EM It w to Fix And Ho

Forum invites authors to share their opinions on environmental issues with EM readers. Opinions expressed in
Forum are those of the author(s), and do not reflect official A&WMA policy. EM encourages your participation
by either responding directly to this Forum or addressing another issue of interest to you.
How An Environmental
Law Encourages Pollutiont
And H
by Susan Helms and Jennifer Sullivan
May 2001
A few little-known stipulations of
the Clean Air Act may actually
encourage some companies to
emit more pollution than they
normally would. According to the
authors of this article, however,
an alternative to these “perverse”
incentives can be found in a
creative EPA experiment called
Project XL, which could turn out
to be the commonsense blueprint
for air permits of the future.
The first Clean Air Act (CAA) with guts was enacted in 1970 to
clamp down on pollution by setting emissions limits for many
categories of new stationary sources (unlike the Clean Water
Act, which requires pollution-limiting permits from all new
and existing sources and requires them to be updated regularly, the CAA only requires pollution-limiting permits for new
or modified sources, and once permits are issued, they are
never revisited), setting health-based ambient levels of criteria
pollutants, establishing national emission standards for some
existing sources of hazardous air pollutants, and requiring
vehicle emissions reductions of 90% by the 1976 model year.
The 1977 amendments subsequently imposed additional
new source standards, such that facilities built or modified in
an area in nonattainment with the National Ambient Air
Quality Standards (NAAQS) had to meet stricter standards
than those built or modified in areas that were in attainment.
Specifically, facilities in attainment areas were subject to the
Prevention of Significant Deterioration (PSD) program designed
to keep such areas in attainment by installing Best Available
Control Technologies (BACT), while those states with areas in
nonattainment had to impose stricter limitations on new or
modified sources (Lowest Achievable Emission Rates, or LAER)
and use tools such as Reasonably Available Control Technology (RACT) on existing sources to achieve attainment.
In 1990, new amendments created the national operating
permit program, known as Title V, which requires all “major
sources,” new and existing, to pay fees based on the size of
their emissions and to undergo periodic monitoring. (Some
consider this permit toothless in that it does not limit emissions; limits are imposed by the still-applicable New Source
Review [NSR] program.)
During its 30 years of existence, the CAA has worked wonders. Note the dramatic effect it had on emissions of criteria
air pollutants (with the exception of nitrogen oxides) between
1970 and 1990 as illustrated in Figure 1.1,2 Despite such
progress, however, certain preferential conditions built into
the CAA create incentives for some companies to emit even
more than they would without the law. How did this come
about? And how can the law be fixed so that it does not
actively encourage pollution?
There are three “perverse” incentives in the CAA that result in
unnecessary emissions (see Table 1). These incentives have been
reported to us both by people working for regulatory agencies
who observe them firsthand and by people working in industry who have taken advantage of them. The extent to which
these loopholes are exploited is not known for the obvious
reason that such activities happen below the U.S. Environmental Protection Agency’s (EPA) radar. However, the fact
that they exist at all raises the question of how proactive EPA is
about pollution prevention. Pollution prevention should be
EPA’s rallying cry, not an idea undermined by archaic policies.
What are the three incentives? First, because all existing
equipment was grandfathered or exempt under the original
Clean Air Act, firms today tend to keep such older equipment
running rather than replacing it with new, nonexempt
Index value (1970=100)
NOx = nitrogen oxides; SOx = sulfur oxides; VOCs = volatile organic compounds; CO = carbon monoxide; TSP = total suspended particulates
Figure 1. Trends in U.S. air emissions of five criteria pollutants following enactment of the CAA.
May 2001
equipment. The advantages of purchasing more efficient and
more cost-effective equipment are offset by the costs and delays associated with needing to apply for a new permit, which
then has the added disadvantage of bringing that source
under the regulatory umbrella for new and modified sources
(i.e., the NSR program). Why does this situation persist? Some
of the most polluting sources are coal-burning power plants
and small coal-fired boilers. We know that EPA would prefer
to have fewer coal-burning boilers around. Nationally, emissions from coal-burning power plants comprise 32% of all
carbon dioxide (CO2), 67% of all sulfur dioxide (SO2), and
30% of all nitrogen oxides (NOx) emissions.3,4 Coal boilers
operated by nonpower-producing plants create the same
emissions, though on a smaller scale.
