NOv BUDGET TRADING PROGRAM
-------
Office of Air and Radiation
Clean Air Market Programs
EPA-430-R-04-010 "
1200 Pennsylvania Ave, NW
(6204J)
Washington, DC 20460
www. epa. gov/ai rmarkef s
August 2004
-------
IMOx Budget Trading Program:
2003 Progress and Compliance Report
Introduction
The NOX Budget Trading Program (NBP), is a market-based cap and trade program cre-
ated to reduce emissions of nitrogen oxides (NOX) from power plants and other large
combustion sources in the eastern United States. NOX is a prime ingredient in the for-
mation of ground-level ozone (smog).
Building on prior efforts to reduce summertime NOX emissions in the Northeast, eight north-
eastern states and the District of Columbia implemented the NBP in 2003. Eleven additional
states joined the program in May 2004. NOX emission levels and baselines for states that
complied with the program in 2003, as well as for eleven additional states beginning compli-
ance in 2004, are presented in this report. When fully implemented the NBP is expected to
achieve significant reductions in summertime NOX emissions across much of the eastern U.S.
This report finds that, in states that participated during the first year of the program, ozone
season (May through September) NOX emissions from power plants and other large combus-
tion sources were reduced by more than 30 percent from 2002 levels. These emission
reductions occurred despite an increase in heat input (a measure of power generation) at
affected sources. Emissions have also been reduced by 70 percent from 1990 levels due to
the combination of the NBP and other NOX control programs. In 2003, of the total affected
population of approximately 1,000 units, all but 7 were in compliance. NOX emissions were
reduced on days with peak emissions. In addition, the NOX allowance market has been active.
In anticipation of entering the NBP and in response to other NOX control programs, particu-
larly annual NOX reductions under the Acid Rain Program, the eleven states that did not
participate in the program until 2004 have also made progress in reducing NOX emissions.
NOX emissions in these states were approximately 50 percent below 1990 levels. In addition,
sources successfully monitored and reported emissions for the first time in 2003.
Ground-level ozone, or smog, is formed from oxides of nitrogen (NOX) and volatile organic
compounds (VOCs) in the presence of sunlight and heat. Levels are highest during the hot
summer months when sunlight is strongest. Ozone aggravates asthma, increases susceptibil-
ity to respiratory illnesses, and contributes to permanent lung damage. It can also damage
forests, reduce the productivity of agricultural crops, and lead to the decay of monuments
and buildings.
Ozone continues to be a pervasive air pollution problem. In April 2004, EPA released a list of
126 areas that do not meet the new health-based 8-hour ozone standard. In most cases, the
Agency's findings are based on air quality data from 2001 through 2003. The ozone nonattain-
-------
8-Hour Ozone Standard (promulgated in 1997]:
The 8-hour standard is met when the 3-year average of the annual fourth highest daily maximum 8-hour average
concentration is less than 0.08 ppm (parts per million).
What Does It Mean?
EPA collects ozone data on an hourly basis. Essentially, 8-hour average ozone concentrations at a monitor cannot
exceed 0.08 ppm more than three days per year. For compliance purposes:
Hourly ozone measurements are used to compute 8-hour average concentrations.
• The daily maximum 8-hour average is recorded for each day.
• For each year, the fourth highest daily maximum concentration is calculated.
These annual fourth highest daily maximum concentrations are averaged over three-year periods.
l If the average exceeds 0.084 ppm (0.085 rounds up), the area is designated as a "nonattainment area."
ment areas, shown in Figure 1, include 474 counties that are home to 159 million people—more
than half of all Americans. The majority of these areas are in the NOX SIP call states. As many of
the states with nonattainment areas plan for the future, the NBP emission reductions will be an
essential component of their strategies for attaining the 8-hour ozone standard.
This report presents the results of the first year of the NBP for affected sources in states with
compliance requirements in 2003, as well as emissions data for NBP sources in other states
that reported their emissions in 2003. Along with the results of the first year of the NBP, this
report presents baseline NOX emission levels prior to the program (for 1990 and 2000). As the
program matures and as EPA continues to assess progress, these baselines will help EPA
analyze emission trends and the impact of NOX reductions achieved by NBP sources.
8-hour Ozone Standard Attainment and Nonattainment Areas in the U.S. as of
April 2004
Attainment or Unclassif iabte Areas (2,668 counties]
Nonattainment Areas (432 entire counties]
Nonattainment Areas (42 partial counties]
Figure 1
Many counties in the NBP region do not meet the 8-hour ozone standard.
Source: EPA
-------
What Is the IMOX Budget
Trading Program^
EPA and the States Have Taken Steps to
Address Regional Transport
Created after years of scientific research and air quality data showed that
upwind NOX emissions can contribute significantly to ozone nonattainment in
downwind states, the NOX Budget Trading Program (NBP) follows several other
major efforts to reduce NOX from large, stationary sources. These initiatives
include the Acid Rain Program, the Ozone Transport Commission's (OTC) NOX
Budget Program, New Source Review, New Source Performance Standards,
application of Reasonably Available Control Technology to existing sources, and
other state and local efforts.
Title IV of the 1990 Clean Air Act Amendments created the Acid Rain Program.
This program, which achieved large SOe reductions from power plants through
a cap and trade program, also required coal-fired power plants throughout the
country to reduce their NOX emission rates (NOX emissions per unit of heat
input). The goal of the Title IV NOX program was to achieve and maintain an
annual 2 million ton reduction in NOX emissions from what emissions would have
been in 2000 without the program. This goal has been surpassed. In 2002, due
A Quick Snapshot of National and Regional IMOx Control Programs
Acid Rain NOx Reduction Program (ARP)— Annual, national program controlling NOX emissions from electric generating units.
Sources are required to meet certain rates of NOX emissions. There is no cap on emissions or allowance trading. The program
began in 1996 with a second phase beginning in 2000.
Ozone Transport Commission (OTC) NOx Reduction Programs— States in the Northeast collaborated to achieve ozone-season
NOX reductions in several phases. In Phase I, sources were required to reduce their annual rates of NOX emissions to meet
Reasonably Available Control Technology requirements. In Phase II, states participated in a cap and trade program, the OTC
NOX Budget Program, to achieve additional reductions during the ozone season. In 2003, the OTC NOX Budget Program was
replaced by the larger NOX Budget Trading Program.
tate Implementation Plan (SIP) call— Building upon analyses done by the Ozone Transport Assessment Group (OTAG),
this rule was finalized by EPA in 1998. It required states significantly contributing to ozone nonattainment problems in other
states to reduce their NOX emissions during the ozone season beginning in 2003. This rule gave states the flexibility to reduce
emissions through various means and gave them the option to participate in the NOX Budget Trading Program.
