Order Code RL33799
Climate Change: Design Approaches for a
Greenhouse Gas Reduction Program
Updated June 2, 2008
Larry Parker
Specialist in Energy Policy
Resources, Science, and Industry Division

Climate Change: Design Approaches for a
Greenhouse Gas Reduction Program
Summary
With the passage of the 2005 Sense of the Senate climate change resolution
calling on the Congress to enact a mandatory, market-based program to slow, stop,
and reverse the growth of greenhouse gases, the issue of related costs has taken on
increased importance. Indeed, the resolution itself states that the program should be
enacted at a rate and in a manner that “will not significantly harm the United States
economy” and “will encourage comparable action” by other nations. Facets of the
cost issue that have raised concern include absolute costs to the economy,
distribution of costs across industries, competitive impact domestically and
internationally, incentives for new technology, and uncertainty about possible costs.
In general, market-based mechanisms to reduce greenhouse gas emissions, the
most important being carbon dioxide (CO ), focus on specifying either the acceptable
2
emissions level (quantity) or the compliance costs (price), and allowing the
marketplace to determine the economically efficient solution for the other variable.
For example, a tradeable permit program sets the amount of emissions allowable
under the program (i.e., the number of permits available limits or caps allowable
emissions), while allowing the marketplace to determine what each permit will be
worth. Likewise, a carbon tax sets the maximum unit cost (per ton of CO2
equivalent) that one should pay for reducing emissions, while the marketplace
determines how much actually gets reduced. In one sense, preference for a carbon
tax or a tradeable permit system depends on how one views the uncertainty of costs
involved and benefits to be received.
Market-based mechanisms attempt to address the cost issue by introducing
flexibility into the implementation process. The cornerstone of that flexibility is
permitting sources to decide for themselves their appropriate implementation strategy
within the parameters of market signals and other incentives. That signal can be as
simple as a carbon tax or comprehensive credit auction that tells the emitter the value
of any reduction in greenhouse gases, to a credit marketplace that is constrained by
a ceiling price (safety valve) and includes incentives for new technology. As
illustrated here, the combinations of market mechanisms are numerous, allowing
decision makers to tailor the program to address specific concerns.
In a sense, the options discussed here represent a continuum between
alternatives focused on the price side of the equation (e.g., carbon taxes) through
hybrid schemes (e.g., safety valves) to alternatives focused on the quantity side (e.g.,
banking and borrowing). They are tools to assist in the assessment of potential
greenhouse gas reduction approaches, leaving any policy decision on balancing the
price-quantity issue to the ultimate decision makers.

Contents
Introduction: The Price Versus Quantity Debate . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Five Dimensions of the Cost Issue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Approaches for Addressing Cost Concerns . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Tonnage Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Dynamic Tonnage Target . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Expand Supply Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
Carbon Tax . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
Timetable Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
Economic-Based Circuit Breaker . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
Technology-Based Timetable . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
Technique Options . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Banking and Borrowing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
Auctioning Permits . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
Safety Valve . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
Illustrative Approaches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
Addressing Costs Through Market Mechanisms:
Resolving the Price-Quantity Issue . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
Appendix A. Summary of Selected Options To Address
Cost Uncertainty of Greenhouse Gas Reduction Programs . . . . . . . . . . . . . 24
List of Tables
Table 1. 2005 U.S. Greenhouse Gas Emissions . . . . . . . . . . . . . . . . . . . . . . . . . . 4

Climate Change: Design Approaches for a
Greenhouse Gas Reduction Program
Climate change has been a continuing policy issue since the United States
ratified the 1992 United Nations Framework Convention on Climate Change
(UNFCCC). An integral part, and sometimes driving part, of the ensuing debate has
been the issue of cost — in several manifestations.1 For the George W. Bush
Administration, the Kyoto Protocol was “fatally flawed in fundamental ways,”
including requiring compliance with mandates that “would have a negative economic
impact with layoffs of workers and price increases for consumers.”2 This concern
about cost can also be seen in the Senate’s most recent resolution on climate change
in 2005 (S.Amdt. 866). Echoing the language of its 1997 resolution (S.Res. 98) on
the same subject, the 2005 Sense of the Senate resolution on climate change declared
that a mandatory, market-based program to slow, stop, and reverse the growth of
greenhouse gases3 should be enacted at a rate and in a manner that “will not
significantly harm the United States economy” and “will encourage comparable
action” by other nations.4 Facets of the cost issue that have raised concern include
absolute costs to the economy, distribution of costs across industries, competitive
impact domestically and internationally, incentives for new technology, and
uncertainty about possible costs.
Because a stalemate has persisted on strategies to control greenhouse gas (GHG)
emissions, particularly because of cost uncertainties, attention has increasingly
focused on options to address these concerns and to move the debate forward. These
options range from incremental mechanisms within a tradeable permit program, such
as banking and borrowing of credits, which minimally affect overall emissions
reduction targets, to more fundamental proposals, such as a carbon tax, which would
take climate change policy in a new and somewhat uncharted direction.
This paper explores these options to address the cost issue in four parts. First,
the basic economic tradeoff between controlling the quantity of GHG emissions and
the program’s compliance costs is introduced and explained. Second, the five
1 For an analysis of federal policy and congressional debate since ratification of UNFCCC,
see CRS Report RL30024, Global Climate Change Policy: Cost, Competitiveness and
Comprehensiveness
, by Larry B. Parker and John E. Blodgett.
2 President George W. Bush, President Bush’s Speech on Global Climate Change (June 11,
2001).
3 The six gases recognized under the Kyoto Protocol are carbon dioxide (CO ), methane
2
(CH ), nitrous oxide (N O), sulfur hexafluoride (SF ), hydrofluorocarbons (HFC), and
4
2
6
perfluorocarbons (PFC).
4 S.Amdt. 866, passed by voice vote after a motion to table failed 43-54, June 22, 2005.

CRS-2
dimensions of the cost issue that have arisen so far in the climate change debate are
identified and discussed. Third, a representative sample of proposed approaches to
address cost concerns is compared and analyzed according to the five cost
dimensions identified previously. Finally, the proposed options are summarized and
opportunities to combine or merge different approaches are analyzed. The paper
does not provide a detailed discussion of allocation and implementation issues that
creating a market-based mechanism (particularly a cap-and-trade program) would
entail.
Introduction: The Price Versus Quantity Debate
In general, market-based mechanisms to reduce GHG emissions, the most
important being carbon dioxide (CO ), focus on specifying either the acceptable
2
emissions level (quantity) or the compliance costs (price) and allowing the
marketplace to determine the economically efficient solution for the other variable.
For example, a tradeable permit program sets the amount of emissions allowable
under the program (i.e., the number of permits available limits or caps allowable
emissions), while permitting the marketplace to determine what each permit will be
worth. Likewise, a carbon tax sets the maximum unit cost (per ton of CO2
equivalent) that one should pay for reducing emissions, while the marketplace
determines how much actually gets reduced. In one sense, preference for a carbon
tax or a tradeable permit system depends on how one views the uncertainty of costs
involved and benefits to be received.
For those confident that achieving a specific level of CO reduction will yield
2
significant benefits — enough so that even the potentially very high end of the
marginal cost curve does not bother them — a tradeable permit program may be most
appropriate. CO emissions would be reduced to a specific level, and in the case of
2
a tradeable permit program, the cost involved would be handled efficiently, though
not controlled at a specific cost level. This efficiency occurs because through the
trading of permits, emissions reduction efforts focus on sources at which controls can
be achieved at least cost.
However, if one feels uncertain of the environmental benefits of a specific level
of reduction and anxious about the downside risk of substantial control costs to the
economy, then a carbon tax may be most appropriate. In this approach, the level of
the tax effectively caps the marginal cost of control that affected activities would pay
under the reduction scheme, but the precise level of CO reduction achieved is less
2
certain. Emitters of CO would spend money controlling CO emissions up to the
2
2
level of the tax. However, because the marginal cost of control among millions of
emitters is not well known, the overall emissions reductions for a given tax level on
CO emissions is subject to some uncertainty.
2
Hence, a major policy question is whether one is more concerned about the
possible economic cost of the program and therefore willing to accept some
uncertainty about the amount of reduction received (i.e., carbon taxes); or one is
more concerned about achieving a specific emissions reduction level with costs
handled efficiently, but not capped (i.e., tradeable permits).

