# Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress > Briefs, arguments, decisions, and more. URL: https://www.frixlaw.com/law-library/documents/crs%3AR40242 ## Record - **Collection:** Congressional research report - **Document type:** CRS Report - **Published:** March 10, 2009 - **Citation:** R40242 ## Text Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress name redacted Analyst in Environmental Policy name redacted Specialist in Energy and Environmental Policy March 10, 2009 Congressional Research Service 7-.... www.crs.gov R40242 CRS Report for Congress Prepared for Members and Committees of Congress Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Summary Market-based mechanisms that limit greenhouse gas (GHG) emissions can be divided into two types: quantity control (e.g., cap-and-trade) and price control (e.g., carbon tax or fee). To some extent, a carbon tax and a cap-and-trade program would produce similar effects: Both are estimated to increase the price of fossil fuels, which would ultimately be borne by consumers, particularly households. Although there are multiple tools available to policymakers that could control GHG emissions—including existing statutory authorities—this report focuses on a carbon tax approach and how it compares to its more frequently discussed counterpart: cap-and-trade. If policymakers had perfect information regarding the market, either a price (carbon tax) or quantity control (cap-and-trade system) instrument could be designed to achieve the same outcome. Because this market ideal does not exist, preference for a carbon tax or a cap-and-trade program ultimately depends on which variable one wants to control—emissions or costs. Although there are several design mechanisms that could blur the distinction, the gap between price control and quantity control can never be completely overcome. A carbon tax has several potential advantages. With a fixed price ceiling on emissions (or their inputs—for example, fossil fuels), a tax approach would not cause additional volatility in energy prices. A set price would provide industry with better information to guide investment decisions: e.g., efficiency improvements, equipment upgrades. Economists often highlight a relative economic efficiency advantage of a carbon tax, but this potential advantage rests on assumptions—about the expected costs and benefits of climate change mitigation—that are uncertain and controversial. Some contend that a carbon tax may provide implementation advantages: greater transparency, reduced administrative burden, and relative ease of modification. The primary disadvantage of a carbon tax is that it would yield uncertain emission control. Some argue that the potential for irreversible climate change impacts necessitates the emissions certainty that is only available with a quantity-based instrument (e.g., cap-and-trade). Although it may present implementation challenges, policymakers could devise a tax program that allows some short-term emission fluctuations, while progressing toward a long-term emission reduction objective. Proponents argue that short-term emission fluctuations would be preferable to the price volatility that might be expected with a cap-and-trade system. Although a carbon tax could possibly face more political obstacles than a cap-and-trade program, some of these obstacles may be based on misunderstandings of the differences between the two approaches or on assumptions that the tax would be set too low to be effective. Carbon tax proponents could possibly address these issues to some degree, but there remains considerable political momentum for a cap-and-trade program. Congressional Research Service Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Contents Introduction ................................................................................................................................1 Cost or Quantity Control: An Overview.......................................................................................3 Economic Theory vs. Uncertainty .........................................................................................3 A Stark Choice or a Policy Continuum?.................................................................................5 A Flexible Emissions Cap ...............................................................................................5 A Flexible Carbon Tax ....................................................................................................6 Limits of the Policy Continuum.......................................................................................6 Potential Advantages of a Carbon Tax .........................................................................................7 Economic Efficiency.............................................................................................................7 Basis for the Argument....................................................................................................7 Underpinnings of the Argument..................................................................................... 10 Modeled Efficiency Gains ............................................................................................. 13 Economic Efficiency Versus Precaution......................................................................... 14 Price Stability ..................................................................................................................... 14 Tax Revenue Applications ................................................................................................... 15 Potential Implementation Advantages.................................................................................. 16 Transparency ................................................................................................................ 16 Administrative Issues .................................................................................................... 17 Policy Modification ...................................................................................................... 18 Potential Disadvantages ............................................................................................................ 18 Uncertain Emissions ........................................................................................................... 18 Political Feasibility ............................................................................................................. 19 What’s in a Name? ........................................................................................................ 20 Support from Industry? ................................................................................................. 21 Support from Environmental Groups? ........................................................................... 22 Consideration of International Efforts and Cooperation ....................................................... 22 Coordination with Existing International Efforts............................................................ 22 Maximizing Participation .............................................................................................. 23 International Implementation Concerns ......................................................................... 24 Implementation of a Carbon Tax ............................................................................................... 24 Point of Taxation................................................................................................................. 24 Where to Impose a Carbon Tax? ................................................................................... 25 CO2 Emissions or All GHG Emissions?......................................................................... 27 Which Emissions Sources to Control?........................................................................... 28 Level of Taxation ................................................................................................................ 32 Tax Based on Estimates of Costs and Benefits............................................................... 33 Tax Based on Meeting an Emissions Target ................................................................... 35 Tax Revenue Distribution.................................................................................................... 37 Estimates of Tax Revenues............................................................................................ 37 Uses of Tax Revenues ................................................................................................... 38 Policy Considerations of Different Revenue Applications .............................................. 40 Distributional Impacts................................................................................................... 41 Conclusions .............................................................................................................................. 44 Congressional Research Service Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Figures Figure 1. Illustration of Price Versus Quantity .............................................................................4 Figure 2. Bridging the Gap between Price and Quantity Control ..................................................6 Figure 3. Illustrative Scenario with a Relatively Flat Marginal Benefits Curve.............................9 Figure 4. Illustrative Scenario with a Relatively Steep Marginal Benefits Curve ........................ 10 Figure 5. Illustrative Scenario with Marginal Costs and Marginal Benefits That Are Higher Than Expected............................................................................................................ 13 Figure 6. “Phase 2” Emission Allowance Prices in the European Union’s Emission Trading System ...................................................................................................................... 15 Figure 7. Illustration of Options for Points of Taxation within the Energy Production-toConsumption Chain ............................................................................................................... 25 Figure 8. Emission Allowance Price Estimates under S. 2191 .................................................... 36 Figure 9. Relative Differences in Efficiency Costs between Different Applications of Tax (or Auction) Revenues and No-Cost Allowance Distribution in a Cap-and-Trade Program................................................................................................................................. 41 Figure A-1. Illustration of Relationship between the Stock of CO2 in Atmosphere and Annual CO2 Emissions........................................................................................................... 46 Tables Table 1. CO2 Emissions Per Unit of Energy for Fossil Fuels ...................................................... 27 Table 2. Selected Sources of U.S. GHG Emissions and Potential Applications of a Carbon Tax ............................................................................................................................ 31 Table 3. Estimates of Potential Tax Revenues from Carbon Tax Proposals from the 110th Congress (in 2005 dollars)...................................................................................................... 38 Table 4. Distributional Effects of Carbon Tax with Different Applications of Carbon Tax Revenues ............................................................................................................................... 42 Table A-1. Comparison of Estimated Carbon Tax-Related Price Impacts to Fossil Fuels and Motor Gasoline from Selected Carbon Tax Rates ............................................................. 47 Appendixes Appendix. Additional Information............................................................................................. 46 Contacts Author Contact Information ...................................................................................................... 