Budget Trading Program (NBP)— An ozone season cap and trade program intended to help states meet their NOX SIP call
required reductions. States in the OTC began to comply in 2003 and many other states across the East and Midwest began to
reduce emissions in 2004. Twenty-one states and the District of Columbia are participating or will participate in the future.
-------
to Title IV and other state actions, annual NOX emissions from Acid Rain
Program sources were more than 3 million tons lower than what they would
have been without Title IV. However, because there is no cap on Acid Rain
Program NOX emissions, NOX emissions may increase in the future as demand
for electricity continues to grow.
The 1990 Clean Air Act Amendments also established the OTC to mitigate
interstate transport of pollution in the Northeast. In September 1994, eleven
states and the District of Columbia signed a Memorandum of Understanding
[MOLJ] committing to reduce NOX emissions throughout the region. In 1995,
the OTC states required existing sources to meet Reasonably Available Control
Technology (RACT) limits, and in 1999 through 2002, most of the OTC states
achieved deep NOX reductions through an ozone season cap and trade pro-
gram for NOX called the OTC NOX Budget Program. The OTC states that
participated in this trading program included Connecticut, Delaware, Maryland,
Massachusetts, New Hampshire, New Jersey, New York, Pennsylvania, Rhode
Island, and the District of Columbia.1
Separate from the activity in the OTC, EPA and the Environmental Council of the
States formed the Ozone Transport Assessment Group (OTAG) in 1995. This work-
group brought together interested states and other stakeholders, including
industry and environmental groups. Its primary objective was to assess the ozone
transport problem and develop a strategy for reducing ozone pollution throughout
the eastern half of the U.S.
The IMOX SIP Call Requires Significant
Summertime IMOX Reductions across Eastern
States
Based on the findings of OTAG, EPA proposed the NOX SIP call in 1997 and finalized
it in 1998. This rule concluded that NOX emissions in twenty-two states and the
District of Columbia contribute to ozone nonattainment in other states, and the rule
required affected states to amend their state implementation plans (SIPs) and limit
NOX emissions. EPA set an ozone season NOX budget for each affected state,
essentially a cap on emissions from May 1 to September 30 in the state. The first
control period was scheduled for the 2003 ozone season.
The NOX SIP call did not mandate which sources must reduce emissions but, rather,
required states to meet an overall cap (or budget) and gave them flexibility to devel-
op control strategies to meet the cap. The NBP was developed to help states
achieve highly cost-effective NOX emission reductions.
1 Vermont and Maine also signed the MOU but chose to meet their NOX reduction goals outside of the trading
program. Northern Virginia is also part of the OTC but did not sign the MOU and has not participated in the
OTC NOX reduction programs. This report, therefore, does not refer to Virginia as one of the OTC states in
terms of trading program implementation.
-------
NOX SIP Call Region
The NBP is a cap and trade program for large electric generating units (EGLJs) and
large industrial boilers, turbines, and combined cycle units. In this type of program,
the emissions budget sets a "cap" on emissions at a
specified level. Sources are provided "allowances"
[each allowance represents one ton of emissions),
and each year the source must hold sufficient
allowances to cover all NOX tons the source emits
during the ozone season. To monitor emissions,
sources use continuous emission monitoring systems
[GEMS] or other approved monitoring methods under
EPA's stringent monitoring requirements (40 CFR
Part 75). If a source's emissions are less than the
allowances it holds, the source can sell the unused
allowances or bank the allowances for use in a future
ozone season.
The NOX SIP call faced many legal challenges.
Although the U.S. Court of Appeals for the D.C. Circuit
largely upheld most of the final rule, the court's deci-
sion resulted in Wisconsin and portions of Georgia
and Missouri being removed from the list of areas
where reductions are required (decreasing the num-
ber of affected states to 21 states and the District of
Columbia).2 In addition, the initial deadline for emission
Figure 2
Participation in the NBP is phased in across the region.
Source: EPA
reductions under the NOX SIP call was delayed until May 31, 2004. The OTC states
had been set to achieve additional reductions in May 2003 under the OTC trading
program. With the exception of New Hampshire, these states instead began to
implement the NBP in May 2003.3 New Hampshire is not affected under the NOX
SIP call and is not part of the NBP.
The OTC states were joined in the NOX SIP call by Alabama, Illinois, Indiana,
Kentucky, Michigan, North Carolina, Ohio, South Carolina, Tennessee, Virginia, and
West Virginia on May 31, 2004 (see Figure 2). All of these states have chosen to
meet the NOX SIP call requirements through participation in the NBP. Affected
portions of Georgia and Missouri will be required to comply with the NOX SIP call in
2007. In Alabama, Georgia, Michigan, and Missouri, only a portion of each state is
subject to the NOX SIP call. Because Georgia and Missouri are not yet participat-
ing in the program, this report does not discuss results for these states.
2 Due to litigation, EPA stayed the NOX SIP call findings with respect to the 8-hour ozone standard for all affected
states (65 FR 56945). On April 91, 9004, EPAs final NOXSIP call Phase II Rulemaking determined that Wisconsin
would not be included in the rule based on the 1-hour ozone standard (69 FR 91609). However in the future,
EPA may still consider lifting the stay with respect to the 8-hour standard for all affected states, including
Wisconsin (69 FR 91608).
3 The overall reduction goals set for the NBP under the NOX SIP call are generally consistent with reduction levels
planned for 9003 in the OTC states, and many program features, such as allowance tracking and reporting, are
consistent between the two programs.
-------
Understanding Baselines
and Budgets: Guidelines for
n Evali
Progress in reducing emissions under a cap and trade program can be evaluated
primarily in two ways. NOX Budget Trading Program (NBP) emissions can be com-
pared to:
A baseline level of emissions (emissions from a period prior to the start of the
program), or
The reduction goals—or budgets—set for the NBP in the rules of each affected
state.
