway, let me go on with my questions now.
Mr. Miller. Will the gentleman now yield?
Mr. Rohrabacher. Yes, I will. Go right ahead.
Mr. Miller. Okay. Mr. Rohrabacher, do you not think the
American people are entitled to know if a witness has financial
interest, if they are employed by an industry whose interests
are at stake in the hearing? I did not attack him personally
but I said that----
Mr. Rohrabacher. Let me just----
Mr. Miller. And those were the questions I asked at our
organizational meeting, why does our form not require the same
law as it has in the past.
Mr. Rohrabacher. Reclaiming my time. The fact is, is that
this witness's background, his financial background is open in
his resume. He stated that very clearly, and he is not hiding
anything, and your line of questioning, which I might add,
questioning the process rather than the arguments, which called
into question our Chairman's integrity was unacceptable as
well, and this is just the type of thing that we see time and
time again. It is the global warming approach as well, dismiss
your opponents, do not confront their arguments, challenge them
on their integrity rather than challenging the positions that
they have taken. This is not acceptable, and that is one of the
reasons we are having this hearing today is because the EPA may
be suffering from that same type of mindset which is not
consistent with good science and is not consistent with finding
the truth.
Dr. Thurston, you had some things you wanted to say, you
wanted to confront, I would be happy to grant you time--you
didn't feel you had the time to confront some of the things,
the points the Chairman was making. Go right ahead.
Dr. Thurston. Well, thank you. Well, I did get a few of
those comments in anyway, but I was trying to refer to Dr.
Smith's testimony saying that, well, below the ambient--the
NAAQS standards, that there is probably no proof that there
were any benefits. But figure one of my testimony is from a
study that I was principal investigator of, and there is the
indications that the benefits do keep going down well below the
standard to levels about seven micrograms per meter cubed. And
the only reason probably we can't show below that is we have no
place in the country that is cleaner than that, that has a
metropolitan area with enough people to study.
Mr. Rohrabacher. Thank you very much.
Dr. McClellan, do you have something to say about that?
Dr. McClellan. Well, again, that is a failure to place the
science in context. If we examine the Thurston graph very
closely, it shows confidence intervals around each one of those
curves, and I think you would be very hard pressed to suggest
that there is even statistical significance in terms of those
effects as we go down at that lower portion of the curve. What
you are arguing, Dr. Thurston, is that the answer to how low is
low enough is zero. I am saying that you have to have context
in terms of making those decisions. I look forward to having an
Administrator of the EPA who will make those policy judgments
and not rely exclusively upon scientists who draw a bright
line.
Mr. Rohrabacher. And Dr. Phalen, did you have something to
comment on that as well?
Dr. Phalen. No, sir.
Mr. Rohrabacher. Okay. Thank you very much.
I would just like to note, Dr. Phalen does come from the
University of California at Irvine and we are very proud of the
work that you do there and very proud of the work that the
university is engaged in.
Dr. Phalen. We are very proud of you as well.
Mr. Rohrabacher. Thank you. God bless.
Chairman Harris. Thank you very much, Mr. Rohrabacher.
I now recognize the gentleman from California, Mr.
McNerney.
Mr. McNerney. Thank you, Mr. Chairman, and I thank the
witnesses for coming in today.
Dr. Honeycutt, I was pretty intrigued by your quote that
correlation is not causation. There is some truth to that.
My question is going to be directed to Dr. Thurston. Dr.
Thurston, how do scientists, in your opinion, determine causal
relationships between pollution and adverse health effects?
What is the right way to go about doing that?
Dr. Thurston. Well, as you pointed out, correlation does
not necessarily mean cause and effect, but when you start to
look at a broad-based knowledge base, you know, as I pointed
out in my testimony, and basically we use the principles of
Austin Hill that he developed to look at cigarette smoke. Back
in the 1960s there were many of those who said well, really you
can't prove that cigarettes cause lung cancer, and so we went
through many decades before people really accepted the fact
that cigarette smoke and smoking cigarettes causes lung cancer.
Believe it or not, people denied that for decades, and based on
these same arguments that sometimes you hear about air
pollution. So then he developed a whole series of things where
you look for coherence between different endpoints, you look
for a time sequence, you look for natural experiments like Dr.
Pope's experiment, well, study, where he followed the fact that
when there was a strike and the pollution levels went down, the
hospital admissions for children also went down, and then when
the pollution--when the plant started up again, the pollution
went up, the hospital admissions went back up again the
following year.
Mr. McNerney. So, I mean, there is no yellow brick road for
science. You have to look at it----
Dr. Thurston. Well, you have to look at all the evidence,
and one study is not enough, and you need to look at many
studies and many different types of studies, and we have done
that, and the evidence is clear that air pollution, increasing
pollution causes increased adverse health effects. And
decreasing pollution causes decreased health impacts and better
public health.
Mr. McNerney. I agree, but that is not what I hear from all
the witnesses here.
Dr. Honeycutt, do you think that the Clean Air Act is worth
enforcing, and if so, who should enforce it?
Dr. Honeycutt. Excellent question, sir. It goes back to
what Dr. McClellan said. It is how low is low enough. What Dr.
Thurston is saying is absolutely correct. But those levels that
they were talking about are extremely high levels. There is no
doubt that, you know, getting air pollution out of the
milligram-per-cubic-meter range into the microgram-per-cubic-
meter range has been a great benefit. But now we have reached
the point of diminishing returns. You know, lowering the
standard a microgram per cubic meter or two or a part per
billion or two, that is in the statistical noise of health
effects. You are not going to be able to see to measure those
health effects, but they are going to cost billions of dollars
to get. So the question now flips back to you. I mean, do you
think that is worth the investment? It is not up to me as a
scientist to make that call.
Mr. McNerney. Well, I don't have enough scientific
knowledge to understand how much one or two micromeasurements
is going to make a difference. Is that within the realm, Dr.
Thurston, of causality? Can we make determinations on those
small differences?
Dr. Thurston. Well, sure. You know, I think that we do
studies where we look at different places and look at the same
place over time and are able to discern those differences. You
need enough power, which is one of the quandaries we have is
that the places where the people live tend to be the more
polluted places, and the less polluted places don't have as
many people, so then it becomes more difficult to study at the
lower levels.
But I did want to pick up on something that Dr. Phalen
pointed out, which is that the components of the pollution are
important, and in fact, the most recent studies are showing
that if you look at the components, like from traffic and from
coal plants and from oil plants, and start looking at the
various pieces and adding them up, you actually see larger
effects than just looking at particulate matter in general
where you just weigh it, and our studies have started to show
this, and some of them we have published doing work for the
Health Effects Institute and right now----
Mr. McNerney. So there is room for more science here?
Dr. Thurston. There is, but what I am saying is, we are
probably underestimating the benefits of clean air, and perhaps
we should be looking at these--and I am agreeing with Bob here
that we should be starting to look at more of the constituents,
the way we did with lead. We looked at lead in particulate
matter and controlled that. Maybe we have to start looking at
the components because right now we are underestimating, I
think, the benefits of clean air.
Mr. McNerney. Thank you. I didn't expect you to end that
quickly.
I just spoke to a friend of mine from Iowa who I grew up
with who has COPD and he said that just goes with being a
farmer in Iowa, the dust that you breathe in, and so Dr. Smith
mentioned, well, what is in the particles makes a difference,
and I agree with that. I mean, if the dust particle has
plutonium in it, that is a lot different than just having dust,
but just the size of the particle by itself makes a difference
and has to be taken into consideration if we want people to
remain healthy and don't want to have to pay continually
increasing health costs for our seniors.
I yield back.
Dr. Thurston. But I don't think that COPD should be a part
of doing a job. I remember in the past people used to say well,
when they smelled pollution, that was the smell of money, and
you know, in reality, we have been able to run our businesses
and clean up the air and, you know, our GNP has been rising at
the same time the pollution levels were going down, so they are
not really as closely linked as you might have been led to
believe today. There are health benefits and also economic
benefits of cleaning the air. I am guessing here, and one
should never guess, but I am guessing that some of the
equipment that was put on those plants were made right here in
America, and we can have a lot of jobs and make a lot of money
selling that pollution control equipment, saving lives around
the world and making money, creating jobs cleaning the air. So
I think that, you know, you have to look at the whole picture.
Mr. McNerney. Thank you.
Chairman Harris. Thank you very much. I couldn't agree
more, you have to look at the whole picture.
I recognize the distinguished gentleman from Florida, the
Chairman, Mr. Hall--I am sorry, from Texas.
Chairman Hall. I will move anytime.
Mr. Harris, I won't ask any questions. I just want to make
a comment here that I probably shouldn't make. You offered me
the chance to ask three EPA witnesses a couple of weeks ago or
so, and I said I wouldn't believe any of them under oath, and I
was with one of them later at a party and I wished I hadn't
said that. So I might say some things that I want to take back,
but when Gina McCarthy came before our Committee and remarked
about people that were out of work, she said ``We are not in
the business of creating jobs.'' I think that is an insult to
every person and every breadwinner that has been a breadwinner
and had to look into the face of his children and tell them why
he couldn't and his wife and tell her why he couldn't send her
children back to school next year because he didn't have a job.
He had no job, very little hope and a whole lot of heartbreak.
I think that is the kind of testimony we are getting from them.
She didn't use any science when she gave us that really smart
aleck remark that hurt a lot of people.
And Mr. Miller, the great lawyer who came here, I think,
because he is a great attorney, but I understand he had the
right to ask for a witness. I don't know if he did or not but
if he didn't, he could always appeal to me, and of course, I
would have told him to go back to you. I think he had an
opportunity to have a witness if he wanted, and he threw the
skunk into the jury box and then left. I didn't get a chance to
tell him that in person.
Chairman Harris. Would the Chairman yield?
Chairman Hall. I have high regard for Mr. Miller, a minimum
high regard for him.
Chairman Harris. Thank you. Would the Chairman yield?
Chairman Hall. Yes, I yield the rest of my time to you.
Chairman Harris. Thank you very much, Mr. Chairman.
Yes, I would like to respond to that. As you know, Mr.
Chairman, the minority always has the opportunity to invite any
witness they would like. They clearly could have invited
someone from EPA. We extended an invitation to the current
Chair of CASAC and they weren't able to attend. I mean, the
bottom line is, we tried. You know, we can't drag them in, as
you know, Mr. Chairman, but we tried.
In response to the consults don't disclose enough
information, this was an astounding line of attack on this
panel of witnesses. Every expert in the field has financial
paths back somewhere, everyone. Look, I was an expert on
obstetric anesthesiology and fetal physiology. I got money from
the NIH. Clearly, I had an interest in that research being done
because I got funding, supported my salary. I suspect Dr.
Thurston gets funding for it. I suspect that you could easily
presume all that from reading anyone's biography that is
submitted to this panel. I just wanted to point for the record
those biographies are part of the record. They are made part of
the record. They were supplied to the minority. Nothing was
hidden at all, just so we clear the air on that, if you pardon
the pun.
Now, Dr. Thurston, I have got to ask you a question here,
because, you know, it is always comfortable to assume well, you
know, if you got a little bit of something that is bad, then if
you have more, it must be worse, and if you have less, it must
be better. Let us take mercury because I think Dr. Honeycutt--I
mean, the science behind mercury is fascinating because I had
read this but had forgot about it years ago, that, you know,
the EPA's blood level for safety is 10 times less than what the
Faroe study showed is safe based on epidemiology, and in fact
there is evidence that as you increase exposure to mercury
because you also increase exposure to the other fatty acids
that are obtained in those same fish that in fact you could
have an improvement in neurologic development. Now, that flies
in the face of examining mercury by itself and saying well, you
know, 58 is good, 5.8 must be better. Well, that is not true.
That is just not what the scientific--so although it is
attractive to say that--I mean, would you agree that that is
one of the stories of the mercury standards?
Dr. Thurston. Well, I can't profess to be an expert on the
toxicity of mercury.
Chairman Harris. Do you disagree with Dr. Honeycutt's
testimony? I mean, you are--I don't know. I guess--so you are
not an expert in other air pollutants besides particulate
matter?
Dr. Thurston. Well, no, but mercury is a toxic metal, and I
have not spent my time studying it, but it is true, you know,
when you look at metals, each one is unique and you really have
to look at that one. Like fluoride is----
Chairman Harris. Sure. Let me just--and I hate to interrupt
you but----
Dr. Thurston [continuing]. Is a nutrient at low levels, and
toxic at high levels.
Chairman Harris. So just like when we look at particulate
matter, PM2.5, I mean, there are numerous different things that
can be a particulate matter of that size, some of which may be
less toxic, some of which may be much more toxic, but the EPA
doesn't distinguish, does it? It says this is the level for any
particle of that size.
Dr. Thurston. It has been a process. They started out
originally collecting all particles in the air, total suspended
particulate matter, and then we moved to PM2.5, which are
particles small enough to pass the trachea. In other words,
they said with the particles that you can't even breathe, we
are not going to monitor those, and then now we have gone to
PM2.5, which are very fine particles that get deepest in the
lung, and we monitor that most closely, although we are still
looking at the other particles that can get into other parts of
the lung, the PM2.5 minus PM2.5 or the coarse particles, that
is something that is being investigated and indirectly can be
controlled through PM2.5 standards. But that is under
consideration by CASAC and EPA what the coarse particle
standards should be.
But then the next step is to, as you say, to go to what the
composition is, and what I am telling you is, when we have
started doing that, what we are finding is, the sum of the
parts is greater than that whole. So it is not an exact measure
of the toxic pollutant but by being so we have uncertainty and
the estimates, therefore it is reducing.
But I did want to comment----
Chairman Harris. Dr. Thurston, we are way, way over.
Dr. Thurston. I know, but you brought my name up----
Chairman Harris. If we have a second round, you will get
it, okay?
Dr. Thurston [continuing]. Vested interest. I have never
taken money from vested interests. That is an important
distinction here.
Chairman Harris. You don't consider government funding a
vested interest? I think the American public would disagree
with you.
Chairman Hall. Mr. Chairman, my time is expired.
Chairman Harris. Thank you for yielding back the balance of
your time, Mr. Chairman.
I recognize the gentlewoman from California, Ms. Woolsey,
for five minutes, and you can take extra.
Ms. Woolsey. Thank you, thank you.
Mr. Cichanowicz, I was going to call you witness number six
until I realized I could say your name. I was taken by your
graph chart picture of how difficult you thought it was for the
utilities to put in control devices. Do you have any kind of
picture or graph or comparison that tracks the results of not
taking care of our clean air, of what happens when our air is
poor, when our kids are getting asthma, when our workers are
unproductive because they have got respiratory disease, because
of the cumulative impact of not just your industry but all
industries and cars and people and airplanes, the effect this
is having and the cost to our health care system? How does that
compare with putting some control devices into some of the
utilities? Do you have anything like that?
Mr. Cichanowicz. Those issues are out of my skill set.
Ms. Woolsey. Well, see, we ought to have that because the
check and the balance of this whole thing is not just about an
industry, it is about the people and the costs of keeping
people healthy in the United States of America, and clean air
is a huge, huge part of that. So I just wanted to say that.
So Dr. Thurston, okay, CASAC got a lot of criticism. As I
was coming in I heard a lot of CASAC this, CASAC that. You were
part of CASAC?
Dr. Thurston. I have served on a panel, yes.
Ms. Woolsey. You served on the panel. You know, there were
public forums. What did the public say when they came before
the commission? Is it a commission?
Dr. Thurston. Well, it's a committee. You know, there is a
public comment period but that is usually people like the
American Lung Association come forward, and they are obviously
very supportive of cleaner air because they are trying to
prevent lung disease.
Ms. Woolsey. Well, then aren't they trying to destroy
business around the country by saying things like that?
Dr. Thurston. I am unaware of that agenda. No, they want to
clean air and bring better health to people, and decrease our
health care costs. In other words, if you clean the air, fewer
people will go to see their doctors, fewer people will have to
end up going to the emergency room, fewer people will have to
check into a hospital, and you actually lower your health care
costs when those events are fewer. And if we don't control the
air, then obviously we are increasing the health care costs
relative to what they would be.
Ms. Woolsey. And isn't the cost of health care also a
burden on business?
Dr. Thurston. Yes.
Ms. Woolsey. So I mean, doesn't that offset the cost of----
Dr. Thurston. And the American people, yes, and that is
something that I think is worthy of spending time trying to
reduce.
Ms. Woolsey. That is clear.
Okay, now, all of you, many utilities have already
installed control devices. If EPA hadn't had those regulations
or if they pulled back on the regulations, what kind of
incentives will there be for them to make these changes? I
mean, why would they do it? Are they going to do it? Mr.
Cichanowicz, is the utility industry going to live up to good-
faith effort in this regard?
Mr. Cichanowicz. The issue I came to discuss was the time
it takes to deploy technology. I don't think the topic of
whether you do it or not is separate. All I am saying is the
time that it takes and the schedule. That is what is key in my
testimony and that is what is key in trying to put these
systems in.
Ms. Woolsey. Okay. How about you, Dr. Smith? Do you have
any idea what would happen if we didn't have these regulations?
Dr. Smith. There are air quality standards that are in
place and those air quality standards will cause power plants
to put these controls in if there are in areas where there is
nonattainment. My point was that the regulations need to be
considered on the merits of their own benefits and their own
costs, and there are a significant number of regulations that
are going into place that don't have benefits anything close to
their costs.
Ms. Woolsey. Well, costs within the industry, not costs for
the heath care costs in the country.
Dr. Smith. Costs compared to their benefits. The benefits
include the health care costs, and those are part of the whole
cost-benefit framework.
Ms. Woolsey. Dr. Phalen?
Dr. Phalen. Yes, ma'am. Thank you. That is an excellent
question. My experience in California has been that utilities
such as Southern California Edison have supported the research,
and I have seen some of the strongest advocates for clean air
as employees of industry.
Ms. Woolsey. Except for Mr. Rohrabacher, aren't we
Californians considered the kooks of the world, that we do
things because they are the right thing to do and we won't have
any business in our state even though it is the eighth economy
in the world and all that because we do have these regulations?
Dr. Phalen. EPA's national air standards are called NAAQS,
and for a while California's were called CAAQS.
Ms. Woolsey. But ours work.
Dr. Phalen. But I am firmly against there being a war
between industry and the public or industry and regulators. I
would like to see a comprehensive evaluation of public health
because we know industry is important to public health and we
know clean air, as you aptly pointed out, is important to
public health. So I see the need to cooperate rather than
battle, because whoever wins----
Ms. Woolsey. I agree with you totally, and Dr. Honeycutt,
if they let me, I will come back to you, but I am going to ask
Dr. Thurston a question.
Chairman Harris. Yes, we will have one abbreviated second
round. Is that okay?
Ms. Woolsey. I know, but I am not supposed to be here after
this.
Chairman Harris. I will give you 30 more seconds.
Ms. Woolsey. Thirty-one seconds?
Chairman Harris. Thirty more.
Ms. Woolsey. Thirty more seconds. So Dr. Thurston, is not--
don't you come to the Science Committee thinking that science
is what we are supposed to rely on for our information, and----
Dr. Thurston. Yes.
Ms. Woolsey [continuing]. Doesn't it seem kind of
counterproductive that we might think we should have a study on
what happens to health care overall? Isn't there a worry that
then the scientist will come to us and say oh, my gosh, what do
we need to do with our air and then this committee will say oh,
well, but we don't want to believe that science.
Dr. Thurston. Well, it is about sound science, and that is
the role of CASAC, to make sure that sound science is leading
the EPA in their decision making, but you have to remember that
after this the OMB, Office of Management and Budget, does a
thorough analysis, economic analysis, that has been required
since the Clinton Administration to show that the economic
benefits outweigh the economic costs, and they would not have
any of these regulations--don't worry about the fact of the
benefits, because they are going to far outweigh the costs or
the regulations are not going to go forward. That is the way it
works. So, you know, we have also got to consider that there is
another process that follows the Administrator's decision,
which is that by OMB that does these economic analyses and does
consider the costs, and the benefits have always outweighed of
clean air. There is something like--you know, it ranges
depending on the decision but a good rule of thumb is maybe
what Benjamin Franklin said: An ounce of prevention is worth a
pound of cure. And oftentimes we see about a 16-to-one
valuation of the benefits to the costs.
Chairman Harris. Thank you very much, Ms. Woolsey.
We have enough time since we are here until noon to have a
second round of questioning for five minutes, and I will
recognize myself for the first second round here.
Let me just ask Mr. Cichanowicz, because you are an expert
in these systems and installing. What percent of these
pollution reduction systems have been installed in the past 30
years as opposed to before then?
Mr. Cichanowicz. Well, the bulk of them have been
installed----
Chairman Harris. Close to 100 percent?
Mr. Cichanowicz. Yes.
Chairman Harris. Okay. When was the last time we had an
unemployment rate of nine percent. Was it 30 years ago? So we
have never had to install pollution control of this magnitude
and cost during a period of recession like we have right now.
Would that be correct? I mean, on the order of magnitude of
what we are talking about.
Mr. Cichanowicz. I believe so.
Chairman Harris. Those are just facts, I mean, and we can
discuss facts. We have not had an unemployment rate of 9.1
percent since the early 1980s, and almost 100 percent of these
pollution controls that we are talking about have had to be
installed in periods when we have not had unemployment. So the
point to making this decision in a vacuum of jobs. Now, if we
have a 16-to-one benefit, I know how to solve our debt problem:
spend $1 trillion on air pollution reduction and we solve our
national debt because we have a $16 trillion benefit. It is
ridiculous on the surface to believe there is a standard of 16
to one, every dollar you spend on this you get $16 worth of
benefit. It is patently ridiculous on the surface. That is my
editorial comment.
Dr. Smith, let us get to mercury because we just had a
hearing about these mercury standards, and my understanding
based on today's testimony how this decision was made in your
testimony almost none of the benefit claimed from that mercury
standard--the heart attacks, the asthma, you know, you hear the
whole litany, same litany over and over for every single one we
have a hearing on this. None of them really come from mercury,
do they? They all come from the co-benefit or the double
counting of the particulate matter. Is that right?
Dr. Smith. Six million dollars of the benefits out of the
$150 billion comes from mercury----
Chairman Harris. That is close enough to nothing. Okay.
Dr. Smith [continuing]. Change of 511 IQ points in total
aggregate across some 30,000--anyway, a large number of
children.
Chairman Harris. And Dr. Smith, if I might just interrupt,
that would presume that the Faroe study is inaccurate and that
in fact by lowering mercury exposure you absolutely get a
reduction.
Dr. Smith. It is presuming that as you reduce the tiny bits
of reduction in mercury and get tiny IQ benefit increases.
Chairman Harris. Could you just briefly, because I have
about two minutes left, just go by this double counting deal
again? Because it always amazed me, the testimony was always
the same. The EPA would come in and say yes, if we do this
rule, we get 34,000 less deaths, 150,000 less asthma attacks,
15,000 less heart attacks, and we kept on hearing the same
thing over and over whether it was ozone or mercury or
whatever. I am thinking this is pretty amazing because when I
went to school, I didn't think mercury caused heart attacks. So
could you go through this double counting for the Committee
once again?
Dr. Smith. If I can just take the Utility MACT that has
17,000 deaths. As I was saying, 4,000 of those are due to
exposures to PM that are between 11 micrograms per cubic meter
and 15 micrograms per cubic meter. All 4,000 of those are due
to attainment of the current safe standard for PM, but if EPA
tightens the standard down as low as 11, which it is
considering doing, those are benefits from the PM standard if
those occur. All the rest, if EPA does not tighten the standard
below 11, it is saying that all the rest of the benefits, the
other 13,000 of them that is in that co-benefits case, aren't
really credible enough in order to protect the public health
from them. Otherwise EPA should just set the PM standard lower,
and EPA is not prepared to do that. It said it is not prepared
to go below 11, and that indicates the degree to which those
extra 13,000 are just not credible.
Chairman Harris. And you also, I think, in your testimony
mentioned not only that but that it counts them in one case but
not the other. It uses the old standard for the PM when it
considers PM deaths but the new standard when it looks at all
the other ones. Was that in your testimony also, in your
written testimony?
Dr. Smith. Well, what is happening here is, they are
putting the cart before the horse. They count the benefits by
putting them into co-benefits for a MACT that has no benefits
of its own. And then later it doesn't really have them perhaps
to count again for the PM rule but it doesn't need them because
they have still got that huge reservoir to keep counting up
against in order to justify the PM rule, so basically taking
benefits from the rich, which are the PM benefits, and
spreading it all around to the other regulations that don't
have any benefits of their own to create a complex web of
regulation, when in fact the right way to do this and the cost-
effective way to do this for our society is to go after the
problem directly and decide where to set the PM standard.
Chairman Harris. Thank you very much, Dr. Smith.
The gentleman from California, Mr. McNerney, is recognized.
Dr. Thurston. I would like to respond to that, if I could.
Chairman Harris. If Mr. McNerney agrees, you will be more
than welcome if he yields time to you.
Mr. McNerney. Thank you, Mr. Chairman.
I just want to say a word or two about the controversy that
was brought up with Mr. Miller's questioning. It has become
standard for vested interests to bring pseudoscientists forward
that are paid for by vested interests to perpetuate the status
quo that started as Dr. Thurston brought out on smoking. It is
being brought out on global warming. We are seeing it brought
out now. So I think it is perfectly appropriate to try to
understand what the witnesses' interests are and how it is
being paid for. So I think he was perfectly justified in those
questions, and I didn't take those as an attack on any
particular witness.
The thing I want to ask about is to follow up with Lynn
Woolsey's question about the benefits or the pollution
equipment that has already been installed in power plants and
the benefits of that installation, and I think Dr. Thurston
sort of hit it on the head. Developing technology to clean up
our air quality is an important economic driver. For example,
how many people were employed when that equipment was put in
the power plants there that you are discussing? We employed
probably thousands of people or more putting in that equipment,
and the United States developed that equipment, and now we are
able to sell that to other countries. So no, I don't buy the
argument that this is entirely detrimental to our economy to
have clean air. I think the opposite is absolutely true. So one
of the things I would like to point out is the cap and trade
that was employed to reduce sulfur dioxide in New England and
the northeastern states, the cost was a fraction of what the
industry was predicting it would be. The benefits were
enormous. We developed technology that put us ahead of the
curve. So I would like Dr. Thurston to respond to that.
Dr. Thurston. I certainly agree. I wanted to bring up the
point, though, that, you know, there seems to be--you know, in
Dr. Smith's testimony, she started saying well, we can't--
shouldn't take credit for the things that happen at the same
time. In other words, when you go to clean up, let us say,
mercury, you put on control equipment which also captures
particulate matter. Well, certainly, they are going to be
controlled and they would not otherwise be controlled. There
are many power plants in the United States. Because of the way
the Clean Air Act was written by Congress, the assumption was
that older plants would go out of service and they were
grandfathered in, and there are many plants, coal-fired power
plants, that are spewing pollution virtually uncontrolled
relative to the technology we have today, and I think what is
happening here is that this rule, by putting on proper
particulate matter controls and vapor controls, is closing the
loophole on these really gross emitting--they are the low-
hanging fruit of air pollution in the United States, these
power plants. In other words, this is the biggest bang for our
buck that we can get. If we are going to control air pollution,
I do think cleaning up these coal-fired power plants that have
been operating under this grandfather clause is a really
efficient way to clean up our air and get the health benefits
that would accrue, and they certainly can because they are not
going to happen anyway.
Mr. McNerney. Are we going to create jobs implementing this
low-hanging fruit that Dr. Thurston is referring to?
Mr. Cichanowicz. It certainly takes people to build and
design the power plants, yes.
Mr. McNerney. Thank you.
I think I will yield back.
Chairman Harris. Thank you very much.
I now recognize the gentleman from California, Mr.
Rohrabacher.
Mr. Rohrabacher. Thank you very much.
Just to get on with our little discussion as to what
standards we should have here in dealing with one another on a
civil and honest basis, I think it is really more important to
try to challenge witnesses based on the arguments that they are
presenting, rather, the facts that they have and their
suppositions they are making rather than trying to discredit
them and dismiss them and perhaps smear them as people who lack
integrity in terms of what they are saying. I found that to be
more important and the best way to go. I have very strong
opinions, and I certainly don't challenge someone's integrity,
and I would not think of asking Dr. Thurston, well, how many
grants have you gotten from the EPA to do your studies, blah,
blah, blah. I wouldn't think about that. I think about what is
he saying, what is he arguing, and this idea that we are trying
to discredit witnesses rather than confront what they say and
challenge them is out of line.
Dr. Thurston. May I respond to that?
Mr. Rohrabacher. No, you can't. You don't run this hearing,
and I have a very limited amount of time, but I will at the end
give you a chance to comment on my comments.
Let me just note, I disagree totally with the idea that the
American people were smoking because somebody told them a lie
and would not testify in Congress that smoking causes cancer.
My brother died of cancer. I was after him for years to quit
smoking. He knew exactly what he was doing. He was addicted to
nicotine, all right? And he couldn't break the habit along with
other people. We all know that. It wasn't that they don't know
and didn't know, and so I deny that supposition.
Dr. Thurston, would you like to comment on that? No? Okay.
Dr. Thurston. Well, I mean, I will just say that there was
a controversy at the time for decades where the cigarette--
maybe you knew and your brother knew, but the cigarette
companies certainly did not admit that their product caused
those effects.
Mr. Rohrabacher. Let me just note, I think the idea that
the American people didn't know that cigarette smoking was
harmful to their health is not something that I believe is the
case.
Now, we are talking about the benefits of cleaner air, and
let us just note that obviously breathing cleaner air is an
important--is something good for your health if that cleaner
air is above--is already above the threshold of being unhealthy
or being healthy if you are going to--in other words, what is
the air at right now? If you breathe more air in and it is
above the threshold that will make you unhealthy, then it will
not have that impact of being healthy. I think that is the
point that we have heard made here.
And one last thing. Mr. Chairman, I ask the kids from my
district every time they come in to see me, and I see every
child that I can from my district, and I always ask them when I
come in the classroom, how many of you think that the air in
southern California was better when I went to high school there
45 years ago or is it better today, and 90 percent of these
kids, maybe 95 percent, say the air pollution in southern
California is so much worse today than it was when you were in
school, how lucky you were to live at a time when you are not
being poisoned by these terrible people in industry, and in
fact, it is just the opposite. It is 180 degrees the opposite.
The air in southern California is so much better now than when
I was their age and yet they are being lied to and they are
being scared to death and they are being--and the people who
are trying to put it in perspective are having their arguments
dismissed and having their credibility challenged rather than
their arguments being challenged. So I would hope that we have
a little more honest discussion, we don't try to frighten
people into behavior that will destroy our economy and put
people out of work and cause a lot more anxiety.
One last point: I have 10 seconds. I do believe in cleaner
air, obviously. We should all believe in cleaner air. I would
hope that we develop the small modular nuclear reactors that
the nuclear industry, which has been opposed by these very same
environmentalists over the years, I would hope that we can get
together and support nuclear energy as a solution which we all
can agree upon. Thank you very much.
Chairman Harris. Thank you very much, Mr. Rohrabacher.
The gentlewoman from California is back, so Ms. Woolsey,
you are recognized for five minutes.
Ms. Woolsey. Thank you, Mr. Chairman.
Dr. Thurston, as you have heard, many members of the
majority party have stated that EPA develops regulations based
upon faulty scientific evidence, and could you explain to us
how the science that underpins EPA regulations is peer reviewed
and the importance of peer-reviewed science?
Dr. Thurston. Well, certainly. I did want to respond, if I
could----
Ms. Woolsey. You may. Go ahead. Use my time.
Dr. Thurston. The Congressman wouldn't ask me about whether
I am funded by EPA for research, when in fact, that is what the
form asks. It asks, have you had federal support. It does ask
that but it does not ask, do you have funding from any vested
interest, and I do think that it is a valid thing to get a
balanced picture of the full information. In other words,
consider the source, and to have not only where people get
their funding from the government but where they get their
funding from other places as well so, you know, have equal
opportunity of information gathering. And right now, the form
did ask where have you gotten your funding from the government,
but not where I have gotten it from, from industry or vested
interests.
Anyway, but in terms of the question of the peer-reviewed
science, yes, you know, it is very important, the process that
EPA goes through where they have their staff along with experts
that they hire, mostly from the academic community, and they go
through the literature and they evaluate the reports that are
out there, and using peer-reviewed things, things that are
published in the literature where the peers of these people
have reviewed it, not just a report that has been put out by an
interest group or by the government. It has to be a peer-
reviewed document, and then that is looked at collectively and
the whole picture for various health outcomes and they evaluate
each health outcome in terms of whether, you know, it is
inconclusive or conclusive or suggestive that there are
effects, so they go through and they rank out the effects and
see what they think is really well supported and conclusive,
and then based on those things----
Ms. Woolsey. Do things change because of their review? I
mean----
Dr. Thurston. Well, that is why they do the review every
five years. Science is constantly changing, and one of the
things that we do find is that as we clean the air, more and
more people are living in cleaner air, as the Congressman
pointed out, although there are parts of California that have
gotten worse--the central valley of California. You know, L.A.
has improved, but the central valley of California has gotten
worse in the recent years as it has become developed. It used
to be an agricultural area. Now it is well developed. And so
the pollution levels are actually rising in California in some
places, but the question--well, I have lost my train of thought
here. But, you know, the science does change over time, and
generally, as I was saying, the pollution levels as they go
down, we are able to have more people living in cleaner areas,
so we are able to study that. If you don't have any people
living in clean areas, you can't study it. I mean, part of the
reason why we don't see effects at very low levels, or unable
to show them definitively, is because we don't have enough
people living in clean air. If we have cleaner air, then we
will be able to test that hypothesis, and that is one of the
limiting things. But over time we are learning more and more,
and we are learning about more health outcomes. Neonatal and
pregnant mother exposures is an active area of interest that we
are finding more and more evidence that suggests that exposures
to mothers and young infants are very important to future
health, not only their health at the time but future health
throughout their life, so and it is the California children's
health study that is following children over time and learning
more and more about what childhood exposure affect. So it is a
learning process, and as time progresses, we learn new things,
and so that is why the standards get evaluated.
But I must point out that the standards don't always go
down. They have gone up based on this assessment. During the
Carter Administration, the ozone standard was actually
increased, and so CASAC does not always recommend, as Dr.
McClellan might sort of imply, that we always are lowering and
lowering. In fact, there is a history of CASAC when the
evidence did not support the present standard saying it should
be higher. So there is a record going both ways on this.
Ms. Woolsey. Okay. I have 16 seconds. Is there anything you
want to say that you didn't get to say in your testimony or in
your answers?
Dr. Thurston. Well, I did appreciate the point that was
made by the Congressman that as you breathe more air, you get
more effects, and we have found that exercising adults, if you
go out on a polluted day and exercise, that increases the
adverse health effects. And so the advice is always, well, stay
indoors on these high air pollution days and don't exercise.
But Americans should exercise more, so I think what we really
have to work toward is getting cleaner air so that we can
exercise every day, not just on the non-polluted days.
Ms. Woolsey. Thank you very much.
Chairman Harris. Thank you very much, and before we
adjourn, I want to thank the witnesses for their valuable
testimony and the Members for their questions. I know there
were issues about potential conflicts of interest. You know, it
is worthy to note that of the seven CASAC members now, five of
them receive funding from the EPA, and we have to decide
whether that in fact is a vested interest that may affect the
members of the CASAC.
Anyway, the members of the Subcommittee may have additional
questions for the witnesses, and we ask you to respond to those
in writing. The record will remain open for two weeks for
additional comments from the Members.
Again, I want to thank you all for a very informative
hearing. The witnesses are excused and the hearing is now
adjourned.
[Whereupon, at 11:51 a.m., the Subcommittee was adjourned.]
Appendix