47 Congressional Research Service Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Introduction A variety of efforts that seek to reduce greenhouse gas emissions (GHG)1 are currently under way or being developed on the international, national, and sub-national level (e.g., individual state actions or regional partnerships). One option (of many, see text box below—“Other Policy Options for Addressing GHG Emissions”) for controlling GHG emissions is to apply a tax or fee on GHG emissions or the inputs that create them. This type of approach is commonly called (and referred to in this report as) a carbon tax,2 whether it would apply to CO2 emissions alone or to multiple GHGs, including some that may have no molecular carbon.3 This report does not provide a comprehensive comparison and analysis of the multiple policy tools available to Congress that would address climate change. Instead, this report focuses on the policy considerations of using a carbon tax to control GHG emissions. Governments may impose taxes for a variety of purposes. The primary reason that governments impose taxes is to raise revenue to fund various objectives or services: e.g., national defense, public education, social security, etc. Generally, governments raise these revenue streams by placing a tax on activities that are recognized as desirable (“economic goods”) such as income, employment, and investment. While this tax placement ensures a relatively steady flow of revenue (often the primary objective of the tax), economists generally describe such taxes as distortionary, because the taxes discourage the “good” activity. For example, many economists have argued that payroll and income taxes discourage employment and investment.4 If these taxes were reduced, the incentives to increase labor and investment would be greater. Economists maintain that levying a charge on pollution (sometimes referred to as a Pigouvian tax)5 would be an efficient way to correct an inherent failure in a particular market. A basic economics principle is that market prices may not reflect the social cost of resource use (e.g., fossil fuel combustion) when economic activities result in pollution (e.g., CO2 emissions). If social costs are not included, the market price of the resources will not reflect their true costs. For example, in terms of climate change policy, the price of using fossil fuels, particularly coal, does not reflect the costs—that is, climate change-related damages—associated with CO2 emissions. 1 The major GHGs discussed include carbon dioxide (CO2), methane (CH4), nitrous oxide (N2O), sulfur hexafluoride (SF6), hydrofluorocarbons (HFC), and perfluorocarbons (PFC). Recent GHG reduction proposals have also included nitrogen trifluoride (NF3). 2 As discussed in this report, terminology is a key issue. Some proponents of the “carbon tax” approach describe the policy instrument as a user fee or user charge. There are multiple reasons that proponents may seek to change the nomenclature. Perhaps the primary concern is the political stigma associated with the word “tax.” Regardless, the term “carbon tax” is the one that is most commonly associated with the GHG control policy instrument discussed in this report. 3 Non-carbon GHGs could still be subject to the tax based on their contribution to global warming in relation to CO2. Global warming potential (GWP) is an index of how much a GHG may contribute to global warming over a period of time, typically 100 years. GWPs are used to compare gases to carbon dioxide, which has a GWP of 1. For example, methane’s GWP is 25, and is thus 25 times more potent a GHG than CO2. The GWPs listed in this report are from: Intergovernmental Panel on Climate Change, Climate Change 2007: The Physical Science Basis (2007), p. 212. 4 See e.g., Gilbert Metcalf, A Green Employment Tax Swap: Using a Carbon Tax to Finance Payroll Tax Relief (2007); Nathaniel Keohane and Sheila Olmstead, Markets and the Environment (2007), Island Press; Ian Parry “Fiscal Interactions and the Case for Carbon Taxes over Grandfathered Carbon Permits,” in Climate Change Policy (Dieter Helm, editor), Oxford University Press (2005). 5 Named after A. Cecil Pigou, author of a landmark economic work, Economics of Welfare (1920). Congressional Research Service 1 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress In economics parlance, the social cost not reflected in the market price is called an “externality.” A pollution tax would internalize the external costs by making the party who profits from the polluting activity include the external costs in the price of the good or service. Policymakers could place a pollution tax on GHG emissions or the inputs that create them. By attaching a price to GHG emissions, a carbon tax would stimulate GHG emission reduction. If the tax were placed on emissions, entities directly subject to the tax, such as power plants, would have an incentive to take actions—for example, energy efficiency improvements or equipment upgrades—to lower tax payments. If the tax were placed on emission inputs—for example, fossil fuels—the price of carbon-intensive energy sources, primarily coal, would increase relative to low-carbon fuels (Table A-1 of this report—located in the Appendix—includes estimates of price increases to fossil fuels and motor gasoline based on different carbon tax rates). Energy consumers—for example, power plants, industry, households, etc.—would be encouraged to (1) switch to less carbon-intensive fuels; (2) use less energy or use energy more efficiently; and (3) prefer products or services that are lower-priced by virtue of incorporating less emission tax. Each of these activities would reduce GHG emissions compared to a business-as-usual track. These expected behavioral changes mirror the activities that are forecast for a potential cap-andtrade program. Both a carbon tax and a cap-and-trade system would place a price on carbon. Both a carbon tax and cap-and-trade system are intended (and expected) to increase the price of coal, oil, and natural gas. Under either program, these price increases would ultimately be borne by energy consumers, both businesses and households. These price increases are integral to a market-based approach to GHG emission reduction, because they send more accurate information to purchasers about the full cost of their choices. This report begins with an overview of the fundamental choices involved between a cost (tax) and a quantity (cap) control instrument. This includes a discussion of policy tools that could be employed to bridge the gap between a carbon tax and a cap-and-trade program. Following this overview, the report analyzes the potential advantages and disadvantages of a carbon tax. In many cases, carbon tax attributes are compared with those of a cap-and-trade program. The next section discusses implementation issues for a carbon tax, including where to apply the tax, at what level to set the tax, and options for distributing tax revenues. The final section provides conclusions. Other Policy Options for Addressing GHG Emissions For policymakers considering actions to address climate change, a variety of policy instruments is available. Although current attention has largely focused on market-based mechanisms, primarily cap-and-trade systems, non-market policy tools may be the most practical option to address some emission sources. Moreover, Congress may consider complementary approaches to market mechanisms to improve their effectiveness. In particular, many experts maintain that the GHG emission targets specified in recent legislation would require development and deployment of improved (low-carbon) technologies. To further this effort, some argue that Congress should address technology stimulation directly or as a supplement to the primary climate change mitigation policy. In addition, Congress has already addressed some specific sectors (cars and light trucks, and government buildings) through performance standards. These standards may be further strengthened, independent of climate change legislation. Efforts are under way to address GHG emissions using the existing Clean Air Act (CAA) statute. The Environmental Protection Agency (EPA) has received at least eight petitions asking it to use its CAA authority to regulate GHG emissions from multiple categories of mobile sources. The agency faces lawsuits seeking GHG regulations (per CAA authority) from power plants, refineries, and other stationary sources. In 2007, the Supreme Court found (Massachusetts v. EPA) that GHG emissions are air pollutants under the CAA, and that EPA has the authority to promulgate GHG emission controls. For more information, see CRS Report R40145, Clean Air Issues in the 111th Congress, by (name redacted); and CRS Report RL34513, Climate Change: Current Issues and Policy Tools, by (name redacted). Congressional Research Service 2 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Cost or Quantity Control: An Overview If policymakers choose to establish a market-based mechanism to control GHG emissions, a fundamental decision would be whether to use a price instrument, such as a carbon tax, or a quantity instrument, such as an emissions cap.6 Economic Theory vs. Uncertainty In an economically efficient market with perfect information, either a price (carbon tax) or quantity control instrument (cap-and-trade system) could be designed to achieve the same outcome. 7 Figure 1 illustrates this basic economic principle. The intersection of marginal costs8 and marginal benefits9 would provide the point (of economic efficiency) at which to set the price or quantity limit. At this point in the figure, the abatement costs10 equal benefits received from abatement. Per economic theory, emission abatement above or below this point would not be economically efficient. 11 The dashed lines indicate the efficient price (tax) and quantity (cap) limits. This figure illustrates that (with perfect information) if either a tax or cap is selected, both the emission abatement level and cost of abatement would be identical. For instance, if a cap were chosen, covered sources (e.g., power plants) would abate emissions until they reach the cap of Q*, at which point the marginal cost of abatement would equal P*. If a tax were chosen, sources would abate emissions until the marginal cost of abatement reached the tax level (P*), at which point the emission abatement level equals Q*. In either case, total abatement is Q*, and the total cost of the program is the shaded area under the marginal abatement cost curve. 6 See also, CRS Report RL33799, Climate Change: Design Approaches for a Greenhouse Gas Reduction Program, by (name redacted); CRS Report RL34513, Climate Change: Current Issues and Policy Tools, by (name redacted). 7 See e.g., William Pizer, Prices vs. Quantities Revisited: The Case of Climate Change (1997), Resources for the Future Discussion Paper 98-02. 8 The marginal cost curve indicates the cost of an additional unit (e.g., one ton of GHG emissions) of emission abatement at different emission levels. A rising marginal cost curve (as depicted in Figure 1) signals that as each incremental unit of GHG emission abatement is made, the cost of abatement (per unit) increases. 9 The marginal benefit curve indicates the benefit of an additional unit (e.g., one ton of GHG emissions) of emission abatement at different emission levels. A declining benefit slope (as depicted in Figure 1) illustrates that with each incremental unit of abatement, the per unit benefit decreases. 10 Abatement may include emission reductions, emissions avoided, and sequestration activities. 11 In other words, for abatement above this level, it would cost more than the value of the benefits received. Society would be paying too much; money would be better spent for other purposes. Congressional Research Service 3 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Figure 1. Illustration of Price Versus Quantity Price Marginal Abatement Costs Efficient Price – P* Marginal Benefits Efficient Quantity – Q* GHG Emission Abatement Source: Prepared by CRS. However, the illustration in Figure 1 depicts economic theory, not economic reality: An economically efficient market with perfect information does not exist. The primary knowledge gaps are the shapes (i.e., slope) and levels (i.e., magnitude) of the marginal abatement cost and marginal benefit curves. These benefit and cost curves are influenced by multiple variables, including uncertain future emission baselines. Baseline estimates are related to projected population levels and economic growth,12 factors that are also unknown and difficult to predict.13 Another uncertain factor, which will have a substantial influence over mitigation costs, is the development and deployment of low-carbon technology. Without accurate information regarding marginal costs and benefits, policymakers cannot know the efficient quantity or price from which to establish an emissions cap or carbon tax. Thus, in one sense, preference for a price (carbon tax) or a quantity limit (emissions cap) depends on one’s preference for uncertainty—either uncertain emissions or uncertain program costs. A tax would set a ceiling on the marginal price of emissions (and thus the total cost of the program), but emissions would be uncertain. A cap would hold emissions to a set level, but the price of emission reductions (and thus program costs) would be uncertain.14 However, as discussed below, the choice is not as simple as it may seem. 12 See CRS Report RL33970, Greenhouse Gas Emission Drivers: Population, Economic Development and Growth, and Energy Use, by (name redacted) and (name redacted). 13 For a analysis of some recent assessments, see CRS Report RL34489, Climate Change: Costs and Benefits of S. 2191/S. 3036, by (name redacted) and (name redacted). 