The power industry prefers, however, to keep coal relatively unregulated. The industry’s efforts 30 years ago, when
enormous amounts had been invested in coal-burning power
plants, created the grandfather clause, allowing existing plants
to be exempt from many clean air laws applicable to new and
modified sources. It was asserted that the shift to cleaner production would be inordinately costly to the power industry
and that the technology would change over time without
legal mandates as equipment became more efficient. But
energy-efficient technology has not been adopted by many
coal burners,5 and the fact that other sources of pre-1970
vintage, including smaller coal boilers, do not trigger NSR
standards either (unless substantially modified), keeps them
immune to cleaner technology requirements.
A second loophole involves companies that have been
known to operate existing equipment so that emissions stay
close to their existing allowable pollution limits even if they
technically could lower their emissions through pollution
prevention or idling unnecessary equipment. The reasoning
here is that in areas of nonattainment, Significant Net Emission Increase standards kick in when the difference between
actual and projected emissions is above a certain threshold.
When this difference is high, the facility will have to purchase
emissions credits from other sources and will be subject to the
very strictest emissions standards (i.e., LAER). However, when
this difference is less dramatic, it may not trigger these stringent standards. Thus, companies are faced with a “use it or lose
it” proposition. Either they use their existing pollution allocation or they suffer the consequences the next time they expand. Thus, to err on the side of caution, some companies have
an incentive to over-emit today so that future modifications
appear to add only a relatively small amount of pollution.
Third, some large companies have been known to ship a
product back and forth, sometimes even across state lines, during the production process in order to take advantage of plants
and processes that are already permitted rather than expand
the capacity of one plant to make it more versatile, and which
may trigger NSR standards. The root cause underlying the
May 2001
excess shipping practice is that obtaining those incremental
permits and complying with NSR standards is often more costly
and more time-consuming than making an end run around
the permitting process by shipping unfinished products from
plant to plant. One EPA representative told us during an
interview that Merck Pharmaceuticals used to legally ship products from one state to another during the manufacturing process to avoid having to get a new permit at any one site, thereby
avoiding a permitting process that could take up to a year.6
Such behavior, according to this EPA representative, is not
unique to Merck, but routine among regulated companies.
Facilities that need the flexibility to make production decisions quickly, like pharmaceutical companies, may be more
likely to engage in this practice. This results in greater air pollution from trucks, trains, or planes needlessly shipping a steady
stream of products back and forth. Further, it is a disincentive
to invest in newer, potentially cleaner technology, because new
technology will always be subject to stringent regulations.
Even if these loopholes have a small environmental
impact in the grand scheme of things, shouldn’t EPA take the
stance of actively encouraging pollution prevention rather
than actively encouraging pollution or, at best, demonstrating indifference? Given these unwanted outcomes of an
otherwise sound environmental law, how can EPA remove the
incentive to pollute? The proposed changes that follow would,
we believe, help EPA achieve this shift.
There are good solutions to each of these disincentives, many
of which are illustrated in the permit designed under the
Merck Pharmaceuticals Project XL completed in 1997. Project
XL, which stands for “eXcellence and Leadership” and is one
of the flagship projects of EPA’s reinvention effort, is a pilot
program aimed at promoting and testing environmental
innovations. The agreements reached between EPA and
participants—who can include industrial facilities, local
governments, and states—can actually bend existing
environmental laws slightly, as long as the project results in
a “superior environmental performance.” The project at
Merck’s Stonewall Plant in Elkton, VA, was launched on March
16, 1995, to demonstrate that a smarter permit can result in
a cleaner environment.7
Through site-specific rulemaking and enforceable permit
conditions, the facility agreed to replace its coal-burning
boilers with substantially cleaner natural gas ones in exchange
for sitewide emissions caps (as opposed to the standard
source-specific caps). This multimillion-dollar technology
upgrade to natural gas boilers is not otherwise required by
regulations and was not necessary from an operational standpoint, yet it resulted in an estimated reduction of more than
900 tons per year (tpy) of criteria air pollutants.
Table 1. The sources and solutions for CAA loopholes.
Incentives to Pollute
Why Does Incentive Exist
and Persist?
What Behavior Does Incentive
Which Items in Merck Permit
Directly Address this Incentive?
