With the United States accounting for over one-fifth of global emissions of greenhouse gases, the U.S. government is facing pressures-from both domestic and international sources-to establish a comprehensive mandatory reduction program to address the risk of global climate change.
If Congress decides to move forward with such a program, it could be creating an environmental regulatory regime of unprecedented scope and impact. Many policymakers are considering innovative market-based approaches to regulation, including a multibillion dollar economy-wide "cap-and-trade" program. The authors evaluate four models for a domestic program against a set of several criteria, including environmental effectiveness, cost, administrative feasibility, distributional equity, and political acceptability.
INTRODUCTION
Until now, U.S. climate change policy at the federal level has consisted of voluntary greenhouse gas (GHG) mitigation programs, research and development, and a subset of energy policies that focus on energy efficiency and renewable energy. However, the U.S. government is facing pressures-from both domestic and international sources-to establish a federal mandatory reduction program to address the risk of global climate change. If Congress decides to move forward with such a program, it could be creating an environmental regulatory regime of unprecedented scope and impact. Sources of greenhouse gases range from electric power plants to every car on the road. In addition, many policymakers are considering innovative market-based approaches to regulation, including a multi-billion dollar economy-wide cap-and-trade program.
This Article identifies issues that must be addressed in the design of a mandatory domestic GHG reduction program. The Article then evaluates a number of proposals, including (1) comprehensive capand-trade programs; (2) a GHG tax; and (3) a "sectoral hybrid" program that combines elements of a cap-and-trade program with product efficiency standards for automobiles and consumer products.
While there is a substantial body of opinion, particularly among economists, that an economy-wide cap-aiicl-trade or GHG tax program may be optimal from a cost-effectiveness point of view, it is possible that a GHG regulatory program will be developed from discrete familiar elements, such as existing Corporate Average Fuel Economy (CAFE) and appliance efficiency standards, plus large stationary source controls modeled on the acid-rain control program. Rather than creating a whole new system, Congress may choose the latter approach because of both familiarity and political sensitivity regarding program designs that result in overt increases in prices for gasoline and home heating fuels. We review the implications of these two fundamentally different approaches.
While this Article focuses on options for federal regulatory policies, it is important to note that a domestic climate change program could enhance its regulatory policies with a range of non-regulatory measures, such as funding for research and development into new technologies, financial and other incentives, public education, and changes in infrastructure and land-use policies. In addition, state and local governments may supplement a federal regulatory program with their own policy initiatives.1
I. U.S. GREENHOUSE GAS EMISSIONS PROFILE
Domestic climate change policy will likely focus on reductions or sequestration of emissions of six GHGs: carbon dioxide (CO2), methane (CH^sub 4^), nitrous oxide (N^sub 2^O), and what have been called the "synthetic gases," hydrofluorocarbons (HFCs), perfluorocarbons (PFCs), and sulfur hexafluoride (SF^sub 6^).2
Because GHGs have long lifetimes in the atmosphere, it matters little where or exactly when GHG emission reductions are made.3 For example, one ton emitted in the United States has the same impact as one ton emitted in Malaysia, and reducing one ton of GHG emissions now, rather than five years from now, will make little difference in atmospheric GHG concentrations in 2050.4 This means that an effective regulatory program can allow flexibility as to where emission reductions occur and substantial but not unlimited flexibility as to when they occur.
Different GHGs vary as to their residence lives in the atmosphere and their heat-trapping, or "radiative forcing," effects.5 Some GHGs have very long atmospheric lifetimes.6 The Kyoto Protocol adopts a weighting formula called "Global Warming Potential" (GWP), which measures the impact of one ton of any GHG with reference to one ton of CO2.7 With such an agreed-upon "exchange rate," policymakers can develop a unitary program objective in terms of "CO2-equivalent" units, which allows regulated firms to pick whatever mix of reductions of different GHGs they believe is most cost-effective.8
A. Carbon Dioxide
Carbon dioxide emissions, resulting almost entirely from combustion of fossil fuels, dominate GHG emissions in the United States and are likely to be among the principal initial targets of any domestic GHG regulatory program. In 2001, energy-related COo emissions accounted for approximately eighty-one percent of U.S. GWP-weighted emissions.9
Within the energy sector, the principal means of abating CO2 emissions are switching from energy sources with high carbon content to those with low or zero carbon content, such as renewables; improving the efficiency of energy conversion or use; reducing energy use; and developing carbon capture and sequestration technologies.10
Annual U.S. CO2 emissions also are affected by land use, land-use change, and forestry (LULUCF) activities.11 Plants and certain other biotic matter remove CO2 from the atmosphere and store or "sequester" it as carbon, at least temporarily, through the process of photosynthesis.12 Hence, forests and agricultural lands are "reservoirs" of carbon and a range of activities can enhance their sequestration potential.13 Conversely, certain land use changes, such as deforestation, can oxidize the carbon stored in biotic matter, thereby leading to CO2 emissions.14
B. Other GHGs
Methane is the second-largest contributor to U.S. GHG emissions, constituting 8.7% of total U.S. GWP-weighted emissions in 2001.15 Methane is emitted from landfills; natural gas and petroleum production, transportation, and processing; agricultural activities; coal mining; stationary and mobile combustion; wastewater treatment; and certain industrial processes.16
Nitrous oxide is a GHG with heat-trapping potential that exceeds that of COo by an order of magnitude.17 Emissions of nitrons oxide made up 6.1% of U.S. GWP-weighted emissions in 2001.18 The primary human activities resulting in emissions of nitrous oxide are agricultural soil management, fuel combustion in motor vehicles, and production processes for adipic and nitric acid.19
Emissions of HFCs and PFCs are primarily associated with their use as substitutes for ozone depleting substances banned under the Montreal Protocol treaty.20 Emissions of HFCs, PFCs, and SF^sub 6^ also result from certain other industrial processes, including production of primary aluminum, certain steps in the manufacture of products in the semiconductor industry, and activities related to the operation of electrical transmission and distribution equipment.21 These gases have very powerful heat-trapping effects.22 They constituted 1.6% of U.S. GWP-weighted emissions in 2001.23
C. U.S. GHG Emission Trends
Eventual stabilization of atmospheric concentrations of GHGs will require very large reductions in GHG emissions worldwide. Notwithstanding a slight decline in 2001,24 U.S. emissions are projected to increase. As discussed above, U.S. emissions were 11.9% higher in 2001 than they were in 1990.25 Between 1990 and 2000, the GHG "intensity" of the U.S. economy-the ratio of total GHG emissions to economic output-declined by 17.5%.26 In a report submitted to the United Nations in 2002, the U.S. government projected that by 2020, U.S. GHG emissions will rise 42.7% from year-2000 levels.27
II. DOMESTIC CLIMATE POLICY FRAMEWORK
The existing federal framework for addressing climate change in the United States is a combination of voluntary programs, tax incentives, energy efficiency standards, and research and development. These programs, and certain Clean Air Act provisions, are described below.
A. Voluntary Prograins
Since 1993, the federal government has established a number of voluntary GHG emission reduction programs to encourage businesses to undertake GHG mitigation actions. This approach began with the Clinton Administration's "Climate Change Action Plan" (CCAP).28 The Bush Administration has adopted a similar voluntary strategy.29 A key supporting element of both the Clinton and Bush Administrations' voluntary programs is the Department of Energy's (DOE) voluntary GHG reporting program under § 1605(b) of the Energy Policy Act of 1992.30 The § 1605(b) program authorizes DOE to develop a system to document voluntary GHG mitigation actions reported by firms and others participating in various voluntary programs.31 Electrie utilities, in particular, have reported numerous projects under the § 1605(b) program.32
While the various voluntary programs have led to a significant number of emission reduction projects, overall emission levels have continued to increase.33 Several factors have contributed to the limited effectiveness of voluntary programs.34 First, while some participants in these programs have committed to taking particular mitigation actions, they have not in many cases committed to limiting their company-wide emissions below a particular baseline; for many, total system emissions increased substantially in response to increased market demand for products and services.35 Second, some participants committed to actions that they might have implemented anyway for business reasons.36 In particular, commentators have asserted that the § 1605 (b) program lacks rigorous reporting standards and verification requirements, and concerns have been raised that some reductions reported under the program have been double-counted.37 The Bush Administration has pledged to address these shortcomings in a planned upgrade to the program to be completed by the end of 2004.38 However, any voluntary program remains subject to a fundamental limitation-it only addresses the emissions of those firms that volunteer to participate.39
For these reasons, current U.S. voluntary programs-while helpful in building awareness, encouraging experimentation, and achieving some company-level emission reductions-are not expected to reduce or even stabilize U.S. GHG emissions in the next decade relative to current levels.40
In addition to the voluntary GHG programs described above, the U.S. government has established a number of non-regulatory programs aimed at increasing energy efficiency.41 Because energy-related GHG emissions make up over eighty percent of total U.S. emissions, these programs contribute to reducing GHG emissions.42 However, like the voluntary GHG reduction programs, they do not impose actual limits on emissions and are incapable of achieving substantial emission reductions with a high degree of certainty.43
Finally, federal tax law provides a range of tax credits and other incentives to encourage use of renewable energy and fuel-efficient vehicles.44 These include: a deduction for a portion of the purchase cost of a "clean-fuel" vehicle, defined to include hybrids;45 a credit for the purchase of an electric vehicle;46 an investment credit for solar or geothermal energy equipment47 and favorable depreciation rates for such equipment;48 and a credit for production of electricity from wind, certain types of biomass, or poultry waste.49 Congress is considering a number of additional tax incentives and modifications to existing tax programs in the context of proposed federal energy legislation.50
B. Product Efficiency Standards
1. Corporate Average Fuel Economy
Existing federal law includes two major mandatory energy efficiency programs: one for automobiles,51 and the other for consumer products other than automobiles.52 Both were established in 1975 under the Energy Policy and Conservation Act (EPCA).53 The program for motor vehicles-known as Corporate Average Fuel Economy or "CAFE"-requires each automobile manufacturer or importer to meet average fuel economy standards for the fleet of new vehicles it manufactures or imports in each model year.54 These standards are expressed in miles per gallon (mpg).55 Separate standards are set for passenger automobiles and "light-duty trucks"-including sport utility vehicles (SUVs) and minivans-currently at 27.5 mpg and 20.7 mpg respectively.56
The statute applies only to new vehicles and does not regulate inuse consumption of fuel.57 More stringent standards improve on-the-road fuel economy only to the extent that new vehicles replace less efficient existing vehicles.58 In addition, for new vehicles, if vehicle miles traveled (VMT) increase faster than average fuel economy, overall fuel use will go up notwithstanding the CAFE requirements.59
The statute contains a number of idiosyncratic features that increase its complexity, while decreasing its effectiveness.60 Trucks and SUVs are subject to far less stringent standards than cars.61 Compliance with the standard is determined separately for vehicles manufactured in the United States, Canada, or Mexico, and those vehicles manufactured elsewhere but used in the United States.62 Special credit is given to electric vehicles and to alternative fuel-capable vehicles.63
While the CAFE program made a significant contribution to moderating U.S. fuel use in the first years after its enactment, its impact has declined over time for a number of reasons.64 First, the standards were frozen for many years. Therefore, the standards have not taken into account the increasing proportions of truck, SUV, and niinivan sales. Starting in 2001, such "light-duty trucks" made up over fifty percent of vehicles sold.65 Congress's decision to freeze the standards throughout most of the 1990s, combined with the change in product mix, has had the effect of decreasing the ability of the program to moderate fuel use.66 Second, real gasoline prices have declined, encouraging more driving and dampening incentives for drivers to demand more efficient vehicles. Accordingly, even though fuel economy for cars has improved since the enactment of CAFE, overall fuel use-and, therefore, GHG emissions-has risen steadily.67
Of course, policymakers did not design CAFE as a domestic GHG regulatory program, and to function as one it would need not only to have the features noted above corrected-removing the freeze on more stringent standards and modifying the electric vehicle and alternative fuel credits68-but also the mpg standard would have to be translated into terms of pounds of CO2 per mile to take into account the carbon content of fuel.69 Additionally, as discussed below, a number of other changes would be needed to integrate such a program into a domestic cap-and-trade program for GHGs.
