 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
Accomplishments of the Clean Air Act, as amended by the Clean Air Act Amendments of 1990
Subcommittee on Energy and Air Quality
Dr. Bernard Goldstein
SUMMARY The 1990 amendments to
Section 112 of the Clean Air Act are an example of the Precautionary Principle,
of which this Committee will hear much more, being used in legislation.
The precautionary principle is evident in: 1) The amendments came about
largely because of frustration with the slowness of the previous risk based
scientific approach. 2) The burden
of proof was shifted away from the requirement that EPA find that an agent was
harmful at ambient levels, and shifted to the requirement that 189 compounds
should be controlled unless proven harmless, and 3) maximum available control
technology (MACT) was required, relegating the previous selective risk and
science based approach to secondary importance. The claim that the new approach
would be faster and cheaper is difficult in retrospect to support in view of the
delays and the cost in establishing the regulations.
But has it worked in terms of providing improved public health?
The good news is that we can certainly expect a decrease in total tonnage
of chemicals released into air. However,
inherent in the precautionary approach is that we really do not know to what
extent we have had an impact on public health.
By requiring MACT, we probably will reduce the emissions of known human
cancer causing chemicals, such as benzene, but these were already regulated
under the previous risk based approach. For
almost all of the newly regulated chemicals there is really no evidence that
they produce harm at ambient levels, and in some cases, such as toluene, there
is sufficient data to suggest that there is no reason for concern.
Certainly, reducing exposure to these compounds can be justified on
prudent public health grounds, but by definition one can not do a cost benefit
analysis when there is no evidence on which to claim benefit.
Requiring that EPA perform a cost benefit analysis for HAPs, in contrast
to NAAQS pollutants for which benefit can be documented, is contradictory to the
regulation of HAPs on a precautionary basis and should be abandoned.
Other problems related to the
precautionary approach to HAPs include the lack of an incentive to improve
control technology once MACT is established; the potential inefficiencies in
treating all chemicals alike, e.g., requiring the same level of control
technology for benzene as for toluene; and, perhaps of greatest long term public
health importance, the disincentive to invest in research to find out the truth
about chemicals we are regulating. To
accomplish our public health goals, we very definitely need exposure and effect
indicators that are far superior to simply measuring pounds of pollutants.
I urge additional support to EPA's ORD, to the NIEHS and to the CDC
Center for Environmental Health to develop these indicators. Finally, the 1990 CAA HAP amendments raise two public health issues unrelated to the precautionary principle. On a public health basis it is hard to justify the emphasis on outdoor air when the highest human exposure levels to most of these compounds occurs indoors. Secondly, the residual risk provision of the CAA are based solely on risk to the Maximally Exposed Individual (MEI) rather than incorporating standard population based approaches. That this is nonsensical is evident, among other reasons, by the calculation that there are circumstances in which we could be protecting an isolated family of four against a risk that would occur once every 17.5 million years, or about ten times longer than humans have been on the planet. Introduction Mr.
Chairman, Members of the Committee. Thank
you for inviting me to give a public health viewpoint on the impact of the
hazardous air pollutant provisions of the 1990 Clean Air Act Amendments.
With your permission, I would like to submit written remarks for the
record and to summarize them in my oral testimony.
My
name is Bernard Goldstein. I am a
physician and an environmental health scientist, and am currently Dean of the
University of Pittsburgh Graduate School of Public Health, one of the nation's
largest schools of public health. I
have had more than thirty years of experience in studying and commenting on the
health effects of air pollutants, including serving as Assistant Administrator
for Research and Development of the US Environmental Protection Agency under
William Ruckelshaus and Lee Thomas, and chairing the Congressionally mandated
Clean Air Scientific Advisory Committee.
Issues
related to the Precautionary Principle In
discussing the public health impact, I believe it helpful to frame the 1990 HAP
amendments in terms of the Precautionary Principle. This is a relatively new term, embodying an evolving and as
yet not well-defined set of concepts that is increasing in prominence among
environmental and public health advocates.
One definition, provided in the 1989 Rio Declaration, is: "Nations
shall use the precautionary approach to protect the environment.