Grandfather Exemption
Facilities prefer to keep inefficient
sources that are grandfathered
because new sources would be
subject to stiffer emissions limits
Keep inefficient equipment longer
than would otherwise
XL agreement allowed Merck to trade
outdated equipment for flexible
permit; facilitywide caps encourage
firm to minimize all emissions to make
room for future expansion; less
pollution is rewarded with less
monitoring and reporting
Over-emit to Make Actual-toPotential Difference Small
Facilities fear being subject to
stricter requirements if proposed
expansion appears to increase
emissions substantially
Deliberately pollute up to
allowable limit
Lifetime limits that cannot be changed
even if firms expand; lower emissions
are rewarded with less monitoring and
Incentive to Transport
Product Excessively
Getting new permits is too costly
and time-consuming
Ship product back and forth in
course of manufacturing to take
advantage of existing emissions
limits; do not invest in newer
technology that might be cleaner
Facilitywide caps allow expansion
at site without additional permit
The facility’s total emissions of criteria pollutants (not including lead) was capped below the level at which the plant
had operated over recent years (approximately 1500 tpy). As a
result, Merck reduced its total emissions cap by 20%, thereby
permanently retiring at least 300 tpy of criteria pollutant
emissions. Within the sitewide total emissions cap, the facility is subject to individual pollutant caps, or subcaps. Merck
reduced the pollutant-specific subcaps for SO2 and NOx by
25% and 10%, respectively. The subcap for particulate matter
under 10 microns in diameter (PM10) was maintained at the
same level as before.8
The new permit did not penalize Merck for updating its
boilers, but instead created a facilitywide emissions cap that
encouraged the company to buy the cleanest, most efficient
equipment. Any savings in emissions that Merck reaped from
its boilers could then be shifted, like an internal emissions
trade, to another process that might otherwise be exceeding
compliance. But because the net result is far lower emissions,
changes at the facility that might otherwise be considered
to result in emissions increases would no longer need prior
approval by the permitting authority under the PSD program
or minor NSR.9 As one Merck representative put it, because of
the plant’s proximity to a pristine Class I air quality area, Merck
used to have to apply for a PSD permit for every little modification, even to put in a small backup pump “a little bigger
than a lawn mower” that operates once or twice per year.10
Under the new permit, the company avoids such requirements.
Another feature of the Project XL permit is that the emissions caps are fixed for the life of the plant; they cannot be
raised or lowered. This removes the incentive to emit right up
to the limit for fear of triggering stricter controls during the
next permit round. As an alternative to the current PSD permitting system, the total emissions cap and subcap system
provides an incentive for Merck to identify and promptly
implement ongoing emission reductions at the facility to
make operating room under the cap for future modifications
and expansions.11 As long as Merck operates under this PSD
permit, it will no longer be able to obtain permits to increase
emissions above the cap, since an exceedance of the total
emissions cap is a basis for termination of the permit. Under
the current permitting system, Merck would not be constrained
by a sitewide emissions cap, and could continue to increase
emissions as long as the proper permits were obtained.12
Lower emissions are further encouraged in that the less
Merck emits, the less it is required to report. In other words,
if emissions are far below the emissions cap, then there is
less need for precise and frequent monitoring. This system
of “tiered” monitoring, record keeping, and reporting requirements provides Merck with another built-in incentive to
minimize emissions and to find opportunities to implement
emission reductions.13 Note that Merck is still subject to
existing and future rules that are not included in this
agreement, such as lead emissions standards and Maximum
Achievable Control Technology (MACT) standards for the
pharmaceutical industry.
In one fell swoop, the Merck project dismantled all three CAA
loopholes. While this was a monumental achievement, it is
clearly impractical for EPA, state agencies, and facilities to write
a site-specific rule in the Federal Register for every plant with
old boilers to upgrade. While Merck and EPA representatives
May 2001
have said they believe they are seeing a reduction in maintenance costs under Merck’s new permit system, they also note
that the initial negotiation period of the permit was expensive, calling it an “incredibly time-intensive process.”12 Yet the
solutions themselves—facilitywide caps, long-term permits,
and less reporting for lower emissions—are workable options
on a wider and more economical scale. To recap, EPA can act
to make the following practices the norm:
• Facilitywide caps that allow within-facility trading.