CRS-3
A model for a tradeable permit approach is the sulfur dioxide (SO ) allowance
2
program contained in Title IV of the 1990 Clean Air Act Amendments (42 U.S.C.
7651). Also called the Acid Rain Program, the tradeable permit system is based on
two premises. First, a set amount of SO emitted by human activities can be
2
assimilated by the ecological system without undue harm. Thus the goal of the
program is to put a ceiling, or cap, on the total emissions of SO rather than limit
2
ambient concentrations. Second, a market in pollution licenses between polluters is
the most cost-effective means of achieving a given reduction. This market in
pollution licenses (or allowances, each of which is equal to 1 ton of SO ) is designed
2
so that owners of allowances can trade those allowances with other emitters who
need them or retain (bank) them for future use or sale. Initially, most allowances
were allocated free by the federal government to utilities according to statutory
formulas related to a given facility’s historic fuel use and emissions; other allowances
have been reserved by the government for periodic auctions to ensure market
liquidity.
There are no existing U.S. models of an emissions tax, although five European
countries have carbon-based taxes.5 The closest U.S. example is the tax on ozone-
depleting chemicals (ODCs). To facilitate the phaseout of ODCs under the Montreal
Protocol and subsequent amendments, the United States imposed a tax on specific
ODCs in 1990. This tax was designed to supplement the allowance trading program
that the EPA had designed to implement the international agreements. Several
activities trigger the tax, including the production and/or importation of the
chemicals, or the importation of products that contain them or used them in their
production processes. In addition, inventories of certain ODCs held on January 1 of
each year are subjected to a “floor stocks tax.”6
Five Dimensions of the Cost Issue
Five dimensions of costs associated with reducing GHG emissions are discussed
in this section: (1) absolute costs, (2) distribution of costs, (3) long-term costs, (4)
price signal and stability, and (5) uncertainty of costs.
The absolute costs of a GHG reduction program are a function of the interplay
among the tonnage reduction required, the timetable imposed on that reduction, and
the techniques available and used to achieve that reduction (the “three Ts”).
Variables involved with the tonnage requirement include the magnitude and firmness
of the reduction requirement and the number of gases and sectors involved in the
program. Variables involved with the timetable include its length and number of
phases, along with the number and extent of any deadline extensions allowed and on
what basis. Finally, variables involved with techniques include promotion and
availability of new technology, the degree of flexibility permitted in complying with
5 Finland, the Netherlands, Sweden, Denmark, and Norway.
6 For CFC-11 and 12, the current (2006) tax is $10.30 per pound. The floor stocks tax is
$0.45 per pound (2006). For more specifics on the current tax level, see IRS Form 6627,
Environmental Taxes.

CRS-4
the program, and any ceiling on compliance costs. All these program design
parameters influence the absolute cost of the program and the timing and extent of
any benefits received.
A second concern with costs is their distribution across the various sectors of
the economy. As indicated by Table 1, GHG emissions are spread throughout the
economy, with about 81% emitted by the electric power, transportation, and industry
sectors. Restricting participation by any group could increase the absolute cost of the
program and would certainly increase the costs to the remaining participants.
However, numerous rationales have been put forward to justify excluding one group
or sector from a reduction requirement, or to provide some other special
consideration. Rationales offered include a sector or industry’s concern about (1)
international competitiveness, (2) lack of cost-effective control options, (3) inability
to make necessary capital investments, (4) economic disruption, (5) credit for
previous efforts that reduced emissions, and (6) the “minor” contribution that
industry or sector makes to the overall problem. It is the multitude of such variables
that make constructing an acceptable reduction allocation scheme very difficult.
Table 1. 2005 U.S. Greenhouse Gas Emissions
Million Metric Tons of
Economic Sector
CO equivalent
Percentage of Total
2
Electric power industry
2,430
33.7%
Transportation
2,009
27.9%
Industry
1,353
18.8%
Agriculture
595
8.3%
Commercial
431
6.0%
Residential
381
5.3%
Total
7,199
100.0%
Source: U.S. Environmental Protection Agency, Inventory of U.S. Greenhouse Gas Emissions and
Sinks: 1990-2005
, EPA 430-R-07-002 (Washington, DC), p. ES-14
Note: The total does not include 62 million metric tons from U.S. territories.
A third concern is the long-term cost considerations of a GHG reduction
program. Climate change policy has to be thought of in decades, not years.
Ultimately, a successful climate change program would involve a long-term
transition to a less carbon-emitting economy. Generally, studies that indicate the
availability and cost-effectiveness of emerging new technologies to achieve this
transition include an economic mechanism to provide the necessary long-term price
signal to direct research, development, demonstration, and deployment efforts.7
Developing such a price signal involves variables such as the magnitude and nature
7 For example, see Interlaboratory Working Group, Scenarios for a Clean Energy Future,
ORNL/CON-476 (November 2000).

CRS-5
of the market signal, and the timing, direction, and duration of it. In addition, studies
indicate combining a sustained price signal with public support for research and
development efforts is the most effective long-term strategy for encouraging
development of new technology.8 As stated by Morgenstern: “The key to a long term
research and development strategy is both a rising carbon price, and some form of
government supported research program to compensate for market imperfections.”9
A fourth consideration is the stability of the price signal in whatever form it
takes (e.g., allowance prices, carbon taxes, auction prices). A stable and reliable
signal is necessary to minimize economic disruption and to encourage new
technology. Experience with existing emissions markets suggests that short-term
price spikes and troughs occur that have at least short-term economic effects, either
disrupting the market (in the case of high prices) or discouraging new technology (in
the case of low prices). Causes of this volatility can include (1) lack of trading
volume, (2) illiquidity in the market, (3) external events, and (4) regulatory
uncertainty. History with previous emissions trading programs suggests that if a
greenhouse gas program is based on a market-based implementation strategy, the
inclusion of flexibility mechanisms to ensure reasonable market stability is
desirable.10
A final cost consideration is the cost uncertainty presented by the wide range of
projected costs of GHG reduction. To the extent one understands the variables that
create the range presented by different forecasting models, one can design a program
to address those variables. Projected costs of a proposed greenhouse gas reduction
program will differ among models, based on the various economic and technological
assumptions either embedded in the particular model’s processes (endogenous
variables) or assumed externally and inserted into the model. Weyant has identified
five assumptions that explain many of the differences in greenhouse gas reduction
program cost estimates11:
8 For example, see CERA Advisory Service, Design Issues for Market-based Greenhouse
Gas Reduction Strategies; Special Report
(February 2006), p. 59; Congressional Budget
Office, Evaluating the Role of Prices and R&D in Reducing Carbon Dioxide Emissions
(September 2006).
9 Richard D. Morgenstern, Climate Policy Instruments: The Case for the Safety Valve
(Council on Foreign Relations, September 20-21, 2004), p. 9.
10 For example, see Dallas Burtraw, David A. Evans, Alan Krupnick, Karen Palmer, and
Russell Toth, Economics of Pollution Trading for SO and NOx (Resources for the Future,
2
March 2005); David Harrison, Jr., Ex Post Evaluation of the RECLAIM Emissions Trading
Program for the Los Angeles Air Basin
(Organization for Economic Co-operation and
Development, January 21-22, 2003); and Andrew Aulisi, Alexander E. Farrell, Jonathan
Pershing, and Stacy VanDeveer, Greenhouse Gas Emissions Trading in U.S. States:
Observations and Lessons from the OTC NOx Budget Program
(World Resources Institute,
2005).
11 John P. Weyant, An Introduction to the Economics of Climate Change Policy (Pew Center
on Global Climate Change, July 2000).