EPA believes both assessment approaches can provide valuable insights on emis-
sion reductions achieved by the program, and the following sections clarify the
assumptions and approaches used to derive baseline and budget levels for pur-
poses of program evaluation. EPA plans to assess emission changes under the
NBP on a regular basis.
Understanding Emission Levels Prior to the
Program Can Help Us Measure Progress
To measure progress in reducing ozone season NOX emissions, it is helpful to under-
stand how emissions under the program compare to emissions prior to the program.
EPA has chosen 1990 as a baseline year because it represents emission levels before
the implementation of the 1990 Clean Air Act Amendments, and can document
progress achieved under the amendments. The 1990 baseline period also was used
throughout the Ozone Transport Assessment Group (OTAG) process, which led to the
development of the NOy SIP call.
For the 1990 ozone season emission estimates, data were derived from annual 1990
National Emission Inventory (NEI) data. The Ozone Transport Commission (OTC) states
used the 1990 NEI data to develop a refined 1990 ozone season inventory as part of
their efforts to reduce ozone in the region4 In this report, EPA is using the OTC's
refined NEI inventory for the OTC states, and EPA has developed an ozone season
inventory directly from the NEI for the non-OTC states.
Emissions have been reduced greatly since 1990, but many of these reductions result
from control programs other than the NBP To better reflect the reductions that can
4 1990 OTC NOX Baseline Emission Inventory, U.S. EPA, EPA-454/R-95-013, July 1995.
-------
be attributed to the NBR emissions from the 2000 ozone season are also presented
here. The year 2000 was chosen because most of the reductions due to the imple-
mentation of earlier NOx regulatory programs under the 1990 Clean Air Act
Amendments had already occurred by 2000, but sources were not yet implementing
the NBP at that time. Emissions in 1990 and 2000 both provide valuable information
for evaluating the extent to which the NBP and other control programs are providing
reductions in NOx levels.
In 2000, the stringent emissions reporting requirements for Acid Rain Program and
DTC NOX Budget Program sources provide accurate ozone season NOX data for most
of the NBP units, and those units account for well over 90 percent of all the emissions
from affected sources in most NBP states in 2000. There is no comparable, quality-
assured emissions monitoring data for units that are not affected by either of these
programs. These units include industrial units and non-Acid Rain Program electric gen-
erating units outside the OTC region. Therefore, to approximate these 2000 ozone
season NOX emissions for industrial units, EPA has used the most recent NEI data
available (1999). To approximate 2000 emissions for non-Acid Rain Program electric
generating units, EPA used data developed to set the NOX SIP call state budgets.5
Figure 3 shows the total 1990 and 2000 baseline ozone season NOX emissions.
Between 1990 and 2000, affected sources in the
region reduced ozone season NOX emissions by approxi-
mately 600,000 tons (34 percent), mainly due to the
requirements of the annual Acid Rain Program and sea-
sonal OTC program.
States Were Required to
Reduce Emissions to Meet
a Budget
States were given budgets, or caps, on their ozone sea-
son emissions. These are the target levels they are
expected to achieve under the NBP. The actual number
of allowances a state allocates to sources for a cer-
tain year under the program may not be equal to the
state's budget. States may hold back some of the
budget to set aside allowances for new units, energy
efficiency or renewable energy projects, or other pur-
Total Ozone Season IMOX Emissions
in NBP States in Baseline Years
(All Sources]
<_i 2,000,000 -
Tn"
.0
t, 1,500,000-
U)
o
'
-------
poses. Another factor that affects the amount of allowances allocated by each
state is the Compliance Supplement Pool (CSP). The CSP is a pool of extra
allowances included in the NOX SIP call to help sources comply with the trading
budgets in the first two years of the program. EPA established the CSP pool to
address concerns about electricity reliability at the beginning of the program.
States may distribute their respective CSP allowances based on early reductions,
directly to sources based on a demonstrated need, or by some combination of the
two methods. In the OTC states, CSP allowances generally were distributed to
sources based on a pro rata share of banked allowances that the sources held in
the OTC NOX Budget Program. Allowances from the CSP in a given state may be
used to cover emissions during the first two control periods in that state. For the
states that began to comply in 2003, CSP allowances can be used only in the
2003 and 2004 ozone
seasons, while in most of
the remaining NBP states,
the allowances can be
used only in the 2004 and
2005 ozone seasons.6
Figures 4 and 5 help illus-
trate the levels of
reductions that would be
needed to meet the emis-
sion reduction goals set by
the NBP. Emissions for
1990 and 2000 show
progress in reducing NOX
prior to the program.
Budgets are presented to
demonstrate the overall
reduction goals of the pro-
gram, and budgets with
CSP allowances demon-
strate the reduction goals
in the early years of the
program. By comparing
baselines and budgets, it is
easy to see how much fur-
ther each state would be
expected to reduce emis-
sions to meet their target
levels of emissions.
Baseline Emissions and Target Reduction Levels in OTC States
Participating in the NOx Budget Trading Program*
200,000 —i
(0
o
1
E
150,000 —
100.000 -
1
50,000 —
11990 Emissions T 1 ™B! Budget Plus Compliance
I I Supplement Pool
12000 Emissions Budget
,l\rm,
n
71,
Ti
CT DC DE MA MD NJ NY
States
PA Rl
Figure 4
Through the OTC NOX program, the Acid Rain Program, and other Clean Air Act
requirements, the OTC states have made significant progress in reducing ozone sea-
son NOX emissions from 1990 levels.
* New Hampshire participated in the OTC program but is not affected by the NBP.
Source: EPA
3 In North Carolina, the state allowed certain sources to use CSP allowances to meet state reduction requirements in 9003, but those
allowances would be deducted from their 9004 allowance allocations under the NBR
-------
Baseline Emissions and Target Reduction Levels in Non-OTC States
Participating in the NOx Budget Trading Program as of May 2004*
250,000 —I
" 200,000 —
(0
I 150,000-1
_en
ill
n? 100,000 —
50,000 —
Figures 4 and 5 show that all states in the NBP were required to achieve sub-
stantial reductions from 1990 levels to meet their budgets. By 2000, the OTC
states had made significant progress toward that goal. They reduced emissions
approximately 55 percent in the region through the OTC NOX program, the Acid
Rain Program, and other Clean Air Act requirements. Because of this, overall, the
DTC states generally only had to achieve small reductions between 2000 and
2003 to meet the target reductions for the NBP.