Permits that involve facility-level caps pool all the equipment that emits a pollutant under one cap regardless of
the technology used. The Merck permit, as well as some
other experimental permits, such as EPA’s Pollution
Prevention Pilot Program (P4) permits, include pollutant-specific caps for a facility as a whole. Facilitywide
limits set up a system of internal pollution credits trading, whereby, for example, a facility has the flexibility to
increase emissions from Stack A, as long as it decreases
emissions of the same pollutant from Stack B. Such a
change may prove economical for the facility, both in
terms of technology requirements and because it avoids
the costly NSR process. Of course, such a system requires
careful implementation. Regulators need to be prudent
about where they set facilitywide caps. If they are set
by summing current permit limits, which are often well
above actual emissions, then emissions caps could simply be a license to pollute more. If, however, they are set
closer to potential emissions achievable through better
technology, then the agency gains something in return
for the flexibility it offers, namely, an incentive for firms
to prevent pollution.
• Lifetime limits. Most permits have a life of five years.
But absent a technical error in the sampling or analysis approach (or the company’s decision to voluntarily lower a cap in lieu of meeting a new requirement) none of the principal partners can raise or lower
the cap. Permits that last indefinitely not only save
facility and agency staff time by avoiding repeated renewal processes, but they guarantee that the total emissions from the facility will not exceed that cap as long
as the permit is valid. The environmental benefit in
this case is that, in order to grow, the facility must
reduce emissions per unit of product—a direct incentive for pollution prevention.
• Monitoring and reporting requirements that decrease
as emissions decrease. The Merck permit sets up a protocol whereby, as the facility’s emissions approach the
cap, increased monitoring and reporting is required.
When emissions are very low, precise information is not
needed to ensure that a violation does not take place.
Less precise, but also less time-consuming practices,
such as annual emissions estimates rather than monthly
May 2001
or daily, suffice. Not requiring detailed monitoring if
emissions are low eliminates an unnecessary burden. In
addition, it is a powerful and effective incentive to stay
far below the cap, which in turn is another incentive for
pollution prevention.
Other creative solutions may exist, but, in our opinion,
EPA should, at a minimum, aggressively seek the three policy
changes described above and summarized in Table 1.2
The benefits of these changes to industry are the flexibility
they allow and the savings in time and money they provide
from less red tape. Administrative requirements at this point
include operating permit reviews every five years and a
tedious application process each time a new piece of equipment is introduced or the use of an existing piece changes. It
could be argued that these changes would reward industry for
dragging its feet in replacing old, inefficient equipment. However, this view does not take into account that industry has
simply responded to the incentives that the system put into
place, and that the benefits of these changes to the public
would be at least as large as those accruing to industry. On the
other hand, if lawmakers and agency personnel want to
reward only responsible corporate citizens, they may consider
introducing such changes incrementally, beginning only
with the good actors—those without a history of emissions
violations—who have more plainly earned the flexibility of
facilitywide caps. Merck was one such facility, as all Project XL
participants are required to be.
While EPA, state environmental agencies, and industrial
facilities would benefit from reduced paperwork and the
decreased emissions resulting from these permit changes,
the public stands to gain perhaps the most of all. Because
air emissions are associated with respiratory disease, poor
visibility, and damaged ecosystems through acid rain and
ground-level ozone (which is known to damage vegetation),
incentives that actively reduce emissions improve the quality of life for everyone.
Whether it is EPA or Congress that should take the lead
depends on the change in question. For example, it would
take an act of Congress to extend the life of the permit or
eliminate the grandfather clause and a decision by EPA to
shift reporting requirements.
Criteria pollutants have been demonstrated to have a
significant adverse effect on the environmental quality of the
Shenandoah National Park, near the Merck plant. Merck’s conversion from coal to natural gas boilers, however, achieved an
up-front reduction of these pollutants: SO2 emissions decreased
by approximately 94% and NOx emissions decreased by 87%
from baseline actual emission levels.13 In addition, the total
It has been 10 years since Congress passed sweeping changes
to the CAA, and no one harbors any illusions that new changes
would be easily approved. But the nature of the changes proposed here to stop encouraging pollution and to achieve reductions in red tape for business and agencies are just the kind
of issues that could rally bipartisan support. The key lies in
the creative solutions outlined in Merck’s Project XL agreement and committed leadership from EPA and Congress.
able at the enviroNET Web site, http://environet.policy.net/cleanair/more/
(accessed February 2001).