----------

Answers to Post-Hearing Questions

Responses by Dr. Roger O. McClellan,
Advisor, Toxicology and Human Health Risk Analysis

Questions submitted by Representative Andy Harris, Chairman,
Subcommittee on Energy and Environment

Q1. During the hearing, Dr. Thurston stated that ``there is a history
of CASAC when the evidence did not support the present standard, saying
it should be higher.''

Q1a. In your view and experience with CASAC, is Dr. Thurston correct
that there is a history of recommending less stringent standards?

A1a. The quote attributed to Dr. Thurston is correct. Unfortunately,
he did not specifically identify the situation to which he was
referring. If he had done so his quote would likely have conveyed a
different picture. My experience with the setting of National Ambient
Air Quality Standards (NAAQS) goes to the early 1970s and pre-dates the
Clean Air Act Amendment of 1977 which called for the EPA Administrator
to seek the advice of a committee of scientists, a committee which is
now known as the Clean Air Scientific Advisory Committee (CASAC).
The situation that Dr. Thurston refers to is probably based on the
actions of an ad hoc Sub-Committee of EPA's Science Advisory Board
chaired by the late Dr. James Whittenberger that advised on the
revision of the Ozone NAAQS that occurred in 1979. The Whittenberger
Subcommittee pre-dated the organization of CASAC. The original NAAQS
for Ozone set in 1971 had a one-hour averaging time and was set at 0.08
ppm measured as photochemical oxidants. The Whittenberger Subcommittee
advised that the numerical level for the one-hour averaging time be
increased to 0.12 ppm measured as ozone. It was the opinion of the ad
hoc Committee that with the change in the measurement method, the
numerical level should be changed to provide equivalent health
protection. The standard with the new indicator, ozone instead of
photochemical oxidants, and set at 0.12 ppm ozone averaged over one
hour was issued on February 8, 1979. The point at which the Standard
was attained was also revised to ``When the expected number of days per
calendar year with maximum hourly average concentrations above 0.12 ppm
is equal to or less than one.'' A complete exposition on the matter by
Dr. Thurston would have conveyed a different impression than his
``sound bite'' statement. I think his statement is misleading.