While the OTC states
were well on their way
to meeting their budg-
ets by 2000, the
non-OTC states gener-
ally still had more
work to do to meet
their budgets.
Between 1990 and
2000, the non-OTC
states had reduced
NOX emissions by
approximately 27 per-
cent through the Acid
Rain Program and
other Clean Air Act
requirements, but
they still had to
reduce emissions sub-
stantially to meet
their target reduc-
tions for the NBP.
Note that the scale in
Figure 5 reflecting
NOX emissions in
states outside the
OTC is different from
the scale in Figure 4
for the OTC states.
j 1990 Emissions
] 2000 Emissions
Totsl Budget Plus Compliance
Supplement Pool
Budget
I
NC OH
TN VA WV
Figure 5
Through the Acid Rain Program and other Clean Air Act requirements, the non-OTC states
have achieved more than a 25 percent reduction in ozone season NOX emissions from
1990 levels and are poised to realize further significant reductions.
Georgia and Missouri will not enter the NBP program until 9007
Source: EPA
Based on the information in Figures 4 and 5, the overall NOX allowance budgets with
and without Compliance Supplement Pool NOX allowances are:
OTC states: 139,593 (plus 25,031 CSP] allowances = 164,624 allowances
Non-OTC states: 366,475 (plus 141,152 CSP] allowances = 507,627 allowances
Regional total: 506,068 (plus 166,183 CSP] allowances = 672,251 allowances
-------
The NBP Affects a Significant Number of Units
Based on data reported to EPA, there are nearly 2,600 affected and operating
units in the NBP states, including the states that joined the program in 2004.
About 1,000 of these units are in states in the OTC that complied in 2003, while
about 1,600 are in non-OTC states that did not have to comply in 2003.
The term "unit" means a fossil fuel-fired combustion boiler; turbine, or combined
cycle unit. At a given facility, there may be multiple units. Electric generating units
[EGLJs] provide electricity for sale. Many EGUs in the NBP are also covered by the
Acid Rain Program, although the NBP also includes other EGUs (such as some
simple combustion turbines, cogenerators, and independent power producers]
that are exempt from the Acid Rain Program. About 62 percent of all units in the
NBP are also affected by the Acid Rain Program, while other EGUs comprise
about 25 percent of the total units, and non-EGUs make up the other 13 percent.
Industrial units are sources that provide electricity or
steam for use at their industrial facility, with little or no
electricity generated for off-site use. The Acid Rain
Program covers EGUs but not industrial units.
However; the OTC trading program included industrial
units, and EPA also decided to include industrial boilers,
turbines, and combined cycle units in the NBP. Of the
total affected NBP population of nearly 2,600 units,
approximately 350, or about 13 percent, are industrial
units. This is an increase from the OTC trading pro-
gram, where about 6 percent of the sources were
industrial units. Figure 6 provides a detailed breakdown
of the classification of NBP units.
Units Affected by the NBP—
A Comparison of Electric Generating
Units and Industrial Units for OTC
and Non-OTC States
2,000 -i
.i
1,500 -
500-
O-1-
The NBP includes boilers, turbines, and combined cycle
units from a diverse set of industries. Some of the OTC
states also have included other combustion units, such
as cement kilns and process heaters. This diversity has
contributed to a wide range of compliance strategies
as facilities faced varying control costs and compliance
needs to meet the requirements of, first, the OTC trad-
ing program and, later; the NBP. The various industrial
sources participating in the program have also creat-
ed additional trading flexibility for EGU compliance. So
fan the inclusion of industrial sources appears to have
been beneficial to the trading programs without creating disproportionate burdens
on these sources. These sources have followed the same monitoring and report-
ing requirements as EGUs and have participated in the market, buying and selling
allowances.
OTC
I Industrial Units
Non-OTC
NorvARP Electric Generating Units
Acid Rain Program (ARP) Units
Figure 6
About 60 percent of the units affected by the NBP are
also affected by the Acid Rain Program. In addition, the
NBP incudes many other electric generating units,
which were not included in the Acid Rain Program.
Source: EPA
10
-------
Performance under the IMBP
in 2003 Already Shows
Positive Resu
In 2003, only sources in the Ozone Transport Commission (OTC) states were
required to hold allowances to cover NOX emissions during the ozone season. In
most of the other states affected by the NOX SIP call, states required sources to
begin monitoring and reporting
ozone season emissions and heat
input data in May 2003, even
though the first control period
did not begin for them until May
31, 2004.7 Thus, the following
sections generally focus on
results in the OTC region,
although there is a brief review of
the reported 2003 emissions in
the other states.
Summertime IMOx
Emissions Have
Declined across
the Region
Figure 7 shows the combined
1990 baseline, 2000 baseline,
trading budget, and 2003 emis-
sion levels for both the OTC states
and the other NOX Budget Trading
Program (NBP) states. The
2003/2004 trading budget levels
in Figure 7 are presented with and
without the Compliance
Supplement Pool (CSP) allowances.
The budgets for OTC states repre-
sent the 2003 budgets, while the
budgets for the other states rep-
resent the first control period in 2004.
Regional Baselines, Trading Budgets, and 2003
Emissions*
Trading Budgets with
Complience Supplement
Pool Allowances
source: EPA
In North Carolina, sources were not required to monitor and report data in 9003, although many sources did
so voluntarily (over 75 percent of the sources provided data).
11
-------
Despite the fact that the 2003 ozone season was not a control period in the non-OTC
states, the universe of sources in the entire NBP area reduced emissions in 2003 by
more than 50 percent from year 1990 baseline levels and by about 33 percent from
year 2000 baseline levels8 In 2003, OTC states had reduced emissions 70 percent
from 1990 levels and 33 percent from 2000 levels. While many of the reductions
from the 1990 baseline represent other programs implemented under the Clean Air
Act (such as NOX reductions under the Acid Rain Program, the OTC trading program,
and other state rules), the significant decrease since the 2000 baseline documents
additional reductions that, at least to some degree, reflect early reductions as
sources begin to implement controls and other operating changes in anticipation of
the NBP.
OTC States Continue to Reduce below Total
Allocated Allowances
During the OTC trading program, emissions were less
than allowances in every year of the program (1999-
2002). That trend continued under the NBP in 2003. In
the participating OTC states, ozone season NOX emis-
sions in 2003 were approximately 134,000 tons, 18
percent less than the number of NBP allowances allo-
cated in 2003. In addition, 2003 emissions in these
states were more than 30 percent less than their emis-
sions in 2002.