2. Appliance Standards
EPCA also established an energy efficiency program for consumer products other than autos-usually referred to as the "appliance efficiency program."70 It includes mandatory energy labeling and energy efficiency standards for a wide range of consumer products, including air conditioners, washers, dryers, kitchen ranges, and furnaces.71 Standards also cover some equipment used in industrial applications, such as most industrial motors.72 According to DOE, the standards program has resulted in a greater than one quad reduction of en erg)' use annually, equivalent to roughly one percent of energy use or about seventy-five million tons of CO2.73 It aims at requiring for each type of consumer product the maximum energy efficiency that is technologically feasible and economically justified; but its complex regulatory framework makes prompt action to promulgate stringent new standards quite difficult.74
While the standards program in its present form could be used for GHG regulatory purposes, it would be better adapted to that purpose if the standards were expressed in the form of direct or indirect GHG emissions per unit of output, and if a trading feature could link it to GHG regulation in other sectors.75
C. Clean Air Act
Aside from a requirement that electricity generators, who account for about one-third of U.S. GHG emissions, monitor and report their CO2 emissions, the Clean Air Act (CAA)76 does not directly address control of GHG emissions, much less explicitly authorize GHG regulation. The question of whether EPA has implied authority under the CAA to regulate GHGs-by virtue of its CAA authority to regulate "air pollutants"-is the subject of vigorous debate.77
This debate is beyond the scope of this Article, which contemplates action by Congress to establish a GHG regulatory program by statute, rather than action by EPA using its existing CAA authorities. Nevertheless, it is worth observing that the acid rain provisions of the CAA present a useful model for a cap-and-trade program applicable to CO2 emissions from electricity generators-which is one of the models for GHG regulation considered below.78 The acid rain program imposes a national limit on SOo emissions from electricity generators-currently set at 8.9 million tons per year-allocates allowances to existing sources to emit specified quantities of SO^sub 2^, and allows sources to trade and bank allowances, so that they can pursue least-cost compliance strategies.79
D. Options for a Domestic Program to Secure Greenhouse Gas Reductions
While voluntary programs, the CAFE program, tax incentives, and product efficiency standards have contributed to reductions in GHGs that would not otherwise have occurred, they neither individually nor collectively are likely to achieve significant economy-wide reductions in GHG emissions from current levels.80 Substantial attention has been given to formulating and evaluating a range of alternative mechanisms for controlling U.S. GHG emissions.81 For example, several bills have been introduced that would establish a CO2 cap-and-trade program for electric utilities, modeled on the SO^sub 2^ program under Title IV of the CAA.82 In January 2003, Senators John McCain (R-AZ) and Joseph Lieberman (D-CT) introduced legislation that would establish an economy-wide GHG cap-and-trade program.83 In March 2004, a companion version of the McCain-Lieberman bill was introduced in the House.84
The principal options for a mandatory GHG reduction program, and the ones evaluated below, are:
Cap-and-Trade: A comprehensive cap-and-trade program, similar in many respects to the acid rain program, that allocates or auctions a fixed number of tradable allowances to emitters and requires them to surrender allowances equal to their emissions in a particular compliance period-known as "downstream" cap-and-trade.85 A variant of this program requires firms to surrender allowances equal to the carbon content of the fuel and the GHG content of certain other products they sell each year-known as "upstream" cap-and-trade.86
GHG tax: A tax either on GHG emissions or on the carbon content of fuel and the GHG content of certain other products.87
Sectoral Hybrid: A program that combines a large-source cap-and-trade program with product efficiency standards, that is, standards for consumer products and equipment that prescribe emissions per unit of output-pounds of CO2 per mile, for example-or energy efficiency levels.88
This Article also discusses in general terms additional options such as stationary source emission standards, stand-alone product efficiency standards, and a stand-alone large-source cap-and-trade program.
III. DESIGN CRITERIA FOR A DOMESTIC GHG REGULATORY PROGRAM
Evaluating different GHG regulatory program options involves a number of considerations. The first design decision is establishing the program's emissions reduction objective. Once an emissions reduction objective is set, policymakers have to design a regulatory program to meet it. Key design criteria include environmental effectiveness, cost, administrative feasibility, distributional equity, and political acceptability. The sections that follow elaborate on each of these criteria.
The emissions reduction target for a domestic program establishes the level and timing of reductions at the national level. The target can be set for purposes of compliance with an international obligation or could be established as a matter of domestic policy, independent of any international obligations. Moreover, it could take the form of a cap on domestic GHG emissions or a limit on GHG emissions per unit of output, also referred to as an "emissions intensity" target. It could establish a GHG reduction target for an initial compliance period, or it could establish a long-term emissions reduction path, phasing in progressively more stringent targets over an extended period of time. This Article does not address the issues of whether or how to set a target, or what target to set. Instead, it evaluates different designs for a program that will meet whatever target is decided upon.89
The criteria for evaluating design options are described below.
A. Environmental Effectiveness: How Effective Is the Program in Meeting Its Emissions Reduction Target?
A regulatory program's effectiveness in meeting its target is a function of a number of factors, including its coverage of sources throughout the economy, its certainty in meeting a particular emissions target, and its provisions for enforcement.
1. Coverage: Are All Sources and Gases Covered?
A program's coverage refers to the extent to which it directly or indirectly regulates sources of GHG emissions throughout the U.S. economy and applies to the full range of GHGs. Broad coverage is preferable from an environmental perspective, but may have to be balanced by considerations of administrative cost. Compared to a program with full coverage, a program with only partial coverage either will reduce emissions less, or will attain the same emission reductions at much higher cost because it excludes opportunities for inexpensive reductions in uncovered sectors or gases. Programs with only partial coverage also risk "leakage."90 Leakage occurs when a regulatory program encourages shifting of emission-generating activities from regulated to non-regulated firms.91
2. Environmental Certainty: Will the Program Ensure That the Emissions Reduction Target Will Be Met?
Some program designs provide greater certainty that total emissions from regulated firms will not exceed a particular level. For example, a "quantity-based" approach, such as a conventional cap-and-trade program, enforces an overall limit on emissions from the covered firms.92 By contrast, "price-based" approaches, such as emission taxes or trading programs with a safety valve, do not place a precise limit on total emissions, but instead impose a particular price or price limits per ton of emissions.93 While establishing an emissions charge or tax has the effect of reducing emissions, the approach does not ensure that emissions will be reduced to a precise level.94 In addition, as explained below, a standards approach that limits emissions per unit of output, as opposed to tons per year-often referred to as a "carbon intensity" approach-will not achieve a particular emissions reduction target with certainty.95 However, because it is cumulative rather than annual emissions that are important, taxes or standards should be able to provide almost equivalent environmental certainty if there is political will to adjust them over time.
3. Enforcement: Is the Program Enforceable?
Any regulatory program's overall success in reducing emissions also is a function of its enforcement mechanisms. Enforcement is, in turn, a function of clear rules, precise and effective measurement of emissions, pursuit of violators, and having non-compliance penalties high enough to exceed any benefits associated with non-compliance.96
B. Cost-Effectiveness: Will the Program Design Allow Cost-Effective Compliance?
A key consideration in evaluating a GHG regulatory program is whether it permits compliance with the program's target at the least cost to the U.S. economy-what is referred to as "cost-effective" compliance. The first cost-related issue is the direct cost of complying with the program. A program designed to meet a particular target minimizes compliance costs to the extent that it maximizes flexibility to adopt a least-cost compliance strategy-that is, flexibility as to what, where, and when emission reductions are attained. In addition, some program designs can cap compliance costs, but do so at the risk of missing the program's target. Another key cost-related consideration is administrative cost. Finally, some program designs raise revenue, which, as explained below, could be used to offset part of the overall cost of the program by reducing "distortionary" taxes on capital and labor.