Q1b. What motivates members of CASAC to recommend a lower or higher
standard?

A1b. In responding to this question it is important to provide some
background information on what has been referred to as CASAC. The Clean
Air Act specifies that a committee to provide advice on the setting of
the NAAQS shall consist of seven individuals. Further, it specifies
that one individual shall be a member of the National Academy of
Science. This has been broadly interpreted to mean an individual who is
a member of the National Academy of Sciences, National Academy of
Engineering or the Institute of Medicine. The Act also specifies that
one member shall be a physician and one member shall represent State
air pollution control agencies.
The CASAC members are appointed by the EPA Administrator for a two-
year term that may be renewed for a second two-year term. The EPA
Science Advisory Board website described an orderly and transparent
process for making appointments to advisory committees. In actual
practice the selection process is conducted behind closed doors by the
EPA Science Advisory Board staff. It is not known to what extent
personnel from the EPA's Office of Air Quality Planning and Standards
participate in the selection process or other EPA officials from
outside of the Science Advisory Board office, including the
Administrator's senior science and policy advisors.
In actual practice, the seven members of CASAC rarely meet,
deliberate or offer written advice solely as a seven-person committee.
Most often, the seven CASAC members are supplemented by an additional
six to 15 individuals who are appointed as consultants for a specific
review, i.e., the CASAC Ozone Review Panel served through the review
that concluded in March 2008 or the Particulate Matter Review Panel
that served through the review that concluded in October 2006. The EPA
and many others regularly refer to the CASAC Panels as though they were
the seven-person CASAC. It is necessary to carefully review minutes or
letters related to CASAC activities to discern whether the functions
were carried out by CASAC or a specific CASAC Panel consisting of both
official Committee members or CASAC augmented by consultants.
Let me now turn to the question of what motivates CASAC members and
consultants. First, I can only relate that my service as Chair of CASAC
(1988-1992) and as a member of numerous CASAC Panels was motivated by a
desire to serve the public by offering advice to the EPA Administrator
on scientific matters of air quality of which I was knowledgeable and
which I recognized were of enormous scientific and societal importance.
I always viewed the scientific advice I and my colleagues offered as
being of substantial value to the EPA Administrator in making difficult
policy decisions informed by the relevant science.
I suspect many CASAC members and consultants were motivated in the
same manner as I was motivated. In other cases, and especially over the
last several decades, it is my impression that some CASAC members and
consultants have also been strongly motivated by a desire to have
scientific publications they authored used in the setting of the NAAQS
and, in some cases, a desire to see that more stringent standards were
set. In my opinion, during some reviews a strong anti-industry bias has
been evident. In some cases, individuals use the ``cigarette industry''
and how knowledge of the health risks of cigarette smoking developed
over the decades as being prototypical of ``all industry.'' In some
cases, the view has been expressed that if standards are set lower,
industry will find a way to meet them. In my opinion, posturing these
complex issues as science versus industry is totally inappropriate. The
issue is really how can the science and the settings of NAAQS best
serve Society as a whole.
It is also important to recognize that a substantial ``clean air
research enterprise'' has developed, especially over the past two
decades. Many individuals are motivated to see the research aspects of
the enterprise continued with sustained and, perhaps, even increased
funding. Thus, some like to point to the science used in the setting of
the various NAAQS as justifying both past as well as additional future
research support. The research endeavors in many ways have shifted from
the conduct of ``issue-resolving'' research to ``issue-perpetuating''
research to be continued through a career of current investigators and
their trainees.
I have a personal concern that the focus on air pollution, at
current levels in the USA, may in a perverse way be negatively
impacting on our efforts as a society to have a positive overall impact
on public health. It can be argued that for a number of health
endpoints air quality accounts for less than 10% of the attributable
risk. Would we as a society achieve greater progress by focusing
attention on the 90% or more of attributable risk related to other risk
factors? I think so. My concern is that a narrow focus on a single risk
factor, air quality may be misguided. Concern over asthma is a great
example. There is no question that asthma rates in children, especially
in inner cities and minorities, have been increasing in recent years.
At the same time air quality has been dramatically improved. In my view
it does not make sense to keep trumpeting air pollution as a major
concern for influencing asthma and to continue to sponsor research on
the link between air pollution and asthma. I think our scarce national
resources, both dollars and scientific expertise, would yield a better
return by focusing on what causes the disease rather than on a single
risk factor--air quality.

Q2. There was some discussion about the independence and impartiality
of CASAC during the hearing. You previously served as the Chair of
CASAC, as well as on individual panels.

Q2a. What are the major strengths and weaknesses of the current CASAC
process?

A2a. The major strength of the NAAQS setting process, including the
role of CASAC, is that it should provide an orderly process for the
periodic review of the science under-girding the policy judgments that
must be made in the setting of the four elements of each NAAQS [(a) the
indicator, (b) averaging time, (c) numerical level and (d) statistical
form)]. I have several concerns with the NAAQS process as it has
evolved, especially during the past two decades. I am very concerned
that some individuals have overstated the role of science in the
process and understated the role of the EPA Administrator in making the
ultimate policy judgments required to set NAAQS. In my opinion, the
best contemporary science should inform all the policy judgments that
are inherent in the setting of each NAAQS.
However, scientists and others should appreciate that there is no
scientific methodology for specifying the precise level, averaging time
and statistical form of each NAAQS. When scientists state these either
as specific numbers or ranges they are offering advice that is a blend
of science and their own personal preference for a policy outcome. In
stating a range of numerical levels, the highest value in the range has
a dominant role in any further deliberations. In essence, the CASAC
Panel or CASAC is saying ``thou shall not set the NAAQS higher than the
upper value in the range.'' The lower end of the range is most likely a
statement of the policy preference of some individuals on the CASAC
Panel. By specifying an upper-bound numerical level for the NAAQS below
the existing NAAQS, the CASAC has clearly offered a policy judgment
that the standard must be lowered. This is a policy judgment
exclusively reserved by the Clean Air Act to the EPA Administrator.
The CASAC Panel review of the Ozone Reconsideration NAAQS serves to
illustrate the point I have made. The CASAC Ozone Panel Chair, Dr.
Jonathan Samet, in his March 30, 2011, letter to EPA Administrator Lisa
Jackson, stated that establishing a margin of safety was inherently a
blend of science and policy. It follows then that identifying a
numerical ceiling for the Ozone NAAQS of 70 ppb (eight-hour average) is
a blending of science and policy preferences.

Q2b. EPA often cites the public health benefits in its Regulatory
Impact Analyses to argue for more stringent standards. Does CASAC or
any other scientific body review these analyses?

A2b. The Regulatory Impact Analyses are typically released to the
Public after the Administrator's decision on each NAAQS is released.
This is done allegedly so that costs of achieving the NAAQS do not
influence the Administrator's decisions on each NAAQS in keeping with
the Supreme Court decision in the case of Whitman v. American Trucking
Associations. In actual practice, I suspect the Administrator is very
knowledgeable of the contents of the Regulatory Impact Analysis when
the final proposed NAAQS is sent to the Office of Information and
Regulatory Impact Analysis, Office of Management and Budget for review
before the final rule is released.
To the best of my knowledge, the Regulatory Impact Analyses are not
routinely reviewed by CASAC or any other scientific committee. I
suspect that EPA would argue that each Regulatory Impact Analyses is
conducted using the basic methodology recommended by various National
Research Council/National Academy of Science Committees and various EPA
Science Advisory Board Committees. I think the Regulatory Impact
Analyses would benefit from critical review by a Committee that
included scientists and engineers as well as economists. In my opinion,
a conceptual review of the approach used to develop regulatory impact
analyses is not adequate. There is a clear need for critical review of
multiple Regulatory Impact Analyses to understand the strengths and
weaknesses of how the Analyses are conducted and how the results are
presented.
In my view, the Agency and society at large would benefit from
having several teams, working at arm's length from the Agency, prepare
Regulatory Impact Analyses using what they view as best practices. It
would be of interest to compare the results of the analyses prepared by
the different teams. The current Regulatory Impact Analyses have had,
for several decades, primary input from a single EPA contractor. A key
consideration in comparing analyses prepared by different teams is how
they address issues of uncertainty. In my opinion, EPA's analyses
typically understate the uncertainties and systematically overestimate
the monetized health benefits of the NAAQS. This issue for the Ozone
NAAQS has been clearly illustrated by Dr. Anne Smith in her testimony.

Q2c. What recommendations do you have for improving the CASAC process
and ensuring that panels are independent, transparent, balanced, and
impartial?

A2c. I offer multiple recommendations for improving the CASAC process:

(a) There is a need to critically evaluate the process by
which CASAC members and consultants are appointed. The appointment
process should recognize that biases originate in many ways. The
current EPA view appears to be view biases as originating with
employment by or serving as a consultant to industry. I argue that
academic scientists who are supported by funding from the EPA, NIH and
other government agencies also bring their biases to the advisory
table; this needs to be recognized. I argue that the biases are
reflected in multiple ways (i.e., biases toward seeing specific papers
authored by the member or consultant and close colleagues cited and
used, biases toward noting the need for more research) and, most
importantly, the existence of a strong anti-industry bias.

(b) The committee process can be improved in multiple
ways. As a starting point, the CASAC Panels should maximize the use of
face-to-face meetings held in public view and minimize the use of
teleconferences.

(c) CASAC Panel meetings should be of sufficient length
(including multiple meetings if necessary) to allow adequate time to
discuss the report at hand and elaboration on the strengths and
weaknesses of the material prepared by the Agency and its consultants.

(d) All reports presented to CASAC Panels should have the
authors of each section identified, not merely a listing of numerous
contributors at the front of the Report.

(e) Sufficient time should be set aside during CASAC
Panel meetings for interested parties to provide oral comments, to
complement previously submitted written comments, on each document. The
current practice of allowing each commentor three to five minutes for
oral comments should be abandoned. The Committee Chair needs to clearly
acknowledge the value of public comments. All too often in the past
decade, public comments have been dismissed as an obligation for the
CASAC Panel to hear in the shortest possible time and for the EPA staff
to deal with.

(f) All deliberations of major issues should be carried
out in public open meetings. The practice that has developed in recent
decades of having key discussions and the drafting of key conclusions
carried out by two or three designated ``lead'' authors outside of
public view should be discontinued. Alternatively, key conclusions and
recommendations should be drawn from the ``provisional'' written
comments prepared by individual CASAC members and consultants and made
available for public consideration at least two weeks prior to any
scheduled meeting.

(g) Letters from the CASAC Panel Chair to the EPA
Administrator should, in general, be much shorter, summarizing key
conclusions and recommendations with the comments of individual Panel
members attached as an Appendix. This approach would require each Panel
member to clearly articulate their views on key issues and
recommendations in their written comments. This approach will allow for
a range of opinions on the science to emerge and avoid the common
practice of hiding behind the ``consensus'' opinion of the Panel. In my
opinion, consensus defined as (a) general agreement, (b) the judgment
arrived at by most of those concerned, or (c) group solidarity in
sentiment and belief is best used by religious, fraternal or other
social groups. In my view, excess emphasis on reaching consensus views
on complex scientific issues can obscure underlying uncertainties and
ambiguities in data and conclusions. It is my view that CASAC should
focus on the science and avoid offering judgments that are inherently
statements of a desired policy outcome.

Recommendations from CASAC with regard to specific numerical levels
and forms of the Standard should always be identified as a blend of
science and policy (See Samet letter of March 30, 2011, to
Administrator Jackson). To the extent feasible, Panel members should
attempt to differentiate between their views on the science and their
personal policy preference as outcomes. It is crucial that the CASAC
never preclude the Administrator's policy option of reaffirming the
existing NAAQS. Science should always inform the policy judgments made
in the setting of each NAAQS, but the science should not be framed as
though it dictates a specific numerical outcome or lower.

Q3. The Federal Advisory Committee Act requires that panels be
``fairly balanced in terms of points of view,'' and General Services
Administration regulations guiding FACA implementation (41 CFR 102-
3.60) require agencies to, in establishing advisory committees,
``ensure that, in the selection of members for the advisory committee,
the agency will consider a cross-section of those directly affected,
interested, and qualified, as appropriate to the nature and functions
of the advisory committee.'' In your view, is EPA CASAC membership
``fairly balanced in terms of points of view,'' and is an appropriate
``cross-section of those directly affected, interested, and qualified''
represented on the Committee?

A3. As I discussed in my answers above, I do not believe the
membership of CASAC or recent CASAC Panels have been ``fairly balanced
in terms of points of view'' and have not included a ``cross-section of
those directly affected, interested, and qualified.'' To the contrary,
the membership has been excessively dominated by scientists that to a
large extent have developed the scientific information contained in the
documents. In some cases, the individuals have already offered opinions
as to how the science should be used to set NAAQS, a more stringent
standard based on their science.
In my opinion, a strong argument can be made for excluding, or at
least limiting, the portion of individuals from this class (authors of
key input) that serve on a specific panel. The subject matter being
reviewed should be presented in a manner that would allow broadly
knowledgeable scientists to offer a scientific view on the material and
its use in Standard setting without the scientist having been directly
involved in developing the science under consideration. I have advanced
the view elsewhere in these comments that the estimated portion of risk
for specific diseases linked to air pollution is sufficiently small
that greater societal benefits might be gained by looking more broadly
at other factors influencing the risk of these diseases. This broader
view is likely to come from scientists including medical practitioners
who can look beyond ``air pollution'' as the dominant risk factor for
the disease of concern. The focus needs to be on improving the health
of society, not single risk factors.

Q4. Guidance from both OMB and the National Academies indicates that
peer reviewers should not review work products that they were involved
in. It appears, however, that 22 of the 25 members of the 2008 ozone
reconsideration CASAC panel were reviewing EPA documents that
specifically cited their work. Is this appropriate, and if not, what
recommendations do you have to better account for and avoid such
situations?

A4. I do not take exception to this statement as fact. To be blunt, a
``clean air research enterprise'' has been created over the past
several decades. The ``enterprise'' wants to be sustained and it wants
to be heard and to have influence. One measure of success expressed by
some individuals is whether the NAAQS. On the positive side,
substantial new scientific information has been developed. However, the
approach has become excessively narrow focusing many times on a single
risk factor, a specific pollutant, or a broad class of risk factors,
air pollution. In my opinion, the utility of this narrow approach may
be reaching the point of diminishing returns. I am personally convinced
the time has come for CASAC members and consultants to be selected
based on a broader health orientation with limited participation of
individuals whose focus is on single risk factors (pollutants or air
pollution in general).

Q5. Several witnesses mentioned significant health effects and
premature mortality associated with socioeconomic status, joblessness,
and other economic factors.

Q5a. Please describe the recent literature that suggests a correlation
between socioeconomic status and health outcomes.

A5a. In my opinion, recent EPA documents have been seriously deficient
in not explicitly acknowledging that the diseases of concern arise from
multiple risk factors and that specific air pollutants, and air
pollution in general, is not the major causative factor. Supreme Court
Justice Stephen Breyer in Whitman v. American Trucking Associations
noted the importance of using a ``comparative health'' orientation in
deciding how low is low enough in setting NAAQS. I have been
disappointed that EPA and CASAC have not followed up on this excellent
advice.
It is my view, drawing heavily on the opinion of Justice Breyer,
that the EPA Administrator should take a broad view of multiple factors
when making policy decisions, informed by science, on the level (with
or given statistical form) of a specific criteria pollutant that will
be protective of public health with an adequate margin of safety. In my
opinion, this should definitely include consideration of background
levels of the pollutant arising from non-anthropogenic sources
including spatial and temporal dimensions.
Two papers come immediately to mind with regard to the role of
socioeconomic factors on health. The recent paper, ``Losing Life and
Livelihood: A Systematic Review and Meta-Analysis of Unemployment and
All-Cause Mortality,'' [D.J. Roelf, E. Shor, K.W. Davidson and J.E.
Schwartz, Social Science and Medicine 72: 840-854 (2011)] provides a
comprehensive review of the health impacts of unemployment. The finding
that ``the risk death was 63% higher among those who experienced
unemployment than among those who did not, after adjustment for age and
other covariates'' is sobering in view of current unemployment in the
United States and around the world. The 63% increase in all cause
mortality stands in stark contrast to the increased risk of less than
10% estimated for air pollution.
An earlier paper, ``All-Cause and Cause-Specific Mortality by
Socioeconomic Status Among Employed Persons in 27 U.S. States, 1984-
1997,'' [K. Steenland, S. Hu and J. Walker, Am. J. Public Health 94:
1037-1042 (2004)] compares the impact of socioeconomic status on
multiple health outcomes. A key statistic is the Mortality Rate Ratio
which is the ratio of the lowest quartile of socioeconomic status over
the top quartile of socioeconomic status. In short, a comparison of the
poorest one-quarter of the population with the one-quarter most well
off. The ratio for all-cause mortality for men was 2.02 and for women
it was 1.29. These increases of 102% and 29%, like the 63% increases
estimated by Roelf et al. for unemployment, are sobering. Heart disease
has been increasingly cited as being of concern for particulate matter.
Steenland et al. found that the Mortality Rate Ratio for socioeconomic
status was 1.88 for men and 1.84 for women. These increases associated
with socioeconomic status of 88% and 84% are substantially greater than
those observed for particulate matter.
In my paper, ``Role of Science and Judgment in Setting Ambient Air
Quality Standards: How Low is Low Enough?'' [R.O. McClellan, Air
Quality and Atmospheric Health, published online 01 June 2011], I
express the view that socioeconomic impacts should be considered by the
Administrator as context for setting the NAAQS. It would be appropriate
for papers such as those by Roelf et al. (2011) and Steenland et al.
(2004) to be discussed within the EPA's various documents undergirding
each NAAQS and in the Regulatory Impact Analyses. It would certainly be
appropriate for CASAC to review these papers and comment on their
scientific quality. In my opinion, the scientific methodology and
quality of these papers are certainly equivalent to that found in the
various papers that focus on air pollution as a risk factor.
In a few instances, those papers such as the reanalysis conducted
by Krewski et al. [Krewski, D., M. Jerrett, R.I. Burnett, R. Ma, F.
Hughes, Y. Shi, M.C. Turner, A.C. Pope III, G. Thurston, F.E. Calle,
M.I. Thun, ``Extended Follow-Up and Spatial Analysis of the American
Cancer Society Study Linking Particulate Air Pollution and Mortality,''
Health Effects Institute, Cambridge, MA, Report No. 140] with support
from the Health Effects Institute, factors such as socioeconomic status
are examined as risk factors. Many times the results for the covariate
such as socioeconomic status are not reported. Krewski et al. (2009)
did include the results in their report to the Health Effects
Institute. Not surprisingly, the relative risks of socioeconomic status
are greater than for the individual pollutants. Unfortunately, these
results are not highlighted and directly compared to the results for
individual air pollutants. In short, these important contextual
findings are buried in the reports and never surface in the EPA's
documents reviewed by CASAC.
I have suggested, only partially tongue in cheek, that the CASAC
review process might be strengthened if the CASAC Panels were to
include several unemployed and several underemployed scientists. My
view is that personal experience with socioeconomic impacts might help
CASAC Panels fully appreciate the enormous impact of their advice and
the need to draw a clear line between comments on the science and any
personal ideological preferences for tightening the NAAQS.
There should be no argument that a healthy population is dependent
on a healthy economy--jobs do count! The offering of scientific advice
that will inform policy judgments on the setting of NAAQS clearly
carries with it substantial responsibility.