Emissions in many of the OTC states were below alloca-
tions in 2003 (see Figure 8). Exceptions included
Delaware, where emissions were essentially the same
as allowances allocated by the state (emissions were
only about 20 tons higher than allocations). In addition,
emissions in Maryland and New Jersey in 2003 were
higher than allocations by about 1,750 tons in Maryland
and about 1,250 tons in New Jersey. However; Maryland
chose to only allocate about 2,200 of its CSP
allowances, while New Jersey allocated about 5,000
fewer allowances than the budget authorized in the NOX
SIP call. New Jersey is using a smaller budget to ensure
that local areas will reach attainment with the ozone
standard. While sources in Maryland and New Jersey acquired allowances from
sources in other states to comply with the program in 2003, emissions in all states,
including Maryland and New Jersey, were significantly lower than 2002 levels.
3 Note that the information available in the non-OTC states excludes several North Carolina units that did not moni-
tor in 9003. The reported emissions for North Carolina in 9003 reflect roughly 90 percent of the total ozone
season emissions for affected sources, so the reductions from baseline levels shown in Figure 7 still generally hold
even if all North Carolina sources had reported ozone season emissions in 9003.
OTC States in the NOX Budget Trading
Program: Ozone Season NOX 1990
and 2000 Baselines, Budget, and
2003 Emissions,
Figure 8
NOX emissions from NBP units in the 2003 ozone
season in the OTC region were below allocations.
Source: EPA
-------
Other NBP States Will Need to Reduce
Emissions in 2004
In 2003, ozone season emissions were substantially
lower than the 1990 and 2000 baseline levels even for
the states that did not begin to comply with the pro-
gram until 2004. Figure 9 shows budget levels for
2004 and beyond in these states. A comparison of
2003 emissions with 2004 budgets demonstrates that
some additional reductions will be necessary for these
states to eventually reach their budgets.
Due to litigation, the 2004 control period for these
states began on May 31, instead of May 1. The
allowance allocations for 2004, however; are based on
a full five-month ozone season. Because of the shorter
control period in 2004 and CSP allowances distributed
in 2004 to help sources comply with the program, EPA
anticipates that these states will have to achieve only
modest reductions in 2004 to comply with the pro-
gram. In 2005 and subsequent years, the control
period will begin on May 1, and deeper reductions will
be necessary.
Daily Emission Trends in OTC
States Continue to Show
Significant Decreases
Non-OTC States in the NOX Budget
Trading Program: Ozone Season
1990 and 2000 Baselines, Budget,
and 2003 Emissions
Source: EPA
Studies indicate that many of the health effects associ-
ated with ozone are attributable to short, peak
exposures. The ozone standard was developed to pro-
tect against such short-term exposures. The NBP, however; is a seasonal program
that ensures significant average regional NOX reductions in the ozone season, and
there have been concerns that a seasonal cap would not sufficiently reduce short-
term, peak NOX emissions that may occur on hot, high electricity demand days,
when ozone formation often is a concern.
In the OTC states, the data from the OTC trading program in 1999 through 2002
and the NBP results for 2003 indicate that these trading programs have reduced
average daily emissions, as well as the highest daily NOX emissions in the ozone
season. Average daily emissions and highest daily emissions from affected units
declined significantly in 2003 as sources began to comply with the NBP (see
Figure 10). This decline provides evidence that a seasonal trading program can
Figure 9
In the non-OTC states, 2003 ozone season NOX
emissions were far below 1990 and 2000 baseline
levels, although further reductions will be necessary
in 2004 to achieve compliance.
13
-------
2500
2000-
1500-
1000-
500
1997 1998 1999 2000
• Highest Daily Emission
reduce peak daily emission levels. Although ozone forma-
tion is a complicated process that is strongly influenced
by the weather; reducing peak daily emissions is expect-
ed to help reduce peak ozone concentrations.
These findings are consistent with a 2003 analysis of
DTC emissions during ozone episodes between 1999
and 2002.9 The study found that, while NOX emissions
from OTC sources tended to be higher during ozone
episodes, these emission increases were related to
increased electricity demand, and such increases would
also have occurred even under a rate-based or com-
mand and control program. That study found that plant
operators did not change or relax control strategies
during high ozone periods.
Emissions Reductions Were
Achieved Despite Increases
in Generation in the Control
Region
Under a cap and trade program, there are many ways that sources can reduce
emissions. One method is to utilize units with high emissions less and to shift
generation to lower-emitting units within the program
or to sources that are not affected by the program.
For example, a plant operator could choose to run a
unit with high emissions less often. The generation
could be taken up by a lower emitting unit at the
plant, by a plant outside of the control region that
would not be affected by the program, or by a genera-
tor in another sector of the power industry (e.g., a
new gas-fired unit). In the NBR if generation shifts to
units upwind of the control area with high rates of
NOX emissions, ozone would be transported into the region and the effective-
ness of the program would be reduced. In 2003, with only a portion of the NBP
states complying with the program, the potential for this type of shifting was
a concern.
Heat input is the heat derived from the combustion of fuel in electric generation.
It is a way to track utilization of affected units. Heat input levels from affected
sources in the OTC states increased between 2002 and 2003 without the addi-
8 Farrell, Alex. [9003] Temporal Hotspots in Emission Trading Programs: Evidence from the Ozone Transport
Commission's NOX Budget. Presented at Market Mechanisms and Incentives: Applications to Environmental
Policy Conference. Washington, D.C.
Daily Ozone Season Emission Trends
for Former OTC Trading States Only*
2001 2002 2003
Average Daily Emission
Figure 10
Emissions data for NBP units show that daily emis-
sions have been reduced, even on the days with the
highest emissions.
Source: EPA
For the 2002 ozone season, OTC sources reported
1.6 million mmBtu of heat input. In 2003, that num-
ber increased to 1.7 million mmBtu. By comparison,
heat input for Acid Rain Program units in non-OTC
states declined from 4.9 million mmBtu in 2002 to
4.7 million mmBtu in 2003.