CRS-6
! Basecase projections of future GHG emissions and climate damages.
! The specifics of the reduction program examined (particularly the
amount of flexibility permitted in complying with its mandates).
! How dynamic the model is, representing substitution possibilities by
producers and consumers, including the turnover of capital
equipment.
! How the rate and processes of technological change are modeled.
! How benefits are modeled.
Figure 1, below, illustrates how these and other variables (such as type of model
used) can influence the estimated costs of climate change legislation. Measured by
impact on GDP, the figure indicates impacts generally ranging from a positive 2%
increase in GDP to a 4% decrease. Interestingly, the variables used in projecting cost
and benefits are sufficiently robust to obscure a strong correlation between cost and
reduction requirements.
The range indicated also reflects the perspectives and parameters assumed by
the forecast authors. In a previous report, CRS noted that cost analyses are
influenced by the perspective (or lens) through which one views the problem.12
Analysts viewing climate change policy through a technological perspective see it as
an impetus for improved efficiency through technology improvements in the
economy, consistent with concepts such as life-cycle costs. Analysts viewing policy
through an economic lens work through the boundaries of market economics and
cost-benefit considerations. Finally, analysts viewing the issue through an ecological
lens look to the benefits of controlling greenhouse gases and are suspicious of
“baseline” scenarios that suggest that “business as usual” is an acceptable yardstick
from which to measure policy changes. Each of these lenses implies fundamentally
different ways of assessing policy actions and modeling potential costs and benefits.
The quantitative results are cost estimates that range from actual savings to the
economy (from GHG reductions) to substantial costs.
12 See CRS Report 98-738, Global Climate Change: Three Policy Perspectives, by Larry
Parker and John Blodgett.


CRS-7
Figure 1. The Predicted Impacts of Carbon Abatement on the
U.S. Economy (162 Estimates from 16 Models)
Source: Robert Repetto and Duncan Austin, The Costs of Climate Protection: A Guide
for the Perplexed
(World Resources Institute, 1997), p. 12.
Approaches for Addressing Cost Concerns
The following analysis of options to address the cost concerns identified above
is loosely arranged by the focus of the specific option: (1) the tonnage requirement,
(2) the time frame, and (3) the techniques allowed for compliance. It should be noted
that several options examined affect more than one of the “Ts.” Also, the options are
not mutually exclusive — many can be combined to create more refined options.
Tonnage Options
Much of the discussion on GHG reductions has focused on a historic baseline
as the starting point for reductions. Assuming that the emissions inventory for a
specific year is adequate to support a regulatory program (whether market-based or
not), such a baseline is reasonable. Most existing emissions trading programs are
based on a historic baseline with modifications. However, there are options to
calculate a baseline that responds to economic events over time without necessarily
compromising the tonnage cap. Also, the historic baseline can be eliminated in favor
of different methods of achieving specific reductions.

CRS-8
Dynamic Tonnage Target. Another approach to address some of the
concerns identified above is to calculate the tonnage target based on economic or
other indexes or measures rather than strictly on a historic or other static baseline.
For example, the National Commission on Energy Policy recommended that the
tonnage requirements for a GHG reduction program begin with year 2000 emissions,
with the future trajectory of emissions based on the product of a progressively
declining limit on the country’s GHG intensity times projected economic growth.
Over time, the progressively more stringent carbon intensity index would produce
progressively more stringent emissions tonnage caps, despite projected increases in
economic growth. The actual steepness of that path would depend on the rate of
decline in carbon intensity mandated by the program and actual economic growth.
Of course, the dynamic tonnage target could be indexed to just about any relative
variable (e.g., energy prices).
Depending on the specifics of the methodology and measures used in creating
it, the dynamic target could be more responsive to some unforeseen events, such as
substantial economic growth, than a static baseline. At least in the short term, this
could reduce costs and economic disruption if a sharp spike in economic growth were
to occur. In contrast, slower-than-anticipated growth would reduce the available
emissions credits and thereby reduce the potential for “hot air” credits (i.e., credits
“created” by a slowdown in the economy rather than by control efforts).13
By potentially mitigating some effects of a static, historic emissions baseline,
a dynamic tonnage methodology allows flexibility in distributing reductions and the
resulting costs among different sectors of the economy. Growth, GHG intensity,
production, and other variables could be tailored for sectors, states, or regions based
on specific concerns, such as competitiveness. For example, an industry growth
index could be used to calculate reduction requirements rather than an aggregated
index such as GDP. Like most schemes, a dynamic target scheme could completely
exclude some industries, with the obvious result of a shift in cost to the ones
remaining in the program.
The effect of a dynamic target on long-term costs would depend on the slope of
reductions mandated by the program. For example, the recommendation of the
National Commission on Energy Policy called for an annual 2.4% reduction in
allowable GHG intensity increasing to 2.8% annually after 10 years. This declining
curve would be multiplied by a projection of presumably increasing economic
growth. A steeper slope in GHG intensity mandates and/or an overly pessimistic
projection of economic growth would strengthen the need for less carbon-intensive
technology, but at the risk of increasing cost if those technologies did not arrive in
a timely manner. A weak GHG intensity mandate and/or an overly optimistic
projection of economic growth could reduce necessary emissions reductions and
provide a weak incentive for new technology.
13 Vicki Arroyo and Neil Strachan, Addressing the Costs of Climate Change Mitigation,
presented at the Aspen Workshop: A Climate Policy Framework: Balancing Policy and
Politics.

CRS-9
A dynamic target would not necessarily prevent short-term fluctuations in the
price signal, depending on the frequency of adjustments. If a target were based on
macro-economic trends, such as GDP, it would not respond much to short-term or
localized events, such as the 2000-2001 electricity shortage in California. Also, the
mixture of indices with different vectors (e.g., GHG intensity reducing targets while
economic growth is increasing targets) may create some uncertainty in markets
regarding the appropriate price of credits.
Finally, a dynamic target would not increase the certainty of cost estimates.
Uncertainty about the future trajectory of economic growth would be reflected in cost
estimates (just as they are now, with emissions capped at historically determined
levels). Likewise, benefit certainty would not improve for the same reason.
Expand Supply Options. The breadth of options permitted under a
reduction program can have a significant effect on absolute costs. Legislation
introduced in recent Congresses has ranged from programs based on one economic
sector (e.g., electric utilities) and one greenhouse gas (e.g., carbon dioxide) to several
sectors (including opt-in provisions) and all six greenhouse gases covered by the
Kyoto Protocol.14 Also, some proposed programs have included international trading
of emissions credits and biological sequestration offsets among the permissible
means of complying with reduction requirements.15 Some of these options,
particularly international trading and sequestration, have included limits on their
applicability. For example, the Regional Greenhouse Gas Initiative16 has put control
cost triggers (characterized as “safety valves”) on the availability of some supply
options, such as sequestration.
Numerous analyses were done on the impact of global trading after the signing
of the 1997 Kyoto Protocol. For the United States, the cost of complying with the
Kyoto Protocol was estimated at $23-$50 per ton of carbon if global trading were
included, versus $61-$119 if only trading among developed (Annex 1) countries were
permitted. Cost estimates of “No trading” scenarios ranged from $193-$295 per
ton.17 Studies have suggested that, beyond international trading, including non-
14 Carbon dioxide (CO ), methane (CH ), nitrous oxide (N O), sulfur hexafluoride (SF ),
2
4
2
6
hydrofluorocarbons (HFC), and perfluorocarbons (PFC).
15 For more on sequestration approaches, see CRS Report RL33801, Direct Carbon
Sequestering: Capturing and Storing CO
, by Peter Folger.
2
16 The Regional Greenhouse Gas Initiative (RGGI) is an initiative of currently seven
northeastern states to reduce GHG emissions. A signed 2005 memorandum of agreement
(MOU) requires the parties to stabilize and then reduce CO emissions from powerplants,
2
implemented through an allowance-based cap-and-trade program. If the allowance price
rises above $7, offsets from outside the region may be used for compliance purposes at a 1:1
ratio, with the generator able to cover up to 5% of its emissions. (If below $7, such offsets
are discounted 50% and the compliance limit is 3.3% of a generator’s emissions. If the
allowance price exceeds $10, offsets from international projects could be used to cover up
to 20% of a generator’s emissions.) For more information, see [http://www.rggi.org].
17 For a review of these estimates, see CRS Report RL30285, Global Climate Change:
Lowering Cost Estimates through Emissions Trading — Some Dynamics and Pitfalls
, by
(continued...)