1. Flexibility: Will the Program Provide Flexibility as to How, Where, and When Emission Reductions Are Attained?
A cost-effective program will provide wide flexibility to regulated firms in determining how to reduce emissions to meet the program target ("what" flexibility), where to reduce them ("where" flexibility), and within limits, when to reduce them ("when" flexibility).97 "What" flexibility' implies that a firm can comply by implementing any of the full range of GHG mitigation measures, including increasing energy efficiency; switching fuels; reducing consumption; adopting LULUCF measures, including agriculture; or taking other action to reduce or sequester GHGs. Second, it implies that firms can comply through reductions in any of the major GHGs. Third, it implies that firms that can achieve low-cost reductions will undertake a greater proportion of emission reductions than firms that achieve reductions at higher costs. Many different kinds of firms and activities generate emissions of different GHGs; their costs of reducing those emissions and the means of reduction available to them vary widely. A program with maximum "what" flexibility has the effect of equating marginal costs of mitigation across all firms subject to the program, thereby generating the lowest-cost distribution of abatement activities throughout the economy.98
The other critical benefit of building "what" flexibility into the U.S. climate policy architecture from the beginning is that it spurs technological innovation. Achieving the long-term aim of stabilizing atmospheric concentrations will not be possible without the development and widespread deployment of a range of next-generation approaches to climate protection, including new clean energy technologies. Policy approaches that prescribe the use of particular technologies, such as design standards, provide little incentive for developing such next-generation approaches. By contrast, approaches that specify environmental outcomes or place a price on environmental damage without prescribing the means of compliance can stimulate the kind of innovation that ultimately will be needed to achieve deeper emission reductions over time.
"Where" flexibility implies that the program will recognize reductions achieved throughout the world. A domestic GHG program that is integrated with the emerging international market in GHG emission reductions almost certainly will have lower compliance costs than a program that credits only reductions made within the United States.99 Studies have suggested that opening up a U.S. climate program to trading even with just the industrialized countries that are subject to Kyoto Protocol emission limits could reduce a U.S. program's marginal abatement cost by anywhere between thirteen percent and sixty-eight percent.100 Gains from trade would be far greater if the U.S. program credited reductions achieved in developing countries, where low-cost abatement options are in abundant supply.101 For these reasons, the ultimate cost of a U.S. climate change program will depend in great measure on the extent to which it provides for international emissions trading.
"When" flexibility provides the regulated firm with choices as to the timing of emission reductions. Even before the regulatory program becomes binding, policymakers can establish a "credit for early action" policy to assure firms that any pre-program efforts to reduce emissions will be recognized. Such early reduction efforts would have the same environmental value as reductions made after the regulatory program has commenced.102 Policymakers also can set an ultimate compliance deadline for the regulatory program that gives firms sufficient lead time to develop cost-effective control strategies and that allows a market for emission reductions to evolve. Further, in establishing a program's emissions target, consideration can be given to determining compliance on the basis of a multi-year emissions average, rather than the level of emissions in a single year. A multi-year approach gives firms the flexibility to manage their emissions over time and avoids penalizing them for emission changes caused by difficult-to-control fluctuations in business cycles and weather.
Other "when" flexibility measures include "banking" and "borrowing."103 Programs can be designed so that firms that over-comply can "bank" emission credits and use them in a subsequent compliance period or sell them at a later date when prices in the trading market might be higher. A "borrowing" provision would allow a firm to comply with its obligations in one compliance period in part by committing to even deeper-than-required reductions in the subsequent compliance period. With a limited borrowing provision, a regulatory program could obtain a greater overall level of emission reductions from those firms that could benefit from additional time to modify their operations or invest in new technologies. A multi-year compliance period approach would offer similar temporal flexibility as a borrowing provision. A firm's ability to borrow has to be limited, however, lest it become a means of simply avoiding reductions.
2. Cost Predictability: Are Costs of Compliance Reasonably Predictable?
A regulatory program also can be designed so that total compliance costs are capped.104 As discussed above, "price-based" approaches, such as emission taxes, do not provide assurances that a particular level of emission reductions will be achieved. On the other hand, such programs do provide assurances that the costs of compliance will not rise above a particular per-ton level. This kind of certainty about costs generally is not possible with a quantity-based program, such as a traditional cap-and-trade program, where it is implied that the quantitative limit on emissions will be enforced regardless of compliance costs. To address the risk of spiraling compliance costs associated with a cap-andtrade program, some have proposed a "safety valve" mechanism, in which additional allowances would be made available at a pre-set price representing the maximum acceptable cost.105
3. Raising Revenue: Will the Program Raise Revenues That Can Be Used to Offset a Portion of Its Costs?
Some program designs that raise revenue, such as GHG taxes or allowance auctions, offer an opportunity to offset economic costs of the program borne by particular sectors through financial assistance programs or reduce the overall cost of the program through a reduction in federal taxes.106 Economic analysis indicates that programs that recycle the revenue to reduce distortionary taxes on capital, labor, or income have significant potential to reduce overall costs of a GHG regulatory program to the economy.107 However, it may prove politically difficult to implement tax cuts that increase economic efficiency. The revenues raised could just as easily be spent on activities that reduce or have no impact on economic efficiency as on activities that improve it.
4. Long-Term Incentives: Will the Program Induce Key Sectors to Begin Investing in Low-Emission Technologies and Practices?
Most climate change analysts agree that moderating the increase in atmospheric concentrations of GHGs ultimately will require a substantial transformation in the way that industrialized countries like the United States produce and use energy.108 Near-term policy choices will have a major impact on the cost of such a long-term effort. The reason is that energy-producing and energy-using technologies involve long-term capital investments that are not readily converted to other uses. Therefore, a domestic program needs to send a credible long-term signal to key sectors of the economy that encourages a shift toward lower-carbon technologies and lower-emitting practices. A domestic program that leaves certain sectors uncovered could result in those sectors "locking in" higher-emitting technologies and practices, potentially increasing the cost of achieving more substantial economy-wide GHG reductions in the future.109
C. Administrative Feasibility: Can the Program Be Administered and Docs It Minimize Administrative and Transaction Costs?
A key consideration in designing any regulatory program is whether it is feasible to administer. A program that is infeasible to administer will be both environmentally ineffective and economically inefficient. One key feasibility consideration is minimizing administrative costs-including the costs of designing the program and the costs of implementing it, both for the regulated firm, which must bear reporting or other costs, and for the regulator. Administrative costs are a function of the number of regulated firms, the availability of needed data about those firms, and the complexity of the regulatory program.110 In addition, program designs that build upon existing and familiar programs will impose smaller implementation costs and less difficulty for the regulator and the firms to be regulated than programs that represent a new departure. Finally, in designing market-based regulatory programs, careful attention needs to be given to avoiding unnecessary program complexities and uncertainties that run up participants' transaction costs.111
Another particularly important administrative criterion for a climate change policy is adaptability, given the necessary duration of any effort to stabilize concentrations of GHGs in the atmosphere. A U.S. climate change policy framework needs to be able to evolve over time to accommodate adjustments in the emission reduction commitments as new information becomes available and as the U.S. economy changes. In addition, because stabilization of GHG concentrations ultimately will require global efforts, the policy framework will have to be flexible enough to provide for coordination with other countries.
D. Distributional Equity: Is the Burden of Compliance with the Program Fairly Apportioned ?
Another consideration in designing a regulatory program is how its costs are distributed across society.112 Even the most cost-effective program design may be unacceptable if its costs are distributed in such a way that is perceived to be unfair.
All other things being equal, a regulatory program that aims to reduce GHG emissions will tend to impose its largest costs on firms and households that produce fossil fuels or are heavily dependent on them.113 A GHG regulatory program also will tend to be relatively more costly for low-income individuals because they spend a greater proportion of their total income on energy.114
Some regulatory programs provide opportunities for modifying these distributional impacts. For example, in an emissions trading program, the government could allocate allowances on a cost-free basis to firms that would bear the brunt of regulatory compliance costs. Alternatively, the government could auction allowances and use the revenue to compensate those particularly burdened by the regulatory program through targeted tax breaks or lump-sum payments. Emission tax programs hold similar revenue recycling potential.
E. Political Acceptability: Are There Elements of Program Design that Affect Its Political Acceptability ?
Program designs that promise relatively greater environmental effectiveness, lower costs, and a more equitable distribution of regulatory burdens will be more likely to obtain political support than other designs. However, the U.S. experience with environmental and energy policy suggests that other factors also affect a program's political acceptability. Indeed, considerations of political acceptability may lead policymakers away from what could otherwise be an optimal program design with respect to environmental effectiveness, cost, and equity.115
For example, twenty-five years of environmental and energy policy experience suggests that it is difficult to gain public support for a program that relies principally on direct increases in the price of energy-either through taxes or regulatory measures-even where such a program arguably is more cost-effective or will result in a more equitable distribution of regulatory burdens than other approaches.116 Even in times of most compelling national circumstances, such as the 1973 Arab oil embargo, Congress was unwilling to vise energy price increases to rein in consumer demand.117 On the other hand, program designs involving emissions trading or emission charges offer the opportunity to develop what may be a politically attractive policy package-using the revenue raised from regulation of GHG emissions as a basis for reducing taxes on income.118
IV. EVALUATING DIFFERENT APPROACHES TO REGULATING DOMESTIC GHG EMISSIONS
Using the criteria developed above, we evaluate three principal approaches to regulating domestic GHG emissions: (1) an emissions trading-or cap-and-trade-program; (2) a GHG tax program; or (3) a sectoral hybrid program combining a large-source cap-and-trade program with product efficiency standards. Each approach presents its own design choices, For example, a cap-and-trade program could be upstream or downstream.