14
-------
IMOX Allowance Reconciliation Summary
in 2003 (OTC States Only)
tion of a significant number of sources. This indicates that, on average, sources in
the OTC region were able to increase their generation while still complying with
the NBR Meanwhile, in the non-OTC NBP states, total heat input levels declined
between 2002 and 2003 (based on data available for Acid Rain Program units
only). While some sources in the OTC may have shifted their generation to
sources in non-OTC states in response to the first year of the NBP, it appears that
this generally was not a preferred compliance strategy. Sources in the OTC
appear to have achieved the emission reductions from 2002 levels through
means other than reducing utilization of affected sources and increasing power
generation outside the region.
Sources Achieved a High Rate of Compliance
in 2003
Under the NBP, sources must hold sufficient allowances to cover their ozone season
emissions each yean Sources can maintain the allowances in compliance accounts
[established for each unit) or in an overdraft account
[established for each facility).10 The overdraft account
allows greater flexibility in "bubbling" between units,
managing banked allowances from previous years,
managing transferred allowances from other sites,
or managing allowances purchased from other NBP
participants. The sources have a two-month window
after the end of the control period to move
allowances between accounts [and buy or sell addi-
tional allowances) so that they can ensure their
emissions do not exceed allowances held. Once that
period ends, allowances may not be transferred into
or out of these accounts while EPA reconciles emis-
sions with allowance holdings and identifies the
appropriate allowance deductions from the accounts
for program compliance.
Nearly all of the NBP sources that participated in
2003, both EGUs and industrial units, held sufficient
allowances to cover their emissions at the time that
EPA performed reconciliation. There were seven
sources that had allowance deficiencies [a total of
75 allowances). In cases where the source does not
hold enough allowances to cover their emissions, the
program requires a penalty deduction [3 allowances for each excess ton of emis-
sions) from these sources' allocations for the next control period. Figure 11
summarizes the allowance reconciliation process for 2003.
10 New Jersey does not use overdraft accounts.
Total Allocated 2003 Allowances
Allowances Held in Compliance and
| Overdraft Accounts
Allowances Held in Other Accounts*
Allowances Deducted for 2003 Emissions
Banked Allowances* *
Allowances Held in Compliance and
Overdraft Accounts
Allowances Held in Other Accounts*
Penalty Allowances Deducted* * *
(from future year allowances)
162,152
148,938
13,214
133,659
28,493
15,279
13,214
225
* Other Accounts refer to general accounts in the NOX
Allowance Tracking System (NATS) that can be held by any
source, individual or other organization, and state accounts.
** Does not reflect take back of 1,315 allowances by
Pennsylvania for underutilization of specific sources.
* * * These penalty deductions are made from future vintage
year allowances, not 2003 allowances.
Figure 11
Source: EPA
15
-------
Market Activity, Banking,
and Compliance Costs
At the start of a cap and trade program, it is worthwhile to evaluate how different
components of the program are beginning to perform because a new market is
becoming established, new sources are beginning to comply with the program, and
many sources are facing more stringent requirements than ever before. This sec-
tion examines some of these features, including how the NOX allowance market is
maturing, how sources are complying with the monitoring requirements, and what
types of control technologies sources are beginning to install to meet long-term
program requirements.
A Healthy Market in IMOx Allowances Continues
under the IMOx Budget Trading Program
Allowance transfer activity can involve three main types of transfers:
Transfers to or from the state as allowance allocations or allowance
surrenders;
Transfers within a company or between related entities (holding company
transfers to an operating subsidiary, for example); and
Transfers between separate economic entities. These transfers are cate-
gorized broadly as "economically significant trades."
In 2003, economically significant trades represented approximately 40 percent of
the total transfers between entities other than a state. The economically signifi-
cant trades provide the strongest indicator of true market activity because they
represent an actual exchange of assets between unaffiliated participants.
Since 2003 is the first year of the NOX Budget Trading Program (NBP), there are
no previous data for all participating states to which the volume of trades can be
accurately compared. The 2003 trading activity was higher than any year under the
Ozone Transport Commission (OTC) program, and EPA expects a significant
increase in 2004 as more sources comply with the program. As in the OTC trading
program, industrial sources have actively traded allowances, and industrial sources
represent about 8 percent of the economically significant trades at this time.
.1
16
-------
Vintage Year IMOX Allowance Prices by Month of Sale
Figure 12
NBP allowance prices have fallen sharply after some early price spikes.
Source: Evolution Markets, LLC and Cantor Environmental Brokerage
The price for NOX allowances continues to fluctuate as companies evaluate
ongoing trends in control installations, energy demand, and other factors that
affect the overall costs of control under the NBP. Recent prices are down appre-
ciably from early 2003 (see Figure 12). This suggests that, as the program
progresses and the uncertainty of allowance availability decreases, further price
reductions may occur. This result is consistent with price behavior observed dur-
ing implementation of the OTC trading program.
There may have been other factors that account for the drop in allowance
prices. Uncertainty of natural gas prices may have led to higher allowance
prices early in 2003. Prices then may have dropped due to increases in the sup-
ply of allowances. Extra allowances were available because CSP allowances were
given for early reductions and because the start of compliance in the 2004 sea-
son was delayed from May 1 to May 31
Banking Occurred in 2003 but Did Not
Trigger Flow Control
Under the NBP, banking provisions allow companies to decrease emissions
more than required early in the program, and save unused allowances for
future use. This creates an economic incentive for sources to achieve deeper
reductions early in the program. Banking allows for earlier environmental and
health benefits and provides a pool of allowances available to address unexpect-
ed events or smooth the transition into deeper emission reductions.
17
-------
If sources use a large number of banked allowances in one yean the elevated
emissions could potentially reduce the environmental effectiveness of the NBR
The NBP's "progressive flow control" provisions were designed to discourage
extensive use of banked allowances in a particular ozone season. Flow control is
triggered when the total number of allowances banked for all sources exceeds
10 percent of the total budget for the next year. When flow control is triggered,
EPA calculates the flow control ratio by dividing 10 percent of the total budget
by the number of banked allowances (a larger bank will result in a smaller flow
control ratio). The resulting flow control ratio indicates the percentage of
banked allowances that can be deducted from a source's account in a ratio of
one allowance per ton of emissions. The remaining percentage of banked
allowances, if used, must be discounted and deducted at a rate of two
allowances per one ton of emissions.
Because emissions were below allowable levels in the
first year of the NBR participating sources banked
over 28,000 allowances, nearly 18 percent of the
allowances they were allocated in 2003. Banking of
greater than 10 percent of allowances would normally
trigger flow control. However, in 2003, only the OTC
states participated in the NBP control requirements.