14 See also, CRS Report RL33799, Climate Change: Design Approaches for a Greenhouse Gas Reduction Program, by (name redacted). Congressional Research Service 4 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress A Stark Choice or a Policy Continuum? There are multiple design elements that can be included with a cap-and-trade program that blur the distinction between price and quantity control. Similarly, a carbon tax program could include flexible design mechanisms allowing policymakers to alter the tax rate, if they determine that emission reductions are not proceeding at a desirable pace. With these design elements available, Congress is presented with a policy continuum, rather than a stark policy dichotomy. A Flexible Emissions Cap Congress could enact a cap-and-trade program that includes mechanisms that provide flexibility in terms of controlling costs. Figure 2 illustrates the options available. As the design elements in Figure 2 (the “bridge pieces”) are added to a cap-and-trade program, the program more closely resembles a carbon tax. These elements are briefly described below: • Banking: Covered sources may save (bank) emission allowances15 and submit them in future years. Banking reduces the absolute cost of compliance by making annual emissions caps flexible over time. In addition, banking reduces price volatility to some degree. 16 • Borrowing: Covered sources may borrow (typically at a relatively high interest rate) emission allowances allotted to future years. The ability to borrow allowances from future years should help control price spikes. • Market oversight: Congress could create a regulatory mechanism (e.g., an oversight board) to observe the allowance market and implement cost-relief measures if necessary. Obviously, the authority granted to the delegated entity would determine its ability to affect program costs. • Safety valve: Generally triggered by prices in the allowance markets, safety valves may include (1) a set price alternative to making reductions or buying allowances at the market price, (2) a slowdown in tightening the emissions cap, and (3) lengthening of the time allowed for compliance. The first type would move the program closest to price control by setting a ceiling on allowance prices and limiting their volatility. • Auctioning allowances: Auctions may be used to allocate some or all of the emission allowances. Like a carbon tax, auctions would raise revenues to support various objectives. In addition, auctions would support price discovery. 17 • Reserve price: A reserve price is a price below which the seller refuses to part with the item for sale. A reserve price would all but guarantee a revenue stream in 15 Emission allowances are essentially the currency in a cap-and-trade program. An allowance is generally defined as a limited authorization by the government to emit 1 ton of pollutant. In the case of GHGs, an allowance generally refers to a metric ton of carbon dioxide equivalent (tCO2-e). 16 A 2008 Resources for the Future (RFF) report estimated the effects banking provisions had on price volatility. The authors concluded that banking eliminates about 20% of the cost differences between a price control (tax) system and cap-and-trade without banking provisions. Harrison Fell et al., Prices Versus Quantities Versus Bankable Quantities (2008), Resources for the Future Discussion Paper. 17 See CRS Report RL34502, Emission Allowance Allocation in a Cap-and-Trade Program: Options and Considerations, by (name redacted). Congressional Research Service 5 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress an emissions allowance auction, thus operating much like a minimum tax or price floor. Figure 2. Bridging the Gap between Price and Quantity Control Auction Allowances Auction Reserve Price (Cost Floor) Safety-Valve (Cost Ceiling) Banking Emissions Cap Borrowing Market Oversight Carbon Tax •Certain Emissions •Certain Price •Uncertain Price •Uncertain Emissions Source: Prepared by CRS. A Flexible Carbon Tax Policymakers could establish a carbon tax that includes mechanisms for addressing long-term emission uncertainty. For example, emission sources would be required to submit emission data periodically, so that policymakers could assess the performance of the carbon tax. If Congress determined that emission reduction was inadequate, the tax rate could be adjusted to achieve a desired emissions level. The decision authority could be retained by Congress or delegated to an existing federal agency—for example, Department of Treasury or Environmental Protection Agency—or another entity (e.g., a newly created independent board) in order to meet predetermined emission reduction objectives. 18 Considering the consequences of altering a carbon tax rate, Congress may be hesitant to delegate this authority. In addition, some may question whether the federal government is nimble enough to modify the carbon tax rate on a periodic basis, particularly considering the potential size of the tax (Table 1 below). Moreover, a flexible carbon tax approach may sacrifice long-term price certainty, depending on the authority of the entity to adjust the tax rate. To address this concern, Congress may consider providing tax rate increase parameters for the delegated entity. Limits of the Policy Continuum Regardless of whether policymakers employ these additional design elements, the gap between price control and quantity control can never be completely overcome: The closest resemblance would be a cap-and-trade program that employs both a price safety-valve and an emission allowance auction with a reserve price. Even if all of the above design options are part of a capand-trade program, the price of emission allowances would still fluctuate to some degree. Likewise, a more flexible tax system would still yield short-term emission fluctuations. 18 See Dieter Helm et al., “Credible Carbon Policy,” in Climate Change Policy (ed. Dieter Helm) (2005). Congressional Research Service 6 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress The policy continuum demonstrates that a comparison between a carbon tax and an emissions cap is not a straightforward exercise. Design details are critical for an appropriate comparison. Depending upon program design, cap-and-trade programs can vary greatly in their ability to control price fluctuations and total program costs. Some of the advantages or disadvantages (depending on one’s perspective) of either system can be addressed through program design. Indeed, recent federal cap-and-trade proposals have included one or more of the design elements listed above. Potential Advantages of a Carbon Tax This report examines four potential advantages of a carbon tax approach to controlling GHG emissions. First, the report compares a carbon tax to cap-and-trade approach from an economics theory perspective, analyzing the assertion (of many economists) that a carbon tax would be more economically efficient than a cap-and-trade system. Second, it discusses the merits of price stability that a carbon tax would provide. The third discussion highlights the advantage of generating carbon tax revenues and examines various revenue applications and the trade-offs that would ensue. Finally, the report identifies potential implementation advantages, primarily practical considerations, of using a carbon tax approach. Economic Efficiency A primary argument supporting the use of a carbon tax is that in the presence of cost uncertainty—that is, the marginal abatement cost curve (Figure 1)—a tax would potentially be more economically efficient than an emissions cap. However, this argument is grounded in assumptions that are the subject of considerable debate and controversy. This section outlines the basis for the argument and provides an analysis of the underlying assumptions. Basis for the Argument As Figure 1 demonstrates, if the marginal cost and benefit curves are known to policymakers, either a tax or a cap instrument could be used to achieve the same result. As discussed above, both the costs and benefits of GHG emission reduction are uncertain. However, some economists argue that the relative slopes of the cost and benefit curves are better understood. This information—the slope of the marginal benefit curve in comparison to the slope of the marginal cost curve—is the foundation for an argument that supports a tax (price control) over an emissions cap (quantity control). This argument was presented in Weitzman’s landmark 1974 study “Prices vs. Quantities.”19 Weitzman’s primary conclusion (explained in some detail below) was that price instruments (e.g., carbon tax) would be preferred when the marginal benefit curve is relatively flat; a quantity limit (e.g., emissions cap) would be preferred when the marginal cost curve is relatively flat.20 19 M. L. Weitzman, “Price vs. Quantities” (1974), Review of Economic Studies, 41(4): 477-491. This conclusion is often referred to as the “Weitzman Rule.” Nathaniel Keohane and Sheila Olmstead, Markets and the Environment (2007), Island Press 20 Congressional Research Service 7 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Intuitively, a relatively flat marginal benefits curve suggests that each additional unit of abatement provides approximately the same benefit. In contrast, a relatively steep marginal cost curve suggests that each additional unit (i.e., ton of CO2) of emission abatement entails a greater cost than the previous unit.21 Some economists argue that the slope of the marginal benefits curve is flat relative to the marginal cost curve. 22 This notion is the subject of intense debate, but it is grounded in the scientific processes (discussed below) through which GHG emissions generate climate change impacts. Based on this (assumed) relationship between the expected slopes of costs and benefits, a carbon tax would be more efficient than an emissions cap. Figure 3 and Figure 4 below illustrate this concept. Figure 3 provides a scenario in which the marginal benefits curve is flatter than the marginal costs curve. Without precise information regarding marginal costs and benefits, policymakers would base a tax or cap on estimates of the marginal cost and benefit curves. These are represented in Figure 3 as the expected marginal costs and benefits of GHG emission abatement. 23 The intersection of the expected marginal abatement cost and benefit curves would be the estimated economically efficient point of regulation. From this efficient point, policymakers could apply either an emissions cap or carbon tax—at Q* or P*, respectively. Either option would produce the same outcome, if the expected marginal cost curve matches the actual marginal cost curve.24 However, the actual cost curve is unlikely to match the expected cost curve. Therefore, some level of economic inefficiency will exist with either a tax or an emissions cap. However, the policies are expected to yield different levels of inefficiencies. For example, consider a scenario (Figure 3) in which actual marginal abatement costs are higher than expected. If policymakers were to apply a carbon tax (P* in the figure) directly to emissions, emission sources would abate emissions until the marginal cost of abatement equals the carbon tax. This occurs at Point A, at which time covered sources would pay the tax for any additional emissions. On the other hand, if policymakers employed an emissions cap (Q*), covered sources would abate emissions until Point B, when the cap is met. 21 In other words, not all emission reductions (or sequestration activities) are equal in terms of cost: there may be “lowhanging fruit” opportunities as well as actions that would require more substantial capital investments. 22 Richard Newell and William Pizer, “Regulating Stock Externalities under Uncertainty,” Journal of Environmental Economics and Management 45: 416-432 (2003); Warwick J. McKibbin and Peter J. Wilcoxen, “The Role of Economics in Climate Change Policy,” (2002) Journal of Economic Perspectives 16(2): 107-29. 23 The “curves” in this report’s figures are depicted as straight lines for illustrative purposes. 