Q5b. In your view, does EPA or CASAC adequately take these regulatory
consequences into account?

A5b. To the best of my knowledge, the EPA has never in any
documentation related to the setting of NAAQS taken account of the
impact of socioeconomic status or unemployment on the health of the
U.S. population. Moreover, the EPA has not in any documentation related
to the setting of NAAQS clearly acknowledged that other risk factors
have a substantially greater impact on the health endpoints under
consideration than does the specific criteria pollutant or air
pollution in general. I am not aware that CASAC has ever advised EPA to
take account of the role of socioeconomic factors, unemployment or
other risk factors influencing the health endpoints under
consideration. In developing the documentation for setting the NAAQS,
EPA has ``blinders'' on with regard to providing any context for the
policy judgments that must be made in setting the NAAQS. Likewise,
CASAC has had a similar narrow focus as though the setting of the
specific NAAQS was the only concern. Indeed, the EPA and CASAC have
even given credence to the use of the results from use of a single
pollutant models when the results of the use of multipollutants models
showed diminished effects from the specific criteria pollutant under
consideration. The EPA and CASAC regularly use the language of the
Clean Air Act and the Supreme Court decision in Whitman v. American
Trucking Associations (2001) to keep a narrow focus on setting the
NAAQS under consideration, leaving the impression that providing
context for policy judgments would detract from the setting of the
NAAQS. Indeed, recent CASAC letters that advise that the NAAQS must be
set at some specific level lower than the current NAAQS appear to have
the goal of assuring the NAAQS will be lowered.

Q6. You discussed the President's decision to withdraw EPA's proposed
reconsideration of the ozone standard. What principles from that
decision should be applied to other EPA standards?

A6. I applauded President Obama's decision directing EPA to not
proceed with setting of the ``reconsideration; Ozone NAAQS. It was the
right ``common sense'' decision. It is unfortunate the President did
not have a conversation with EPA Administrator Jackson in early 2009
indicating the need for using common sense in the setting of NAAQS as
well as all other regulatory decisions.
The OIRA/OMB memorandum signed by Cass Sunstein that undergirded
the President's decision emphasized several points:

(a) Decisions on each NAAQS must use the latest science
to inform the policy judgments inherent in setting each NAAQS.

(b) Decisions on each NAAQS should take account of other
related regulatory actions. In this specific case in considering the
potential for developing a ``reconsideration'' Ozone NAAQS,
Administrator Jackson should have recognized that setting of the Ozone
NAAQS in March 2008 had already triggered the next review. In short,
the memo emphasizes the need be efficient in use of resources and avoid
needless efforts.

(c) Be respectful of the guidance in the Clean Air Act
for periodic review of each NAAQS. A corollary is to avoid arbitrary
and capricious deviations from that schedule.

(d) Remember to consider context in taking regulatory
actions including the setting of the NAAQS. In this specific instance,
the President pointed to the need for considering the current dire
state of the economy that among other factors requires predictable
regulatory actions. I can only assume the President sought legal
counsel in deciding that the language the Clean Air Act provided
latitude for the decisions of OIAR/OMB and his own actions.

I think these four principles offer a sound foundation for future
EPA actions in reviewing and setting NAAQS and making other regulatory
decisions.

Q7. Dr. Phalen and Dr. Thurston both mentioned the role of particulate
matter constituents in determining health effects, as opposed to
particulate matter mass. In your view, what is the validity of
developing regulations or estimating regulatory health benefits based
upon particulate matter mass?

Q7a. Has your research or research by others been able to identify and
rank the relative toxicities of the known components of PM2.5?

A7a. In my opinion, the use of a particulate matter mass based metric
based on particle size, i.e., Total Suspended Particulate Matter, PM2.5
and PM2.5 was appropriate at the time each of those indicators was
adopted. The science available at the time informed the policy decision
to use them. However, even when initially adopted it was recognized
that these were relatively crude indicators of the potential toxicity
of particulate matter. In short, they provided blunt tools for guiding
development and implementation of particulate matter control
strategies. However, the science available today clearly indicates that
not all particulate matter has equal toxicity potency irrespective of
the chemical compositions. It is crucial to recognize this range of
potency, especially when EPA with CASAC concurrence has moved toward
lower and lower standards as though all Particulate Matter being
regulated was as potent as the most potent particulate matter studied.
There is clearly both a temporal and spatial pattern to particulate
matter potency. For example, the strongest signals of a particulate
matter causing morbidity and mortality are in the Northeast USA while
it appears that particulate matter levels in California have not
resulted in statistically significant increases in health risks.
In my opinion, a strong case can be made for reaffirming the
particulate matter NAAQS put in place in 2006 and offering guidance for
imposing more stringent particulate matter standards only when there is
clear evidence linking an increase in adverse health effects to (a)
specific chemical composition on a size selection basis (b) or to
particulate matter emitted from specific sources. A PM2.5 NAAQS based
only on mass and set at lower and lower levels is not consistent with
current scientific knowledge.

Q7b. Could you provide your ranking of the toxicity of PM2.5
components on either a qualitative or quantitative basis?

A7b. It is not possible in the limited time and space available to
provide a complete response to this broad and important question.
Suffice it to note that the several EPA documents that undergirded the
1997 and 2006 revisions of the NAAQS for particulate material and the
current review described the substantial spatial variability in the
composition of particulate matter in the several size ranges. For
example, the most recent Integrated Science Assessment for Particulate
Matter (December 15, 2009) notes that the contribution of sulfate in
the east (16 to 46% of particulate matter, 2.5 m exposures) is
substantially greater than in the west (about 4%) while motor vehicle
emissions and secondary nitrate are greater sources of exposure in the
west (about 9%) as compared to the east (about 4%). Previous documents
have shown as much as a thousandfold difference in the concentration of
specific elements (such as vanadium, nickel and lead) in particulate
matter samples from across the USA. The same EPA documents report the
results of toxicity assays with both ambient particulate matter and
specific aerosols such as Carbon Black. Many of these studies have
yielded negative results. I would speculate that the differences in
toxic potency for aerosols (in the PM2.5 size range) are at least as
great as the acknowledged difference in the potency of PM2.5 versus
PM2.5 aerosols.
The simple fact is that EPA, in company with CASAC, seems to have a
focus on using a ``one size fits all'' approach in creating a NAAQS for
Particulate Matter, 2.5 m, and driving the standard to lower and lower
concentrations. In my opinion, if CASAC were doing its job it would
have urged a ``time out'' and asked--does the current approach make
scientific sense? I do not think it does, and from Dr. Robert Phalen's
testimony it is clear that he does not think the current exclusive
focus on a mass-based NAAQS for PM2.5 is appropriate. Likewise, it
appears that Dr. George Thurston might agree. I submit that a more
scientifically sound approach to promoting public health would be for
the EPA to look at complementary (a) PM NAAQS and (b) source specific
standards based on the potency of actual emissions from different
sources. The National Ambient Air Quality Standards are not the only
tool EPA has in the Clean Air Act tool box!
A serious problem with EPA and CASAC's approach to evaluating
scientific evidence is the current excessive emphasis on ``causality.''
With this approach, causality is evaluated exclusively based on PM2.5
mass; all the studies are placed in the same bin. Indeed, one can argue
that the approach is directed at having an increasing number of health
endpoints identified as causally related to PM2.5 exposure,
irrespective of the levels of exposure. It is disappointing that CASAC
did not offer EPA scientific advice to use the ``causality'' approach
to evaluating key individual constituents in particulate matter.

Q8. Both ozone and particulate matter occur naturally. Could you
describe the role of background levels of ozone and particulate matter
in setting standards for these pollutants? Have EPA and CASAC properly
accounted for these background levels in establishing regulations or
estimating health benefits?

A8. The EPA has not acted in a consistent, science-based manner in
dealing with background levels of any criteria pollutants including
ozone and particulate matter. Some CASAC Panel members and consultants
have urged that greater attention be given to background levels. Other
individuals have expressed the view that background levels of ozone
should not be considered in the setting of the ozone NAAQS, that
background should only be considered during implementation of the
NAAQS. I strongly disagree with that view.
In my opinion, the EPA needs to do a much better job of creating a
scientifically sound understanding of the spatial and temporal
dimensions of background levels for all criteria pollutants, especially
for ozone and particulate matter, and then using that scientific
information to inform policy decisions in the setting of NAAQS. I note
the need for acknowledging spatial and temporal dimensions. I make this
point because there are clear differences in ozone background across
the USA; one size does not fit all. A failure to acknowledge those
differences in the setting of NAAQS can penalize certain areas of the
USA with naturally occurring higher levels of background ozone.
The temporal pattern of ozone background levels is also very
important. This was ignored by EPA and CASAC in the setting of the
March 2008 ozone NAAQS. In that case, EPA did a very poor job of
characterizing what it called ``policy relevant'' background ozone. EPA
understated the background levels of ozone, especially as they would
occur with an eight-hour averaging time and a standard set at about the
98th percentile for exceedances. The result was to substantially
overstate the potential benefits of reducing the level of the Standard
(see testimony of Dr. Anne Smith). These points were emphasized in a
report I provided to EPA in October 2007 as part of the public comment
process. This is a very specific example of how EPA and CASAC regularly
ignore public comments that do not support their position, a lower
NAAQS.
The issue of background levels for PM2.5 is also important,
especially as regards the 24-hour averaging time standard. The
Particulate Matter Integrated Science Assessment (2009) noted that
Policy Relevant Background levels, levels in the absence of U.S.
anthropogenic sources, had a maximum daily range of 3.1 to 20 mg/m3
with a peak of 63 mg/m3 at the nine National Parks across the USA. It
is hoped that the EPA Administrator will consider these levels when
making a policy decision to reaffirm or revise the NAAQS for PM2.5.

Q9. As you know, influential studies based on data from the American
Cancer Society and the Harvard Six Cities Study provide the basis for
major EPA regulations and determines how EPA develops its ``deaths
avoided'' estimates for particulate matter. These data sets were
developed with government funds, but are not publicly available so they
can be analyzed by other scientists. Do you support making this and
similar federally-funded highly influential scientific data and
information transparent and publicly available so they can be analyzed
by other scientists? Do you support making this and similar federally-
funded highly influential scientific data and information transparent
and publicly available?

A9. It is my understanding that the Harvard Six Cities Study,
initiated by the late Professor Ben Ferris, was funded largely by
grants from the National Institutes of Health with perhaps some limited
supplemental funding from the EPA. The study also made substantial use
of air monitoring data collected with support from the EPA. Individuals
in six different studies were enrolled in the study with the explicit
understanding that it was being conducted to evaluate the health
effects of air pollution. Thus, it is correct that this study was
funded largely by U.S. government funds. The Harvard University
investigators have indicated in the past that individuals enrolled in
the study with the understanding that their identities would not be
revealed and, thus, one basis for not releasing the data to other
investigators is that the identity of specific subjects might become
known.
The American Cancer Society (ACS) studies present quite a different
situation. It is my understanding that the ACS cohort of individuals
from across the USA were enrolled for the purpose of conducting studies
to better understand the occurrence of cancer in the population. It is
my understanding that individuals self-enrolled and, thus, it was not a
random sample from the U.S. population and is not likely a
representative sample of the U.S. population. The study, from its
beginnings, has been managed and funded by the ACS. The survival of the
enrollees has been followed using National Death Records. It is my
understanding that what is known about each enrollee is based on what
they provided at the time of self-enrollment, including place of
residence and the matched information at death including when and where
they died. Obviously, information about each individual's place of
residence, life style, occupation, etc., between time of enrollment and
death is unknown. The ACS very closely controls access to the ACS data.
Decades ago I urged that these extraordinarily valuable data sets
that have a pivotal role in EPA's setting of NAAQS should be made
available to other investigators for evaluation. I expressed this view
not out of concern for Harvard University investigators to conduct the
studies in a competent manner, but rather related to a view that
different analytical teams with varied scientific backgrounds might
identify alternative approaches to analyzing the data sets. After much
negotiation the Health Effects Institute (HEI), jointly funded largely
by the EPA and the automotive industry, would sponsor a ``re-analysis''
of the Harvard Six Cities Study and ACS Study data sets. I was pleased
that their re-analysis was conducted. I was very disappointed that a
second or third team of analysts were not funded to do parallel
analyses.
By and large, the single re-analysis did verify the core findings
of the original analyses. Most importantly, the re-analysis revealed
additional information over and above that of the original analyses.
It is my personal opinion that any research study conducted with
U.S. government funding should, at some reasonable time interval, have
provision for the data (stripped of personal identifiers) to be
released for evaluation by other competent investigators. This only
seems reasonable as an approach to realizing the best return on the
expenditure of U.S. government funds. It is obvious that such an
approach will pose many logistical challenges. I am confident the
hurdles can be overcome if there is a will to serve the public good.
My own personal preference is to require that a procedure be
established that will require that for any study results to be used by
the EPA in the setting of NAAQS or other standards, the base data from
the study be made available (stripped of any personal identifying
information) for further analysis by the EPA or other interested
parties. It is important to recognize that decisions with multi-billion
dollar impacts (both benefits and costs) are being based on the
analysis of these very large and complex data sets. Critics of my
proposal will likely argue that the standard peer review process used
by prestigious scientific journals should be sufficient to assure the
quality of the analyses of the original investigators. I am a strong
proponent of peer review and fully recognize both its strengths and
weaknesses. A comparison of the peer review process used by HEI in
conducting the re-analyses of the Harvard Six Cities Study and the ACS
Study with the original peer review given the original publications
should be sufficient to convince skeptics if the need for something
beyond the usual peer review when policy decisions informed by science
are used to set NAAQS that have enormous impact on society.
Responses by Dr. George Thurston, Professor,
New York University School of Medicine

Questions submitted by Representative Andy Harris, Chairman,
Subcommittee on Energy and Environment

Q1. Do recent studies report an association between particulate matter
and mortality in all cities and areas of the U.S.?

Q1a. If not, how do you account for this fact that PM is associated
with mortality in some cities, but not in others?

Q1b. Similarly, are there cities and areas of the U.S. where no
association between ozone and acute mortality are reported? If so, what
explains this lack of an association?

A1a-b. While there is a nationwide association found between exposure
to air pollution and mortality (e.g., see Pope et al., 2002; Bell et
al., 2004; Dominici et al., 2005), there is not always enough
statistical power to detect such a size of effect in single cities, due
to lower population size (and therefore in numbers of observations) in
smaller cities. Thus, the most consistent air pollution-health effects
relationships are usually found in large cities, such as New York and
Los Angeles, where there are sufficient numbers of people to provide
enough statistical power to discern, in a single city, the air
pollution effect that is also seen overall nationwide.
As I stated in the hearing on October 4, 2011: ``You need enough
power, which is one of the quandaries we have is that the places where
the people live tend to be the more polluted places, and the less
polluted places don't have as many people, so then it becomes more
difficult to study at the lower levels.'' And then, again, later in the
hearing, I also stated: ``part of the reason why we don't see effects
at very low levels, or unable to show them definitively, is because we
don't have enough people living in clean air. If we have cleaner air,
then we will be able to test that hypothesis.''

Q2. You wrote in 2009 that access to vital ``records that have
informed . . . pivotal research has recently been curtailed sharply,
threatening the continuation of the type of research necessary to
support future standard setting.'' You have been a co-author on
influential studies based on data from the American Cancer Society and
the Harvard Six Cities Study. These data provide the basis for major
EPA regulations and determine how EPA gets their ``deaths avoided''
estimates. These data sets are not publicly available so they can be
analyzed by all scientists. Do you support making these data sets
transparent and open to outside scrutiny if they are being used to
justify regulatory decisions?

A2. While making data available for independent scientists to analyze
is an important goal, a major complication is the countervailing
priority of protection of patient and/or study subject privacy. To
resolve these two important countervailing concerns, these two
particular data sets have been released to an independent body (The
Health Effects Institute), and their respective results and findings
validated independently (see Krewski et al., 2003).

Q3. In both your previous work and your testimony, you have indicated
substantial differences in the effects of components of particulate
matter, such as PM from Seattle and PM in Detroit. You have stated that
``PM composition has an appreciable influence on the health effects
attributable to PM'' and that ``we should be starting to look at more
of the constituents.''

Q3a. If these constituents are the critical determinant of health
effects, does it make sense to develop regulations or to estimate
regulatory health benefits based upon particulate matter mass?

Q3b. Have you considered how EPA might develop some type of toxicity-
weighted NAAQS?

Q3c. Has your research or research by others been able to identify and
rank the relative toxicities of the known components of PM2.5?

Q3d. Could you provide your opinion as to the appropriate ranking of
PM2.5 constituents by toxicity on either a qualitative or quantitative
basis?

Q3e. You mentioned the 1989 Pope study about PM and health effects.
How did the composition of PM vary between strike and non-strike days?