CRS-10
carbon dioxide greenhouse gases and sequestration in the supply mix can play an
important and cost-effective role in any climate change program.18
Expanding supply sources could help industries that do not have readily
accessible means of reducing greenhouse gases on their own by providing them with
additional options and making the credit market more liquid. To the extent the
expanded supply sources help create an integrated market with a true market price
for credits, industries could avoid very high compliance costs and lessen the impact
of those costs on their profitability. However, if competitors in other countries do not
have to reduce emissions at all (as is currently the case with the Kyoto Protocol),
competitive disadvantage would remain in some cases.
The degree to which expanded supply options would contribute to a long-term
and stable price signal would depend on how integrated these sources are in the
overall permit market. For example, with the European Trading System (ETS), there
are separate markets for credits created within the 15 members of the European
Union (EU) covered by the EU bubble, credits created by Joint Implementation with
eastern European countries, and credits created via the Clean Development
Mechanism (CDM) with Third World countries.19 One type of credit cannot be
traded for another. The result is a range of credit prices, reflecting the relative risk
and availability of the various credit types. Thus, the long-term signal being
delivered is currently unclear, and may take time to develop. Likewise, substantial
fluctuations in the EU credit market have not been stabilized by the existence of the
other two credit types.
In some ways, expanding supply options may increase the uncertainty of cost
estimates, not only because of disparity in assumed reduction costs, but also in
assumed availability and penetration of the options themselves. For example,
emissions reductions via the Clean Development Mechanism could be substantial
and very cost-effective. However, the mechanism itself creates uncertainty with
respect to availability, as does the willingness of foreign governments to participate.
It is difficult to quantify the effect such an option could have on costs without some
track record, as is slowly being built by the ETS.
Carbon Tax. The most radical approach to controlling costs and addressing
the concerns identified above is to impose a carbon tax in lieu of proposed allowance
17 (...continued)
Larry Parker (available from the author).
18 For example, see John Reilly, Marcus Sarofim, Sergey Paltsey, and Ronald G. Prinn, The
Role of Non-CO2 Greenhouse Gases in Climate Policy: Analysis Using the MIT IGSM
, MIT
Joint Program on the Science and Policy of Global Change, Report No. 114 (August 2004);
MIT Joint Program on the Science and Policy of Global Change, Multi-gas Strategies and
the Cost of Kyoto
, Climate Policy Note 3 (April 2000); Vincent Gitz, Jean-Charles
Hourcade, and Philippe Ciais, “The Timing of Biological Carbon Sequestration and Carbon
Abatement in the Energy Sector Under Optimal Strategies Against Climate Risks,” 27 The
Energy Journal
3 (2006), pp. 113-133.
19 For more on the ETS, see CRS Report RL33581, Climate Change: The European Union’s
Emissions Trading System (EU-ETS)
, by Larry Parker.

CRS-11
trading programs. As discussed in the introduction to this report, under a carbon tax,
the costs are fixed by the legislation and the quantity of emissions reduced becomes
the variable. Carbon taxes are generally conceived as a levy on natural gas,
petroleum, and coal, according to their carbon content, in the approximate ratio of 0.6
to 0.8 to 1.0, respectively. However, the levy would not have to be imposed on the
fuels themselves; proposals have been made to impose the tax downstream at the
point where the fuel is converted into heat and CO . In addition, there is no reason
2
why the tax could not be expanded to include all greenhouse gases in appropriate
carbon equivalents.
A carefully designed carbon tax could potentially address all five of the
concerns identified above. A carbon tax puts a limit on absolute cost by capping the
marginal costs that participants should pay to reduce GHG emissions. Participants
would receive a firm price signal with respect to the upper value of GHG emissions,
and respond in the most cost-effective manner — that is, reduce emissions up to the
cost of the carbon tax and pay the tax on any remaining emissions that are more
expensive to eliminate.
A carbon tax can be tailored to address distributional concern in two ways. The
first would be to exempt, either partly or completely, whatever sectors or industries
were felt to be threatened, either competitively or otherwise, by imposing the tax.
The current tax code provides numerous exemptions from various taxes for a variety
of reasons. However, such an approach would create economic distortions and
complicate the tax structure. The second approach would be to use some of the
revenue generated by the tax to provide appropriate relief to targeted sectors or
industries. This could involve increasing funding for existing programs for such
sectors or industries, or creating new ones. In some ways, this approach might be
more transparent than an approach that involves a potentially complicated tax
structure. These approaches are not mutually exclusive; they could be combined if
considered appropriate.
Likewise, a carbon tax can be employed to address long-term concerns in two
ways. First, the carbon tax would create a long-term price signal to stimulate
innovation and development of new technology. This price signal could be
strengthened if the carbon tax were escalated over the long run, either by a statutorily
determined percentage or by an index (such as the producer price index). Second,
some of the revenue generated by the tax could be used to fund research,
development, demonstration, and deployment of new technology to encourage the
long-term transition to a less-carbon-intensive economy.
A carbon tax’s basic approach to controlling GHG emissions is to supply the
marketplace with a stable, consistent price signal — the fourth cost concern.
Designed appropriately, there would be little danger of the price spikes or market
volatility that can occur in the early stages of a tradeable permit program.
Finally, a carbon tax basically places an upper boundary on projected economic
cost uncertainty. However, it increases uncertainty with respect to environmental
benefits by making emissions reductions a dependent variable. This is the basic
tradeoff that a price-based control system presents. One way that might mitigate the
problem to some extent would be to combine the carbon tax with some form of

CRS-12
quantity controls. As noted earlier, the CFC program attached a tax to its trading
program with beneficial results. However, it is the trading program, not the CFC tax,
that is the primary regime for control. In this manner, a carbon tax would be more
of a revenue raiser than a control regime. A second hybrid would be a “safety valve”
that capped allowance prices such as proposed by the National Commission on
Energy Policy.20 That approach is discussed later in this report. The degree to which
the problem is mitigated (and others created) depends on the interplay between the
quantity control and the carbon tax.
Timetable Options
Similar to the country’s three-plus decades effort to reduce smog, climate
change promises to be an effort measured in decades, not years. Unlike conventional
pollution control efforts, the environmental benefit of mitigating climate change
would come from a reduction in the stock of greenhouse gases that have built up in
the atmosphere for decades, whereas the economic costs of control are related to the
current flow of additional gases into the atmosphere. Thus, in a situation similar to
protecting the stratospheric ozone layer, there would be a substantial delay between
control costs and environmental benefits. Indeed, if short-term reductions in the
stock of greenhouse gases were the focus of climate change policy, control efforts
would be focused on controlling methane, which has a 20-year lifetime, compared
with CO , which has a 200-year lifetime. Likewise, temporary measures, such as
2
biologic sequestration, would be accelerated with the assumption that new
technology would be available in the future to capture the biologically sequestered
carbon dioxide when it is released decades from now.
This situation leads to disputes over how time should be managed under a GHG
reduction program. One argument is that modest cuts (or slowing of the increase)
early, followed by steeper cuts later, is the most cost-effective. Generally, three cost-
related arguments are made in favor of this approach. First, over the long-term,
sustained GHG reductions involve a turnover in existing durable capital stock — a
costly process. If the time frame of the reduction is long enough to permit that
capital stock to be replaced as it wears out, the transitional costs are reduced.
Second, increased time to comply would permit the development and deployment of
new, less carbon-intensive technologies that are more cost-effective than existing
technology. Third, assuming a positive rate of return on current investment, less
money needs to be set aside today to meet those future compliance costs.21
A counter argument to the above focuses on the risks of delay, both in terms of
scientific uncertainty and technology development. In terms of scientific uncertainty,
there is no consensus on what concentration of greenhouse gases should not be
exceeded in order to avoid undesirable climate change. If the stabilization level
needed is relatively low, any delay in beginning reductions could be costly, both
20 The National Commission on Energy Policy, Ending the Energy Stalemate: A Bipartisan
Strategy to Meet America’s Energy Challenges
(December 2005), p. 21.
21 Robert Repetto and Duncan Austin, The Costs of Climate Protection: A Guide for the
Perplexed
(World Resources Institute, 1997), p. 21.