24 Although marginal benefit information is also uncertain, the relative-slope principle generally depends only on the marginal cost curve. The reasoning behind this approach and circumstances that may undermine this approach are discussed below. Congressional Research Service 8 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Figure 3. Illustrative Scenario with a Relatively Flat Marginal Benefits Curve Higher Marginal Abatement Costs Expected Marginal Abatement Costs Cap Efficiency Loss Price Tax Efficiency Loss Lower Marginal Abatement Costs B Tax Efficiency Loss P* A Expected Marginal Benefits Cap Efficiency Loss Q* GHG Emission Abatement Source: Prepared by CRS. With a tax instrument in Figure 3, under higher-than-expected costs, covered sources would abate less than the optimal level—the intersection of benefits and actual costs; with an emissions cap, sources would abate more than the optimal level. However, Figure 3 illustrates that the difference in efficiency losses—the shaded triangles—from the two policies is substantial. Figure 3 also illustrates the scenario of lower-than-expected marginal abatement costs. With lower-than-expected marginal abatement costs, an emissions cap would lead to abatement that is less than optimal, whereas a tax would stimulate abatement beyond the optimal level. As with higher-than-expected costs, a tax instrument would yield less inefficiency than an emissions cap. The different efficiency losses are depicted by the striped triangles in Figure 3. Figure 4 illustrates the alternate scenario, in which the marginal benefits curve is relatively steep. In this situation, an emissions cap is the preferred instrument. Whether the marginal abatement costs are higher or lower than expected, the efficiency losses from a carbon tax are greater than those from an emissions cap. Congressional Research Service 9 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Figure 4. Illustrative Scenario with a Relatively Steep Marginal Benefits Curve Expected Marginal Benefits Higher Marginal Abatement Costs Price Tax Efficiency Loss Cap Efficiency Loss Expected Marginal Abatement Costs Lower Marginal Abatement Costs P* Cap Efficiency Loss Tax Efficiency Loss Q* GHG Emission Abatement Source: Prepared by CRS. Underpinnings of the Argument The argument that a carbon tax would minimize economic inefficiency in the face of uncertain costs contains several assumptions. A primary assumption concerns the relative steepness/flatness of the cost and benefit curves. In addition, there are other underpinnings of the argument that may be questioned. These issues are discussed below. Relative Flatness of Benefits Curve? At the crux of the argument that a tax would maximize efficiency is the assumption that the slope of the marginal benefits curve is flat compared to the marginal costs curve (Figure 3). Many would assert this is the case, based on the process by which GHG emissions generate damages. In the context of climate change, the damages related to GHG emissions are due to their concentration (referred to as “stock”) in the atmosphere, as opposed to their annual flow.25 By comparison, more traditional air pollutants (e.g., sulfur dioxide, nitrogen oxide) impose damages through their annual emission flows. With a “stock pollutant” like CO2, the processes involved in adding or removing the gas from the atmosphere are measured, not in years, but in decades (or centuries for some gases). The current stock of CO2 in the atmosphere is approximately 3,000 gigatons;26 in 2004, the entire world 25 26 Other GHG emissions may yield damages unrelated to climate change impacts. 1 gigaton = 1 billion tons. Congressional Research Service 10 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress emitted roughly 30 gigatons of CO2. Because emission quantities would have only a minor effect on the accumulation of CO2 in the atmosphere, the benefits from controlling the stock (or the damages from business-as-usual activities) are thought to occur slowly and steadily.27 Figure A-1 in the Appendix illustrates this concept. However, the assertion that the benefits curve is flat assumes that there are no climatic “tipping points.” Some scientists maintain that at a certain stock level—concentration of GHG emissions in the atmosphere—particularly drastic events might occur.28 By definition, a tipping point implies that when it is reached, the course of climate change is abrupt and the marginal benefits curve becomes very steep. The climate changes may even be irreversible and catastrophic, at least for some populations and environments. Examples often cited include collapse of Greenland’s ice sheet; and dieback of the Amazon rainforest.29 If such tipping points exist, they would dramatically alter the shape of the benefits curve, because abating emissions to avoid large adverse consequences would yield large benefits. As argued by the Stern Review: “over the long term, as the stock of GHGs grows, marginal damages are likely to rise and – as the stock reaches critical levels – marginal damages may rise sharply. In other words, the damage function is likely to be strongly convex.”30 However, the Stern Review also notes that: “to the extent that damages may relate to the rate of climate change, the relationship is more complex, but it remains true that the damage curve is likely to respond most to cumulative emissions over several years or even decades.”31 Depending on whether the tipping points are “several years” or “decades” away, an emissions cap may be preferred. However, scientists and policymakers have very poor information regarding the GHG concentration levels (or rates of concentration changes) at which tipping points may occur. A 2003 economic study found that emissions caps would be “optimal only when there is a welldefined threshold with clearly catastrophic consequences on the other side.”32 Relative Steepness of the Marginal Cost Curve? The marginal cost curve is perhaps better understood (and less controversial) than benefits estimates. Analysis of recent cap-and-trade proposals introduced in Congress suggest that with the current portfolio of technologies the marginal cost curve does not appear flat.33 Although the first units of emission abatement opportunities may be relatively inexpensive, the costs of abatement are likely to increase with each additional unit. However, if a carbon market spurs 27 William Pizer, Prices vs. Quantities Revisited: The Case of Climate Change (1997), Resources for the Future Discussion Paper 98-02. 28 See e.g., James Hansen, Testimony before the House Select Committee on Energy Independence and Global Warming (June 20, 2008). 29 See Timothy M. Lenton, et al., “Tipping Elements in the Earth’s Climate System,” Proceedings of the National Academy of Sciences 105:6 (February 2008). 30 Nicholas Stern, The Economics of Climate Change: The Stern Review (2006), p. 314 (emphasis in original). 31 Nicholas Stern, The Economics of Climate Change: The Stern Review (2006), p. 314, footnote 12. 32 William Pizer, Climate Change Catastrophes (2003), Resources for the Future Discussion Paper. See also, CBO, Policy Options for Reducing CO2 Emissions (2008). 33 See CRS Report RL34489, Climate Change: Costs and Benefits of S. 2191/S. 3036, by (name redacted) and (name red acted). Congressional Research Service 11 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress technology development, the marginal cost curve may flatten. This could reverse the relationship between costs and benefits, and make a cap the more economically efficient policy. 34 Short-Term Time Frame In general, Weitzman’s economic principle regarding the relative slopes of the marginal cost and benefit curves applies to a static or fixed time-frame. The principle is best suited to compare the costs and benefits of controlling a “flow pollutant” (such as SO2), which would yield short-term damages. However, as noted above, CO2 is a stock pollutant: its potential damages are related to total accumulation in the atmosphere; one year of emissions would likely have negligible effects. Thus, the static analysis is over-simplistic, because (by definition) it assumes that activities in different time periods are unrelated.35 However, equipment upgrades or efficiency improvements made in one time period would affect the emissions generated in future years as well. Other timerelated factors need to be considered, including the discounting of future benefits and the atmospheric decay of GHGs. Several economic studies have addressed this issue of time, expanding Weitzman’s basic principle within a dynamic framework (i.e., a multi-year time-period). In general, these studies concluded that Weitzman’s basic principle carries over to a dynamic setting.36 Only Cost Uncertainty is Considered Although marginal benefit information is also uncertain, the relative-slope principle generally depends only on the marginal cost curve. The rationale behind this is that when covered sources make decisions regarding emission abatement—for example, reduce onsite emissions or pay the carbon tax—benefits of abatement are not part of their economic calculation. Economists point out that if covered sources did consider benefits, there would be no externality in the first place. However, in some circumstances an emissions cap may produce a more economically efficient outcome. As mentioned (in a footnote) by Weitzman (1974) and re-examined by Stavins in 1996,37 if the uncertainty of costs and benefits are positively correlated—for example, both the marginal benefit and cost curves are actually higher than expected—quantity control provides a more economically efficient result. This scenario is illustrated in Figure 5. The advantage for quantity control over price control under these circumstances depends on the degree of correlation and the magnitude of uncertainty with respect to the cost and benefits curves – the greater the difference between anticipated and actual cost and benefits values, the greater the advantage for quantity control measures. Stavins points out that this reversal in preference for price over quantity control could occur with plausible values for the two variables; however, he 34 CBO, Policy Options for Reducing CO2 Emissions (2008). 35 William Pizer, Prices vs. Quantities Revisited: The Case of Climate Change (1997), Resources for the Future Discussion Paper 98-02. 36 Lawrence Goulder and Ian Parry, Instrument Choice in Environmental Policy (2008), Resources for the Future Discussion Paper, implicitly referencing: Michael Hoel and Larry Karp, “Taxes Versus Quotas for a Stock Pollutant,” Resource and Energy Economics (2002) 24:367-84; Newell and Pizer (2003); and Larry Karp and Jiangfeng Zhang, “Regulation of Stock Externalities with Correlated Abatement Costs,” Environmental and Resource Economics (2005) 32:273-299. 37 Robert Stavins, “Correlated Uncertainty and Policy Instrument Choice,” Journal of Environmental Economics and Management 30:218-232 (1996). Congressional Research Service 12 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress argues that “the advantage of price over quantities remains unless the true benefits of carbon mitigation are many orders of magnitude greater than our best estimate.”38 Figure 5. Illustrative Scenario with Marginal Costs and Marginal Benefits That Are Higher Than Expected Higher Marginal Abatement Costs Expected Marginal Abatement Costs Cap Efficiency Loss Tax Efficiency Loss B Price Higher Marginal Benefits P* A Expected Marginal Benefits Q* GHG Emission Abatement Source: Prepared by CRS. Modeled Efficiency Gains Several economic studies have examined the efficiency gains of using a carbon tax versus an emissions cap. While the analysis is more complicated and involves more than simply the relative slopes of marginal abatement and damage curves, the analyses consistently find that taxes dominate cap-and-trade systems for a broad range of parameter values consistent with scientific understanding of the global warming problem [emphasis added].39 For example, Pizer found that taxes are more efficient than an emissions cap by a factor of five to one. 40 Although this result is from 1997, similar estimates were made in more recent studies that 38 Newell and Pizer (2003). Gilbert Metcalf et al., Analysis of U.S. Greenhouse Gas Tax Proposals (2008), MIT Joint Program on the Science and Policy of Climate Change, Report No. 160. 40 Pizer (1997). 39 Congressional Research Service 13 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress applied different parameters.41 The magnitude of these efficiency gains were estimated to be in hundreds of millions to several billions of dollars per year.42 However, a 2008 Massachusetts Institute of Technology (MIT) report points out that the modeled efficiency gains may be smaller in practice. 43 As discussed above, policymakers can design capand-trade programs with flexible components, such as banking and borrowing, that could alleviate some of the inefficiency loss. But a 2008 Resources for the Future (RFF) study indicates that banking may not help “as much as proponents might suggest.” The study found that banking provisions in a cap-and-trade program would improve efficiency by 20% over a cap-and-trade without banking. 44 Economic Efficiency Versus Precaution Despite the economic efficiency argument in favor of a carbon tax, some argue that the downside risks of tipping points—that is, irreversible climate changes—demand that economic efficiency take a backseat to precise control of GHG reductions, achievable only with a quantity-based emissions cap. This viewpoint could be described as an application of the precautionary principle: i.e., keeping one’s options open in the face of uncertainty.45 This approach argues that, while GHG emissions are held at an exact level, scientists and economists can continue to gather information regarding abatement costs and climate change benefits. As (presumably) better information is garnered, climate change policy can be amended as necessary. Price Stability Depending on the chosen GHG control instrument—carbon tax or cap-and-trade—either emissions or prices would fluctuate over short-term time periods (months to years). Proponents of a carbon tax often highlight the volatile price swings that have been observed in the U.S. sulfur dioxide (SO2) emissions cap-and-trade program.46 Between 2001 and 2006, the price of a SO2 allowance has varied by a factor of 12.47 The inclusion and substantial use of banking has not been sufficient to overcome other factors creating price volatility, such as uncertainty regarding proposed regulatory changes in the program (e.g., The Clean Air Interstate Rule). Although the price volatility in the SO2 market did not translate into electricity price spikes for consumers, CO2 emission allowances could play a more integral role in the U.S. economy: Fossil fuels accounted for 85% of energy consumption in 2006.48 Nordhaus states that strict quantity 41 See, Hoel and Karp (2002); Newell and Pizer (2003); Karp and Zhang (2005). Ibid. 43 Metcalf et al. (2008). 