Where there are threats of serious or irreversible damage, scientific
uncertainty shall not be used to postpone cost-effective measures to prevent
environmental degradation" The
more recent formulations have tended to extend the precautionary principle to
public health. In some cases there
has been a weakening of the emphasis on cost effectiveness and on the extent of
adverse impact needed to invoke the precautionary principle.
You
will hear much more about the Precautionary Principle, not the least because it
is being heavily advocated by the European Community, perhaps as a rationale for
trade barriers on US products. Although
not discussed as such at the time, the 1990 amendments to Section 112 of the
Clean Air Act governing the control of hazardous air pollutants contain a
classic use of the precautionary principle.
First, the amendments are derived from a sense of frustration with the
slowness of a risk based scientific approach - relatively few of HAPs had been
regulated. Second, the burden of
proof is shifted as is evident from Congress listing 189 pollutants and limiting
EPA's role to one of removal of a pollutant from the list based upon proof of
no harm, which replaced the previous dependence upon a finding of harm to be
listed for regulation. Third,
there is a requirement of maximum available control technology for pollutant
control of all sources, relegating the previous selective risk-based approach to
secondary importance. But
has it worked? In 1990 advocates of
this new approach claimed it would be "faster, better, cheaper".
It is hard to argue that it has been faster or cheaper, given how long it
has taken to write the regulations and how much it has cost to do so, as well as
the toll that uncertainty always has on the market place.
However, the key question is whether the 1990 CAA HAP Amendments have
improved public health. Let
me start with the positive. We do
know that many tons of HAPs have been or will be removed from the air.
However, we know little about how much of a health difference this has
made. One of the guiding
concerns in the control of HAPs is that of human cancer.
Almost all of the pollutants that were known or reasonably anticipated to
cause cancer or to have other adverse effects at ambient air levels had been
regulated before the 1990 CAA amendments. These
amendments have led to more stringent contol on at least some of these
pollutants through requiring maximum available control technology.
To the extent that these controls can be quantified (e.g., reduction in
the emissions of benzene, a known cause of human leukemia), some estimate can be
made of the additional benefit of the MACT provision.
However, this will not be possible for almost all of the newly regulated
pollutants on this list. The
reason there is so much uncertainty about health benefit for almost all of the
pollutants listed by name as HAPs in the 1990 CAA Amendments is inherent in the
precautionary principle. Simply
put, you can not know what harm has been averted if you regulate pollutants
without some degree of proof that they are harmful.
As a corollary, the congressional requirement in the Clean Air Act that
EPA perform cost-benefit analysis, which can be done for NAAQS pollutants, can
not be achieved for those HAPs for which there is no evidence of benefit,
despite the substantial cost in controlling these pollutants.
And almost all of the HAPs for which there is sufficient evidence of harm
to provide the basis for a benefit analysis were regulated under the pre-1990
Section 112 rules. If
we really want to know the public health benefits of regulating this broad list
of agents, we must develop benefit indicators that go beyond the simple measure
of tons of pollutants. Using
pollutant weight, as does our HAP regulations, rather than pollutant effect, can
be problematic, particularly where there is a clear differential among these
pollutants in their potential for toxicity.
Removing a ton of toluene from air emissions is probably meaningless in
terms of public health, while removing a ton of benzene is very likely to be of
direct health benefit. The current
approach to HAPs focuses the same attention on both. Better indicators of the potential for adverse health effects
are needed if we are to develop cost-effective approaches to hazardous air
pollutants, particularly as for many of these pollutants the outdoor exposures
regulated by the CAA are relatively trivial compared to indoor exposures. Another
area of concern about the precautionary principle is often is antithetical to
scientific research that gets to a true understanding of cause and effect
relationships. The central
principles of toxicology are that chemicals have very specific actions within
the human body and that they vary greatly in the dose that causes the action.
They are the basis for two of the components of risk assessment - hazard
identification and dose response assessment.
These principles are no more complicated than saying that aspirin works
for a headache but not for constipation, and that a very tiny grain of aspirin
will have no effect while too much can kill.
By treating all of the 189 chemicals on the list as exactly the same in
terms of specificity of action and dose responsiveness, the CAA Amendments
simply ignore these toxicological principles.