CRS-13
economically and environmentally.22 Secondly, given the sometimes long lead times
for technology development, both a long-term price signal and research and
development funding may have to be initiated quickly to encourage technology
development and deployment in time to hold GHG concentrations to a level that
avoids unacceptable damages. In the same vein, an early signal with respect to
climate change policy is necessary to discourage investment in durable long-lived
(50-60 years) carbon-intensive technologies.23 As stated by Jaccard and
Montgomery:
The window of opportunity for reducing cost implies a need for immediate and
continuing action to develop new low-carbon technologies and to begin shifting
long-lived investment decisions toward alternatives that lower carbon emissions.
Absent these actions, the rapid future emissions reductions included in the
delayed emissions scenario may be more costly than more evenly paced, and
earlier reductions.24
Economic-Based Circuit Breaker. Delaying or suspending compliance
with environmental mandates because of energy and economic reasons is not a novel
idea. The Clean Air Act contains provisions permitting the President, in response to
a petition by an affected state’s governor, to temporarily suspend any part of a state
implementation plan or enforcement of the SO trading program, to address a severe
2
national or regional energy emergency.25 For example, during the 2000-2001
California energy crisis, President Clinton directed all federal agencies to do their
part to assist the state in meeting its electricity demand. For its part, the EPA revised
its guidance on emergency generators to allow backup generators to be used to avert
a power blackout.26 Previously, backup generators could be used only when the
power was actually interrupted. The increased flexibility permitted by the EPA
during the emergency meant more power at the expense of more pollution
(particularly of carbon monoxide and nitrogen oxides).
Likewise, market-based systems are not immune to being suspended if
economic or energy conditions turn severe. A contributing factor in the California
power crisis was the Regional Clean Air Incentives Market (RECLAIM), a credit
trading system for reducing nitrogen oxide (NOx) emissions. RECLAIM was
established in 1994 to provide flexibility for companies in the South Coast (Los
Angeles) area as controls on NOx, a major contributor to smog formation, were
tightened. Because of record electricity demand in 2000, electric generators in the
22 CERA Advisory Service, Design Issues for Market-based Greenhouse Gas Reduction
Strategies: Special Report
(February 2006), pp. 54-55.
23 CERA Advisory Service, Design Issues for Market-based Greenhouse Gas Reduction
Strategies: Special Report
(February 2006), pp. 54-55; Robert Repetto and Duncan Austin,
The Costs of Climate Protection: A Guide for the Perplexed (World Resources Institute,
1997) p. 22.
24 M. Jaccard and W.D. Montgomery, “Costs of Reducing Greenhouse Gas Emissions in the
USA and Canada,” 24 Energy Policy 10/11 (1996), pp. 889-898.
25 The Clean Air Act (42 U.S.C. 7401-7626), Section 110(c)(5)(C).
26 U.S. Environmental Protection Agency, Letter to California Independent System Operator
Corporation (August 12, 2000).

CRS-14
South Coast area generated more power than they did in the base period, resulting in
utilities buying RECLAIM trading credits in unprecedented quantities. As a result,
the price of credits rose from less than $1 per pound of NOx in January 2000 to more
than $60 per pound of NOx by March 2001.27 To solve this problem, in March 2001,
the South Coast Air Quality Management District amended RECLAIM to remove
large power plants from the trading system and required owners of such facilities to
reduce emissions under a mandated command-and-control regime. Such facilities
returned to the trading system in 2007.28
Proposals have been made to formalize a “circuit breaker” into any GHG
reduction program. In general, proposals envision a declining emissions cap system
where the rate of decline over time is determined by the market price of permits. If
permit prices remain under set threshold prices, the next reduction in the emissions
cap is implemented. If not, the cap is held at the current level until prices decline.29
Such a cap could be implemented on an economy-wide basis or by sector or other
relevant grouping.
Because the conditional reduction approach attempts to turn both the price and
the quantity of reductions into variables solved by the trading market, its effect on
cost depends on a host of variables — most obviously the profiles of the emissions
reduction targets and the price triggers. For example, the price trigger could be based
on the spot-market price, the long-term market price, or some hybrid price
mechanism. Also, the reliance on the market to either directly or indirectly determine
price and quantity puts pressure on regulators to oversee operations and prevent any
market manipulation designed to slow emissions reductions.
A conditional tonnage target could address distributional issues if its tonnage
targets and timetable triggers are tailored for specific sectors or industries. This
would substantially increase the complexity of the scheme and potentially risk
bifurcation of the permit market. As with other permit schemes discussed here,
another approach to addressing distributional concerns under a conditional tonnage
target would be to simply exempt certain sectors from its mandates.
When the control regime responds to relatively short-term events, it may not
provide the long-term price signal necessary to promote long-term solutions. The
elastic time frame also gives ambiguous signals for planning the appropriate pace and
scope of research and development efforts. In contrast, the regime’s focus on short-
term economic disruption may help in damping short-term volatility in the allowance
market. As noted, the responsiveness of the price and timetable triggers would
determine how effective the program would be in avoiding such disruption.
27 “RECLAIM Poised for Major Changes,” Executive Brief (New York: Evolution Markets),
pp. 1-2.
28 South Coast Air Quality Management District, Governing Board Meeting (January 7,
2005).
29 See Clean Power Group website, [http://www.eea-inc.com/cleanpower/index.html].

CRS-15
In reducing the uncertainty for cost estimates, the scheme introduces about the
same number of new uncertainties as it does in reducing others. The circuit breaker
prevents ever increasing costs, although with some undetermined lag time. However,
it is difficult to estimate absolute costs because one does not know how often it will
be used.
The scheme also increases uncertainty about the future trend in benefits by
making the quantity of emissions reduced a variable. A short-term break might not
make much of a difference, particularly if participants were required to make up the
emissions later. However, it introduces uncertainty into the system that would be
difficult to quantify.
Technology-Based Timetable. Another approach to increase flexibility in
the system and encourage long-term technology development would be to provide
special compliance schedules for entities deploying innovative, less carbon-intensive
technologies. An example of this possibility is Section 409 of Title IV of the 1990
Clean Air Act Amendments.30 Under Section 409, utilities choosing to meet their
sulfur dioxide reduction requirements by installing a qualifying clean coal technology
receive a four-year extension on the program compliance deadline. During the
extension, the affected emitter is allowed to operate under existing regulations and
operating conditions. If the technology fails to operate as designed, the affected unit
may be retrofitted with another qualifying technology or with an existing control
technology. For a GHG reduction program, a qualifying technology could include
geologic sequestration, emerging energy efficient technology, or advanced solar
power.
A technology extension could reduce costs in two ways. First, the delay in
compliance itself would reduce cost by allowing the affected company more time to
gather resources and optimize a compliance plan. Second, to the extent the delay
encourages more cost-effective approaches to GHG reductions, compliance cost and
long-term cost would be reduced. Of course, the risk is that the delay will not result
in successful technology development. Indeed, it is likely that at least some of the
projects would fail — that is the nature of innovation. However, because technology
development is crucial to long-term reductions in greenhouse gases, some may feel
the risk is worth it.
Assistance with distributional costs under this option would depend on the
opportunities for new technology in given sectors of the economy. Although some
industries may have potentially cost-effective technology-fixes, such as geologic
sequestration, others may involve long-term structural changes.
Of course, the focus of a technology-based timetable is to provide a long-term
signal to the market encouraging new technology. Such a signal could be
strengthened significantly with increased government funding of projects.
Because this option is focused on new technology, it would seem likely to have
little effect on short-term price volatility. However, there may be a risk that the
30 P.L. 101-559, Title IV, Section 409 (1990).