42 44 Harrison Fell et al., Prices Versus Quantities Versus Bankable Quantities (2008), Resource for the Future Discussion Paper. 45 See Alan Ingham and Alistair Ulph, “Uncertainty and Climate Change Policy,”in Climate Change Policy (Dieter Helm, editor), Oxford University Press (2005). 46 Indeed, the Carbon Tax Center lists price volatility control as one of the fundamental reasons taxes are superior to emissions caps. Carbon Tax Center, Tax vs. Cap-and-Trade, at [www.carbontax.org]. 47 William Nordhaus, “To Tax or Not to Tax: Alternative Approaches to Slowing Global Warming” (2007), Review of Environmental Economics and Policy, Vol. 1 (1): 26-44. 48 Energy Information Administration, Annual Energy Outlook 2008 (2008). Congressional Research Service 14 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress limits might have “major disruptive effects on energy markets and investment planning, as well as on the distribution of income across countries, inflation rates, energy prices, and import and export values.”49 Volatile emission allowance prices may discourage investment, because firms would have less reliable cost information upon which to plan for capital improvements.50 A GHG cap-and-trade program may or may not have similar price swings to those experienced with the SO2 allowance market. The largest GHG cap-and-trade program currently operating is the European Union’s Emission Trading System (EU ETS). During the four years of its existence, both Phase 1(2005-2007) and Phase 2 (2008-2012) EU ETS allowance prices have experienced significant volatility. Phase 1 allowance prices from the EU ETS are arguably not an appropriate comparison, because the fluctuations were largely related to program design issues. 51 Building on the experience gained during Phase 1, Phase 2 of the ETS began in 2008, and these emission allowance prices may provide a better example of the potential effects of market forces and the fluctuations that could occur under a U.S. cap-and-trade program.52 Figure 6 illustrates the market prices for Phase 2 allowances between 2005 and January 2009. Figure 6. “Phase 2” Emission Allowance Prices in the European Union’s Emission Trading System 35 Euro/tCO2 30 25 20 15 10 5 Jan-09 Oct-08 Aug-08 May-08 Feb-08 Nov-07 Aug-07 Jun-07 Mar-07 Dec-06 Sep-06 Jun-06 Apr-06 Jan-06 Oct-05 Jul-05 Apr-05 0 Source: Prepared by CRS with data from European Climate Exchange, at http://www.ecxeurope.com. Tax Revenue Applications Carbon tax proponents often cite the generation of carbon tax revenues as a primary advantage of a carbon tax. However, policymakers could design a cap-and-trade program that could deliver 49 Nordhaus (2007). Ian W.H. Parry and William Pizer, Emissions Trading Versus CO2 Taxes Versus Standards (2007), Resources for the Future Issue Brief. 51 See CRS Report RL34150, Climate Change and the EU Emissions Trading Scheme (ETS): Kyoto and Beyond, by (name redacted). 52 Although Phase 2 began in 2008, the market for Phase 2 allowances was created in 2005. Unlike Phase 1 allowances, Phase 2 allowances may be banked for future use in a planned Phase 3 anticipated to begin in 2013. 50 Congressional Research Service 15 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress similar, if not identical, results. For example, if policymakers establish a cap-and-trade program and distribute 100% of the allowances through an auction process, the auction proceeds would function as tax revenues. 53 Thus the strength of this advantage depends on the design of the capand-trade program to which the tax is compared. From a public finance perspective, neither carbon tax nor auction revenues would be a reliable source of funding, because an effective carbon price (established through a tax or a cap) is expected to alter consumer behavior (e.g., demand for carbon-intensive goods). However, a carbon tax revenue stream would have a more predictable source than an allowance auction revenue stream, because the tax would be a known value, while the emission allowance price would fluctuate (as described above). Different options for applying the tax revenue are discussed in the section “Implementation of a Carbon Tax.” Potential Implementation Advantages Depending on the design, a carbon tax may offer several implementation advantages over a capand-trade program. These are discussed below. Transparency Arguably, the concept of a carbon tax is easier to explain and understand than a cap-and-trade program. A cap-and-trade system may have more moving parts, which may be challenging to follow, particularly the emission allowance allocation scheme. Moreover, as policymakers include more flexible design elements—primarily to improve efficiency and control price volatility—a cap-and-trade program would increase in complexity and potentially become less transparent. Some contend that the relative transparency of a carbon tax would help garner support for its enactment. 54 This argument may gain strength in the context of the recent financial climate, as many have grown more skeptical of seemingly complex financial structures.55 Although the concept of a carbon tax is arguably a simpler approach, many argue that the U.S. tax code is complex. Congress could establish a carbon tax framework that rivals the complexity of a cap-and-trade program. For instance, a carbon tax that only applies to CO2 emissions from fossil fuel combustion may be more transparent than a carbon tax that address non-CO2 GHG emissions. Moreover, policymakers could provide subsidies or exemptions to the fossil fuel industry that would run counter to a carbon tax. In addition, policymakers could allow for tax credits for carbon sequestration projects, similar to carbon offsets in a cap-and-trade regime. As with carbon offsets in a cap-and-trade program, this would require a further level of administrative responsibilities, and potentially weaken the program if the sequestration projects lack credibility. 56 53 See CRS Report RL34502, Emission Allowance Allocation in a Cap-and-Trade Program: Options and Considerations, by (name redacted). 54 Carbon Tax Center, “Tax vs. Cap-and-Trade,” at http://www.carbontax.org/issues/carbon-taxes-vs-cap-and-trade/. 55 See CRS Report RL34488, Regulating a Carbon Market: Issues Raised By the European Carbon and U.S. Sulfur Dioxide Allowance Markets, by (name redacted) and (name redacted). 56 See CRS Report RL34436, The Role of Offsets in a Greenhouse Gas Emissions Cap-and-Trade Program: Potential (continued...) Congressional Research Service 16 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Ironically, transparency, particularly in regards to costs, could be a political liability for a carbon tax. Although both a carbon tax and a cap-and-trade program would impose higher energy costs, the costs from a cap-and-trade program would be more difficult to estimate, because the market would determine the price of emission allowances (and thus the overall costs of the program). In other words, cap-and-trade proponents may have an advantage in promoting their approach, because they can cite studies that estimate relatively low program costs.57 Administrative Issues A well-developed administrative structure for collecting taxes already exists in the United States. Moreover, fuel sales are well-documented and are currently taxed (for various reasons) to some degree. 58 Thus, a carbon tax on fossil fuels—if applied upstream in the economy (discussed below)—could utilize this existing framework. However, a carbon tax that included other GHG emissions or their inputs would not have the advantage of this existing administrative structure. Point of taxation issues are discussed later in this report. Some argue that an emissions trading system would require the establishment of new institutions to administer the program. [T]he upfront costs of creating institutions to administer trading are significant and likely to produce entrenched bureaucracies that clamor for ever-tighter controls on carbon emissions.59 Others maintain that these concerns may be overstated. “Experience with existing trading programs, such as the U.S. SO2 trading program, has shown that these institutions can arise quickly and for the most part inexpensively.”60 However, comparing the SO2 trading program with a potential GHG emission trading program highlights appreciable differences. First, the scale of the SO2 trading program is considerably smaller than the proposed cap-and-trade initiatives. Second, the SO2 emission sources are more concentrated than GHG-emitting sources, which cut across multiple economic sectors. As with other comparisons, the relative advantage of a carbon tax would depend on the designs of the instrument alternatives under scrutiny. For example, a cap-and-trade system with a more downstream structure may present more of a challenge, because more sources would be subject to the cap. In addition, if a cap-and-trade program were to provide emission allowances to covered sources at no cost, policymakers would need to devise a system for determining allotment. Emission allowance auctions would alleviate this responsibility, but would require an additional administrative responsibility, and some have raised concerns that auctions may impose a risk of market manipulation. (...continued) Benefits and Concerns, by (name redacted). 57 See CRS Report RL34489, Climate Change: Costs and Benefits of S. 2191/S. 3036, by (name redacted) and (name red acted). 58 Metcalf (2008). 59 AEI, 2007. 60 Parry and Pizer, 2007. Congressional Research Service 17 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Another potential administrative advantage of a carbon tax approach is that it carries less risk of being hindered by litigation. Compared to a carbon tax, cap-and-trade legislation may delegate more authority to EPA or another government agency to implement certain components—for example, specific allocation decisions, offset protocols, etc.—of the program. 61 Such decisions may be vulnerable to legal challenges from stakeholders. Policy Modification Some have argued that one of the advantages of a carbon tax is the relative ease—compared to a cap-and-trade program—in which the program’s stringency could be modified. 62 In contrast, they assert that policymakers would face difficulties if they sought to adjust an emissions cap after the program’s initiation.63 The rationale for this assertion is that covered sources that made or purchased emission allowances beyond those needed in a given year would lose some of the value of these allowances if Congress raised (i.e., loosened) the cap at a later time. Similarly, a covered source may make capital investments based on the assumption of a stringent cap. If policymakers subsequently loosened the cap, these covered sources would take longer to recoup their investments.64 However, this concern could also apply to a carbon tax. For example, energy producers and consumers may make investments based on an expected carbon tax. If the tax is subsequently altered, the value of such investments may change. Potential Disadvantages A GHG emission reduction program that employs a carbon tax may face several disadvantages, compared to a cap-and-trade program. As with the potential advantages, the size and/or relevancy of the disadvantages are dependent on the design of the programs being compared. The potential disadvantages are discussed below. Uncertain Emissions The primary disadvantage of using a carbon tax to control GHG emissions is that the level of emissions would be uncertain. 65 For some this concern may present a non-starter, precluding a carbon tax as an option to control GHG emissions. As discussed above, some argue that the potential for irreversible climate change impacts necessitates the emissions certainty that is only available with a quantity-based instrument (e.g., cap-and-trade). 61 As with other comparisons, this would depend upon the design of the cap-and-trade program. See Kenneth Green, et al., Climate Change: Caps vs. Taxes (2007), American Enterprise Institute. 63 Ibid. 64 This potential problem could be alleviated by allowing covered sources to bank emission allowances. 