A. Emission Trading (Cap-and-Trade) Programs
1. Overview of Emission Trading Programs
A conventional cap-and-trade program establishes an economy-wide or sectoral "cap" on emissions in terms of tons per year or other compliance period, and allocates or auctions tradable allowances, such as the right to emit one ton of GHGs, to GHG emission sources or to fuel suppliers.119 The total number of allowances is equal to the cap. A downstream cap-and-trade program applies to sources of GHG emissions and requires them to surrender allowances equal to their emissions.120 An upstream program applies to fuel suppliers and requires them to surrender allowances equivalent to the carbon content of fossil fuels they supply.121 Cap-and-trade programs are best suited to regulation of emission sources that can be readily measured and monitored, hi the GHG context, such sources include almost all sources of CO2 emissions from fossil-fuel combustion as well as many sources of other GHG emissions.122 Other types of sources can be regulated on an "opt in" or project basis, or through supplemental regulation.123 The trading feature of a cap-and-trade program authorizes regulated firms-and anyone else-to buy, sell, or hold allowances.
In a well-functioning emissions trading market, allowances will end up distributed among firms that need them in a way that minimizes the cost of reducing emissions. For example, in a conventional downstream cap-and-trade program, firms subject to the program buy allowances if their costs of reducing emissions-referred to as their costs of "abatement"-exceed the allowance price.124 Firms sell allowances if their abatement costs are lower than the allowance price.125 Trades continue in this way until firms are indifferent between buying and selling allowances-or, in other words, between abating one more ton of CO2 or emitting an additional ton.126 At this point, the program has equalized marginal abatement costs across the economy, and, in theory, the final distribution of allowances and abatement throughout the economy reflects the least-cost outcome.127
A GHG emissions trading program could incorporate all forms of "what," "where," and "when" flexibility, discussed above. Each firm affected by a GHG emissions trading program could reduce its need for allowances or exposure to higher energy costs by adopting its lowest-cost means of abatement. Firms also would have an incentive to develop new technologies or practices to reduce emissions or increase their energy efficiency. A U.S. domestic cap-and-trade program also could be integrated with emerging cap-and-trade programs in other countries and, if the parties so provided, with an international regime such as the Kyoto Protocol.128
A cap-and-trade program can be extended beyond energy-related sources of CO2 emissions by directly regulating: (1) sources of nonCO2 GHGs and/or (2) LULUCF activities that emit or remove CO2. Some GHG sources and sinks, however, may not be amenable to regulation through such an approach because their emissions may be too difficult to measure for purposes of setting a cap and allocating allowances, or to monitor for purposes of enforcement.
In some cases, these sources and sinks could be incorporated into the cap-and-trade program on a project-by-project basis, known as "project-based crediting." Under project-based crediting, a firm could earn emission credits by undertaking a climate change mitigation project at a source or sink not otherwise subject to the cap-and-trade program. To earn credits, a project would have to meet certain criteria. For example, the firm would have to provide for adequate measurement and monitoring and demonstrate that the project achieves reductions or removals beyond a baseline or "business-as-usual" scenario. The firm also would have to establish that the project would not simply shift emitting activities from the project site to another, unregulated site, an effect commonly referred to as "leakage." Credits earned for projects could be fully fungible with allowances in the emissions trading market. An example of this kind of project-based crediting mechanism is the Kyoto Protocol's "Clean Development Mechanism."129 From a cost-effectiveness standpoint, project-based crediting is inferior to a cap-and-trade approach because it entails higher transaction costs. Project-based crediting, however, may be the only way to incorporate certain difficult-to-measure sources into a market-based program.
In addition to these forms of "what" and "where" flexibility, a GHG emissions trading program could provide for "when" flexibility by allowing for banking and borrowing. Firms required to surrender allowances to cover their emissions or the carbon content of fuel supplied could be authorized to bank surplus allowances for use in a later compliance period. Some form of limited borrowing, using future allowances to cover current emissions, also could be considered. Borrowers could be required to repay with "interest," i.e., additional allowances.
2. Designing an Emissions Trading Program
Creating a GHG emissions trading program involves three fundamental design decisions that build upon this basic model.150 PoIicymakers need to determine which firms will be required to hold allowances for compliance, how allowances initially will be allocated, and whether the program will enforce a strict quantitative emissions target or adopt a price-based "safety valve" approach-an approach that provides that permits will not exceed a specified cost threshold. Each design decision has various implications for the trading program's effectiveness, cost, administrative feasibility, distributional consequences, and political acceptability.151
3. Who Is the Regulated Firm?
A key step in designing a GHG emissions trading program is determining who are to be the regulated firms-that is, the firms that will be required to hold allowances for compliance purposes.152 As noted above, there are two basic options: a downstream approach and an upstream approach.153 A downstream program would require firms to hold allowances to cover their GHG emissions.154 An upstream approach, by contrast, would limit emissions by requiring fuel suppliers to hold allowances for the carbon content of fuel they sell to downstream emitters.155 A limit on the carbon content of fuel equates to a limit on CO2 emissions because, with a few minor exceptions, all of the carbon in fuel sold downstream is fully combusted as CO2.156 Programs that combine downstream and upstream approaches also are possible.
a. Downstream Cap-and-Trade
A downstream program has the political and administrative advantages of familiarity. The CAA acid rain provisions for electricity generators is widely regarded as a success and could be relatively easily adapted for GHG trading for those firms. A number of extant proposals for a domestic GHG regulatory program have focused on the establishment of a CO2 cap-and-trade program covering the electricity-generating sector.157 In addition, the European Council has approved the establishment of a downstream cap-and-trade program for the member countries of the European Union.158
However, a pure downstream approach to regulating U.S. GHGs has a fundamental drawback: it could not feasibly be applied on an economy-wide basis. Sources of CO2, the primary GHG, number in the hundreds of millions. The sources include not only large facilities, such as those in the electricity generating sector, but also households and vehicles. The administrative costs of allowance allocation, monitoring, and enforcement for so many sources, especially the small ones, would likely be prohibitive.
Realistically, a downstream trading program could encompass only a subset of emission sources, such as electricity generators and other large stationary sources. While such a large-source downstream program would not be hindered by the administrative impediments associated with an economy-wide downstream program, it could reach, at most, less than half of the nation's CO2 emissions, primarily because it would not reach emissions from the transportation and building sectors.159 In terms of dollars per ton, a limited downstream program would likely be more costly than a more comprehensive emissions trading program that met the same reduction target. The full burden of achieving the emissions objective would fall on electricity generators and large industrial sources. Low-cost abatement opportunities in other sectors could be lost.160 In addition, a limited downstream program could lead to leakage-that is, firms would have incentives to shift production from regulated to exempt facilities.161 For example, if the program applied only to industrial sources above a certain size, output-and therefore emissions-might shift to sources below the size cutoff. Finally, opting for a large-source downstream cap-and-trade program instead of a program with economy-wide coverage would raise the long-term cost of achieving more substantial emission reductions because the sectors left unregulated would lack incentives to begin investing in low-carbon technologies, and instead might lock in higher-emitting technologies and practices.
A downstream cap-and-trade program that focused on electricity generators and large industrial sources still could be designed to achieve substantial emission reductions. The electricity-generating sector accounts for approximately forty percent of the U.S. CO2 emissions and ten percent of world emissions.162 The choice to start with a limited downstream program would not necessarily preclude moving to a more comprehensive upstream program later. The second stage of the program could be an upstream program for other sectors of the economy. In the alternative, policymakers could shift the point of regulation from electricity generators to upstream fuel suppliers, in which case, the former could sell any of their banked allowances to the latter. However, such a transition may be difficult because program participants may develop vested interests in the persistence of the program in a particular form.
b. Upstream Cap-and-Trade
While a realistic downstream emissions trading program could reach at most about fifty percent of U.S. emissions, it would be feasible to address virtually all sources of U.S. CO2 emissions through an upstream emissions trading program.163 The Center for Clean Air Policy has found that an upstream program involving fewer than 2000 regulated facilities-approximately the same number of regulated facilities that are subject to the CAA acid rain program-could reach virtually all of the CO2 emissions in the U.S. economy.164 These 2000 facilities would include a combination of petroleum refineries, oil importers, natural gas pipelines, natural gas processing plants, coal preparation plants, and certain coal mines where the production bypasses preparation plants. Fuel data is generally available for these firms, thereby easing the reporting burden on the firms and the monitoring and enforcement burden on the government.165 Like a downstream system, an upstream emissions trading program would give downstream energy users the incentives and the flexibility to implement the most cost-effective means of reducing their emissions.166 However, the incentive would take a different form. Instead effacing limits on their emissions, downstream sources would face limits on the physical availability of carbon-based fuels, which, in turn, would be reflected in fuel price increases. Theoretically, downstream firms and consumers should respond to this price signal in the same way as they would to a requirement to hold allowances directly-that is, under an upstream emissions trading program, the cap on fuel carbon would induce downstream sources to adopt the least-cost mix of emission reduction measures. Whether in practice the impacts on fuel use, technical innovation, and efficiency will be the same is not possible to predict. However, because an upstream emissions trading program feasibly, though indirectly, would reach all sources of CO2 emissions, such a program arguably could achieve any given emissions reduction objective at less cost than a large-source downstream program.