Interview with Cecil Rodriguez, attorney for U.S. Environmental Protection Agency (EPA) Region 3, September 9, 1999.
Reinvention Pilot Programs. Fed. Regist. 1995, 60 (99), 27282 (60 FR 27282
Interview with Merck representative (prefers anonymity), September 9,
60 FR 27282 (1995).
Interviews with Merck representative (prefers anonymity) and Cecil
Rodriguez, attorney for U.S. Environmental Protection Agency (EPA)
Region 3, September 9, 1999.
60 FR 27282 (1995).
About the Authors
Susan Helms ([email protected]) is an associate scientist and
Jennifer Sullivan is a research associate, both of Tellus Institute,
Boston, MA, who analyze and design environmental policies at the
local, state, and federal level. They recently completed a study of
innovative permitting programs for the National Academy of Public
Administration as part of NAPA’s report to Congress, entitled Learning from Innovations in Environmental Protection. Helms holds a
Master’s degree in Environmental Studies from the Yale School of
Forestry and Environmental Studies, and Sullivan holds Masters
degrees in Environmental Engineering and Technology and Public
Policy from MIT.
Crow, M.; Pfeiffer, A.; Susskind, L. Seeking Operational Flexibility and Pollution Prevention Under Title V of the Clean Air Act: ETP99-02; U.S. Environmental Protection Agency’s P4 Program; Environmental Technology and
Public Policy Program; Massachusetts Institute of Technology, Department of Urban Studies and Planning: Boston, MA, 1999.
Helms, S.; Sullivan, J.; White, A. The Potential and the Pitfalls of Innovative
Permits: Learning From New Jersey’s Facility-Wide Permitting Program. Submitted to National Academy of Public Administration: Washington, DC,
May 2000.
Percival, R.C.; Miller, A.S.; Schroeder, C.H.; Leape, J.P. Environmental Regulation: Law, Science and Policy; Little, Brown & Co.: New York, NY, 1992.
National Air Pollutant Emission Trends, 1900-1998; EPA-454/R-00-002; U.S.
Environmental Protection Agency; Office of Air Quality: Research Triangle Park, NC, 2000.
EPA Project XL, Fact Sheet: Merck Stonewall Plant, September 1998; EPA-100F-98-020; U.S. Environmental Protection Agency: Washington, DC, 1998.
EPA Project XL, Progress Report: Merck Stonewall Plant, March 1999;
EPA-100-F-99-007; U.S. Environmental Protection Agency: Washington,
DC, 1999.
May 2001
Photograph courtesy of the U.S. National Park Service.
emissions cap and subcaps will
ensure a continuing, permanent
reduction of these pollutants, as
well as provide an ongoing incentive to minimize actual emissions
to preserve the operating margin
under the caps. Besides the significant reduction in criteria pollutants resulting from the project,
the conversion to natural gas
also was projected to result in a
reduction of approximately 65%
of hazardous air pollutants (HAPs),
specifically hydrogen chloride and
hydrogen fluoride, which result
from the burning of coal. These
two HAPs are also associated with
the formation of acid rain; thus,
reducing the emissions of these
chemicals is contributing to efforts Blue Ridge Mountains, Shenandoah National Park, VA, located near the Merck plant.
to improve air quality in the Shenandoah National Park and
1. 21st Annual Report; Council on Environmental Quality: Washington, DC,
the surrounding community.
1991; Table 39.
In addition, the year-plus permit process included numer2. National Air Quality & Emissions Trends Report, 1990; U.S. Environmental
Protection Agency: Washington, DC, 1991.
ous opportunities for public input. EPA, Merck, and the Com3. National Summary Percent Contribution by Unit Fuel Type, 1998; Acid Rain
Program; U.S. Environmental Protection Agency: Washington, DC; availmonwealth of Virginia believed that the outcome would be
able at the U.S. Environmental Protection Agency Web site, http://www.
useless if it were not supported by its citizens. This element is
epa.gov/acidrain/emission/us_sum.htm (accessed February 2001).
4. National Air Quality and Emission Trends Report, 1997; U.S. Environmental
crucial to maintain in any successful large-scale changes in
Protection Agency: Washington, DC, 1998; Tables A-4, A-8; pp 114, 117.
5. Clean Air FAQs; National Environmental Trust: Washington, DC; availthe future.