A3a-e. (3a.) Since the evidence links PM2.5 mass with health effects,
it is important to continue to regulate that, but other constituent-
specific regulations can also be set, as done in the past with lead
(Pb) in particulate matter. This could potentially allow regulations
for PM2.5 control to be more efficiently focused on controlling the
particulate matter with the greatest toxicity, rather than merely those
contributing the greatest mass. (3b.) As noted above, it might take the
form of regulating PM2.5 constituents, as has previously been the case
with lead (Pb). (3c.) That goal is the subject of ongoing research by
me and others in my field of research. (3d.) I don't think it is
possible to quantitatively rank the toxicity of PM constituents at this
time, and it likely varies from health endpoint to health endpoint, but
generally speaking we have found that the particulate matter air
pollution from fossil fuel combustion is among the most strongly
associated with adverse health effects, including mortality. (3e.) This
has been investigated in depth by intramural researchers at the U.S.
EPA. They have discovered that:

``the 1986/1988 (Utah) extracts contained more sulfate,
cationic salts (i.e., calcium, potassium, magnesium), and certain
metals (i.e., copper, zinc, iron, lead, strontium, arsenic, manganese,
nickel). Although total metal content was (3/4) 1% of the extracts by
mass, the greater quantity detected in the 1986 and 1988 extracts
suggests metals may be important determinants of the pulmonary toxicity
observed.'' (Dye et al., 2001).

Moreover, these U.S. EPA scientists were also able to replicate key
biological effects of this PM using a mixture of such metals,
reporting:

``The parallel epithelial injury induced by the extracts
and their surrogate Zn + Cu + V mixtures suggests that these metals are
mediating the acute airway epithelial effects observed; however, metal
interactions appear to play a critical role in the overall cellular
effects induced by the PM-derived extracts. These experimental findings
are in good accord with epidemiologic reports of adverse airway and
respiratory health effects in Utah Valley residents.'' (Pagan et al.,
2003)

In related research by this group of U.S. EPA investigators, it has
been found that the co-presence of sulfates, such as sulfuric acid, can
make these transition metals more ``bio-available'' to cells in the
body, and therefore more damaging. Thus, the co-presence of acidic
sulfates in such metals containing particles makes them even more
potent at damaging the lung (e.g., Gavett et al., 1997). This
conclusion is also supported by studies of human respiratory cells
(e.g., Veronesi et al., 1999), and by the conclusions of a recent
report from the Committee on the Medical Effects of Air Pollutants
(COMEAP, 2009).

Q4. Figure 3 in your written testimony appears to indicate that
relative risk for PM2.5 drops below 1.0 at 14 micrograms per cubic
meter (mg/m3) with no data for PM2.5 levels lower than 13 mg/m3. In
your oral testimony you referred to this graph as an indicator that
premature deaths due to PM2.5 occurred well below the NAAQS.

Q4a. Could you explain if this graph or any related study would
support EPA's calculating PM2.5-related death down to the lowest
measured value of approximately 4 mg/m3?

Q4b. In light of your testimony about the ``independent expert
advice'' provided by CASAC, why have they failed to recommend a NAAQS
down to this lowest measured value?

Q4c. This figure shows that much of the data is consistent with a
threshold (relative risk of 1.0 or below) at about 15 mg/m3. In your
mind, what relative risk would suggest a causal relationship?

A4a. The graph is not inconsistent with EPA's assumption, and is
consistent in that effects are suggested to extend well below the
present long-term standard for PM2.5.

A4b. There are different health outcomes to consider, and differing
strengths of association for the various outcomes at various
concentration levels. Also, the setting of a standard is a different
process from that used for a risk analysis, so they are not necessarily
going to agree on levels of applicability.

A4c. This question is based upon a misinterpretation of Figure 3. The
relative risks (RRs) provided in that graph are relative to the mean
long-term PM2.5 concentration in the study (about 15 mg/m3), so the
fact that the RRs go below 1.0 only means that, as the concentration
goes below the mean, the risk of mortality is proportionally reduced
from the mean risk. It does not mean there are no effects below 15 mg/
m3. Instead, it means that there are mortality risks of PM2.5 exposure
to concentrations below 15 mg/m3, but they are smaller than if the
concentration were 15 mg/m3. Thus, this would indicate that there are
PM2.5 mortality risk benefits of reducing PM2.5 that extend below 15
mg/m3.

As I stated on the day of the hearing: ``there is the indication
that the benefits do keep going down well below the standard to levels
about seven micrograms per meter cubed. And the only reason probably we
can't show below that (concentration) is that we have no place in the
country that is cleaner than that, that has a metropolitan area with
enough people to study.''

Q5. You discussed the value of peer review in ensuring that scientific
information is ``not just a report that has been put out by an interest
group or by the government.'' Would you support EPA's Regulatory Impact
Analyses being subjected to peer review by CASAC or another body?

A5. Proper peer review is an important aspect of scientific research
evaluation, so I strongly support that occurring for all scientific
research. U.S. EPA RIAs are public documents that already undergo a
significant amount of scrutiny under the present process. It is also
important to note that the RIAs are not research, per se, but instead
they are documents that rely on published, peer-reviewed literature.
Moreover, I am uncertain whether the legislative authority for the U.S.
EPA RIA allows for that process to occur under CASAC, in the case of
such a regulatory document that is required, not by the Clean Air Act,
but by a separate Executive Order.
If such a review were deemed to be in the purview of CASAC, I also
am also uncertain whether CASAC presently has the proper external
environmental economist expertise to review such an economic document.
For example, the EPA's recent Air Toxics Rule RIA reportedly examines
changes in employment in the directly regulated industry (utilities),
and the increased demand for labor directly stemming from the
construction and installation of pollution abatement and control (PAC)
equipment resulting from this regulation (see Bivens et al., 2011,
attached to this letter for inclusion in the record), which is not, to
my knowledge, a topic covered by the present expertise of EPA's CASAC.
Thus, I see possible legal and logistical barriers to this proposed
additional RIA peer review.

References

Bell ML, McDermott A, Zeger SL, Samet JM, Dominici F. 2004. Ozone
and short-term mortality in 95 US urban communities, 1987-2000. JAMA.
2004 Nov 17;292 (19):2372-8.

Bivens, J. (2011). A lifesaver, not a job killer: EPA's proposed
``air toxics rule'' is no threat to job growth. Briefing paper #312.
Economic Policy Institute, Washington, DC.

COMEAP. (2009). Long-term exposure to air pollution: effect on
mortality. London: Committee on the Medical Effects of Air Pollutants
(COMEAP), Department of Health.

Dominici F, McDermott A, Daniels M, Zeger SL, Samet JM. 2005.
Revised analyses of the National Morbidity, Mortality, and Air
Pollution Study: mortality among residents of 90 cities. J Toxicol
Environ Health A. 2005 Jul 9-23;68(13-14):1071-92.

Dye JA, Lehmann JR, McGee JK, Winsett DW, Ledbetter AD, Everitt JI,
Ghio AJ, Costa DL. Acute pulmonary toxicity of particulate matter
filter extracts in rats: coherence with epidemiologic studies in Utah
Valley residents. Environ Health Perspect. 2001 Jun;109 Suppl 3:395-
403.

Gavett SH, Madison SL, Dreher KL, Winsett DW, McGee JK, Costa DL.
(1997) Metal and sulfate composition of residual oil fly ash determines
airway hyperreactivity and lung injury in rats. Environ Res. 1997
Feb;72(2):162-72.

Krewski D, Burnett RT, Goldberg MS, Hoover BK, Siemiatycki J,
Jerrett M, Abrahamowicz M, White WH. 2003. Overview of the reanalysis
of the Harvard Six Cities Study and American Cancer Society Study of
Particulate Air Pollution and Mortality. J Toxicol Environ Health A.
2003 Aug 22-Oct 10;66(16-19):1507-51.

Pagan I, Costa DL, McGee JK, Richards JH, Dye JA. 2003. Metals
mimic airway epithelial injury induced by in vitro exposure to Utah
Valley ambient particulate matter extracts. J Toxicol Environ Health A.
2003 Jun 27;66(12):1087-112.

Pope, CA III, Burnett, RT, Thun, MJ, Calle, EE, Krewski, D, Ito, K,
and Thurston, GD. Lung cancer, cardiopulmonary mortality and long-term
exposure to fine particulate air pollution. J. Am. Med. Assoc. (JAMA)
287(9):1132-1141 (2002).

Veronesi B, Oortgiesen M, Carter JD, Devlin RB. (1999). Particulate
matter initiates inflammatory cytokine release by activation of
capsaicin and acid receptors in a human bronchial epithelial cell line.
Toxicol Appl Pharmacol. 1999 Jan 1;154(1):106-15.
Responses by Dr. Michael Honeycutt, Chief Toxicologist,
Texas Commission on Environmental Quality

Questions submitted by Representative Andy Harris, Chairman,
Subcommittee on Energy and Environment

Q1. During the hearing, there was some discussion about the
independence and impartiality of CASAC. You have frequently been
involved in the CASAC process and you were on EPA's ``short list'' for
the most recent particulate matter panel.

Q1a. What are the major strengths and weaknesses of the current CASAC
process?

Q1b. EPA often cites the public health benefits in its Regulatory
Impact Analyses to argue for more stringent standards. Does CASAC
review these analyses?

Q1c. What recommendations do you have for improving the CASAC process
and ensuring that panels are independent, transparent, balanced, and
impartial?

A1. This question appears to have been meant for Dr. Thurston, but I
will also address it.
Strengths

Formed from non-EPA scientists, engineers, and economists
and other social scientists recognized as experts in their fields.

Should have a wide range of representation from academia,
industry, federal, state, and tribal governments, research institutes
and non-governmental organizations.

Chartered to:

Review the criteria and standards promulgated by EPA,
and provide other related scientific and technical advice;

Recommend to the EPA Administrator any new NAAQS and
revisions of existing criteria and standards as may be appropriate;

Advise the EPA Administrator of areas in which
additional knowledge is required to appraise the adequacy and basis of
existing, new, or revised NAAQS;

Describe the research efforts necessary to provide the
required information;

Advise the Administrator on the relative contribution
to air pollution concentrations of natural as well as anthropogenic
(human-caused) activity;

Advise the Administrator of any adverse public health,
welfare, social, economic, or energy effects which may result from
various strategies for attainment and maintenance of NAAQS.

Chartered Member Composition:

At least one member of the NAS;

At least one physician;

At least one person representing State air pollution
control agencies.

Weaknesses

Members recommended by EPA staff and appointed by EPA
Administrator.

Financial and administrative support given solely by EPA.

There are only seven chartered members.

Current Chartered Member Composition:

None of the current members appears to be a member of
the NAS--a charter requirement;

While there are two members with an M.D., neither
appears to be practicing physicians;

One member represents the eight northeastern states
through a non-profit organization, accounting for approximately 14%
(one out of seven members)--most states and tribes are not represented;

Academia is represented by approximately 57% of the
panel (four out of seven members);

Research organizations represent approximately 29% (two
out of seven members);

Industry, federal and tribal governments, and non-
governmental organizations have 0% representation on the current panel.

A1a. It does not appear that the CASAC panel reviews EPS's Regulatory
Impact Analyses.

A1b. Appointed membership to the CASAC panel should be an impartial
and transparent process that allows for a balanced representation for
all interested parties. To achieve this, it would be advisable to have
the membership appointed by neutral parties. It is a conflict of
interest for a person to pick the members of the review committee that
will review their work.

A1c. I believe the CASAC review process should not be run by EPA. It
should be run either by another federal organization such as the
Council on Environmental Quality or by an independent group such as the
NAS. The review panels should be broader than they currently are, with
multiple experts in each subject area with varied and balanced views.
For example, the ozone CASAC should consist of, at a minimum, two
epidemiologists, two toxicologists, two human health risk assessors,
two practicing physicians (e.g., pulmonologists), two atmospheric
chemists, and two environmental engineers. These experts should
represent a variety of interests, including federal government, state
government, academia, industry, and non-governmental organizations.
Charge questions should not be so pointed as to lead the peer reviewers
to a pre-determined conclusion. Of course, peer reviewers should not be
reviewing their own work as either bias or the perception of bias could
influence the process. EPA should model their peer reviews after those
conducted by Toxicology Excellence for Risk Assessment (www.tera.org),
a non-profit group recognized worldwide for their independent
scientific peer reviews.

Q2. The Federal Advisory Committee Act requires that panels be
``fairly balanced in terms of points of view,'' and General Services
Administration regulations guiding FACA implementation (41 CFR 102-
3.60) require agencies to, in establishing advisory committees,
``ensure that, in the selection of members for the advisory committee,
the agency will consider a cross-section of those directly affected,
interested, and qualified, as appropriate to the nature and functions
of the advisory committee.'' In your view, is EPA CASAC membership
``fairly balanced in terms of points of view,'' and is an appropriate
``cross-section of those directly affected, interested, and qualified''
represented on the Committee?

A2. No, the CASAC membership is not fairly balanced and is not an
appropriate cross-section of those directly affected, interested, and
qualified, as noted in the response to question 1. The states and
tribes are grossly underrepresented, and no CASAC members represent
industry.

Q3. Guidance from both OMB and the National Academies indicates that
peer reviewers should not review work products that they were involved
in. It appears, however, that 22 of the 25 members of the 2008 ozone
reconsideration CASAC panel were reviewing EPA documents that
specifically cited their work. Is this appropriate, and if not, what
recommendations do you have to better account for and avoid such
situations?

A3. It is not appropriate for scientists to peer review their own
work; it presents a clear conflict of interest. When a panel reviews a
document in which a member's work is cited, that member should recuse
himself from review of the document. Prior to each review, panel and
committee members should sign a legally-binding document stating their
conflicts of interest, which should be made available for transparency.

Q4. In recent testimony before our Committee, EPA Assistant
Administrator Gina McCarthy stated that the Cross-State Air Pollution
Rule would prevent ``up to 34,000 premature deaths'' and Administrator
Lisa Jackson recently stated before another Committee that ``if we
could reduce particulate matter to healthy levels, it would have the
same impact as finding a cure for cancer.'' Do you agree with these
statements from the Environmental Protection Agency? In your view, do
current levels of particulate matter and ozone cause hundreds of
thousands of premature deaths?

A4. I do not agree with the statements that Ms. McCarthy and Ms.
Jackson made. The data that EPA used to base these statements cannot be
used scientifically to quantitatively ``count lives'' like EPS did. In
my view, current levels of particulate matter and ozone do not cause
hundreds of thousands of premature deaths.
EPA relies on studies that take mortality data from ecological
epidemiology studies to calculate the number of theoretical deaths that
would be avoided with a lower standard. Not only do these studies
suffer from severe limitations, but estimates of theoretical lives
saved are also meaningless from a scientific and practical standpoint.
It is not possible to verify either the current number of deaths due to
exposure or the future change in deaths if the standard is lowered. All
estimates of lives saved are estimates, not factual. There is no
guarantee of increased life expectancy or degree of confidence in such
an estimation, since some degree of risk is present in all aspects of
daily life. It is impossible to tease out the miniscule risks from low
levels of air pollution from the overwhelming risks of diet, genetics,
smoking, etc.

Q5. Dr. Phalen and Dr. Thurston both mentioned the role of particulate
matter constituents in determining health effects, as opposed to
particulate matter mass. In your view, does it make sense to develop
regulations or to estimate regulatory health benefits based upon
particulate matter mass?

Q5a. Has your research or research by others been able to identify and
rank the relative toxicities of the known components of PM2.5?

Q5b. Could you provide your ranking of the toxicity of PM2.5
components on either a qualitative or quantitative basis?

A5. Particulate matter (PM) in the atmosphere contains both primary
(i.e., emitted directly by sources) and secondary (i.e., formed in the
air from combustion processes) components, which can be anthropogenic
or natural in origin. PM can contain inorganic as well as organic
components, including acids (such as nitrates and sulfates), metals,
and soil or dust particles.
In general, the potential for adverse health effects depends on the
mass concentration, size, shape, and composition of the particles.
Differences in the composition of ambient PM introduce uncertainty into
estimates of health effects. EPA uses a generic mass concentration of
PM for risk (health effects) estimates, in which case they assume all
varieties of PM have identical potential for toxicity, which we know is
not the case. The estimates, if based solely on the same PM mass level,
may result in over- or underestimates, depending on spatial variability
in the PM components. Therefore, PM mass concentration most likely is
not the primary factor in the reported association. For example, coarse
PM (PM2.5) in urban or industrial areas is likely to be enriched by
anthropogenic pollutants that tend to be inherently more toxic than the
windblown crustal material which typically dominates coarse particle
mass in arid rural, agricultural, and mining areas.
In its 2009 Integrated Science Assessment (ISA) for PM, EPA
recognized that epidemiological studies evaluating health effects
associated with long- and short-term fine particle (PM2.5) exposures
have reported heterogeneity in responses both within and between cities
and geographic regions in the U.S. This heterogeneity may be
attributed, in part, to differences in the fine particle composition.
The ISA concludes ``that many constituents of PM2.5 can be linked with
multiple health effects, and the evidence is not yet sufficient to
allow differentiation of those constituents or sources that are more
closely related to specific health outcomes.'' Many different
constituents of the fine particle mixture as well as specific source
categories of fine particles are linked, at some level, to adverse
health effects.
Furthermore, in the Health Effects Institute (HEI) Research Report
161 (Assessment of the Health Impacts of Particulate Matter
Characteristic), Dr. Michelle Bell analyzed data on 52 chemical
components of PM2.5 and found significant associations of PM2.5
elemental carbon, nickel, and vanadium content with cardiovascular
hospital admissions, although these components contain relatively low
percentages of PM2.5 total mass. \1\ In summary, it is not appropriate
to develop regulations or to estimate regulatory health benefits solely
based on PM mass without considering chemical composition.
---------------------------------------------------------------------------
\1\ 1 Health Effects Institute (HEI). 2011. Fine Particles and
Health: Which Components Might Matter? Fall 2011 Update. Available
from: http://pubs.healtheffects.org/getfile.php?u=668.

A5a. In a study by Valberg (2004), the chemical components of ambient
PM2.5 were ranked according to their individual toxicities as a
comparison to those using the epidemiological PM-mortality ``effect-
function'' (toxicity per unit mass) methodology. In addition, the HEI
launched a National Particle Component Toxicity (NPACT) Initiative to
evaluate the comparative toxicity of specific components of PM in 2009.
\2\ Specifically, HEI NPACT will examine relationships between PM
components and health in toxicological studies in 12 locations across
the U.S. and companion epidemiological studies in over a hundred
cities. The results of these studies could provide information for
future PM NAAQS to focus on different PM components.
---------------------------------------------------------------------------
\2\ Health Effects Institute (HEI). 2007. The HEI National
Particle Component Toxicity (NPACT) Initiative: Answering Key Air
Pollution Questions. Available from: http://www.healtheffects.org/Pubs/
NPACT.pdf.