65 A cap-and-trade program, depending upon its design, may carry some risk of uncertain emissions as well. For example, if offsets are allowed as a compliance option, there is some concern that they would not represent real emission reductions. Emissions leakage is another threat to the certainty of the emissions cap. See CRS Report RL34436, The Role of Offsets in a Greenhouse Gas Emissions Cap-and-Trade Program: Potential Benefits and Concerns, by (name redacted); and CRS Report R40100, “Carbon Leakage” and Trade: Issues and Approaches, by (name redacted) and (name redacted). 62 Congressional Research Service 18 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress However, uncertain emissions do not necessarily equate with no emission control. Multiple models have estimated the carbon tax rate that would be required to achieve certain emission targets, targets that are comparable with the emission reduction goals of recent cap-and-trade proposals.66 Regardless, these models can only provide estimates based on the best information available about the marginal costs of abatement (Figure 1). Although uncertain emissions are inherent with a carbon tax approach to emission control, there are policy options available to Congress to enhance the emission control certainty of a carbon tax. In theory, policymakers could devise a carbon tax program that yields only short-term emission fluctuations, as it progresses towards its long-term emission reduction objective. To achieve this goal, Congress would need to enact a mandatory GHG emission reporting regime to act in parallel with a carbon tax. Data from such a regime could be used to track the impact and performance of a carbon tax. If policymakers determine emission reduction is not occurring at a desired pace, the tax rate could be amended. As discussed above, carbon tax proponents would argue that short-term emission variations would not undermine efforts to control climate change. Indeed, they would assert that short-term emission fluctuations are greatly preferred to the price volatility that a cap-and-trade program would potentially impose. This argument is supported by the notion that CO2 is a stock pollutant, which generates damages through its overall concentration in the atmosphere, not its annual flow. Critics of such an approach may question whether modifying the carbon tax rate on a periodic basis would be a politically palatable option. For example, it may be difficult for policymakers to increase the tax rate, especially during downturns in the economy. To address these concerns, some have suggested that Congress authorize an independent board or agency with the ability to modify the tax rate, in order to meet pre-determined emission reduction objectives.67 Although this approach would improve emission certainty, long-term price certainty would be sacrificed to some degree, depending on the authority of the delegated entity to adjust the tax rate. Political Feasibility A common argument against a carbon tax is that it would present more political challenges to enact than an emissions cap, and thus policymakers should focus on the latter instrument. The Council on Foreign Relations recommended a cap-and-trade system over a carbon tax, stating: The Task Force finds that, assessed from a domestic economic perspective, the relative advantages of cap-and-trade and carbon tax approaches are easily overstated. Either, designed with a mix of efficiency and fairness in mind, would be an acceptable approach to reducing emissions. The Task Force finds, however, that the political momentum behind capand-trade makes its near term adoption much more likely.... [emphasis added]68 Although a carbon tax would likely face more political obstacles than a cap-and-trade program, some of these obstacles may be based on misunderstandings regarding the differences between the two approaches or on assumptions that the tax would be set too low to be effective. Carbon 66 See e.g., Metcalf (2008) and Shapiro (2008). See Dieter Helm et al., “Credible Carbon Policy,” in Climate Change Policy (ed. Dieter Helm) (2005). 68 Council on Foreign Relations, Confronting Climate Change: A Strategy for U.S. Foreign Policy (2008), Independent Task Force Report No. 161. 67 Congressional Research Service 19 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress tax proponents may be able to address these issues to some degree, but the recent political momentum for a cap-and-trade program may trump these efforts. This section examines some of the political concerns associated with a carbon tax. What’s in a Name? Enacting a “carbon tax” to control GHG emissions would pose a political challenge for policymakers simply due to the word “tax.” In contrast, the term “cap-and-trade” does not carry the same stigma. To some degree, the terminology advantage of a cap-and-trade program is arguably unfair, because it would act very much like a carbon tax. By placing a price on carbon, both a cap-and-trade and a carbon tax would likely impose costs on the economy. Indeed, some Members have argued that a cap-and-trade approach should be labeled as a “cap and tax,” because it would act effectively as a tax.69 Proponents of enacting a carbon tax understand the political ramifications of their instrument’s name. To overcome this hurdle, some have described carbon tax programs as “user fees” or “tariffs”or if they are highlighting revenue distribution opportunities—carbon tax “swaps”70 or “shifts.” There may be procedural consequences of a name change—see the text box “The Role of Committee Jurisdiction,” below. Proponents of a carbon tax could potentially make inroads in the debate by framing the debate within the larger context of tax policy, particularly tax reform. The Role of Committee Jurisdiction The instrument’s name and function would play an important role in terms of committee jurisdiction. A carbon tax proposal would likely be referred to the House Committee on Ways and Means and/or the Senate Committee on Finance; a cap-and-trade proposal would likely be referred to a committee that covers environmental policy (e.g., the Senate Committee on Environment and Public Works, House Committee on Energy and Commerce). In general, environmental policy committees may have different perspectives, expertise, or priorities than tax committees. These differences could influence the design of a GHG control instrument that originates in a particular committee, as well as subsequent program oversight and adjustments: e.g., the use of revenue from a carbon tax or a cap-and-trade system. The committee jurisdiction factor may play some role in the conceptualization of, and the debate over GHG emission control. Some may contend that if the policy were called a “user fee” or “user charge” then the primary committee of jurisdiction would not necessarily be a tax policy committee. Standing committees of the House, other than the Appropriations and Budget Committees, may report legislation creating or modifying user fees. However, this procedural point would likely be the subject of intense debate: A ruling on whether a policy instrument is a user fee or a tax measure may depend on the nature of the charge rather than its label. For more information on this issue see U.S. Office of Management and Budget, Analytical Perspectives, Budget of the United States Government, Fiscal Year 2009 (2008); and Congressional Budget Office, The Growth of Federal User Charges (1995). Some Members have argued that a cap-and-trade proposal would ultimately act as a tax on GHG emissions. They question whether the potential of a cap-and-trade program to raise revenue would have implications for committee jurisdiction. See Letter from Representatives Sensenbrenner and Cantor to Representatives Conyers and Nadler (February 11, 2009), at http://republicans.globalwarming.house.gov/index.shtml. 69 See Letter from Representatives Sensenbrenner and Cantor to Representatives Conyers and Nadler (February 11, 2009), at http://www.eenews.net/features/documents/2009/02/11/document_pm_03.pdf. This letter also argues that if a cap-and-trade bill were to raise revenue, the bill should originate in the House of Representatives. 70 See e.g., Gilbert Metcalf, A Green Employment Tax Swap: Using a Carbon Tax to Finance Payroll Tax Relief (2007). Congressional Research Service 20 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Support from Industry? Certainly the term “industry” covers a wide and diverse array of economic actors. Their motivations for support of, or opposition to, a carbon tax may vary as well. Support may depend on the anticipated design of the alternative—that is, a cap-and-trade system. Cap-and-trade proposals have evolved in recent years. Industry stakeholders may have supported a cap-andtrade approach because, if enacted, they expected to receive some or perhaps all of the allowances at no cost.71 This may not be an unrealistic expectation, considering the emission allocation strategies in the sulfur dioxide (SO2) emissions cap-and-trade program and the European Union’s Emissions Trading Scheme (EU ETS),72 both of which provide almost all of their allowances to covered sources at no cost. The no-cost allowance allocation to EU emissions sources generated windfall profits for some industries. 73 However, the more recent cap-and-trade proposals from the 110th Congress would have provided considerably fewer allowances to covered sources at no cost.74 Industry stakeholders have voiced concern over the apparent evolution of allocation strategy.75 If industry expects to receive fewer allowances at no cost than previously thought, their support of a cap-and-trade system may erode, which could increase interest in a carbon tax. Some argue that if industry is acting in its own self-interest, taxes would be the least-preferred option, even behind auctioned allowances in a cap-and-trade system. 76 The explanation for this preference is that auctions may allow room for industry to manipulate the price, driving down the overall cost of compliance (as opposed to a tax, which would be fixed). But some industries would prefer the cost to be fixed and known in advance (as a carbon tax would do), so they can better assess investment options (e.g., equipment upgrades, efficiency improvements, etc.).77 In addition, some industry stakeholders may lend their support to a tax, if the alternative were to resemble recent cap-and-trade proposals (from the 110th Congress), which would have applied directly to upstream entities (i.e., energy producers). This approach would resemble a tax— 71 For example, the U.S. Climate Action Partnership (USCAP), a group of large companies (and environmental organizations) recommends that a “significant portion of allowances should be initially distributed free to capped entities and to economic sectors particularly disadvantaged by the secondary price effects of a cap.... ” U.S. CAP, A Call to Action, at http://www.us-cap.org/. 72 See CRS Report RL34150, Climate Change and the EU Emissions Trading Scheme (ETS): Kyoto and Beyond, by (name redacted). 73 See IPA Energy Consulting, Implications of the EU Emissions Trading Scheme for the UK Power Generation Sector (2005), Prepared for the United Kingdom Department of Trade and Industry; Jos Sijm et al., “CO2 Cost Pass-Through and Windfall Profits in the Power Sector,” Climate Policy 6 (2006): 49-72; Point Carbon Advisory Services, EU ETS Phase II—The Potential and Scale of Windfall Profits in the Power Sector (2008), Prepared for World Wildlife Fund. 74 See CRS Report RL33846, Greenhouse Gas Reduction: Cap-and-Trade Bills in the 110th Congress, by (name redacted), (name redacted), and (name redacted). 75 For example, the President and CEO of Duke Energy, who initially was an advocate a cap-and-trade approach (and remains a member of USCAP), described the Lieberman-Warner bill (S. 2191) as a “bastardized” version of cap-andtrade due to its allocation scheme. See Jim Rogers, “Climate Change Legislation Should Not Be Punitive” (February 2008), Energy Daily. 76 Hepburn (2006), citing P.D. Klemperer, Auctions: Theory and Practice (2004), Princeton University Press. 