CRS-16
temporary removal of significant emitters from the market-system in response to the
incentive could increase short-term volatility and uncertainty by diminishing permit
demand and trading volume.
It is difficult to determine the effects of this option on cost estimates. It would
depend on how widespread the assumed participation rate is.
Technique Options
Most current GHG reduction proposals assume a market-based implementation
strategy — generally a permit trading program. This is not surprising, as flexibility
and new technologies are considered the keys to a cost-effective implementation
strategy over the long run. Generally, technique options range from making a
tradeable permit program more flexible through mechanisms like banking, to creating
a hybrid program where the regime shifts from a quantity-based permit program to
a carbon tax, depending on defined circumstances.
Banking and Borrowing. Most existing trading programs include provisions
for banking credits for either future use or future sale. Indeed, the absence of
effective banking in the RECLAIM program (discussed earlier) is credited with
contributing to RECLAIM’s suspension during the California energy crisis. As
summarized by Resources for the Future (RFF):
Allowance banking has been an essential component of the SO program. Its
2
absence is a costly feature of the NOx programs, eroding the opportunity for cost
savings from interannual trading and contributing directly to the suspension of
trading in RECLAIM.31
Banking and borrowing reduces the absolute cost of compliance by making
annual emissions caps flexible over time. The limited ability to shift the reduction
requirement across time allows affected entities to better accommodate corporate
planning for capital turnover and technological progress, to control equipment
construction schedules, and to respond to transient events such as weather and
economic shocks. Generally, banking and borrowing would not have any direct
impact on distributional concerns, which are more directly determined by initial
allocation decisions. Banking and borrowing can help provide a long-term market
signal by supporting credit prices when costs are lower than expected.32
The flexibility provided by banking and borrowing, as noted, can help dampen
short-term volatility. The degree that they help is disputed. As discussed later, some
argue that banking and borrowing may provide sufficient flexibility in some cases to
31 Dallas Burtraw, David A. Evans, Alan Krupnick, Karen Palmer, and Russell Toth,
Economics Pollution Trading for SO2 and NOx, RFF Discussion Paper 05-05 (March 2005),
p. 45.
32 Henry D. Jacoby and A. Denny Ellerman, The Safety Valve and Climate Policy, MIT Joint
Program on the Science and Policy of Global Change, Report No. 83 (July 2002), p. 9.

CRS-17
keep market disruptions to a minimum.33 However, others argue that if a program
involves more than modest reductions, a more robust “safety valve” is preferable.34
In estimating costs, banking and borrowing help smooth out the reduction
requirement, as witnessed by the current acid rain program. This economically
desirable effect does not necessarily reduce the uncertainty in cost estimates because
estimators will make different assumptions about the extent to which banking and
borrowing are used by emitters. The smoothing effect, however, has no effect on the
reduction requirement (in contrast with several of the other alternatives discussed
here). This is a major reason why this alternative is generally favored by those whose
priority is to achieve specific reductions.
Auctioning Permits. Auctions can be used in market-based pollution control
schemes in several different ways. For example, Title IV of the 1990 Clean Air Act
Amendments uses an annual auction to ensure the liquidity of the credit trading
program. For this purpose, a small percentage of the credits permitted under the
program are auctioned annually, with the proceeds returned to the entities that would
have otherwise received them. Private parties are also allowed to participate. A
second possibility is to use an auction to raise revenues for a related (or unrelated)
program. For example, the Regional Greenhouse Gas Initiative (RGGI) is exploring
an auction to implement its public benefit program to assist consumers or pursue
strategic energy purposes.35 A third possibility is to use auctions as a means of
allocating some, or all, of the credits mandated under a GHG control program. In
examining a modified auction program, a Resources for the Future (RFF) analysis
found that an auction scheme is “dramatically more cost-effective” in allocating
credits than either a grandfathered allocation method36 or a generation performance
standard37 (GPS) approach.38 Obviously, the impact that an auction would have on
the cost dimensions identified earlier would depend on how extensively it was used
in any GHG control program, and to what purpose the revenues were expended.
The cost-effectiveness of an auctioning system results from allowing the
marketplace to allocate credits. However, unlike a carbon tax, the market-clearing
33 Henry D. Jacoby and A. Denny Ellerman, The Safety Valve and Climate Policy, MIT Joint
Program on the Science and Policy of Global Change, Report No. 83 (July 2002), p. 1.
34 Richard D. Morgenstern, Climate Policy Instruments: The Case for the Safety Valve,
Council on Foreign Relations (September 2004), p. 10. This option is discussed further
below.
35 See Dallas Burtraw and Karen Palmer, Summary of the Workshop to Support
Implementing the Minimum 25 Percent Public Benefit Allocation in the Regional
Greenhouse Gas Initiative
, RFF Discussion Paper DP 06-45 (September 2006).
36 Used in the SO trading program, credits are allocated gratis to entities in rough
2
proportion to their historic emissions.
37 Also called an output-based allocation, credits are allocated gratis to entities in proportion
to their relative share of total electricity generation in a recent year.
38 Dallas Burtraw, Karen Palmer, Ranjit Bharvirkar, and Anthony Paul, The Effect of
Allowance Allocation on the Cost of Carbon Emission Trading
, RFF Discussion Paper 01-30
(August 2001).

CRS-18
price for credits is not limited (unless the system is combined with a safety valve, as
discussed below). Hence, an auction for credits would be more expensive for
specific industries than under a historically based grandfathered system, where they
would receive their credits free. Likewise, the price consumers pay may be greater,
depending on the companies’ ability to pass on their additional costs to them.
However, when the substantial revenues received by the auctions are considered,
auctions are more cost-effective than grandfathered or GPS systems. As stated by
RFF:
The bottom line is that the AU [auction] approach weighs in at substantially less
economic cost to society than either of the two gratis approaches to allocating
allowances.... AU also provides policymakers with flexibility, through the
collection of revenues that can be used to meet distributional goals or to enhance
the efficiency of the AU even further by reducing pre-existing taxes. Because the
AU approach is so cost-effective, a corresponding a [sic] carbon policy will have
less effect on economic growth than under the other approaches. This attribute
provides the most significant form of distributional benefit.39
As noted by RFF, the revenues from an auction can be used to address a host of
distributional concerns. Indeed, as noted earlier, the auction could be tailored to raise
only as much as necessary to address those concerns (as with RGGI funding of public
benefits programs) or made more comprehensive to address credit allocation.
In terms of a long-term price signal, the type of auction employed would have
some effect. For example, the program could implement a price floor to facilitate
investment in new technology via a reserve price in the allowance auction process.
In addition, the stability of that price signal could be strengthened by choosing to
auction allowances on a frequent basis, ensuring availability of allowances close to
the time of expected demand and making any potential short-squeezing of the
secondary market more difficult.40
An auction could provide substantial incentive for new technology if the auction
is structured to encourage a long-term and stable price signal and if revenues received
are at least partly directed toward research, development, and demonstration
programs.
An auction would probably not reduce the uncertainty with costs, because
differing assumptions could be made about the actual operation of the auction, its
efficiency, and the effectiveness of the recycled revenues. However, an auction
would not have any effect on benefits received by the program, unless it were joined
with a safety valve or other limit on auction prices.
Safety Valve. The purpose of a safety valve is to limit the costs of any climate
change control program (price) at the potential expense of reductions achieved
(quantity). Safety valves encompass a variety of carbon tax-tradeable permit hybrid
39 Burtraw et al., Allowance Allocation, p. 30.
40 Karsten Neuhoff, Auctions for CO Allowances — A Straw Man Proposal, University of
2
Cambridge Electricity Policy Research Group (May 2007), pp. 3-6.