Given the entry of many additional states in 2004, the
overall regional budget grew substantially, above
500,000 tons. Therefore, the number of allowances in
the bank is less than 10 percent of the budget and flow control will not apply in
2004. If the broader universe of sources bank a comparable percentage of
allowances in 2004, however, flow control is likely to be triggered in 2005.
Nearly All Sources in NBP States Began
Monitoring and Reporting in 2003
NBP units are required to comply with monitoring provisions specified in the
Code of Federal Regulations (40 CFR Part 75, subpart H). The original focus of
Part 75 was on electric generating units (EGLJs), but EPA has broadened the
rule over time to encompass industrial combustion units as well. The industrial
units affected under the NBP have successfully met these monitoring require-
ments as part of their NBP compliance efforts.
Several options are available for NBP units to meet their monitoring require-
ments. These options are based on the type of unit, the type of fuel combusted,
its operating status, and its level of emissions. NBP units are generally required
to use a NOX continuous emission monitoring system (OEMS). OEMS sample,
analyze, and directly measure flue gas components on an ongoing basis. In addi-
tion to measuring NOX concentration, units also must measure heat input to
Why No Flow Control in 2004?
Total number of banked allowances/Total NOX
Budget for 2004 season =
28,493/506,068 = 0.06
Because ratio is < 0.10, flow control is not triggered.
18
-------
Use of Monitoring Methods as a Percent
of Total (number of units]
Low Mass
Emissions Method
Appendix E
Figure 13
The vast majority of NBP units use NOX Continuous
Emission Monitoring Systems [GEMS] to comply with
the monitoring requirements.
Source: EPA
calculate NOX mass. To calculate heat input, any
unit can use a stack flow GEMS, but oil- and gas-
fired units instead can use fuel flow meters under
Part 75. In addition to the basic NOX GEMS option,
alternative methods of quantifying NOX emissions
are available for certain types of units or for moni-
toring systems that meet specific criteria,
including:
Part 75, Appendix E, which may be used only by
gas and oil-fired peaking units (i.e., units that
operate principally when electricity demand is
at its highest). Under Appendix E, the NOX emis-
sion rates and the heat input rate for the
peaking unit are determined at a minimum of
four loads covering the unit's operating range,
and the test results are used to establish a
correlation curve. Then, when the unit is run-
ning during the ozone season, NOX emission rates are estimated based on
the values on the curve that correspond to the unit's measured heat input
rates. These Appendix E units all use fuel flow meters to measure heat input
under Appendix D so that they do not use any GEMS.
The low mass emissions (LME) methodology in section 75.19, which allows
certain small or infrequently-operated gas and oil-fired units to use conserva-
tive fuel-specific default emission rates and estimates of hourly heat input to
calculate the hourly NOX emissions.
Other monitoring alternatives approved by EPA on a case-by-case basis
[subpart E).
As Figure 13 shows, of all the NBP units that are currently operating and that
have submitted a monitoring plan, the majority (71 percent) use a NOX GEMS to
comply with the monitoring requirements (these units also use either a stack
flow GEMS or Appendix D fuel flow meters to calculate heat input). Nine percent
use Appendix E (with fuel flow meters), and 20 percent use the LME methodolo-
gy. Less than 1 percent (only 4 units) currently use an approved alternative
monitoring system under subpart E.
19
-------
Source: EPA
In general, although many units use the non-CEMS
monitoring options, the highest emitting sources tend
to use GEMS. About 96 percent of all ozone season
emissions can be attributed to units that use GEMS as
a monitoring method (see Figure 14).
Sources Are Investing in IMOX
Controls fc Jnits
To meet the emission reduction targets of the NBR
sources can choose from a variety of compliance
options. These options include decreasing generation
from units that emit NOX, modifying the basic combus-
tion process to control the formation of NOX, optimizing
boiler operation to minimize NOX production, using add-
on controls, or purchasing allowances from other
market participants. Sources can use any one or a com-
bination of these options in a way that best fits their own circumstances.
To meet the NOX emission limits of the Acid Rain Program, many electric gener-
ating units installed combustion controls, including low NOX burner and overfire
air technologies, which modify the combustion process to reduce formation of
NOX from the nitrogen present in the boiler combustion air and fuel. Advances in
combustion control technologies continue to provide a cost-effective means of
reducing emissions even further for many units.
Add-on control technologies, such as selective catalytic reduction (SCR) or selec-
tive non-catalytic reduction (SNCR), are frequently applied for NOX control. SCR
is typically used on larger units in the power sector that can achieve significant
emission reductions in a highly cost-effective way. SCR and SNCR are control
technologies that achieve NOX reductions by injecting ammonia (or urea for
SCR) into the flue gas within or downstream of the combustion unit to react
with NOX, forming nitrogen and water. SCR uses a catalyst to improve the effi-
ciency of NOX removal and to allow reactions to occur in a lower temperature
range. For units that cannot use other methods to control NOX, reburning of
gas or coal is also an option. In this technique, gas or coal is injected down-
stream of the primary combustion zone to remove NOX.
EPA analyses in support of recent NOX reduction initiatives assume a 90 per-
cent reduction efficiency for SCR on coal-fired boilers (down to O.OB Ib/mmBtu]
and an 80 percent reduction efficiency for oil and gas units. For SNCR, EPA
assumes a 35 percent reduction for coal-fired boilers and a 50 percent reduc-
tion for oil and gas units."
11 Documentation of EPA Modeling Applications (v2.1 ] Using the Integrated Planning Model, U.S. EPA,
EPA 430/R-02-004, March 9009.
20
NBP Ozone Season IMOX Emissions Based
on Monitoring Methodology
4%
-------
Sources report pollution control information, including installation dates, in moni-
toring plans submitted to EPA. EPA examined these data to determine which
units had installed controls. While it is difficult to iso-
late the reason that a source installed a control, EPA
assumed that most, if not all, installations in the last
few years that were not in response to other pro-
grams (such as New Source Review permitting for new
facilities or the Ozone Transport Commission trading
program) were likely to be in response to the NOX
Budget Trading Program. Based on that review, there
appear to be 75 coal-fired units that report using SCR controls to meet the
NBP requirements. Nineteen coal-fired units (only 4 of which are industrial units]
appear to have installed SNCR for the NBP. Most of this activity has been in
states outside the OTC region, which will require the most significant reductions
to meet the NBP requirements. However, since October 2002, when the OTC
program was replaced by the NBP, sources in the OTC states have installed
SCR controls on 5 units with approximately 4,300 MW capacity, and SNCR on 5
units with about 300 MW capacity. These data indicate that the implementation
of the NBP appears to have been an impetus for many units to reduce their NOX
emissions through the use of add-on controls, especially in the states where sig-
nificant reductions are needed to comply with the NBP.