Some commentators argue that an optimal domestic program would combine an upstream cap-aiid-tracle program with enhanced product standards.167 Their rationale is that, from a societal view, consumers often do not respond efficiently to changes in the price of energy.168 For example, studies suggest that drivers do not take into account fuel costs savings over the entire useful life of a vehicle in deciding what level of fuel economy they want from a new vehicle.169 This potential failure of some end-users to respond efficiently to a price signal does not affect the environmental effectiveness of an upstream cap-and-trade program because such a program imposes an absolute cap on the carbon content of fuel used in the economy. On the other hand, if consumers do not respond efficiently to the price signal, a disproportionate share of the burden of meeting an emissions cap could fall onto firms in the electricity-generation and industrial sectors, potentially diminishing the overall cost-effectiveness of the program.
These commentators argue that supplementing the upstream program with efficiency standards-such as modified CAFE requirements-could address these market failures by forcing more energyefficient products into the marketplace.170 For example, the program originally proposed by Senators McCain and lieberman in January 2003 would have established an upstream cap on transportation sector emissions, but also would have incorporated incentives for automakers to sell more fuel-efficient cars.171 The latter element was removed from the version of the McCain-Lieberman bill that was voted on in the Senate in October 2003.
Of course, any economy-wide upstream approach implies that households will see price increases in gasoline and home heating fuels. Policymakers concerned about shielding households from such price increases might prefer alternatives to an economy-wide approach, such as a downstream cap-and-trade program, which would shield consumers from fuel price but not electricity price increases or a program that combines a downstream program with product efficiency standards. In assessing these alternatives, however, it is important to keep in mind that program designs that shield households from overt price increases for gasoline and home heating fuels do not necessarily shield them from higher costs. Alternative programs would put greater pressure on other sectors to achieve the emissions target; their compliance costs would come back to households in the form of higher prices for electricity and other goods and services. Indeed, because alternative designs are less efficient, the overall costs faced by households likely would be higher under such designs. Additionally, any program design that fails to provide a key sector with incentives to start shifting to loweremitting practices and products increases the costs of achieving deeper emission reductions in the future. Finally, it should be noted that the impact of an upstream program on fuel prices can be controlled, either by starting with a moderate emissions cap or, as discussed in Part IV.A.4 below, by incorporating a safety valve approach.
c. Upstream/Downstream
Another approach is an upstream/downstream program that would use the familiar design of the Clean Air Act acid rain program for electricity generators, but would cover other sectors, such as the transportation sector, with an upstream program. The program proposed by Senators McCain and lieberman reflects such an upstream/downstream design.172 An upstream/downstream program would require upstream suppliers of fuel, such as refiners, gas pipelines, and processors, to hold allowances sufficient to cover the carbon content of fuel they deliver, subject to an exemption for deliveries to firms, such as electricity generators, that are subject to downstream regulation. These downstream firms, in turn, would be required to hold allowances for their emissions.
An upstream/downstream cap-and-trade that subjected electricity generators to downstream regulation and made all transactions for other uses subject to upstream regulation would end up with a somewhat greater number of regulated firms and would require a significantly more complex administrative system.173 For example, because electricity generators' fossil fuel usage would be subject to a downstream allowance requirement, refiners' sales of fuel oil to those generators would be exempt from the upstream allowance requirement. This arrangement would make fuel destined for electricity generators less expensive than fuel destined for non-generators, such as truckers and building owners, thus creating an incentive for generators to buy fuel and resell it to others. Regulatory controls would be needed to prevent such behavior.
d. Sectoral Hybrids
A fourth approach would combine a cap-and-trade program covering large sources with efficiency standards for smaller sources in the transportation sector and the residential and commercial buildings sector. This hybrid approach is discussed in Part IV.C.
4. How Should Allowances Be Allocated?
Once a cap is set for the cap-and-trade program and it is determined what firms will be regulated, then a number of allowances equal to the cap must be distributed for use within the economy. The process for this distribution-the allowance allocation methodology-is likely to be the most difficult and potentially contentious issue in designing a cap-and-trade program.
There are two fundamental choices for allowance allocations: (1) distribution of allowances at no cost to firms affected by the regulatory program, or (2) an auction under which the government sells allowances to the highest bidder, and uses the proceeds to compensate affected firms, workers, or communities, to reduce taxes, or some combination of the above. If free distribution is chosen, then policymakers will have to decide how to allocate allowances to firms or individuals. Several allocation methods have been suggested, based on experience with existing electric generator cap-and-trade programs. These include "grandfathering," where allowances are allocated based on emissions prior to the start-up of the regulatory program, and a "generation performance standard," which allocates allowances based on post-start-up electric output, measured either as of a certain date or on the basis of a periodic updating.174 If an auction is chosen, policymakers will have to decide on the disposition of the revenues from the auction (revenue recycling). Recycling alternatives include direct compensation to affected firms, workers, communities, or consumers, and reductions in taxes on labor and capital.
The choice between auction and free distribution, and the subsidiary choices respecting allocation method and revenue recycling, have important implications both for the one who bears the cost of the program and the program's overall cost. These choices and their implications are explored below.
a. Free Distribution
Under conventional "command-and-control" environmental regulation, the regulated firm bears the direct costs of limiting emissions to an allowable level, but is not required to purchase its entitlement to allowable emissions.175 Free distribution of allowances reaches a comparable result under market-based regulation by providing an initial free allocation of allowances to firms affected by the program. This approach was used in the CAA acid rain program.176
A number of recent studies argue for a departure from the acid rain model in the context of a GHG cap-and-trade program, for several reasons.177 First, the studies indicate that the allowances created by a U.S. GHG trading program could have substantially greater value than acid rain allowances under the CAA.178 Second, these studies-if correct-indicate that free distribution of 100% of allowances to regulated firms would overcompeiisate them for their lost profits, because these firms can pass through to customers much of their costs of compliance. One study found that for a stand-alone electric generator cap-and-trade program, free distribution of all allowances to the electricity generators would increase the sector's net worth by fifty percent,179 implying-if the analysis is correct-that power producers in the aggregate would be better off with mandatory GHG regulation modeled on the acid rain program than they would be with no GHG regulation at all. Of course, the impacts on allowance recipients would depend on the method of free distribution and would vary on a firm-by-firm basis. That is, utilities heavily reliant on coal would fare worse than utilities with natural gas, nuclear, or renewable power plants.180 In addition, overcompensation might not be an issue for utilities subject to cost-of-service regulation if, for purposes of setting customers' rates, regulators required that utilities value the allowances at their cost-zero-rather than at their market value.181
In any event, distribution of all of the allowances to firms subject to the cap would do nothing to alleviate the financial losses borne by firms and consumers not subject to the cap. For example, absent some compensation mechanism under any GHG cap-and-trade program, both coal producers and owners of coal-fired power plants would suffer a substantial proportion of the financial losses resulting from the emissions cap. Yet, under a downstream program where allowances were allocated only to electricity generators subject to the cap, coal producers would receive no relief, even though, according to one study, their projected equity losses could be more than sixty percent.182 Nor would this approach to allowance allocation provide any relief to coal miners who might face significant losses in income. Similarly, under an upstream trading program, distribution of allowances to fuel transporters and processors subject to the cap would do nothing to address the financial losses of the electricity generators downstream from the point of regulation, which would be paying more for coal and natural gas. Moreover, neither approach would address the impacts on other firms and on households, both of which would face significantly higher energy prices as a result of either an upstream or downstream trading program,183 or the likely reduction in federal tax revenues because of reduced levels of economic activity attributable to the program.184
For these reasons, these recent economic studies urge policymakers to de-link the allocation of allowances from the incidence of regulation, and to link it instead to economic losses attributable to the regulatory program. In this regard, an important finding of the allowance allocation studies is that the government might need to distribute only a relatively small percentage-six to thirteen percent-of the total allowances to energy sector firms to eliminate their equity losses from an efficient upstream cap-and-trade program.185 If correct, this means the government could distribute the value of the balance of the allowances to achieve other ends-for example, to assist burdened firms outside the energy sector, to help consumers, to aid particularly hard-hit workers or their communities, or to prevent a decline in government revenues.186 It should be noted, however, that these conclusions are critically dependent on the details of the regulatory program and on modeling techniques. As discussed below, if the regulatory program were less efficient than an upstream cap-and-trade program, or if the period for allocation of allowances to compensate affected energy firms were limited to ten years for example, then the percentage of allowance value allocated to the firms would have to be higher and the percentage available for other uses would be, at least initially, much smaller.
These studies also add an important perspective on the longstanding debate on whether, in a free distribution regime, allowances should be allocated on a grandfathering or generation performance standard basis. If a policy objective of allowance allocation is to compensate firms affected by the cap-and-tracle program for their lost profits, allowances should be distributed to firms on the basis of their projected financial losses from the emissions cap, not past emissions or current output. This is because a firm's economic losses are not necessarily related to its output or emissions. Determining the amount of such losses on a firm-by-firm basis could be complex, but it could be done administratively in the same way "stranded investment" is determined in electric restructuring proceedings.187
b. Allowance Auction and Revenue Recycling
A number of economists and policy analysts advocate for the governmental distribution of allowances through an auction or, alternatively, through a fiduciary.188 They cite two advantages of this approach.189 First, it could potentially provide a less cumbersome mechanism for distributing the value of the allowances to groups suffering financial losses from a GHG emissions cap. Instead of giving consumers and others allowances to sell, the government itself could sell the allowances and recycle the revenue to the economically vulnerable groups through lump-sum payments or aid programs; the government also could retain some of the revenues to prevent erosion of the federal tax base.