77 In a January 2009 speech, the CEO of ExxonMobil (Rex Tillerson) spoke in favor of a carbon tax over a cap-andtrade system, stating a carbon tax would be “the most efficient means of reflecting the cost of carbon in all economic decisions—from investments made by companies to fuel their requirements, to the product choices made by consumers.” This quotation was taken from a UPI news report, at [www.upi.com]. Congressional Research Service 21 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress without a price certainty—for downstream energy consumers (e.g., power plants).78 Moreover, the upstream entities subject to direct compliance—that is, required to submit allowances—would have little or no direct emissions. The gap between compliance and emissions could pose challenges in a situation without a known price (i.e., a cap-and-trade program). Support from Environmental Groups? Supporters of stringent GHG emission control may oppose a carbon tax, because they assume that if Congress enacts a carbon tax, Members would set the rate too low to induce emission mitigation: e.g., energy conservation, fuel substitution, or technology development. This may be more of an argument against the tax rate, than the instrument itself. A similar concern of stringency may exist if policymakers decide to employ a cap-and-trade system. For example, the cap might be set at undesired levels, or the program may allow for the use of questionable offsets, which could pose a risk of undermining the objective of emission reduction. Likewise, the pressure to delay the strengthening of an established cap that is considered ineffective or expensive could be considerable. Regardless, an emissions cap—at least on paper—offers more of a guarantee that a specific emission level would be met. Some environmental groups may assert that this perceived guarantee overshadows other concerns. Consideration of International Efforts and Cooperation It is generally recognized that mitigation of human-induced climate change will ultimately require action by all major emitting countries. Therefore, when assessing different mechanisms of GHG emission reduction, Congress may want to consider how an market-based instrument—for example, a cap or a tax—would 1) complement existing or future international efforts and/or 2) maximize participation among major emitting nations. For policymakers and negotiators who are attempting to create an international GHG reduction regime, establishing either instrument on an international scale would present substantial and unique challenges. A carbon tax may be at a disadvantage, primarily because the most prominent international activities currently involve cap-and-trade programs. However, some assert that a carbon tax approach may present a stronger opportunity to maximize participation between the United States and developing economies—for example, China and India—that emit substantial portions of GHG emissions. Some of the issues are discussed below. Coordination with Existing International Efforts At the core of the international GHG mitigation efforts under way—for example, the Kyoto Protocol and the EU ETS—is a cap-and-trade program among nations with quantitative commitments. If Congress enacted a cap-and-trade system in the United States, the program could be designed to allow trading between systems. In terms of overall economic efficiency, linked trading programs would be ideal, because covered sources would have access to more 78 There may be some industry stakeholders who support a carbon tax on the assumption that it will be politically challenging for Congress to subsequently raise carbon taxes to the level needed to stimulate the stringent emissions reductions targeted (e.g., 80% below current levels by 2050) in recent cap-and-trade proposals. Congressional Research Service 22 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress emission reduction opportunities. However, linked programs could yield unintended consequences if they differ in scope and/or stringency.79 A U.S. carbon tax program could conceivably have a linkage of sorts with international cap-andtrade programs. For example, a carbon tax system could allow for tax credits for emission reductions made outside of the United States.80 However, this trade would only flow in one direction: Covered sources in the EU could not purchase reductions from covered sources in the United States. Thus, this type of linkage would not equate with a union of cap-and-trade programs. Sources in the United States that were subject to the carbon tax would find this opportunity worthwhile only if the carbon tax rate were higher than the emission allowance price (EU emission allowances or EUAs) or, if eligible, offset project prices (traded as certified emission reductions or CERs) in the EU ETS. However, if policymakers were to provide tax credits for international emission abatement projects, concerns similar to those involving offsets in a cap-and-trade system would be raised. If the illegitimate projects qualify as tax credits, the primary objective of the program—GHG emission control—may be compromised. Maximizing Participation Some have raised the argument that encouraging maximum participation among nations should be a higher priority than instrument choice and the potential consequences of choosing a cap over a tax (e.g., loss of economic efficiency). 81 Would a carbon tax or a cap-and-trade program be the most effective approach to bring the most nations, or more precisely, emissions, under a mitigation umbrella? Neither strategy would necessarily require equal commitments between nations. For example, the United States and China could each establish unilateral carbon tax programs—as opposed to an international, harmonized system—with different tax rates. Indeed, an international carbon tax system would encounter substantial implementation challenges, particularly regarding issues of national sovereignty. It would be (and has proven) difficult to achieve wide and diverse participation—namely, agreement from both the United States and developing nations—in a GHG control regime. Some may favor a cap-and-trade approach, because of the buildup of existing programs and the political momentum behind the cap-and-trade approach. However, this reasoning may be questioned on several counts. First, although a substantial percentage of the world’s CO2 emissions are covered under the Kyoto or EU ETS cap, an even larger percentage are not covered. 82 This latter group includes the United States, China, and India, which accounted for approximately 54% of the CO2 emissions from the top-25 CO2 emitting nations in 2005.83 Second, the mere existence of a program (and its political momentum) does not indicate it is the optimal solution. Indeed, the 79 For example, different treatment of offsets could raise concern, particularly if covered entities in the United States gain access to questionable offset projects. See CRS Report RL34436, The Role of Offsets in a Greenhouse Gas Emissions Cap-and-Trade Program: Potential Benefits and Concerns, by (name redacted). 80 Qualified credits could range from EU emission allowances (EUAs) from the EU ETS to certified emission reductions (CERs)—i.e., international offsets—issued by the Clean Development Mechanism of the Kyoto Protocol. 81 Scott Barrett and Robert Stavins, “Increasing Participation and Compliance in International Climate Change Agreements,” International Environmental Agreements: Politics, Law and Economics, Vol. 3:349-76 (2003). 82 However, Kyoto participants account for the majority of 1990 emissions (the Kyoto baseline). 83 Data from World Resources Institute, Climate Analysis Indicators Tool (CAIT), at http://cait.wri.org/. Congressional Research Service 23 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress existing international cap-and-trade systems, which are still in their early stages, have encountered substantial problems. 84 However, a carbon tax approach would encounter obstacles similar to those observed in the capand-trade discussions. For example, it would be difficult to reach agreement on the equitable or fair tax rates for each nation. Not only would this agreement be difficult between developed and developing nations—for multiple reasons that are beyond the scope of this report—but it would also pose a challenge between developed nations. For example, a 2000 study found that European nations were (at the time of the study) effectively taxing carbon at a rate of approximately $27/ton of CO2.85 If European nations and the United States sought to establish their own carbon tax rates, pre-existing carbon taxes (and subsidies) may complicate the discussions. International Implementation Concerns Regardless of the chosen instrument, either would impose its own set of logistical challenges. Some assert that an international cap-and-trade system would be more susceptible to corruption. 86 However, it could be argued that monitoring and enforcement of carbon taxes in individual nations might pose similar difficulties. A primary obstacle would be determining a nation’s “net carbon tax,” because nations or political subdivisions could offset the carbon tax through energy subsidies or loopholes for the entities most impacted by the tax.87 Moreover, some nations may not have transparent tax systems in place now. Adding a carbon tax to such a framework could be problematic. Implementation of a Carbon Tax If Congress were to use a carbon tax to control GHG emissions, policymakers would face several implementation decisions: (1) the point of taxation—where to impose the tax and what to tax; (2) the level of taxation; and (3) how to apply the tax revenue. These considerations are discussed below. Point of Taxation A point of taxation debate is analogous to a debate over the scope or coverage of a potential capand-trade program. Congress would face similar considerations of emissions coverage with either instrument. Describing coverage issues, EPA stated: Ideally, all sources, sectors, and emissions would be included for full coverage and maximum environmental effectiveness and economic efficiency. However, measurement 84 See CRS Report RL34150, Climate Change and the EU Emissions Trading Scheme (ETS): Kyoto and Beyond, by (name redacted); and CRS Report RL33826, Climate Change: The Kyoto Protocol, Bali "Action Plan," and International Actions, by (name redacted) and (name redacted). 85 William Nordhaus and Joseph Boyer, Warming the World: Economic Modeling of Global Warming (2000), MIT Press. 86 William Nordhaus, “To Tax or Not to Tax: Alternative Approaches to Slowing Global Warming” (2007), Review of Environmental Economics and Policy, Vol. 1 (1): 26-44. 87 David Victor, The Collapse of the Kyoto Protocol and the Struggle to Slow Global Warming (2001), Princeton University Press. Congressional Research Service 24 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress capabilities and costs, available control options, administrative burdens, political considerations, and other constraints may limit participation to a subset of emission sources.88 A comprehensive discussion of these issues is beyond the scope of this report,89 but the following sections provide options for policymakers. Note that these sections overlap to some degree. Where to Impose a Carbon Tax? Regarding where in the economy to impose a carbon tax, the options—at least for fossil fuel combustion-related CO2 emissions90—are often couched in terms of their location in the energy production and consumption chain: upstream or downstream (Figure 7): Figure 7. Illustration of Options for Points of Taxation within the Energy Production-to-Consumption Chain Upstream Oil wells Natural gas wells Coal mines Importers Midstream Natural gas processors/pipelines Oil refineries Electric utilities Downstream Vehicles Households Commercial buildings Industry Source: Prepared by CRS. Note: Electric utilities could be listed as either downstream entities—because they are direct sources of emissions—or midstream, because their emissions are tied to the electricity consumption of their customers, the further downstream consumers. 88 EPA, Tools of the Trade: A Guide to Designing and Operating a Cap and Trade Program for Pollution Control (2003), p. 3-5. 89 For additional reading on these considerations, see e.g., Gilbert Metcalf and David Weisbach, The Design of a Carbon Tax (June 2008), Tufts University and the University of Chicago; EPA, Tools of the Trade (2003). 90 Non-CO2 GHG emissions come from a wide range of sources, most of which are outside the energy sector. Congressional Research Service 25 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress • Upstream—An upstream approach would apply a carbon tax to fossil fuels when they enter the U.S. economy, either at the mine, wellhead, or another practical “chokepoint” in the production chain. Imported fuels would be addressed at their point of entry into the United States (Figure 7). Under this approach, a carbon tax would be levied before emissions have occurred. The tax would be based upon the carbon content of the fossil fuel produced. An advantage to the upstream approach is that the number of sources directly subject to the tax would be minimized, while covering a broad spectrum of the economy. • Downstream—A downstream approach would apply the tax on the emissions themselves. For example, emission sources, such as power plants or industrial facilities,91 would pay a fee for each ton of CO2 released into the atmosphere. In some economic sectors, there may be advantages to imposing a carbon tax downstream, even if more entities would be directly subject to the tax. For instance, Congress may consider a downstream approach for the electric utility industry, because power plants already continuously monitor their CO2 emissions to the EPA. 92 There may be other advantages to applying the tax directly to emissions from coal-fired power plants. For example, some argue that this approach would provide a greater stimulus93 to the development of carbon capture and sequestration (CCS) technology. 