The combined megawatt capacity of the add-on
control installations implemented to meet the NBP
requirements is over 47,000 MW (out of over
270,000 MW capacity for affected EGUs in the
NBP region].
J
-------
Levels
asini IMOx
Reductions Will Help
Meet the Ozon
Ozone Levels Have Been Decreasing In Most
Areas since 1990
EPA released a report on ozone trends in April 2004, finding that ozone levels nation-
wide were lower in 2003 than they have been since 1980 (The Ozone Report:
Measuring Progress through 2003, www.epa.gov/airtrends/ozone.html). EPAs
Ozone Report concluded that ozone improvements in 2003 were primarily due to
favorable weather conditions across many parts of the nation. In addition, national
NOX and volatile organic carbon (VOC) emissions were at their lowest levels since
1970, due to successful programs controlling NOX and VOCs.
Figure 15 shows national trends
in the fourth highest daily maxi-
mum 8-hour ozone concen-
tration (ppm, parts per million]
from 1990 to 2003. Nationally,
this measure of ozone exposure
has been reduced by 9 percent
since 1990.
In the East, many metropolitan
areas have exhibited an overall
improvement in ozone levels
since 1990. In most areas, a
temporary increase in ozone lev-
els occurred during the
mid-1990s, but this increase
was followed by decreases in
ozone levels beginning in 1998.
The improvement in ozone levels
in the late 1990s corresponds
temporally with reductions in
NOX emissions from stationary
sources (mainly through the
annual NOX requirements under
National 8-Hour Ozone Air Quality Trend, 1990-2003, Based
on 3-Year Rolling Averages of Annual Fourth Highest Daily
Maximum Ozone Concentrations
o.ao-1
480 Monitoring Sites
90th percentile
National Standard
Mean
10th percentile
1990-2003: -9%
90 91 92 93 94 95 96 97 98 99 00 01 02 03
Year
Source: EPA
22
-------
the Acid Rain Program and the ozone-season NOX
reductions in the OTC states), along with the NOX and
VOC reductions from mobile sources that occurred
during this time period.
A closer look at trends in measured ozone values on a
regional level shows differences in progress made
since 1990 in the NOX Budget Trading Program
[NBP] states. Figure 16 shows trends in fourth high-
est daily maximum 8-hour ozone concentration in EPA
Regions that include NBP states. The greatest
progress in reducing ozone concentrations since
1990 in the NBP states was achieved in the
Northeast and Mid-Atlantic states (Regions 1 through
3). This may be partly due to the NOX reductions
achieved by the Ozone Transport Commission (OTC]
states from 1999 through 2003. Region 1 (13 per-
cent decrease from 1990 levels] and Region 2 (11
percent decrease] reduced ozone concentrations
greater than the national average, while Region 3 (7
percent decrease] has significantly decreased ozone
concentrations. Regions 4 and 5 had reductions in
ozone concentrations of B percent but had lower
ozone levels in 1990 and, therefore, less room for
improvement. Even small improvements in ozone concentrations, however; are
expected to result in substantial benefits to public health when a large population
is exposed.
Additional IMOX Reductions Will Help States
Meet the Ozone Standards
Despite improvements in ozone air quality in many areas of the country, ozone
continues to be a pervasive air pollution problem, with nearly 159 million people
still living in 474 counties across the nation that are in nonattainment areas
that do not meet the 8-hour ozone standard. The reductions anticipated under
the NBP will help reduce emissions of NOX and improve air quality.
New national mobile source regulations will also help local areas meet the 8-
hour ozone standard by reducing NOX from heavy-duty diesel engines, highway
vehicles, and other mobile sources. Finally, to address the regional component
of the residual ozone nonattainment problem, as well as the year-round prob-
lems of fine particles, regional haze, and acid deposition, EPA recently proposed
the Clean Air Interstate Rule, which by 2015 would reduce annual NOX emis-
sions from the power industry in 29 eastern states and the District of
Columbia by approximately 64 percent from 2002 levels.
Trend in Fourth Highest Maximum 8-Hour
Ozone Concentration [ppm] by EPA Region
[1990-2003]
.091
Source: EPA
-------
|U4
lOOs
When fully implemented, the NOX Budget Trading Program (NBP) is expected to achieve
a significant reduction in ozone season NOX emissions across much of the eastern U.S.
In 2003, affected sources already had reduced ozone season emissions by over one
million tons from the estimated 1990 baseline levels and by over 400,000 tons from
2000 ozone season levels, even though the first control period in many states was not
until 2004. In the Ozone Transport Commission (OTC) states where 2003 represented
the first control period, sources emitted almost 20 percent below the required 2003
budget levels and more than 30 percent below their emissions in 2002. Of the total
affected population of approximately 1,000 units, all but seven were in compliance.
These achievements occurred despite a small increase in total heat input (plant utiliza-
tion) in the affected OTC region.
Additional sources joined the program on May 31, 2004, and most of these sources
already have at least one year's experience with NBP monitoring and reporting.
The review of control data shows that many sources have been actively engaged in
installation of pollution control equipment to achieve additional emission reductions. The
NOX allowance market remains active and allowance prices appear to have stabilized
from early price spikes. Meanwhile, ozone levels have decreased in the past two
decades, although ozone nonattainment remains a persistent problem. The NBP, along
with additional control programs being proposed or implemented, should help address
this problem. EPA will continue to evaluate all of these issues as additional NBP states
begin the control requirements of the program in 2004.
-------
-------
o
o
co
CD
CO
CO
!™
O
0)
>
CL
b
0)
Q.
CD
_Q
CD
O
_O
to
O
Q.
CD
O
CD
Q.
O
LO
to
CO
_CD
en
c
-
o
O
to
JZ
CD
Q.
CO
Q.
O
~O
CD
Q_
------- |