CRS-19
schemes. Perhaps the most publicized version is that recommended by the National
Commission on Energy Policy.41 The Commission scheme would be implemented
through a flexible, market-oriented permit trading program. The total number of
permits each year would be based on a mandated decline in GHG intensity and
projected GDP growth. However, the scheme includes a cost-limiting safety valve
that allows covered entities to make a payment to the government in lieu of reducing
emissions. The initial price of such payments would be $7 per ton in 2010. Thus,
if a covered entity chooses, it may make payments to the government at a specific
price rather than make any necessary emissions reductions.
Effectively, a safety valve places a ceiling on compliance costs; in that way, it
acts like a carbon tax. To the extent an entity’s control costs, or the permit market,
remain below the safety valve, the scheme acts like a tradeable permit program. The
degree to which a safety valve reduces costs would depend on the extent to which it
is used by entities (e.g., who do not have a cost-effective alternative). However, the
complex interactions involved in a scheme that includes both price and quantity
controls should not be underestimated. As stated by Jacoby and Ellerman:
The usefulness of the safety valve depends on the conditions under which it
might be introduced. For a time, it might tame an overly stringent emissions
target. It also can help control the price volatility during the introduction of
gradually tightening one, although permit banking can ultimately serve the same
function. It is unlikely to serve as a long-term feature of a cap-and-trade system,
however, because of the complexity of coordinating price and quantity
instruments and because it will interfere with the development of systems of
international emissions trade.42
In contrast, Morgenstern argues that the complexity is worth it in preventing
price spikes, particularly if a substantial reduction in emissions is envisioned: “If
only modest reductions are undertaken, a system of banking and offsets is likely to
be adequate in preventing price spikes. In order to achieve more ambitious targets,
however, the safety valve is clearly preferred.”43
To address distributional concerns, a safety value could be tailored for specific
sectors to address concerns about cost-effective reduction options or competition.
In addition, to the extent the safety valve created revenues, some of the funds raised
could be recycled to affected parties.
The effect of a safety valve on new technologies reflects the complexity
discussion above. If a low safety valve price were chosen (meaning it would keep
compliance costs low), it could have a dampening effect on long-term development
of new technology. By creating a ceiling on the value of GHG reductions, but
providing no floor for those reductions, a weak market signal may be sent. This
might be offset to some degree if funds collected by the safety valve were directed
41 The National Commission on Energy Policy, Ending the Energy Stalemate: A Bipartisan
Strategy to Meet America’s Energy Challenges
(December 2005).
42 Jacoby and Ellerman, The Safety Valve and Climate Policy, p. 1.
43 Richard D. Morgenstern, The Case for the Safety Valve, p. 10.

CRS-20
toward new technology, but marketing of any resulting technology might still be
difficult if the market price is held low.
A safety valve would dampen the possibility of an upward spike in credit prices
— indeed, it is a major reason for considering such an option. However, it would
not affect any volatility occurring below the safety valve value and have no effect on
a collapse in credit prices. By the same token, the safety valve would put an absolute
limit on the projected costs of the program at the level of the safety valve. However,
it would do this at the expense of certainty in terms of reductions achieved.
Illustrative Approaches
The selected options discussed above are summarized in Appendix A. As
suggested, the various options identified have different strengths and weaknesses,
depending on the facet of costs one wants to address. Fortunately, many of the
options are not mutually exclusive, nor do they require complete adoption; parts of
individual approaches can be combined with other parts to meet program
specifications in terms of firmness of the goal (also called the “hardness” of the
emissions cap) and time frame.
To illustrate, a program focused on achieving a specific tonnage reduction with
some flexibility in implementation but not in a manner that threatens the integrity of
the cap could incorporate several of these options. The most obvious mechanism to
include in the quantity-based cap-and-trade would be banking and borrowing options
that would increase flexibility of the program across time without any deterioration
in the tonnage requirement. Flexibility and protection against price increases could
be enhanced by expanding supply options to include all greenhouse gases,
sequestration, and international trading. Depending on one’s confidence in the
individual supply options, use could be restricted to a maximum percentage of
reduction achieved through the option (common in many proposals) or to a more
flexible percentage restraint based on credit prices (as proposed by RGGI). Proper
monitoring and enforcement could minimize any potential effect on the cap.
This illustration would not necessarily provide either the long-term price signal
or funding necessary for new technology. One supplemental option that could help
mitigate this problem would be credit auctions. Auctions would have no effect on
the cap, but would provide the program with a revenue flow that could be at least
partly directed toward research and development. The auction could be designed to
raise revenues only by auctioning a small percentage of the credits (such as the
current acid rain program), or be comprehensive and auction all credits, thus
improving overall economics and providing a clear market signal (as is being
proposed by New York to meet its RGGI requirements). In the latter case,
coordinating the auction with any trigger price mechanism for expanded supply
options would promote harmonious implementation. Depending on the structure of
the auction chosen, the comprehensive auction would also provide a clear market
price for reductions and, with the addition of forward markets, some indication of the
general direction of those prices.

CRS-21
Finally, the auction and its resulting revenues could also be used to address
pressing distributional cost issues. Although the mixture of options used in this
illustration could potentially mitigate several of the cost issues identified here, it
would not provide cost certainty. The quantity side of the equation is the controlling
factor under this illustration; prices could be tempered by the market flexibility
introduced by the options, but actual costs would not capped.
In contrast, a more price-oriented illustration could employ a safety valve to
place an absolute limit on credit prices. In such a hybrid system, the focus of the
program is the safety valve limit as much as any tonnage cap. The quantity-based
limits of the emissions cap determine the probability that the safety valve would be
triggered, assuming a well-functioning market. However, in addition to the supply-
demand dynamic that the credit market will reflect, any market failure or disruption
resulting from external events could trigger the safety valve for participants.
Ultimately, quantity is subordinate to price.
One can potentially reduce the probability that the safety valve would be
invoked by including several of the other options discussed here. Expanding supply
options would enlarge the pool of available reductions and potentially improve the
stability of the credit market if properly integrated. Employing a dynamic tonnage
target or an economic-based circuit breaker could help address any economic growth
spike that might trigger the safety valve. The question of using these options in a
safety valve program is whether they would affect the cap more or less than invoking
the safety valve. In contrast, borrowing and banking would help stabilize markets
without having any effect on the cap.
Like the illustration above, this approach would not necessarily promote new
technology — indeed the safety valve could discourage such development, unless it
generated revenue that was directed toward research and development. If revenues
were deemed insufficient for new technology (and to address distributional concerns
if desired), the safety valve program could be supplemented with an auction.
However, in any case, this illustration is driven by price concerns — concerns that
make coordinating new technology development and minimizing impacts on the
emissions cap difficult.
A final illustration could also be the simplest — imposition of a carbon tax. The
clear focus of the program would be the level of the tax, the steepness of any future
increases in the tax, and who has to pay the tax. As noted earlier, it could be crafted
to address all the cost concerns identified in this report; however, it would represent
a new direction in U.S. climate change and current international efforts.