Recycled revenues could further be used to reduce distortionary taxes that produce a net drag on the economy. Economists argue that existing wage-related taxes create a disincentive to work, and that existing taxes on interest, dividends, capital gains, and corporate income discourage productive investments. According to this argument, using the proceeds of an allowance auction to reduce taxes on income or investment, rather than as a means of direct compensation, would result in overall economic gains that could significantly reduce the cost of GHG regulation to society as a whole.190
A number of recent studies, including a study conducted by Charles River Associates (CRA), described below in Box 3, look at the efficiency implications of different design options for a cap-and-trade program.191 While the quantitative results of these studies are very much dependant on modeling assumptions, they are useful in illustrating the interactions of the design elements of a cap-and-trade program.192 First, compensatory allowance allocations to energy sector firms, if they are made on a permanent basis, appear to require only a small percentage of allowances or allowance revenues.193 However, if these allocations are made over a relatively short transition period-ten years, for example-then the percentage allocated to these firms will have to be much larger in the early years of the program. Second, if the program is designed to be revenue-neutral to the federal government and policymakers assume no offsetting fiscal benefits from the program, then a large proportion of auction revenues-in the CRA analysis of an upstream cap-and-trade program, about fifty percent-would be retained by the government.194 Third, in the CRA analysis of an upstream cap-and-trade program, once prior claims are satisfied, the allowance proceeds that remain are sufficient to reduce the social cost of the trading program by thirty-five percent, if the proceeds are dedicated to reducing marginal tax rates.195 Fourth, if policymakers settle on a program that is less cost-effective than an upstream cap-and-trade program, then the costs imposed by the regulatory program will increase while the total allowance proceeds available to the government to address those costs will decrease.
For all of these reasons, choices respecting allowance allocation have important implications, both for cost-effectiveness and distributional equity. They also raise political feasibility issues. For example, the practice of requiring regulated firms to purchase allowances through a government auction could be characterized by its opponents as a new tax. In addition, obtaining the efficiency benefits of revenue recycling implies taking on not only development of a GHG regulatory program but also tax reform.
5. Emissions Certainty Versus Cost Certainty
The third critical design issue in designing an emissions trading program is determining what balance to strike between certainty about achieving a particular level of emission reductions and certainty about costs of compliance.
Policymakers can limit the costs of complying with an emissions trading program through a safety valve feature, which would authorize the government to sell additional allowances at a predetermined price.205 With a safety valve mechanism in place, the market price of allowances-and therefore the marginal cost of abatement-will rise no higher than the safety valve price.206 The effect is to cap compliance costs.207
Establishing a safety valve, however, implies that emissions are not capped. If compliance costs turn out to be higher than expected, firms can purchase more allowances and total emissions can rise above the cap. This is not to say that emissions would be entirely unlimited-firms would have to pay the safety valve price to increase their emissions-but the safety valve option would mean that there was not a precise and absolute cap. Thus, the safety valve option presents policymakers with a potential trade-off between emissions certainty and compliance cost certainty.
How important is certainty about meeting a particular emissions cap? The atmosphere is not particularly sensitive to small differences in emission levels. Scientists have not identified a particular threshold level over which the potential for damage is great. For these reasons, policymakers might not attach significant value to assurances that the United States will meet a particular, near-term emissions target with precision.
By contrast, assurances that the compliance costs will not rise above a particular per-ton level could be central to building political support for moving forward on climate change. There are significant differences in opinion on how much it will cost to reduce GHG emissions in the United States because the cost would be largely a function of future levels of economic activity, which are difficult to forecast.208 Yet, establishing a cap-and-trade program without a safety valve mechanism means that the cap will have to be met, regardless of cost. A safety valve mechanism can help remove cost uncertainties as a barrier to action.209
Some commentators argue that a safety valve mechanism inevitably would be an "easy out," diminishing incentives for firms to innovate or to build a bank of early reductions, both of which are key factors in driving down the long-term costs of reducing emissions.210 However, other commentators have argued that a safety valve option could make risk-averse households and firms willing to accept a more aggressive emissions cap-and therefore higher emissions price-than otherwise would be the case, because they would have assurances that their costs would not exceed the safety valve level.211
Ultimately, the decision whether to adopt a safety valve approach could depend on the timing and stringency of the regulatory program. Because the United States has elected not to become a party to the Kyoto Protocol, at least for now, U.S. policymakers have flexibility in setting the emissions target and compliance timetable for a domestic climate program.212 They might opt for a gradual approach, that is, an approach that aims to make a start in reducing U.S. emissions while keeping compliance costs low. If policymakers design such a moderate, and therefore relatively lower-cost program, they might reasonably conclude that a safety valve provision is unnecessary and opt instead for certainty in meeting the target. Alternatively, they could incorporate a safety valve in the program's early stages and raise the safety valve price over time.213
6. Evaluation of Cap-and-Trade Programs
Environmental Effectiveness: A cap-and-trade program, if comprehensive in coverage and properly administered, can be highly effective ill meeting its target. A comprehensive upstream cap-and-trade program would be environmentally effective as to CO2, but may not be feasible for other gases or sinks. A large-source downstream program could be equally effective with respect to the sectors it covered, but would have to be supported by other measures to provide full coverage. An all-sector downstream program is likely to be ineffective because it could be administered and enforced only with great difficulty.
Cost-Effectiveness: Cap-and-trade programs, if they include flexibility measures, can attain emission reductions at low cost. Allowance allocation policies could increase or decrease the costs imposed by the program.
Administrative Feasibility: An upstream cap-ancl-tracle program appears to be fully administrable for CO2 and for certain other GHGs. An all-sector downstream cap-and-trade does not appear to be feasible because of the number of regulated firms involved. A hybrid program that combines a downstream cap-and-trade for large sources with an upstream program applicable to suppliers of fuel for smaller sources appears to be feasible, though somewhat more complex than a full upstream program.
Distributional Equity: The distributional consequences of a capand-trade program depend critically on how allowances are allocated, or-if they are auctioned-how the auction proceeds are distributed.
Political Acceptability: Because any all-sector cap-and-trade program, whether upstream or downstream, will drive up consumer costs for gasoline, natural gas, and home heating oil, it is likely to be politically difficult. An all-source downstream cap-and-trade, because it implies regulating millions of sources, is likely to be even more difficult. A downstream cap-aud-trade program limited to electricity generators and other large stationary sources could be more acceptable politically, but to be effective it would have to be coupled with a regulatory program to cover other sectors.
B. GHG Tax
1. Overview of GHG Tax Approach
Another market-based approach to reducing GHG emissions is a GHG tax program. Under such a program, policymakers would impose a per-ton fee on COo emissions or on the carbon content of fuel. Other GHG emissions, to the extent measurable, would also be taxed. In addition, the program could be designed so that firms earn a tax credit for CO2 emissions reduced through land-based sequestration projects, carbon capture projects, or for project-based reductions in GHGs that are not subject to tax. Firms subject to the tax would have an incentive to reduce their emissions-thereby avoiding the tax-until the cost of achieving reductions exceeded the cost of paying the tax. Accordingly, as with an emissions trading program that incorporates a safety valve, a tax program would provide near-term cost certainty but not absolute near-term emissions certainty.
A tax program would offer practically all of the flexibility, and therefore cost-effectiveness, of an emissions trading program. Firms would have the incentive and the opportunity to adopt the lowest-cost means of reducing their energy-related emissions; the "payment" for additional reductions would take the form of tax savings. Just as firms subject to an emissions trading program could bank excess allowances, firms participating in an emissions tax program could literally bank their tax savings from reducing their emissions. Tax credits also could be made available for emission reductions achieved through projects financed in other countries or for valid emission allowances acquired from other countries' regulatory programs.214
Designing a domestic GHG tax program would raise some of the same fundamental issues that arise in designing an emissions trading program. For example, it would be necessary to determine whether the program should tax upstream firms, downstream firms, or some combination of the two. The analysis is largely the same as that for an emissions trading program. A downstream tax would take the form of a tax on CO2 and certain other GHG emissions. Because enforcing the tax would require tracking the emissions of each firm subject to the tax, a downstream emissions tax, like a downstream emissions trading program, could not feasibly reach all of the hundreds of millions of sources of CO2 emissions in the economy. An upstream tax program would take the form of a tax on the carbon content of fuels sold into the energy system. Like an upstream emissions trading program, an upstream GHG tax could be applied to a few thousand firms that produce, refine, and market fuels. The tax on these firms would lead to higher prices for carbon-intensive fuel and higher prices for energ. The program thus could effectively regulate the entire energy system, providing downstream firms with incentives to switch fuels, increase energy efficiency, and reduce energy use.
A tax program would raise revenue in much the same way as would an emissions trading program with an allowance auction. Accordingly, a tax program would offer an opportunity to reduce distortionary taxes on labor or capital. In addition, revenues from the tax could be used for any of the purposes described above with regard to revenues from an allowance auction, such as assisting vulnerable workers and communities.
In addition, it is possible to design a tax program to mimic the effect of free distribution of allowances under an emissions trading program. How would this work? The tax program could offer an exemption from the tax up to a certain fixed amount of tons of carbon supplied (upstream) or emitted (downstream).215 As with free allocation of allowances, a tax program could base the size of the exemption on particular characteristics of the firms, such as output in a base year. The tax program still could achieve its environmental objective so long as firms remain subject to the tax at the margin, that is, for the last tons supplied (upstream) or emitted (downstream). However, as with free allocation of allowances in an emissions trading program, a modified tax would reduce the burden of the program on those firms directly subject to the tax, but would not assist firms and consumers suffering indirect costs from the tax program. They would not pay the tax and therefore would not benefit from the partial exemption, yet they still would bear financial losses as a result of the program. Of course, opting for this kind of modified GHG tax would reduce the total revenues brought in by the tax program and therefore reduce the ability to achieve other objectives with those revenues.