A5b. An example of toxicity rankings of known chemical components in
ambient PM2.5 can be found in the Valberg (2004) study. \3\ The TCEQ
has not undertaken such a ranking study.
---------------------------------------------------------------------------
\3\ Peter A. Valberg. 2004. ``Is PM More Toxic Than the Sum of Its
Parts? Risk-Assessment Toxicity Factors vs. PM-Mortality `Effect
Functions'.'' Inhalation Toxicology, 16(suppl. 1):19-29.

Q6. In your written testimony you describe the importance of
discerning between a statistical significance and actual biological
significance of a given adverse health effect. Is there a plausible
biological explanation for all of the NAAQS levels endorsed by EPA? Is
there a plausible biological explanation for the various health
---------------------------------------------------------------------------
benefits cited by EPA in the accompanying Regulatory Impact Analyses?

A6. I agree with EPA on the biological bases and regulatory levels for
the sulfur oxides, nitrogen dioxide, and carbon monoxide NAAQS. I
disagree with EPA on their biological bases and regulatory levels for
the ozone and PM2.5 NAAQS. I mostly agree with EPA on their biological
basis for the lead NAAQS, although I disagree with their regulatory
level. I do not think there is a plausible biological explanation for
the various health benefits cited by EPA in their ozone and PM2.5 NAAQS
Regulatory Impact Analyses.

Q7. Can you provide the Committee an explanation of an ecological
epidemiological study? How are such studies conducted? What is the
value of an ecological epidemiological study compared with a clinical
or toxicological study? How are these studies used by EPA and CASAC?
Are you aware of how EPA or CASAC weights the evidence derived from
these different studies?

A7. In most ecological epidemiology studies, researchers gather death
certificates for a certain time period for people in a particular city
who died from non-accidental causes, including diseases such as cancer
or liver disease. The assumption is that breathing ozone made them die
earlier than they would have otherwise. For each time of death, the
researchers find out what the outdoor ozone level was at various time
periods before the person died. In the case of ozone, for example,
suppose a 90-year old person died of congestive heart failure at noon
on a particular day. The researchers will find out the eight-hour ozone
average for 4 a.m. to noon. Then they will back up an hour and find out
the eight-hour ozone average for 3 a.m. to 11 a.m. They will back up
another hour, and so on, usually for 36 hours, collecting numerous
eight-hour ozone averages that they will then run through various
statistical models. They will repeat this process for hundreds to
thousands of people, depending on the study. No other patient
information is evaluated other than the time of death.
Usually the researchers will use the outdoor ozone readings from
the monitor with the highest ozone measured in the city. Some studies
average the ozone readings across the city. Using the highest monitor,
or even an average value, in a city is unscientific, since ozone levels
can vary tremendously across a city, as illustrated in the map on the
following page.

The map above is of the Dallas-Fort Worth non-attainment area
showing (as a typical example) eight-hour ozone concentrations for 1
a.m. to 9 a.m. on September 19 of this year. Monitored ozone
concentrations vary by more than 50% across this area, and yet
ecological epidemiology studies do not look at the monitor closest to
where a person actually lived. Using this map as an example, it is
scientifically unsound to assume that a person who had, in reality,
been exposed to 33 ppb was exposed to 63 ppb.
These studies did not look at whether those people that died were
truly outdoors for eight hours just prior to their death to in fact
breathe the ozone. It is most likely that the 90-year old person in our
example was in a hospital or hospice in the days preceding their death,
where the ozone concentrations were most likely near zero. Even if we
assumed that these ill people spent eight hours outdoors just before
they died, were they near the monitor that the researchers assumed they
were? Were they exposed to other pollutants during that day? Did they
take their medications that day? There are a whole host of common sense
questions that go unanswered in these studies. Simply put, ecological
epidemiology studies cannot tell us if ozone caused these deaths or if
these people died prematurely, much less tell us what level of ozone
caused their premature death.
These ecological epidemiology studies are the primary studies EPA
used to base their previously proposed ozone standard. Ecological
epidemiology data has very little, if any, value when compared to
toxicological or clinical studies. There are ample, very well-conducted
toxicological and clinical studies on which to base the ozone standard.
In my opinion, EPA should give ecological epidemiology studies no
weight at all in deriving a quantitative standard. Toxicological and
clinical studies should be given the most weight, taking into account
mode of action, biological plausibility, personal exposure, and what
actually constitutes an adverse effect (e.g., a 5% decrement in FEV1
[Forced Expiratory Velocity at one second] is NOT an adverse effect).

Q8. As you know, influential studies based on data from the American
Cancer Society and the Harvard Six Cities Study provide the basis for
major EPA regulations and determines how EPA develops its ``deaths
avoided'' estimates for particulate matter. These data sets were
developed with government funds, but are not publicly available so they
can be analyzed by other scientists. Do you support making this and
similar federally funded highly influential scientific data and
information transparent and publicly available?

A8. Yes, I strongly support making these types of data and similar
information publicly available. The reproducibility of results by
independent scientists is of paramount importance in science, and this
is a criterion of good science which cannot be tested using data not
publicly available. This is particularly important and relevant to
highly influential data with significant implications for regulation,
public health, and the economy. Independent scientific analyses may
reveal significantly disparate results depending upon the methods and
assumptions utilized in the analyses and/or that initial conclusions
were not adequately scientifically defensible to withstand any
reasonable level of scientific scrutiny. Any medical privacy concerns
may be alleviated through appropriate coding of potentially identifying
information (e.g., names, addresses). The antithesis of governmental
transparency is a highly influential government analysis based on
hidden data, a predicament for EPA only exacerbated by the fact that
the non-publically available datasets were developed using taxpayer
dollars. Clearly, important underlying data such as these should be
publically available so that the resulting scientific findings, by EPA
or any other entity, may be vetted through additional scientific
analyses and peer review.

Questions submitted by Representative Randy Neugebauer

Q1. You discussed in your testimony that ``EPA may have the most
conservative safe level for mercury in the world.'' Could you provide a
comparison of EPA's reference dose for mercury and the blood mercury
concentrations associated with health effects?

A1. The figure below shows levels of mercury in blood associated with
health effects from studies done on people in the Faroe Islands and the
Seychelles Islands, and data from recent surveys of blood mercury
concentrations in U.S. women aged 16-49, as compared to the blood
mercury level that EPA's RfD is set to protect against (5.8 ppb). EPA
is causing unnecessary alarm in the public by using decade-old data in
their assertions that 6% of women of child-bearing age have mercury in
their blood at a level capable of causing adverse effects in the
developing fetus. Newer data documenting decreases in blood mercury
concentrations over time show this statement to be false and
misleading. Additionally, adverse effects are associated with much
higher blood mercury levels.

There are no widespread mercury health effects issues in the United
States. In fact, unwarranted concerns about mercury may be causing
women to avoid eating fish, which itself could lead to adverse health
effects. Researchers from the University of Rochester recently
published a 17-year follow-up to the Seychelles study.&heir conclusions
are as follows: ``At age 17 years there was no consistent pattern of
adverse associations present between prenatal MeHg (methylmercury)
exposure and detailed domain specific neurocognitive and behavioral
testing. There continues to be evidence of improved performance on some
endpoints as prenatal MeHg exposure increases in the range studied, a
finding that appears to reflect the role of beneficial nutrients
present in fish as demonstrated previously in younger subjects. These
findings suggest that ocean fish consumption during pregnancy is
important for the health and development of children and that the
benefits are long lasting.'' \4\
---------------------------------------------------------------------------
\4\ Philip W. Davidson, et al., 2011. "Fish Consumption and
prenatal methylmercury exposure: Cognitive and Behavioral Outcomes in
the Main Cohort at 17 years from the Seychelles Child Development
Study." Neurotoxicology, In Press.

Q2. How has air quality in Texas changed over the last 10 and 20
years? Have particulate matter, ozone, and mercury levels increased or
decreased? If air quality has improved, is this progress due to steps
taken by Texas or is this just the result of federal enforcement of
---------------------------------------------------------------------------
Clean Air Act requirements?

A2. With a vigorous economy, a rich supply of natural resources and a
diverse population, Texas continues to see improvement in air quality
throughout the state. Ambient air monitoring data demonstrate trends of
decreasing concentrations over time for particulate matter and ozone,
and reported mercury emissions have continued to decrease. The strides
that Texas has made in reducing emissions, and more importantly ambient
concentrations of ozone, is more impressive considering the population
increase (meaning more cars and electricity needs) and Texas' position
as an economic engine of the entire country. Texas now has the second
largest population in the country, behind California. Between the 2000
and 2010 census, Texas' population increased by 20.6%, which equates to
4,293,741 people.
Although Texas has some of the most highly industrialized and
populated areas in the nation, air quality in these and other areas of
the state continues to improve and is comparable to or better than that
of similar areas in other states. The state has been especially
successful in reducing ozone air pollution. For example, in the last 10
years (2000 through 2010), ozone levels in Texas have decreased by 27%
statewide. By comparison, the rest of the nation averaged only a 14%
decrease in ozone levels over this same time period. Nitrogen oxides
(NO2), a main precursor to ozone formation, decreased substantially in
Texas from 2000 to 2009. Point source NO2 emissions were reduced from
796,247 tons per year (TPY) in 2000 to 336,417 TPY in 2009, a decrease
of 57.75%. In the Houston-Galveston area alone, one of the most
comprehensively controlled industrialized complexes in the world, over
180,000 tons of ozone-producing NO2 emissions had been reduced by 2009.
This reduction equates to more than the total NO3 emitted
from all Texas power plants in 2009 (145,000 tons).
Until 2006, all the PM2.5 monitors in the Houston area had recorded
design value readings lower than the NAAQS of 15.0 micrograms per cubic
meter (mg/m3)--except for the Clinton Drive monitor, which is on a
heavily traveled road across the street from the entrance to the Port
of Houston Authority (PHA). To determine the cause of the elevated
readings at this monitor, the TCEQ funded a series of in-depth studies.
The studies concluded that the high readings were confined to a small
area near the monitor, which is in close proximity to heavy truck
traffic at the port entrance, unpaved shipyards along the Houston Ship
Channel, and railroad tracks that run parallel to the road. To remedy
the situation, the TCEQ worked in cooperation with the PHA, the City of
Houston, Harris County, and local industry. Subsequent readings, in
2009, at the Clinton Drive monitor showed an annual average of 12.6 mg/
m3 of PM2.5 that translates to a design value of 14.1 mg/m3 for 2007
through 2009. On October 8, 2009, the EPA sent a letter to the governor
concerning violations of the annual PM2.5 standard at the Clinton Drive
monitor in Harris County for the design value years of 2006 through
2008. On February 4, 2010, the governor submitted to the EPA a
recommendation that Harris County remain designated as attainment for
the 1997 annual PM2.5 standard of 15 mg/m3. On April 29, 2010, the EPA
regional administrator signed a letter stating that he concurred with
the governor's recommendation that Harris County remain attainment for
PM2.5.
PM2.5 design values are presented in the figures below to show 10-
year trends. There are two PM2.5 standards, the 24-hour and an annual.
All areas in Texas meet both NAAQS and most areas have shown decreases
in both design values since 2002.

Currently, Texas has one nonattainment area for the PM2.5 NAAQS--El
Paso. In 2010, the El Paso area did not meet the PM2.5 NAAQS.
The El Paso area typically observes exceedances of the PM2.5
standard, but mostly from dust storms created by high winds, and, high
winds in El Paso typically qualify as an exceptional event. Most
likely, the Texas Commission on Environmental Quality (TCEQ) will be
petitioning the EPA for an exceptional event, thereby excusing the
exceedances of the PM2.5 standard. EPA may take up to three years to
consider TCEQ's request, which may delay any potential redesignation of
the El Paso area to attainment.
In the last 10 to 20 years, air quality in Texas has seen
significant improvement, especially with ozone. Specifically the one-
hour and eight-hour ozone design values for two of Texas' most
populated and heavily monitored areas--the Houston-Galveston-Brazoria
(HGB) and Dallas-Fort Worth (DFW) areas have decreased significantly.
In the early 1990s, the ozone design values in HGB were some of the
highest in the nation, 220 parts per billion (ppb) for one-hour and 120
for eight-hour. As of 2010, they were down to 126 ppb and 84 ppb,
respectively. In DFW, the one-hour design value was near 150 ppb in the
1990s and by 2010 the area had attained the standard; for the eight-
hour standard, the area's design value was as high as 105 ppb and in
2010 it was at 86 ppb. From 1990 through 2010, the population of each
area has grown by several million.

Improvements in air quality are not limited to the HGB and DFW
areas. Other areas of Texas have also reaped the benefits of improved
air quality, including Beaumont/Port Arthur, which today is attaining
the 1997 eight-hour ozone standard and was redesignated attainment by
the EPA in 2010. Other areas in Texas with regulatory ozone monitors
have also shown improvements. Austin, San Antonio, and Tyler/Longview
have implemented air quality strategies on a voluntary basis that are
part of the statewide improvements, and have allowed them to remain
attainment. The only areas that are designated in nonattainment of the
1997 eight-hour ozone standard are the HGB and DFW areas.
U.S. anthropogenic mercury emissions have decreased, while mercury
emissions from other parts of the world have increased. \5\ According
to EPA, the U.S. contribution to global anthropogenic mercury emissions
has declined from 10 percent in 1990 to five percent in 2005, due to
reductions in U.S. emissions and increases in emissions from other
countries. Specifically in the U.S., emissions of mercury for coal-
fired units above 25 megawatts (MW) decreased from 46 tons in 1990 to
an estimated 29 tons in 2010. Mercury is naturally present in coal in
trace amounts, which is then emitted when the coal is burned. From 2000
to 2010, mercury and mercury compound releases to the air in Texas
ranged from a minimum of 12,776 pounds in 2009 to a maximum of 16,639
pounds in 2000, according to the EPA's Toxic Release Inventory (TRI).
Just using the two endpoints of reporting year 2000 and reporting year
2010, the percent decrease for mercury and mercury compound emissions
is 11.3%. Over the last 11 years, the reported emissions have
fluctuated from year to year, but all are less than reported emissions
in 2000. TRI changed the reporting status of mercury and mercury
compounds beginning with reporting year 2000, lowering the threshold
reporting criteria for manufacturing and processing from 25,000 lbs to
10 lbs. Previous to reporting year 2000, only one facility reported air
emissions for mercury and mercury compounds. After the change in
reporting criteria for reporting year 2000, the number of facilities
reporting mercury or mercury compounds air emissions has been about 80
each year.
---------------------------------------------------------------------------
\5\ EPA. 2011. Draft National Emission Standards for Hazardous Air
Pollutants From Coal and Oil-Fired Electric Utility Steam Generating
Units and Standards of Performance for Fossil-Fuel-Fired Electric
Utility, Industrial-Commercial-Institutional, and Small Industrial-
Commercial-Institutional Steam Generating Units. May 3.
---------------------------------------------------------------------------
In Texas, protection of air quality predates the Federal Clean Air
Act, and state requirements are often more stringent than what is
required by the federal statute. States are given primary
responsibility for ensuring air quality protection under the Federal
Clean Air Act, with EPA's role primarily supervisory and secondary to
the role of the states. States, including Texas, are responsible for
developing state implementation plans (SIP) which contain the necessary
control strategies for ensuring that states attain and maintain the
NAAQS. SIPs must also contain major and minor permitting programs, and
provisions for public participation. These programs are developed and
managed by the states, with the exception of some states that rely on
EPA to manage their Prevention of Significant Deterioration (PSD)
permitting programs, the programs that permit major sources of air
pollutants. Texas has been delegated authority to manage its own PSD
permitting program from EPA (with the exception of greenhouse gas
permits), and permits both major and minor sources of air pollutants in
the state. In Texas, with the exception of certain activities that
produce de minimis amounts of air pollution, all stationary sources
that produce air contaminants must be permitted. Texas has also
developed a variety of robust rules to set limits on types of air
pollution, particularly in the state's nonattainment areas, to ensure
that those areas meet and attain the NAAQS by the applicable Federal
Clean Air Act deadlines. The following strategies have resulted in
significant reductions in nitrogen oxides (NOx) and volatile organic
compounds (VOC) over the last 10 years in Texas:

The NOx Mass Emission Cap and Trade (MECT) program
required an overall 80% reduction in NOx emissions from sources in the
program and applies to most point sources on the Houston-Galveston-
Brazoria (HGB) emissions inventory and even some minor sources that are
not included in the point source inventory (phased implementation from
2002 to 2008).

A comprehensive suite of rules adopted for the Dallas-
Fort Worth (DFW) 1997 eight-hour ozone nonattainment area requiring NOx
reductions from cement kilns, power plants, industrial boilers,
stationary engines used in the oil and gas industry, and many other
sources (March 2009 to March 2010).

Rules for enhanced monitoring and testing of flares,
cooling towers, and other sources in the HGB area with highly reactive
VOC (HRVOC) emissions known to cause rapid formation of ozone (January
2006).

Annual and short-term limits on HRVOC emissions for
sources in Harris County (January 2007).

More stringent requirements for VOC storage tanks in the
HGB area to address VOC emissions from roof landings on floating roof
storage tanks and from flash emissions on crude oil and condensate
tanks that were found using new technology, like gas-imaging cameras,
that allow the operators to observe plumes of VOC emissions that would
normally not be visible (January 2009).

While not required by federal regulations, Texas has also adopted
regional control strategies that required reductions from certain
sources in counties outside the nonattainment areas. These controls
help improve air quality in areas like DFW by reducing transport of
pollution from outside the area. Examples of such regional control
strategies include:

East and Central Texas Utility Rule (Senate Bill 7):
Required NOx and SO2 reductions from grandfathered power plants in East
Texas (2003 to 2005).

Regional Cement Kiln Rule: Required NOx reductions from
cement kilns in Bexar, Comal, Ellis, Hays, and McLennan counties (2003
to 2005).

East Texas Combustion Rule: Requires NOx reductions from
certain stationary gas-fired engines in 33 attainment counties east and
southeast of DFW, primarily in the oil and gas industry (March 2010).