CRS-22
Addressing Costs Through Market Mechanisms:
Resolving the Price-Quantity Issue
The fundamental policy assumption that changed between the U.S. ratification
of the 1992 United Nations Framework Convention on Climate Change (UNFCCC)
and the George W. Bush Administration’s 2001 decision to abandon the Kyoto
Protocol process concerned costs.44 The ratification of the UNFCCC was based at
least partially on the premise that significant reductions could be achieved at little or
no cost. This assumption helped to reduce concern some had that the treaty could
have deleterious effects on U.S. competitiveness — a significant consideration
because developing countries are treated differently from developed countries under
the UNFCCC. Further ameliorating this concern, compliance with the treaty was
voluntary. While the United States could “aim” to reduce its emissions in line with
the UNFCCC goal, if the effort indeed involved substantial costs, the United States
could fail to reach the goal (as has happened) without incurring any penalty under the
treaty. This flexibility would have been eliminated if the United States had ratified
the Kyoto Protocol with its mandatory reduction requirements; the George W. Bush
Administration cited this lack of flexibility as a major reason for rejecting the Kyoto
process.
With the passage of the 2005 Sense of the Senate climate change resolution
calling on Congress to enact a mandatory, market-based program to slow, stop, and
reverse the growth of greenhouse gases, the need to address the cost issue has arisen
again. Indeed, the resolution itself states that the program should be enacted at a rate
and in a manner that “will not significantly harm the United States economy” and
“will encourage comparable action” by other nations.45 Facets of the cost issue that
have raised concern include absolute costs to the economy, distribution of costs
across industries, competitive impact domestically and internationally, incentives for
new technology, and uncertainty about possible costs.
Market-based mechanisms attempt to address the cost issue by introducing
flexibility into the implementation process. The cornerstone of that flexibility is
permitting sources to decide their appropriate implementation strategy within the
parameters of market signals and other incentives. That signal can be as simple as
a carbon tax or comprehensive credit auction that tells the emitter the value of any
reduction in greenhouse gases, to a credit marketplace that is constrained by a ceiling
price (safety valve) and includes incentives for new technology. As illustrated here,
the combinations of market mechanisms are numerous, allowing decision makers to
tailor the program to address specific concerns.
In a sense, the options discussed here represent a continuum between
alternatives focused on the price side of the equation (e.g., carbon taxes) through
hybrid schemes (e.g., safety valves) to alternatives focused on the quantity side (e.g.,
44 For a review of U.S. climate change policy, see CRS Report RL30024, Global Climate
Change Policy: Cost, Competitiveness, and Comprehensiveness
, by Larry Parker and John
E. Blodgett.
45 S.Amdt. 866, passed by voice vote after a motion to table failed 43-54, June 22, 2005.

CRS-23
banking and borrowing). They are tools to assist in the assessment of potential
greenhouse gas reduction approaches, leaving any policy decision on balancing the
price-quantity issue to the ultimate decision makers.
This balance will not be easy to achieve. By offering flexibility to program
designers and participants, market-based mechanisms can assist in implementing a
GHG reduction program at less cost than more traditional command-and-control
methods.46 However, the complexity of market mechanisms (particularly trading
programs) increases substantially with the scope of emitting sources included
(particularly if international trading is envisioned) and the specificity of any
allocation scheme. Thus, perhaps the most difficult issue to be addressed in designing
a market-based implementation strategy for reducing GHG emissions is determining
who is included, and who is exempted.
46 For background on this point with respect to climate change, see CRS Report RL30285,
Global Climate Change: Lowering Cost Estimates through Emissions Trading — Some
Dynamics and Pitfalls
, by Larry Parker.

CRS-24
Appendix A. Summary of Selected Options To Address Cost Uncertainty of
Greenhouse Gas Reduction Programs
Option
Absolute Costs
Distributional Costs
Long-Term Costs
Price Stability
Cost Uncertainty
Effect on Benefits
Carbon Tax
Allows economics to Distributional
Long-term
Would provide a
Would provide an
Would make
determine ultimate
concerns about costs
development of new
stable, consistent
upper limit on
reductions dependent
emissions reductions. can be addressed by
technology would be
price signal.
potential cost
on the level of the
Costs limited to tax
either partly or
stimulated by
estimates. The lower tax. The quantity
levy. Actual costs
completely
creating a long-term
limit would still be
reduction becomes
would depend on the
exempting specific
price floor on carbon
subject to
the variable while the
level of the tax,
sectors or targeting
and strengthened
uncertainty.
price is fixed.
availability of
sectors with funding
further by targeting
reduction below that
from the received tax R&D with funding
level, and the
revenues.
from the received tax
distribution of the
revenues.
revenues.
Dynamic Tonnage
Depending on
Distributional
Incentive for new
Would not
Would only have
Depending on the
Target
specifics, would
concerns about costs
technology would
necessarily avoid
modest effect on
specifics of the
probably offer some
could be addressed
depend on the slope
short-term
reducing uncertainty
target, benefits could
cost protection
by variety of regional of reductions
fluctuations in
in cost estimates.
be at least slightly
against unforeseen
or sector-specific,
mandated by the
market price.
dependent on
spikes upward in
metrics.
program.
Different metrics for
economic conditions.
economic growth.
different sectors
could also create
market price
uncertainty.

CRS-25
Option
Absolute Costs
Distributional Costs
Long-Term Costs
Price Stability
Cost Uncertainty
Effect on Benefits
Expanded Supply
Can substantially
Can help sectors that
Depends on how well Depends on how well Can increase
Should have no
Options
reduce costs,
do not have cost-
the additional
the additional
uncertainty by adding effect on reductions
depending on the
effective means of
sources are
sources are
new variables to the
achieved, assuming
additional options
reducing emissions
integrated into the
integrated into the
estimates, including
proper safeguards are
included.
on their own.
overall market — a
overall market — a
availability,
taken, but new risks
stratified market can
stratified market may penetration, and costs are introduced with
muddle the long-term result in independent of the additional
some options (like
price signal.
pricing trends.
options.
international
trading).
Economic-Based
Reduces costs by
Could address
Its short-term focus
Depending on the
Scheme introduces
Increases uncertainty
Circuit Breaker
temporarily
distributional
could muddle the
responsiveness of the many new
of benefits by
extending
concerns by tailoring long-term price
tonnage and
uncertainties while
making the quantity
compliance deadlines its tonnage and
signal important for
timetable triggers, it
reducing others in
of reductions
and/or slowing
timetables triggers to developing new
would help mitigate
estimating costs.
achieved a variable.
emissions reduction
specific sectors.
technology.
short-term price
targets. The degree
volatility.
of cost savings
depends on the
specifics of the
program.

CRS-26
Option
Absolute Costs
Distributional Costs
Long-Term Costs
Price Stability
Cost Uncertainty
Effect on Benefits
Technology-Based
Potentially reduces
Would depend on
Arguably, the
May have little effect Scheme introduces
Effect would depend
Timetable
costs by delaying
opportunities for new primary focus of this
on price stability;
new uncertainty to
on how widespread
compliance and
technologies in given scheme is to
indeed, it could
cost estimates.
the assumed
encouraging more
sectors.
encourage new
increase short-term
participation rate is.
cost-effective
technology
volatility and
approaches in the
deployment.
uncertainty by
long-term.
removing demand
and volume from the
market.
Banking and
Reduces costs by
Little effect.
Can help support a
The added flexibility No significant effect. No significant effect
Borrowing
making the emissions
long-term price
can help damp short-
over the long-term.
cap more flexible
signal for new
term volatility, but
over time.
technology by
not eliminate it.
supporting prices
when costs are lower
than expected.
Auctioning Permits
Allows the
Can be used to
Some revenues could Depending on the
Scheme introduces
No significant effect.
marketplace to
address concerns by
be targeted for new
volume of the
new uncertainties to
allocate permits.
tailoring auctions for
technology. Also,
auction, could have
cost estimates.
Actual costs would
specific sector and/or auctions would help
some effect on short-
depend on percentage directing revenues
determine market
term volatility, but
of permits auctioned
toward affected
price of reductions.
not eliminate it.
and distribution of
sectors.
the revenues.

CRS-27
Option
Absolute Costs
Distributional Costs
Long-Term Costs
Price Stability
Cost Uncertainty
Effect on Benefits
Safety Valve
Effect on cost
Safety valve levels
By setting a ceiling
Would place an
Would place an
Would make
depends on level that could be tailored for
but not a floor on
upper limit on price
upper limit on cost
reductions a function
the safety valve is
specific sectors.
prices, could have a
volatility.
estimates.
of the safety valve
set.
damping effect on
level.
new technology
depending on the
level imposed.