Finally, the major problem with a GHG tax is that it is a tax. As noted above, U.S. experience since 1973 indicates that taxes as an instrument of energy or environmental policy, no matter how pressing the need, have not been accepted by Congress or the public.216 Thus, a workable GHG tax system could be devised, but its adoption would appear to contradict conventional political wisdom. On the other hand, a GHG tax system could be politically palatable if it were an integral part of a comprehensive reform of the tax code, in which the GHG tax replaced or reduced other, even less popular taxes.
2. Evaluation of GHG Tax Approach
Environmental Effectiveness: A GHG tax program (upstream) could be highly effective in reducing U.S. GHG emissions because of its economy-wide coverage of CO2 emissions. However, if certainty in meeting a particular short-term emissions target were a priority, a tax program would be less preferable than an upstream cap-and-trade program. As with a trading program, sources and sinks not amenable to direct taxation would have to be addressed through a tax credit mechanism or through standards.
Cost-Effectiveness: A GHG tax program would offer all sources incentives and opportunities to adopt their least-cost mitigation options. As a price-based program, a tax program would offer certainty as to compliance costs. As with allowance allocation, different use of the tax revenues could decrease or increase the total cost imposed by the program.
Administrative Feasibility: An upstream GHG tax program would not present significant administrative complexities. An economy-wide downstream GHG tax program, on the other hand, would be administratively infeasible.
Distributional Equity: The distributional consequences of a GHG tax depend critically on how tax revenues are used.
Political Acceptability: Experience suggests that Americans are reflexively opposed both to tax programs and to gasoline price increases. The GHG tax combines the two.217 A GHG tax approach might have some appeal if introduced as part of a tax reform package that included cuts in income or payroll taxes.
C. Sectoral Hyrid Program
1. Introduction
The final of the three major design options for a U.S. climate change program analyzed in this paper is a sectoral hybrid program. A sectoral hybrid program could combine a downstream cap-andtrade program for large sources in the electricity and industrial sectors with enhanced product efficiency standards to cover small GHG sources-mainly consumer products and equipment-in the transportation, residential, and building sectors.218 This hybrid approach has the potential of avoiding some of the political challenges associated with a full upstream cap-and-trade program or GHG tax.
A sectoral hybrid program would provide a mechanism to reach transportation and household emissions that policymakers may be unwilling or unable to regulate directly-as through a downstream cap-and-trade program, or through regulation of fuels-as through an upstream cap-and-trade program. The standards component of the program would regulate the performance characteristics of newly-manufactured products used in the transportation sector and in the residential and commercial buildings sector. For example, while it would not be administratively feasible to directly regulate every household on the basis of its furnace use or every motorist on the basis of emissions resulting from use of his or her motor vehicle, it would be feasible to implement standards that force more energy-efficient furnaces and more fuel-efficient cars into the marketplace.
One advantage of using product efficiency standards to complement a cap-and-trade program is that policymakers could build upon the energy efficiency standards already in effect under U.S. law. Some of the flexibility benefits of a cap-and-trade program can be attained by establishing "tradable" standards, thus providing a degree of exchange between sectors subject to a cap-and-trade program and sectors subject to standards. While a sectoral hybrid program could be attractive to policymakers because it starts with familiar elements, it would require addressing or accepting a number of problematic aspects of a product efficiency standards program. A sectoral hybrid still would leave noticeable gaps in emissions coverage, unless current efficiency standards were significantly expanded. In addition, transforming conventional standards into tradable standards and coordinating the standards program with the cap-and-trade program would pose considerable administrative challenges. And even if these obstacles could be overcome, a standards program remains inherently less cost-effective than a full upstream cap-and-trade program because standards do not provide any incentives to reduce use, nor do they dictate the rate at which end-users replace their old products for more efficient, new ones.
2. Designing the Standards Component of a Sectoral Hybrid Program
Designing the product efficiency standards component of a sectoral hybrid would involve a number of steps. First, it would be necessary to adapt existing standards to the new purpose of regulating GHG emissions. Second, policymakers may decide that it is necessary to develop new standards for products and processes not now covered by standards. Third, policymakers may want to formulate many of the standards as tradable standards.
Most existing standards are expressed in terms of an energy efficiency requirement, for example, miles per gallon. In a climate program, policymakers would need to translate these standards from energy per unit of output to GHG emissions per unit of output, or at least adjust the standards to reflect the carbon content of different fuels. To achieve broad coverage of emissions, a sectoral hybrid prograin would necessitate the establishment of a range of new standards. While standards currently are in place for most major energy-using consumer products and equipment-including motor vehicles and residential and commercial natural gas and oil-fired equipment-standards do not apply to most commercial and industrial equipment.219 Federal standards also do not apply to building envelopes, that is, heat loss and heat gain from buildings. For example, air conditioner standards will ultimately result in more efficient air conditioners replacing less efficient ones, but they do not deal with energy loss from uninsulated buildings. Most importantly, however, standards are not currently in place for a range of sources in the transportation sector, including locomotives, vessels, aircraft, buses, and heavy trucks; these uncovered sources accounted for nearly fifty percent of GHG emissions in the transportation sector in 2002.220 For these reasons, combining a large-source cap-and-trade program with existing standards only would reach approximately eighty percent of the nation's energy-related CO2 emissions.221
Another design consideration is the inflexibility of conventional standards. Typically, standards reflect a command-and-control approach-that is to say, they prescribe a uniform emissions limit or technology without regard to the varying circumstances of the regulated firms. Accordingly, reliance on conventional standards would mean forgoing the flexibility benefits of emissions trading.
One solution to this problem is to design tradable standards. How would such standards work? A tradable standards program would use estimates of the average life and use of a product to translate overcompliance with a standard into a stream of emission allowances assigned to particular years, known as "vintaged" allowances.222 Conversely, the program would translate a failure to achieve the standard into an annualized deficit of allowances. Box 5 provides a detailed description of how a tradable standards program for motor vehicles-referred to as "Corporate Average Carbon Efficiency" (CAGE) standard-could work.
Tradable standards would increase flexibility and therefore reduce the cost for firms to comply with standards. Such an approach could provide for at least three levels of trading: (1) ultra-firm trading, in which a firm could achieve an average level of efficiency across its product lines, instead of being required to meet the standard for each product line; (2) trading among firms subject to standards; and (3) trading between firms subject to standards and firms subject to the cap-and-trade program.
As discussed in greater detail below, a potential drawback of a tradable standards approach-and, indeed, any approach that relies on standards-is that it does not ensure that emissions will be limited at any particular level. An alternative approach that could address this drawback is a capped tradable standards program.223 Under such an approach, policymakers would set a cap on the total emissions associated with particular types of newly manufactured products. To sell products subject to the capped standard, manufacturers would have to obtain and surrender allowances. In other words, it would not be sufficient merely to produce products that met the standard; manufacturers would have to account for the projected emissions associated with each product they sold. A capped tradable standards program would entail resolving a number of design issues, including issues related to allowance allocation, shutdowns, new market entrants, changes in manufacturer market share, and changes in overall level of output.
It is important to note that either tradable standards or capped tradable standards could raise intra-industry competitiveness issues. Firms with a wide range of product lines may be able to generate internal allowances from efficient product lines that can be used to "subsidize" inefficient products in other product lines-arguably to the competitive detriment of single product line manufacturers.224
3. Integrating Tradable Standards with a Cap-and-Trade Program
Developing a domestic program that combines tradable standards with a cap-and-trade program raises an additional design issue. If trading is allowed between firms subject to standards and firms subject to a cap, how will such trading be regulated so as to prevent double-counting of reductions?
For example, if a firm that manufactured an electrical appliance, such as a refrigerator, exceeded the efficiency standard for that product, the resulting improvement in efficiency would reduce electricity use, and therefore reduce emissions by electricity generators. If the refrigerator manufacturer earned a tradable allowance for its overcompliance and the resultant emissions reduction by the electric generator also created a surplus allowance, the same ton of CO2 reductions would generate two tons of allowances. To deal with this problem, manufacturers of electricity-consuming products that are subject to standards could be precluded from trading outside their own sectors. However, there would be no reason not to allow them to trade between electric product lines and with other firms subject to electric product efficiency standards.
Double-counting would not be an issue for products that emit CO2 directly, such as automobiles or gas appliances. Accordingly, a hybrid program could permit manufacturers of such products to trade freely into the cap-and-trade market.226
4. Sectoral Hybrid Approach Issues
Even a well-designed sectoral hybrid program has some significant drawbacks compared to an economy-wide upstream capand-trade program.
First, standards provide no incentive to adopt what, in many cases, may be the lowest-cost abatement option: reduced use. As explained below, the absence of any incentive to reduce use means that a standards approach-even if the standards are tradable-may be a significantly less cost-effective means of meeting any emissions limit compared to a cap-and-trade program that regulates fuel producers or end-users directly.
In the transportation sector, for example, standards would force lower-emitting vehicles into the marketplace, but they would not provide any incentive for motorists to drive less. Indeed, if gasoline prices were to stay the same, motorists that purchased compliant vehicles might increase their miles traveled because more fuel-efficient vehicles cost less to drive. This "rebound effect" could offset some of the projected emission reductions.229 In addition, standards only apply to new products, not existing products. Accordingly, the effectiveness of standards in limiting emissions would depend on the rate at which consumers replaced their old, unregulated products with the new, regulated ones.230 Gauging this rate is complicated by the "junker effect": subjecting products to standards may increase their price, thereby encouraging consumers to hold onto their unregulated, higher-emitting models.231
A third potential drawback of a standards approach is that it relies heavily on estimates. For each standard, policymakers would have to formulate various estimates of lifetime product use and associated emissions, as well as rates of adoption. Even a capped tradable standards program would cap only projected, not actual, lifetime emissions associated with covered products. The heavy reliance on estimates means that a hybrid program would offer substantially less certainty about meeting emission reduction goals than a cap-and-trade program.