In addition to rules that are required for implementation of the
NAAQS, Texas has also worked to develop innovative permitting
mechanisms to allow flexibility while requiring sources to control
their emissions. Texas has required all major sources of air pollution
that were uncontrolled under the Federal Clean Air Act, because of
grandfathered status, to obtain air quality permits that contain
federally enforceable emissions limitations. In this way, Texas has
gone beyond what is required by the federal statute to ensure that
emission sources in the state will have control requirements that can
be enforced to ensure protection of the state's air quality resources.
Because of innovative programs for point sources, Texas has seen
58% reduction to point source NOx emissions from 2000 through 2009. The
strides that Texas has made in reducing emissions and more importantly
ambient concentrations of ozone is more impressive considering the
population increase (meaning more cars and electricity needs) and
Texas' position as an economic engine of the entire country. Texas now
has the second largest population in the country behind California.
Between April 1, 2000, and July 1, 2009, Texas population increased by
more than 840,000 people, more than any other state, and its mobile
source emissions still decreased. The Federal Government has the
primary responsibility to regulate mobile sources. States have very
little ability to effect change in this area. The Texas Legislature,
however, chose to fund one of the most aggressive, if not the most
aggressive, programs to reduce NOx from mobile sources. The TCEQ has
provided over $900,000,000 in grants through its Texas Emissions
Reduction Plan program to provide financial incentives to upgrade or
replace older heavy-duty vehicles, non-road equipment, locomotives,
marine vessels, and stationary equipment to reduce NOx emissions in
eligible areas. Over $150,000,000 has been provided through the Drive A
Clean Machine program to repair gasoline vehicles that fail emission
tests and replace old vehicles with newer cleaner cars and trucks.
Texas also has requirements for cleaner-burning fuel that are more
stringent than federal fuel requirements in order to reduce NOx and VOC
emissions (Texas Low Emission Diesel and Low Reid Vapor Pressure
Gasoline programs).

Q3. Could you explain the significance of the four-year air quality
study conducted by Texas A&M University and Driscoll Children's
Hospital on the connection between hospital admissions and ambient
ozone levels?

A3. This study was conducted in the Corpus Christi area and provides
Texas-specific data on pediatric asthma patients compared with air
quality indicators (e.g., ozone). The results of this study are
consistent with results of numerous other studies in showing that
hospital emissions for respiratory symptoms such as asthma are not
directly related to ozone levels. As with numerous other studies, more
hospital admissions occurred on days with an ozone air quality in EPA's
good range than on days when air quality for ozone was in EPA's bad
range. Therefore, this suggests that air quality indicators like ozone
do not predict hospital admissions related to respiratory symptoms.
It is widely known that hospital admissions for asthma are much
higher in the winter than in the summer. As an example, see the graph
below the TCEQ developed for ozone concentrations versus asthma
hospital admissions for Dallas County, Texas, for 2005 to 2008.

As can be seen, when ozone levels increase, hospital admissions for
asthma decrease and when ozone levels decrease, hospital admissions for
asthma increase. This phenomenon occurs worldwide.

Q4. EPA recently released several ``technical adjustments'' to their
Cross-State Air Pollution Rule. In the view of the Texas Commission on
Environmental Quality, are these adjustments sufficient to alleviate
the economic and reliability impacts of this regulation in Texas?

A4. The TCEQ believes that Texas should not be included in the CSAPR
for PM2.5. Texas was not included in the rule for PM2.5 at proposal.
The TCEQ has technical concerns with the EPA claim that Texas is
contributing to the monitor in Granite City, Illinois, that we were not
allowed an opportunity to comment on.
The EPA did recently propose on October 6, 2011, revisions to the
CSAPR that would provide an additional 70,067 tons of SO2 allowances to
the Texas CSAPR budget and a delay until 2014 for the implementation of
the assurance provisions limiting interstate trading. Based on TCEQ's
initial review of the EPA's proposed revisions, while the proposal may
lessen some of the impact of the CSAPR on some Texas utilities it does
not address TCEQ's overall concerns regarding the feasibility of such
substantial reductions in sulfur dioxide (SO2) emissions in an
unprecedented short period of time. Even accounting for the additional
allowances proposed for Texas' budget, recent SO2 scrubber startups,
and announced SO2 scrubber startups for 2012, the TCEQ expects that
substantial SO2 reductions will still be needed in Texas for the 2012
control period. While the 2012 control period is an annual compliance,
companies must reduce their SO2 emissions early enough in the year to
avoid running out of allowance mid-year and being forced to shut down.
Companies must certify compliance with the CSAPR and there are
significant penalties associated with a company's actual SO2 emissions
exceeding the allowances held. Therefore, companies are unlikely to
gamble compliance on SO2 allowances becoming available at the end of
the 2012 control period. The EPA's intent for delaying the assurance
provisions until 2014 is to encourage trading in the initial two years
of the CSAPR program. However, Texas is still limited to trading with
Group Two states which still does not appear to be a viable trading
market for SO2 allowances sufficient to address Texas' concerns. In
effect, companies will only have a matter of months to achieve the
large reductions in SO2 emissions that the EPA is mandating with the
CSAPR which leaves some companies with limited options for compliance.
The TCEQ will continue reviewing the EPA proposed revisions to the
CSAPR and plans on submitting comments to the EPA on the proposal.
However, the TCEQ does not consider the CSAPR as finalized or the
proposed revisions to the rule to be cost-effective and certainly not
the $500 per ton claimed by the EPA.
Responses by Dr. Robert F. Phalen, Professor of Medicine, and Co-
Director,
Air Pollution Health Effects Laboratory, University of California,
Irvine

Questions submitted by Representative Andy Harris, Chairman,
Subcommittee on Energy and Environment

Q1. During the hearing, Dr. Thurston said that asthma was being
exacerbated by current levels of air pollution and that ``we are really
underestimating the benefits of clean air.'' In your view, what is the
cause of increased asthma rates over the last several decades?

A1. Although I am not an expert on asthma, the fact that asthma is on
the increase while air quality has significantly improved implies that
other factors are dominant. Such factors include: (1) obesity; (2)
overmedication of children's infections; (3) lack of immune challenge
to allergens during early childhood; (4) increased exposure (after
developing asthma) to indoor allergens, e.g., from insect infestations;
and (5) poverty.

Q2. Both you and Dr. Thurston mentioned the importance of particulate
matter constituents as opposed to mass.

Q2a. Please discuss the research that shows variability in particulate
matter and health effects associations.

Q2b. In light of the importance of particulate matter speciation, does
it make sense to develop regulations or to estimate regulatory health
benefits based upon particulate matter mass?

Q2c. Did the CASAC particulate matter panel on which you served
discuss whether it was appropriate to regulate PM on the basis of mass?

A2a. Numerous studies indicated that PM components, rather than PM
mass, are resonsible for the health effects. A good example is the
study by M.D. Bell et al.; Hospital admissions and chemical composition
of fine particulate air pollution (Am. J. Respir. Crit. Care Med.,
179(12):1115-1120, 2009) in which vanadium, elemental carbon, and
nickel were driving hsopital admissions in persons aged 65 years or
older. Also relevant is the study of 35,789 elderly Californians for
J.E. Enstrom (Inhal. Toxicol., 12(14): 803-816, 2005) that found the
small risk of fine particles found prior to 1982 vanished during the
period of 1983-2002. These study examples imply that PM mass is a poor
indicator of health effects.

A2b. I believe that it does not make sense to regulate fine
particulate mass for health purposes.

A2c. My CASAC-PM panel did discuss the use of PM mass for regulatory
purposes. The apparent reason for sticking with mass was the lack of
enough research to set individual component standards. I believe that
it was a mistake to continue and propose mass-based standards.

Q3. During the hearing, there was some discussion about the
independence and impartiality of CASAC. You served on the most recent
CASAC panel for particulate matter.

Q3a. What are the major strengths and weaknesses of the current CASAC
process?

Q3b. EPA often cites the public health benefits in its Regulatory
Impact Analyses to argue for more stringent standards. Does CASAC
review these analyses?

Q3c. What recommendations do you have for improving the CASAC process
and ensuring that panels are independent, transparent, balanced, and
impartial?

A3a. Please see my submitted testimony on seven weaknesses of the
CASAC-PM process.

A3b. The claimed benefits do not take into account offsetting adverse
consequences that also affect health, such as economic impacts of the
proposed new standards.

A3c. More CASAC representation by experts on the economy, industry,
agriculture, and overall public health would be helpful.

Q4. The Federal Advisory Committee Act requires that panels be
``fairly balanced in terms of points of view,'' and General Services
Administration regulations guiding FACA implementation (41 CFR 102-
3.60) require agencies to, in establishing advisory committees,
``ensure that, in the selection of members for the advisory committee,
the agency will consider a cross-section of those directly affected,
interested, and qualified, as appropriate to the nature and functions
of the advisory committee.'' In your view, is EPA CASAC membership
``fairly balanced in terms of points of view,'' and is it an
appropriate ``cross-section of those directly affected, interested, and
qualified'' represented on the Committee?

A4. My answer to 3c. above applies here.

Q5. Guidance from both OMB and the National Academies indicates that
peer reviewers should not review work products that they were involved
in. It appears, however, that 22 of the 25 members of the 2008 ozone
reconsideration CASAC panel were reviewing EPA documents that
specifically cited their work. Is this appropriate, and if not, what
recommendations do you have to better account for and avoid such
situations?

A5. The funding-related and other potential conflicts of interest of
CASAC-PM members appeared to influence their input, to the detriment of
the public good. I believe that there is adequate scientific expertise
in the U.S. to form a panel of scientists that did not perform the work
that is reviewed, or receive substantial support from regulatory
agencies that was not balanced by other sources.

Q6. How much do we know about the relationship between air pollutants
like particulate matter and human health effects? Are there areas that
need to be studied more?

A6. There are perhaps thousands of studies that relate to the effects
of inhaled air pollutants. The problem is not so much about a lack of
information, but the selection of information on the direct effects of
individual pollutants without due consideration of the totality of risk
factors faced by the public. Driving the levels of individual
pollutants down to very small concentrations can have a net harmful
effect on public health. The logical basis of current PM regulations is
weak; the pulic deserves better.

Q7. As you know, influential studies based on data from the American
Cancer Society and the Harvard Six Cities Study provide the basis for
major EPA regulations and determine how EPA develops its ``deaths
avoided'' estimates for particulate matter. These data sets were
developed with government funds, but are not publicly available so they
can be analyzed by other scientists. Do you support making this and
similar federally funded highly influential scientific data and
information transparent and publicly available?

A7. I do believe that publically funded research must be transparent
and available to the public. However, I'm not strongly supportive of
requiring such disclosure retroactively, except in cases where the data
are directly responsible for setting a standard. My concern is that
retroactive requirements can place an unnecessary burden on the
research community without additional funding (e.g., for personnel) for
this potentially large task.
Responses by Dr. Anne E. Smith,
Senior Vice President, Nera Economic Consulting

Questions submitted by Representative Andy Harris, Chairman,
Subcommittee on Energy and Environment

Q1. During the hearing, Dr. Thurston stated that the benefits
associated with particulate matter reductions ``keep going down well
below the standard to levels about seven micrograms per meter cubed.''

Q1a. What does the literature suggest about these effects?

A1a. The study Dr. Thurston refers to is based on a database first
established by the American Cancer Society (``ACS'') in 1982. At the
time that the individuals were recruited for the ACS study, they had to
be at least 30 years old. The average age of all the individuals at the
time of recruitment was 59 years. The study Dr. Thurston refers to has
tracked the survival outcomes of those individuals since 1982, thus
building up estimates of the average mortality risk at each age level
in dozens of cities across the U.S. Researchers have then assessed
whether a statistical correlation exists between the estimated average
mortality risk in each city and the cities' average ambient PM2.5
concentrations, after attempting to control for all the other major
factors that contribute to mortality risk. With this as background, I
will explain why Dr. Thurston's statement that the PM2.5-mortality risk
association has been observed to levels as low as seven micrograms per
cubic meter is not an appropriate indication of what the literature
suggests.
The estimates of differences in mortality risk across cities are
built up by following the survival outcomes of the people in each city
over many years. This means that the observations of their mortality
risks at each age, if attributable to air pollution at all, could be a
result of exposures they experienced many years in the past, or that
they accumulated over a long period of time. For example, all of the
individuals in the ACS database were exposed to U.S. pollution levels
since at least 1952 (i.e., 30 years before 1982), and the average
individual in the database experienced U.S. pollution levels dating
back to 1923. As researchers using the ACS database have stated, ``In
the 1950s, levels of air pollution in most North American and European
cities were 10 to 50 times higher than those found today.'' \1\ Since
the mortality risk estimated for each city is based on many years of
tracking these people, recent average PM2.5 concentrations such as
those in 2000, cannot be viewed as indicative of the PM2.5 exposure
level that most affected their observed survival outcomes. Those
individuals who had not already died by 2000 would have already lived
at least 48 years of their lives while being exposed to earlier, higher
PM2.5 levels. Yet, Dr. Thurston's statement that the PM2.5-mortality
relationship is observed down to concentrations as low as seven
micrograms per cubic meter is just a statement that the lowest annual
average PM2.5 concentration reported among the dozens of cities in the
ACS database was about seven micrograms per cubic meter in the most
recent update of the study, i.e., concentrations measured during 1999-
2000. \2\ It is misleading to imply that the estimated mortality-risk
relationship has been observed down to that level because PM2.5 levels
in that city were higher in the many earlier years of the study period
during which most of the deaths used to estimate each city's average
mortality risk occurred. That is comparable to assuming that recent
lower levels of PM2.5. accounted for the health outcomes of people who
died as much as several decades ago.
---------------------------------------------------------------------------
\1\ Krewski et al., Reanalysis of the Harvard Six Cities Study and
the American Cancer Society Study of Particulate Air Pollution and
Mortality, Special Report, Health Effects Institute, July 2000, p. 33
\2\ The lowest measured level of 7.5 micrograms per cubic meter
among the ACS cities was for the PM2.5 concentrations that were used in
Pope et al. ``Lung Cancer, Cardiopulmonary Mortality, and Long-term
Exposure to fine Particulate Air Pollution,'' JAMA, Vol. 287(9), March
2002, pp. 11332-1141.
---------------------------------------------------------------------------
While the lowest level in the cities in the most recent update of
the ACS cohort study was about 7.5 micrograms per cubic meter, the
lowest annual average PM2.5 level at the time that those individuals'
survival outcomes were first being tracked (i.e., about 1820) was 10
micrograms per cubic meter, and the levels across all the cities ranged
from 10 to 38 micrograms per cubic meter, with an average of 20
micrograms per cubic meter.
Another problem with Dr. Thurston's assertion that effects have
been observed down to about seven micrograms per cubic meter is that an
observed statistical trend based on a scatter of data across multiple
cities cannot be attributed to any single city within that dataset--it
is an average across all the cities. It is unsupportable to assert that
the association exists at the lowest measured level because the data
establishing that association are sparsest at the ends of the range.
This makes it impossible to assign statistical confidence to any
quantitative estimates of a concentration-response relationship at the
extreme ends of the range of data. The quantitative estimate of the
concentration-response relationship is an average slope based on many
data points (as many data points as there are cities in the analysis),
and the statistical reliability of that average slope estimate weakens
rapidly for concentrations at the far ends of the range of observations
of concentrations in the dataset.

Q1b. What levels have CASAC and EPA considered in setting ambient
standards for particulate matter?

A1b. EPA staff, with CASAC's concurrence, is only considering
tightening the annual PM2.5 standard to a level somewhere between 11
and 13 micrograms per cubic meter. This range does not extent lower
than 11 micrograms per cubic meter in large part because of the lack of
statistical confidence that the same magnitude of effect exists for
cities with PM2.5 concentrations at or near the lower bound of the
studies.

Q2. You outlined in your testimony the problems with EPA's Regulatory
Impact Analyses, which are used to generate health benefits like
premature deaths avoided by a regulation.

Q2a. Are these health estimates peer reviewed by EPA's Clean Air
Scientific Advisory Committee or any other body?

A2a. EPA's Regulatory Impact Analyses are not reviewed by EPA's Clean
Air Scientific Advisory Committee. They are not subject to public
comment and response either, because they are not required under the
Clean Air Act, but only by an Executive Order.

Q2b. Last month, President Obama made the decision to withdraw EPA`s
reconsidered ozone standard last month, citing feasibility concerns and
the need for the ``best available science.'' Prior to that decision,
EPA`s analysis claimed that the new standard would prevent 12,000
premature deaths a year. Was this analysis significantly different from
the analysis conducted for CSAPR or the Utility MACT? Did EPA rely on
particulate matter co-benefits in this analysis as well?

A2b. EPA's supplemental RIA for the ozone standard reconsideration
reported a range of estimates of avoided premature deaths for each of
five different alternative standard levels. These estimates included
premature mortality due to ozone and to PM2.5. The majority of the
avoided premature deaths were due to particulate matter co-benefits
rather than due to ozone. For example, for the 0.070 ppm alternative
ozone standard, EPA estimated that ozone-related mortality would be
reduced by 250 to 1,100 deaths per year, while the benefits of
coincidental reductions in PM2.5 were estimated to be 430 to 4,200
deaths per year. Both sets of mortality reduction estimates were
calculated in the same manner as for other RIAs such as CSAPR and
Utility MACT.
These coincidental PM2.5-related mortality risk reductions under a
regulation that is not designed nor intended to reduce PM2.5 are called
``PM2.5 co-benefits.'' In the case of the ozone RIA, the PM2.5 co-
benefits were based solely on estimates of reduced NOx emissions, which
EPA projected would cause ambient PM2.5 concentrations to decline too,
as a result of efforts to reduce NOx to attain the tighter ozone
standard. EPA is reporting PM2.5 co-benefits in nearly all of its air
regulation RIAs, including the Utility MACT, in which PM2.5 co-benefits
accounted for at least 99.99% of all the benefits of that rule. (In
that case, to co-benefits were due solely to reductions of SO2
emissions that would come from efforts to reduce acid gases required by
the Utility MACT.) Thus the entire cost-benefit case for the Utility
MACT rests on PM2.5 reductions, despite the fact that neither PM2.5 nor
its precursor, SO2, are air toxics that are the target and sole purpose
of the Utility MACT.
PM3-related benefits are also calculated in a comparable
manner in the CSAPR RIA. In the latter case, however, CSAPR is intended
specifically to reduce ambient PM2.5 (to assist in attainment of the
PM2.5 NAAQS) and thus in that RIA, those estimates are categorized as
direct benefits, not co-benefits.

Q3. President Obama recently said that ``I reject the argument that
says for the economy to grow, we have to roll back . . . rules that
keep our kids from being exposed to mercury.'' What percentage of the
benefits claimed in the Utility or Mercury MACT comes from mercury?

A3. Only 0.0004% to 0.011% of the benefits claimed in the Utility MACT
comes from any air toxic reduction, and all of that is due to just one
of the many air toxics in that rule, mercury. Stated in dollars, the
benefits estimated from mercury in the Utility MACT RIA are between
$0.5 million and $6 million per year, relative to the rule's estimated
annual cost of $10,900 million per year. These small benefits are
estimated even though the Utility MACT rule is projected to reduce
utility mercury emissions from 28.7 tons per year to 6.8 tons per year.

Q4. In calculating costs for Regulatory Impact Analyses, EPA
frequently cites single-year annual costs instead of net present value
of cost streams. In your view, is this the appropriate approach to
estimating regulatory compliance costs?