There is no question that avoidance of any possible effect of any of
these agents could be preventive - there is also no question that this shotgun
approach inherent in the precautionary principle is less than a fully efficient
means of dealing with the potential public health consequences of HAPs There
is work under way attempting to develop better indicators of air pollutant
health effects. In fact, EPA
Administrator Whitman provided major leadership in this area while governor of
New Jersey and now at EPA. I wish
to particularly commend the research activities supported by various government
agencies including the National Institute of Environmental Health Sciences, the
CDC Center for Environmental Health and by the EPA's Office of Research and
Development, and to encourage Congress to enhance support for these approaches. There
is another problem inherent in the precautionary principle approach to HAPs.
This is a paradox built into the MACT approach.
The goal is to achieve the lowest possible levels of pollution control by
specifying the best available control technology today (actually defined as the
"best performing 12% of existing sources").
But once having done so, once having spent better than a decade writing
the regulations guiding the use of the specific technology, how do you get
better control technology? Pollution
control technology is an applied field, usually borrowing from advances in basic
technology that are then utilized to meet the demand for pollution control.
But what demand will be left after the regulations are completed?
Thus a potential negative of the MACT approach is that it leaves little
likelihood that there will be continued improvement in control technology.
One
other aspect of the potential public health impact of a precautionary principle
approach is also difficult to quantify. This
is the extent to which the misplaced emphasis on unnecessary public health
actions limit the availability of resources for needed public health activities.
Our national public health infrastructure is under tremendous pressure.
We have taken for granted many of the advances in public health.
These advances have been sustained by a highly efficient workforce that
had been stretched to the limit even before September 11th.
This workforce needs reinforcement, it needs to have a greater level of
support for its activities and our nation needs more focus on how we will
replace those already in the field. In
making your judgments as to where to place needed public health resources, I
urge you to give high priority to the workforce infrastructure. Issues
not directly related to the Precautionary Principle There
are two issues concerning the HAP provisions of the 1990 CAA Amendments that are
directly related to public health but do not clearly fall under the heading of
the Precautionary Principle: the relative lack of emphasis on the major public
health threat of these chemicals, that of indoor air pollution; and in
calculating residual risk, the inappropriate sole focus on the Maximally Exposed
Individual rather than also on the total population. Public
health principles for hazardous chemicals require us to focus on the highest
levels of exposure to the most toxic agents.
In general, for HAPs this is exposure in the home to HAP compounds that
were already regulated before 1990. Thus,
there is a disconnect between the major public health concern about these
chemicals and the emphasis that the CAA puts on their control. Risk
assessment remains a part of the regulation of HAPs in the form of a residual
risk estimation. Unfortunately, the
risk assessment approach specified is not in keeping with public health practice
in that the risk estimate is driven by the Maximally Exposed Individual rather
than the population at risk. This
is both an inappropriate and an inefficient way to protect public health.
Population based approaches should be the primary driver in risk based
approaches with risk to the Maximally Exposed Individual also being calculated
to be sure that no one individual is particularly at risk.
Just one of the many problems in using the MEI as opposed to the
population based approach is that we in essence assume that the MEI lives at the
fence line of the source 24 hours a day, 7 days a week, for 70 years.
This simplifying approach is true if we assume that someone is at that
location as part of a population based risk estimate, but it is a gross and
highly variable overestimate to assume that a single individual gets that full
70 years. Moreover,
one can readily demonstrate the silliness of the reliance on the MEI with a
paper experiment. Assume that there
is a plant at the edge of a rural area such that only one family of four lives
immediately downwind and that it has a life time cancer risk of one in one
million due to these emissions. Assume
further that this plant goes on emitting the same level of pollutants ad
infinitum, and this family of four is replaced every 70 years by another family
of four. The time period during
which one cancer case is expected to occur is 17.5 million years, or roughly one
case during a period about ten times longer than humans have existed on this
planet. This needs to be changed. So
as to be sure that my critique of the public health impact of the HAP provisions
is not misunderstood, let me emphasize that the inability to prove a benefit is
a common issue in public health. We
all agree that prevention is valuable, and that sometimes we have to fall back
on the well-known 16:1 cost benefit ratio for primary prevention.
This is based on an ounce of prevention is worth a pound of cure.
We have learned in public health that to extract that sixteen-fold
benefit requires highly efficient approaches focused on the major threats using
the best available science. In that
regard, the 1990 HAP amendments have unfortunate shortcomings.
|
|||||||||||
|