Fourth, a hybrid program also would be more difficult to administer over time than an upstream cap-and-trade program. With an upstream cap-ancl-trade program in place, gradually phasing in more ambitious national emission targets would involve little more than gradually ratcheting down the economy-wide cap. Achieving the same result with a hybrid program, on the other hand, would involve continuously promulgating adjustments to multiple standards for multiple sectors.
To date, none of the economic studies has modeled a sectoral hybrid with tradable standards. However, analysis of hybrid programs without the trading feature indicates that these programs entail significantly greater cost when compared with an upstream cap-and-trade program. It is possible that a sectoral hybrid program could be substantially less cost-effective, as well as more administratively burdensome, than an economy-wide upstream cap-and-trade or tax program.232
5. Evaluation of Sectoral Hybrid Approach
Environmental Effectiveness: A sectoral hybrid would have higher environmental effectiveness than a downstream program alone, because standards could address emissions from sources that could not be covered by a downstream cap-and-trade program. On the other hand, it would be less effective than an economy-wide upstream capand-trade program because standards would not address the intensity of product use or the replacement rate of new products for old. In addition, not all sources that fall outside a large-source cap-and-trade program could be regulated through standards.
Cost-Effectiveness: A sectoral hybrid program would be a more costly means of achieving any particular emissions target than an economy-wide upstream cap-and-trade program. The ultimate cost of the sectoral hybrid option also would depend on, among other things, whether the standards were tradable standards.
Administrative Feasibility: It would be relatively straight-forward to modify existing efficiency standards for purposes of a sectoral hybrid program. However, transforming such conventional standards into tradable standards would present some new complexities. Capped tradable standards present significant design issues. In addition, hybrid programs are significantly more complex administratively than are any of their individual elements because of the need for coordination. Trading would need to be carefully regulated to prevent double-counting of emission reductions and evasion of allowance requirements. In addition, promulgating new standards to cover products and practices not now subject to standards also would be an administrative burden.
Distributional Equity: A sectoral hybrid program could exclude households from the direct burden of regulation under the cap-and-trade program. However, such an approach would increase electricity prices and would put additional burdens on the manufacturing sector. These burdens would be felt indirectly by households in the form of higher product prices. The ultimate distribution of impacts from a sectoral hybrid program also would depend on how policymakers opted to distribute allowances, or the revenues from an allowance auction.
Political Acceptability: Hybrid programs offer policymakers options for addressing domestic GHG emissions while avoiding gasoline and home heating fuel price increases. In particular, a sectoral hybrid program would offer a means of largely avoiding these price increases-but not electric rate increases-albeit at some cost to environmental effectiveness. Also, the familiarity of the standards component may enhance its acceptability.
CONCLUSION
Policymakers in the United States face a plethora of choices for the design of a domestic GHG regulatory program-upstream or downstream cap-and-trade, GHG tax, product standards, and hybrid programs-as well as the myriad details of program design that must be addressed once the overall approach is chosen.
Using the criteria spelled out in Section IV, we evaluated the principal design options. The results of that evaluation are suinmarized below.
A. Cap-and-Trade Programs
1. All-Source Downstream Cap-and-Trade
An economy-wide downstream cap-and-trade program-because it implies the regulation of millions of individual GHG sources, including cars and homes-would be difficult and costly to administer. It would not be a viable prospect for a domestic GHG regulatory program.
2. Upstream Cap-and-Trade
An economy-wide upstream cap-and-trade program would be environmentally effective, could attain least-cost compliance if it incorporates flexibility measures, and would be administratively feasible. Its distributional consequences would depend on how allowances were allocated and, if auctioned, how the auction revenues were recycled back into the economy. These allocation and revenue recycling decisions would influence overall compliance costs. Some methods of allocating allowances, such as generation performance standards, can be less economically efficient than others, and can be less efficient than an auction. According to some economists, using auction revenues to reduce clistoi tionary taxes could partially offset the costs of the program. Finally, because an economy-wide upstream cap-andtrade program would drive up the cost of gasoline and home heating fuels, it is likely to present a political challenge.
Thus, if policymakers were willing to accept a program that results in visible increases in gasoline and home heating fuel prices, one environmentally effective, efficient, and feasible option would be a comprehensive upstream cap-and-trade program. Such a program could be coupled with limited free distribution of allowances to compensate affected business, auction of the remaining allowances, and the use of auction revenues for tax reductions and other ends.
There are substantial theoretical benefits from such an approach. The near-term environmental outcome is clear, assuming that the government will maintain the emission limits in the face of possibly significant price uncertainty and volatility. Current analysis indicates that it would minimize economic costs to the economy, be manageable administratively, avoid over-compensating existing emitters, and perhaps capture some offsetting benefits from reduction of distortionary taxes.
Nevertheless, an economy-wide upstream cap-and-trade program raises a number of issues. First, critics may characterize it as a large, ambitious, and untried experiment in regulation, and may question how it will work in practice. Second, auction revenues may be difficult to predict, making it difficult to answer the question of whether, and when, Congress will enact such a system. Even in times of most compelling national circumstances, such as the 1973 Arab oil embargo, Congress has not been willing to allow fuel prices to increase sufficiently to bring demand in balance with supply.233 On the other hand, adopting an upstream cap-and-trade program does not inevitably mean accepting a significant and immediate hike in consumer fuel prices. The price impacts could be limited to only a few cents per gallon if the program began with a moderate emissions target and then phased in a more stringent target gradually over time, or if it incorporated use of a safety valve.
A workable variant of the upstream cap-ancl-trade program described above is an upstream/downstream design that combines a downstream cap-and-trade program for electricity generators and other large sources with an upstream cap-and-trade program for other major sectors of the economy. The McCain-Lieberman bill reflects this approach.
3. Large-Source Downstream Cap-and-Trade
A large-source downstream cap-and-trade program-one applicable to electricity generators and large industrial sources of CO2 and of certain other greenhouse gases-is administratively feasible and could be environmentally effective with respect to the sectors it covered. To be fully effective, however, such an approach would have to be coupled with a program to cover other sectors. A large-source downstream program might be more politically acceptable than an upstream economy-wide program because it would not result in price increases for gasoline and home heating fuels, though it still would result in price increases for electricity.
B. GHG Tax
An upstream GHG tax program would allow for adoption of least-cost mitigation strategies, offer short-term cost certainty, and be administratively feasible. A tax program would not provide certainty in meeting a particular short-term emissions target. However, because it is cumulative rather than annual emissions that are important, taxes should be able to provide almost equivalent long-term environmental certainty if there is political will to adjust them over time. The ultimate distributional consequences of a GHG tax would depend on how policymakers distributed revenues from the tax. However, political acceptability is likely to be a major obstacle since a GHG tax combines both new taxes and fuel price increases. A GHG tax may be more politically attractive as part of a larger tax reform program.
C. Sectoral Hybrid Program
A sectoral hybrid program such as the one outlined in Box 4 would consist of a large-source downstream program coupled with product efficiency standards. Such a program would be more environmentally effective than either a downstream program or standards alone, because standards could address emissions from sources-such as automobiles and appliances-that could not feasibly be covered by the downstream cap-and-trade program. Relying on existing standards programs, the first phase of such a program could attain coverage of about eighty percent of U.S. energy-related CO2 emissions. A second phase of the program could address the remaining twenty percent through an upstream cap-and-trade program or through expanded product efficiency standards; the program could cover emissions of other greenhouse gases through other measures.
A sectoral hybrid program has the advantage of building on existing regulation, and in the case of CAFE and appliance standards, potentially improving on it by permitting manufacturers to trade among product lines, with each other, and with other sectors. It would avoid the politically difficult step of attaching a carbon cost to the price of gasoline and home heating. The tradable standard feature would capture some, but not all of the benefits of a full cap-and-trade system.
However, these largely political attractions of the hybrid program could come at some cost. Substituting product efficiency standards for the transportation fuel component of an upstream cap-and-trade program may downgrade the cost-effectiveness of such a program. Even with a trading feature that tries to equate marginal control costs among sectors, a product efficiency standards program lacks incentives to discourage product end-use. Indeed, it might actually encourage greater use-the rebound effect, or encourage consumers to replace their existing inefficient products for the more efficient new ones-the junker effect. The absence of such incentives is likely to make a domestic program that relies on product efficiency standards as an alternative to upstream regulation a more expensive approach to meeting any GHG reduction target. In addition, incorporating tradable standards would present significant administrative challenges because of the need to prevent double-counting of emission reductions and to deal with potential compliance evasion. Finally, any hybrid program is likely to give some beneficiaries of the program a vested interest in retaining it, significantly increasing the difficulty of ultimately converting the hybrid program into a simpler, more efficient economy-wide upstream cap-and-trade program.
In sum, the analysis suggests that an economy-wide downstream cap-and-trade program is difficult to administer, that a stand-alone, large-source cap-and-trade program is incomplete in coverage, and that a GHG tax that is not part of a larger tax reform initiative is unviable politically. The analysis does suggest that the comprehensive, upstream cap-and-trade approach and the sectoral hybrid approach are the most viable alternatives for a domestic GHG reduction program. While an economy-wide cap-and-trade approach may present the best option for low-cost reductions in greenhouse gases, there are a set of existing sector-based approaches that could be built upon to address greenhouse gases, such as the acid rain program for electricity generators, appliance efficiency standards, and motor vehicle fuel economy standards. For a variety of institutional, practical, and political reasons, a U.S. domestic emissions reduction program may evolve in this direction. If policymakers decide on that course, then careful attention will have to be given to minimizing the economic costs and administrative complexity, and assuring that the program can be effectively enforced.
No comments:
Post a Comment