A4. Regulatory compliance costs and benefits should be considered on a
present value basis. EPA's practice of reporting the costs and benefits
for a single year can be misleading, especially if the baseline of
emissions is declining after the single year selected. For example,
PM2.5 and SO2, and NOx can all be expected to keep declining after 2015
even if the Utility MACT rule is not imposed because there are specific
standards already in effect that will take effect between now and 2020.
However, EPA reports its PM2.5 co-benefits only for 2016, at a point in
time where PM2.5 emissions should be on a steady decline through 2019
(which is the latest attainment date for the 2006 PM2.5 NAAQS). Thus,
there must be a declining trend in baseline risks, and hence PM2.5 co-
benefits should be much smaller soon after 2016, yet the annual costs
will not decline. Thus, choosing 2016 as the single year for reporting
the benefits and costs from the Utility MACT gives an overstated
impression of the size of the benefits relative to their costs. If a
single year is assessed, it should be selected as the year in which all
other existing regulations are fully implemented. At a minimum, EPA
should report the trend in annual benefits and costs in future years.
Nevertheless, a present value of costs and benefits would be a more
sensible way of addressing this problem, rather than to report costs
and benefits only for a single point in time.

Q5. Your testimony outlined EPA's reliance on coincidental particulate
matter co-benefits to justify a variety of Clean Air Act regulations.
Executive Order 12866 states that each ``agency shall avoid regulations
that are inconsistent, incompatible, or duplicative with its other
regulations.''

Q5a. Does this reliance on PM co-benefits meet this requirement to
avoid duplicative regulations?

A5a. The practice of relying on PM2.5 co-benefits in RIAs for rules
that are not addressing PM2.5 directly is inconsistent with the purpose
of RIAs. The purpose of RIAs is to provide policy makers and the public
with an understanding of which types of rules are best serving the
public interest. RIAs are useful only if they can help identify laws
and regulations that are not providing benefits in a degree that
warrants the extra regulatory complexity that they impose on our
society. The current reliance on PM2.5 co-benefits is inconsistent and
incompatible with this purpose because it masks the regulatory burden
of new regulations that have few or no direct benefits of their own. At
the same time, any PM co-benefits that are deemed credible would be
gained much more cost effectively by direct regulation of PM2.5 itself,
as already provided for by the Clean Air Act. Thus, reliance on co-
benefits from pollutants that already have regulatory frameworks in
place, such as PM2.5, allows RIAs to encourage growth in regulations
that are unnecessarily duplicative of the existing provisions under the
Clean Air Act to protect the public health in the most cost-effective
manner, which is a serious concern if those regulations cannot be
justified based on their own direct benefits.

Q5b. In your view, has EPA provided adequate evidence that it is not
counting coincidental and incremental PM2.5 reductions more than once
for these various regulations?

A5b. EPA states that it includes all existing regulations in its
baseline, but never provides sufficient evidence in its reports or
associated technical documentation for anyone to confirm that point. In
addition, double-counting is almost surely occurring when multiple
regulations are being analyzed simultaneously, rather than in a
sequence that actually would allow each RIA's baseline to account for
all other regulations that will be in effect by the time the additional
new rule is actually being implemented. Another problem in which
double-counting occurs relates to EPA's practice of reporting benefits
for a single year. As I explained in my response to Question 4 above,
if the year selected for reporting benefits of a new rule is earlier
than the year in which another existing rule would be fully
implemented, then the benefits reported for the new rule will
effectively be overstated because they will be temporary in nature, and
a few years later would be the benefits that were originally attributed
to the existing rule.

Q5c. What are the problems with taking PM benefits and spreading them
around to other regulations that have insignificant benefits of their
own?

A5c. The practice of taking PM benefits that should be attributed to
PM regulations themselves and spreading them around to other
regulations that have insignificant benefits of their own subverts the
entire purpose of RIAs, which is to help policy makers and the public
recognize excessive and non-productive regulatory requirements, and to
be informed enough to decide whether certain laws and regulations might
be better to reform than to continue. The practice of attributing PM
benefits to changes in PM exposures that are in attainment with the
PM2.5 NAAQS also undercuts the motivation of policy analysts and EPA to
grapple with whether those estimates are credible enough to tighten the
PM2.5 NAAQS directly. If they are that credible, then the only cost-
effective way to address those risks is via their direct regulation.
Thus the practice of spreading PM benefits around to other regulations
also leads to cost-ineffective methods of dealing with the main air
quality risks that our public health may be facing.

Questions submitted by Representative Judy Biggert

Q1. On September 21st, NERA Economic Consulting released a new
economic analysis on four major EPA rules affecting electric generating
units: the Utility MACT; the Cross-State Air Pollution Rule; rules on
cooling water intake structures, and the classification of coal
combustion residuals as hazardous. What did this analysis find about
the employment and electricity price impacts of these rules?

A1. NERA's analysis finds that these regulations will decrease
employment and increase electricity rates. On average over the period
2012-2020, when these four regulations are being implemented, NERA
estimates 183,000 fewer jobs in each year. On average over the period
2012-2020, NERA estimates that electricity rates would be 6.5% higher
on a national average basis. This electricity rate impact varies
significantly by region of the U.S., with a range from 0% to 14%. The
largest rate impacts are in regions that rely more extensively on coal-
fired generation. A copy of the full NERA report on this analysis can
be downloaded at http://www.americaspower.org/sites/default/files/
NERA-Four-Rule-Report-Sept-
21.pdf.

Q2. A recent analysis conducted by EPA found that the Clean Air Act
results in $2 trillion of economic benefits. Do you agree with this
analysis? How did EPA arrive at this figure, and what was the role of
``willingness to pay'' surveys that estimate the value of a
``statistical life?''

A2. The EPA analysis mentioned is summarized in an EPA report
sometimes referred to as the Second Prospective Benefit-Cost Analysis
of the Clean Air Act. This analysis attempted to estimate the benefits
that will be derived by 2020 as a result of the 1990 Clean Air Act
Amendments (``CAAA''). About 90% of the $2 trillion estimate is due to
projected reductions in mortality risk from reductions in ambient PM2.5
in 2020, relative to what EPA projected ambient PM2.5 concentrations
would be in 2020 but for the 1990 CAAA. The $2 trillion is not an
estimate of financial benefits, such as would appear in measures of GDP
or average household income. Rather, this estimate is intended to
reflect the improved ``sense of well-being'' of the U.S. population as
a result of EPA's projected large improvement in mortality risk, also
known as willingness to pay (``WTP'').
The WTP measure that EPA used to value the mortality risk
reductions is called the value of statistical life (``VSL''). It is not
the value assigned to an individual life. Rather, it is a summary
statement of the WTP for a very small change in risk of dying, such as
a change in risk of dying in a given year of one in 10,000. If that WTP
is found to be an average of $800 across the entire affected
population, then the VSL would be $8 million, which simply means that
for every 10,000 people benefitting from this amount of risk reduction,
there would be one less life lost per year, and yet the aggregate WTP
among those 10,000 affected people would be 10,000 times $800, or $8
million. Hence, the WTP for this risk reduction would be $8 million per
expected life saved, or per ``statistical'' life saved. One difficulty
with the estimate of VSL is obtaining a sound estimate of what people
are actually willing to pay for those small risk reductions. EPA's VSL
estimate is based on a mixture of two types of evidence: (1) estimates
based on wage data of what workers are willing to give up in annual pay
to work in jobs have lower on-the-job death risks and (2) direct
questionnaires that present a certain hypothetical risk reduction
opportunity to survey respondents and ask them to state what they would
be willing to pay to have that risk reduction. Both of these methods
present significant methodological difficulties. The VSL that EPA has
used in this study is based on a review of 21 wage-risk studies and
five WTP survey studies. All but one of the survey studies asked about
WTP for changes in job and other accidental risks, rather than about
disease risk.
I do not agree with the $2 trillion estimate for several reasons,
which are explained in more detail in a paper I co-authored with W.
David Montgomery in June 2011, which can be downloaded from the website
of the National Taxpayers Union at: http://www.ntu.org/news-and-issues/
energy-environment/macro-vs-wtp-v19-
pdf3.pdf. Three of the primary reasons I consider the $2 trillion
estimate to be far overstated are:

(1) The size of the risk reduction being valued in the
EPA analysis is far larger than the 1-in-10,000 to 1-in-100,000 risk
reductions in the WTP studies from which EPA's VSL assumption is
derived. These values for very small risk changes cannot be simply
linearly increased to represent values for much larger risk changes
such as 1-in-1,000 to 1-in-100, but the latter are the levels of annual
risk change that EPA is attributing to the PM2.5 reductions from the
CAAA. Due to budget constraints, a valid WTP for these much larger risk
changes, if it were to be measured, would likely be much smaller.

(2) I believe that the mortality risk change that EPA has
assumed for each unit of change in average annual PM2.5 concentrations
is overstated to the point of non-credibility. EPA is assuming a range
on PM2.5 concentration-response relationships that implies that there
is a 25% probability that PM2.5 caused more than 25% of all deaths
nationwide in the U.S. when ambient concentrations were like those that
existed in the period around 1980.

(3) EPA's baseline of PM2.5 but for the 1990 CAAA is far
too high. EPA assumes that no PM2.5 NAAQS standard would have been
imposed if the 1990 CAAA had not been enacted, which is not a credible
assumption. All of the authority necessary to issue the PM2.5 NAAQS and
its associated emissions regulations was provided in the Clean Air Act
of 1977. In fact, the 1990 CAAA did not make any changes to Section 109
of the 1977 Clean Air Act, which is what established the NAAQS process
that exists today. Thus, the PM2.5 baseline in EPA's analysis is
grossly overstated, and thus the $2 trillion of benefits due to
reductions in PM2.5 ``due to the 1999 CAAA'' is also grossly
overstated.

Q3. If EPA had not incorporated coincidental particulate matter co-
benefits (including health benefits associated with reductions below
the National Ambient Air Quality Standard), how many EPA Clean Air Act
regulations would have passed a simple cost-benefit test in the last
two years?

A3. I interpret ``pass a simple cost-benefit test'' to mean that the
estimate of annual benefits exceeds the estimate of annual costs in the
year that EPA analyzed in its RIA. The concept of ``particulate matter
co-benefits'' only applies to rules that do not directly aim to reduce
ambient PM2.5. I have identified 13 RIAs released in the last two years
for emission-reducing rules under the Clean Air Act that do not
directly aim to reduce ambient PM2.5.

Of these 13, two were to reduce greenhouse gases. Both of
the greenhouse gas rules pass the cost-benefit test without accounting
for PM2.5 co-benefits, but only because EPA has estimated that these
rules will have negative costs, and so they would pass the cost-benefit
test even with zero direct benefits.

The 11 remaining RIAs are for regulations of air toxics
(``NESHAP''), ambient air quality standards (``NAAQS'') and new source
performance standards (``NSPS''). Of these 11, not one passes the
simple cost-benefit test without incorporating coincidental PM2.5 co-
benefits, yet all but two do pass the cost-benefit test based on their
PM2.5 co-benefits. (One exception is an RIA for a recently proposed
rule for NSPS and NESHAP for the oil and natural gas industry. That RIA
discusses PM2.5 co-benefits qualitatively, but does not provide any
quantitative estimates; if EPA quantifies its co-benefits in a future
draft of the RIA, that rule may also pass a cost-benefit test based on
PM2.5 co-benefits. The other exception is a NESHAP for area sources
from industrial boilers. In that case the cost lies just a bit higher
than the high end of the estimated range of PM2.5 co-benefits.)

Responses by Mr. J. Edward Cichanowicz, Consultant

Questions submitted by Representative Andy Harris, Chairman,
Subcommittee on Energy and Environment

Q1. Could you provide the Committee with your estimate of the number
of electric generating units (EGUs) that will be regulated by the
Cross-State Air Pollution Rule and the Utility MACT? Could you also
provide the number or percentage of these units that will require
additional air pollution control systems to comply, and the number or
percentage of those EGUs that may not be able to acquire and install
these systems in time to comply with the rules?

A1. The Cross-State Air Pollution Rule (CSAPR) will require
approximately 60 generating units to retrofit control technology.
Approximately half of these units will not be able to install the
required equipment by the CSAPR mandate. The Utility MACT, based on EPA
estimates, will require 533 generating units to retrofit fabric filter
control equipment, with 90 units also requiring ``semi-dry'' scrubbers.
Significantly less than half of these units will be able to comply with
the Utility MACT mandate.
In summary, about 600 units--approximately half of the 1,200
generating units in the U.S--will be required to retrofit control
technology for both the CSAPR and the Utility MACT. Less than half of
these 600 units will be able to meet the respective CSAPR or Utility
MACT compliance mandate.

Q2. Short of ceasing operations, are there any options available to
these non-compliant EGUs under the Clean Air Act that can guarantee
these units can come into compliance without facing an enforcement by
EPA or citizen suits?

A2. The only option--other than ceasing operation--for non-compliant
units is to switch the source fuel to natural gas, if adequate supplies
and delivery pipeline are available. The cost for natural gas (in terms
of cost per million Btu of energy content) significantly exceeds the
cost of solid fuel (coal), which will proportionally increase
generating costs. Further, significant capital investment could be
required to provide for natural gas access.
The switch to natural gas is not perceived to be a realistic option
on a national basis.

Q3. In your opinion, what about EPA's assumptions for installation
timelines of pollution control equipment is incorrect? If EPA were to
fix the timeline problem, would you agree with the other technology
assumptions they make?

A3. As noted in my written testimony, EPA's assumptions about the time
required to retrofit the key control technologies--24 and 27 months
depending on the control device--are too optimistic (e.g., presume a
shorter schedule than realistic). Recent experience shows in most cases
these control technologies will require 40-50 months. As a result, it
will not be possible for the complete national inventory of generating
units to comply with the CSAPR.
Some of EPA's key technology assumptions are correct while others
are not. Assumptions about the control capability of the selective
catalytic reduction (SCR) ``catalytic reactors'' for NOx, and the flue
gas desulfurization (FGD) ``scrubbers'' for SO2 are generally accurate.
In contrast, EPA assumptions about the control capability of dry
sorbent injection (DSI) to remove hydrogen chloride (HCI) are
optimistic. These overly optimistic assumptions address the (a) degree
of HCI removal achievable, (b) availability of the highly specialized
sorbent, (c) impact on power generation equipment, and (d) other
environmental impacts such as solid byproduct management.

Q4. How many power plants are currently utilizing dry sorbent
injection (DSI) to meet the proposed Utility MACT requirements? How
does their real world capture rate compare with EPA's proposed
standard?

A4. Only two of the total of 28 generating units reported by EPA to
apply dry sorbent injection utilize this approach at conditions that
reflect the Utility MACT. Even these two units--which utilize western
coal with low chlorine and high inherent alkalinity--do not reflect the
conditions that will be experienced by the national inventory of units
that must comply with the Utility MACT.

Q5. According to your testimony, fast-tracking pollution control in
the manner advocated by EPA could compromise design quality and
equipment construction equipment. Should we also be concerned that
EPA's compliance deadlines could compromise worker safety?

A5. I have spent considerable time at plants during construction
activities and heavy equipment maintenance. Worker safety should be
considered in developing plans to fast-track installation, particularly
conducting field work that requires relocating or installing material
with cranes.

Q6. EPA has cited instances where pollution controls have been
installed in less than the 40-50 month timeline cited in your
testimony. Is EPA correct to use those examples as representative of
the entire electricity industry? Is there anything about those examples
that make them significantly different than normal experience?

A6. EPA is not correct in citing the examples noted for two reasons.
First, the time required to complete a project cited by EPA does not
include the total start-to-finish scope, but reflects only a portion of
the work. Specifically, almost without exception, EPA estimates
``start'' time as when contracts for final engineering are awarded. EPA
ignores the permitting and preliminary engineering steps that require
significant time. Also, some of the installations cited by EPA reflect
less challenging applications than typically encountered.

Q7. According to a report by the Clean Energy Group (CEG), a
consortium of energy companies that stands to profit from EPA's
regulations, the amount of pollution controls required to satisfy CSAPR
is less than the number of controls that industry installed between
2008 and 2010, proving industry can meet EPA requirements. Is the CEG
correct, or are there problems with its analysis?

A7. Although it is true the number of controls required for the CSAPR
is less than those retrofit between 2008 and 2010, the CEG ignores the
difference in start time. In general, much of the work to install the
control technologies that were completed in 2008 and 2010, as
referenced by the CEG, was started in 2005. Many projects started prior
to 2005. The key difference between the start times and deployment
dates for these projects is due to the specifics of the Clean Air
Interstate Rule (CAIR), the relevant mandate for projects installed
during the 2008 to 2010 time period. Specifically, compared to the
CSAPR, the CAIR offered a greater time period between when the rule was
finalized and when units had to be in compliance. Further, the CAIR
contained provisions that offered a financial incentive for owners of
generating units to deploy control technology early. In contrast, the
CSAPR does not contain such provisions; in fact, the usual
disincentives exist that penalize owners for early work for the CSAPR
if the rule is not issued in final form as proposed.

Questions submitted by Representative Judy Biggert

Q1. The Environmental Protection Agency has claimed that they can
provide a one-year extension for the 2015 Utility MACT compliance date.
In your testimony, you outline the 10 steps needed to add pollution
control equipment to an existing power plant. In your view, would a
one-year extension provided on a case-by-case basis be sufficient to
allow utilities to go through all 10 steps?

A1. No. The one-year extension would help alleviate, but not
eliminate, the significant delay. Analysis that I have conducted for
the Utility MACT (described in a reference submitted with my testimony)
shows the extension of one year would enable only about 50% of the
projects to be completed. \1\ This same analysis shows that a two-year
extension is required for about 95% of projects to be completed.
---------------------------------------------------------------------------
\1\ See Figure 1-2 of ``Feasibility of Retrofitting Fabric Filter
Particulate Matter Control Technology to the Electric Generating Unit
Inventory as Projected by EPA,'' July 2011.
Appendix 2

----------

Additional Material for the Record

Additional Material for the Record
Reprint of article by Dr. Roger O. McClellan, Advisor, Toxicology and
Human Health Risk Analysis: ``Role of Science and Judgment in Setting
National Ambient Air Quality Standards: How Low Is Low Enough?'', Air
Quality and Atmospheric Health (published online 01 June 2011).

Letter from Dr. Roger O. McClellan, DVM, MMS, Dsc (Honorary) to
Honorable Lisa P. Jackson, Administrator, U.S. Environmental Protection
Agency, May 6, 2011.

Memo from Honorable Cass R. Sunstein, Administrator, Office of
Management and Budget, Office of Information and Regulatory Affairs,
September 2, 2011.