94 However, in some sectors—for example, transportation or residential—a downstream approach may be challenging, and perhaps impractical to implement because of the comparably large number of emission sources. • Combined Approach—Sometimes referred to as a hybrid strategy, this approach would cover emissions from different economic sectors in different manners. For example, transportation fuels could be taxed upstream at the point of extraction (or midstream at the refinery); coal could be taxed at the emission sources: power plants and specific industrial sectors. Where Congress decided to impose a carbon tax would determine which entities would be required to (1) make tax payments based on emissions or emission inputs, (2) monitor emissions or emission inputs, and (3) maintain records of relevant activities and transactions. A direct tax on emissions may be easier to understand, but considerably more difficult to implement across all economic sectors, due to the large number of emission sources. A complicating factor is that some fossil fuel uses—for example, feedstock in asphalt production— do not generate emissions but sequester the carbon. Policymakers may consider mechanisms to exempt the fuels used for such purposes or allowing for tax credits for these activities. 91 Power plants may be considered midstream, because their emissions are influenced by the demand from electricity consumers: businesses and households. In addition, some carbon-intensive industries, such as steel or cement, might be categorized as midstream. 92 Section 821 of the 1990 Clean Air Act Amendments requires electric generating facilities affected by the acid rain provisions of Title IV to monitor CO2 in accordance with Environmental Protection Agency (EPA) regulations. 93 Although a price signal would be sent in either case—thus (in theory) encouraging CCS development—some contend that if coal is taxed upstream at the extraction point, some of the price signal may be weakened before it reaches coal-fired emission sources. 94 For more information on this technology, see CRS Report RL33801, Carbon Capture and Sequestration (CCS), by (name redacted). Congressional Research Service 26 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Fossil fuels have varying levels of carbon content, thus a carbon tax would impact fossil fuels differently. For example, coal has almost twice the carbon content per unit of energy as that of natural gas (Table 1). Thus, a carbon tax would raise the price of coal more than the price of natural gas. Table 1. CO2 Emissions Per Unit of Energy for Fossil Fuels Fossil Fuel CO2 Emissions Per Unit of Energy (million metric tons/quadrillion BTU) Coal 94 Crude oil 74 Natural gas 53 Source: Prepared by CRS, based on CO2 emission factors from EIA, at http://www.eia.doe.gov/environment.html. Note: CO2 emissions from coal vary by the type of coal being used. The 1993 Energy or Btu Tax A carbon tax is different from an energy consumption tax, like the British thermal unit (Btu) tax that was proposed by President Clinton in 1993. Compared to a carbon tax, a Btu tax would be a less efficient mechanism to reduce CO2 emissions. As highlighted in Table 1, fossil fuels generate different amounts of CO2 emissions per unit of energy. A Btu tax would treat energy sources equally, thus a Btu tax would serve different objectives than a carbon tax. The 1993 Btu tax proposal called for a levy of 25.7 cents per million Btu, with a surcharge of 34 cents/million BTU on petroleum. The goals of the 1993 Btu tax proposal were to promote energy conservation and raise revenue. At the time, the proposed tax would have generated a new revenue stream of about $30 billion per year. The proposal was met with strong opposition and was not enacted; Congress ultimately enacted an (approximately 5-cent) increase in the motor fuels taxes. CO2 Emissions or All GHG Emissions? Another point-of-taxation issue is whether a carbon tax should cover CO2 emissions or a more comprehensive array of GHG emissions. Mitigation opportunities from non-CO2 gases are often less expensive than other reduction options. Economic studies indicate that one advantage of including more or all GHG emissions, as opposed to only CO2 emissions, is that the overall costs of the GHG reductions would be decreased.95 A recent analysis of a carbon tax proposal—H.R. 3416 (Larson)—from the 110th Congress found that by adding other GHG emissions to the tax base, the overall costs of achieving the same amount of emission reduction would fall by 20%.96 However, this finding was based on the inclusion of GHG emissions from all sources. As discussed below, the majority of the non-CO2 GHG emissions come from sources that may present tax implementation challenges. 95 See e.g., John Reilly et al., “The Role of Non-CO2 GHGs in Climate Policy: Analysis Using the MIT IGSM” (2006), The Energy Journal, Special Issue No. 3: 503-520. 96 Metcalf et al. (2008). The overall costs fall, because the tax rate is decreased. By including all GHGs, the tax rate can be lowered if the cumulative target remains the same. Congressional Research Service 27 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress Which Emissions Sources to Control? GHG emissions are spread throughout the economy and are generated by millions of discrete sources: e.g., smokestacks, exhaust pipes, households, and livestock. Although CO2 is the primary GHG, some GHG emissions sources predominantly emit non-CO2 GHGs. When determining which sources and gases to control through a tax, policymakers would need to balance the benefits of inclusion with administrative costs.97 To achieve a specific emissions target, a tax applied on only a subset of economic sectors—thus covering only a percentage of U.S. GHG emissions—would need to have a higher rate than a tax applied to all GHG emissions. Numerous rationales have been put forward to justify excluding one group or sector from a reduction requirement, or to provide some other special consideration. Table 2 lists the top emission sources of six GHGs in the United States. These sources combine to account for approximately 95% of the total (based on 2006 data).98 Table 2 also provides potential points of taxation that would cover the emissions from these sources. CO2 emissions from fossil fuel combustion yielded approximately 80% of U.S. GHG emissions in 2006.99 Policymakers could address CO2 emissions from fossil fuel combustion and nonenergy uses—in aggregate 82% of U.S. GHG emissions—by levying an upstream carbon tax on fewer than 2,500 upstream or midstream entities (Table 2). A portion of these entities are already subject to environmental taxes that support other objectives.100 A downstream approach may be feasible for electric generating facilities powered by fossil fuels, which number approximately 3,300,101 but impractical for other fossil fuel combustion sources— automobiles, offices, homes—which number in the millions. One possible alternative to a complete upstream/midstream strategy would be to address coal, the most carbon-intensive fossil fuel, at its combustion location. In 2006, coal-fired power plants accounted for 92% of U.S. coal combustion.102 As shown in Table 2, there are about 641 coalfired power plants in the United States. Some argue that there would be advantages to applying a point of taxation at electric generating facilities. For example, these facilities may have a stronger incentive and/or capability to seek tax credit opportunities, such as biological sequestration projects (if allowed within the tax system). Moreover, some contend that the development of CCS would be enhanced if coal-fired facilities were directly subject to the tax. The rationale for this argument is that under an upstream approach (that indirectly taxes coal-fired emissions) the price signal may be weaker than if the tax were applied directly to emissions at coal-fired facilities. However, other sources of coal combustion would need to be covered to avoid leakage situations—that is, increased coal (and thus emissions) use in the uncovered sectors. It may be 97 Gilbert Metcalf and David Weisbach, The Design of a Carbon Tax (June 2008), Tufts University and the University of Chicago. 98 There are additional GHGs, but they are not included in the current U.S. GHG inventory. For more information, see CRS Report RL34266, Climate Change: Science Highlights, by (name redacted). 99 Based on 2006 data. EPA, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2006 (April 2008). 100 For example, a 5-cent-per-barrel tax on the oil industry—domestically produced oil and oil imports—supports the Oil Spill Liability Trust Fund. 101 In 2006, there were 3,292 plants that had at least one fossil fuel-fired generator (EIA, 860 data files). 102 EIA, Annual Coal Report 2006 (2007). Congressional Research Service 28 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress burdensome to address the remaining sources of coal combustion at the point of emissions, particularly at residences and commercial entities. 103 Beyond fossil fuel combustion, policymakers could cover more GHG emissions by expanding the coverage to some or all of the remaining sources listed in Table 2. Individually, many of these sources account for relatively small proportions of U.S. GHG emissions, but if some combination of these sources were included in the carbon tax base, the increase in emissions coverage could be appreciable. The remaining source categories vary by (1) the number of entities that would be subject to a carbon tax, and (2) the circumstances of emission release. In general, a large number of entities subject to a carbon tax would present higher administrative costs per unit of emissions. Regardless of entity number, some emissions sources would be more difficult to cover because of logistical challenges, such as emission measurement and/or monitoring. Although GHG emissions from the agriculture sector—farming and livestock—account for approximately 6% of total GHG emissions in the United States, agricultural emissions would be comparatively difficult to cover under a carbon tax. As indicated in Table 2, these emissions are generated by a large number of entities, and there are no upstream measurement points analogous to those that could cover emissions from the transportation sector. However, nitrous oxide (N2O) emissions from the use of artificial fertilizers—as opposed to natural fertilizers like manure— might be an emission source that could be addressed through an upstream point of taxation. EPA estimates that approximately 25% of the N2O emissions from agricultural soils are related to artificial fertilizer use.104 This represented approximately 0.8% of U.S. GHG emissions in 2006, so it would be included in Table 2 if counted separately. A tax on fertilizers may lead to less fertilizer use, but some argue that it could stimulate other activities—for example, increased use of natural fertilizers—that would release N2O emissions (a form of emissions leakage).105 This issue is beyond the scope of this report. In many cases, the emissions from agricultural activities may be difficult to monitor and/or measure. For practical concerns and other reasons—for example, political considerations—the agriculture sector has thus far been excluded as a capped source in the cap-and-trade proposals that have specifically designated covered sources. 106 If not subject to a carbon tax, agriculture emissions could be addressed with other policy tools. For example, tax credits could be provided for qualifying reduction, abatement, or sequestration projects.107 The other emission sources listed in Table 2 may be more amenable to a carbon tax, at least in terms of the number of entities and emission circumstances. Compared with agricultural sources, the other emission source categories—excluding HFCs (discussed below)—comprise a relatively small number of entities, at which emissions are easier to measure and monitor. In fact, certain 103 Industry used 5% of the coal in 2006; residences and commercial entities used 3%. EPA, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2006 (April 2008). 105 Metcalf and Weisbach (2008). 106 Some of the cap-and-trade proposals made in recent years would delegate coverage decisions to the implementing agency, typically EPA. 107 Alternatively, policymakers could develop emission/sequestration standards for specific activities. This latter option would not be a market-based approach, but it was included in one of the cap-and-trade proposals from the 110th Congress—H.R. 6186 (Markey). 104 Congressional Research Service 29 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress sources—for example, some landfills and underground coal mines—currently control their (methane) emissions for air pollution and safety reasons. Methane from landfills—1.8% of U.S. emissions—could be covered by including an additional 1,800 landfills in the tax base. To some degree, large municipal landfills are already required per federal air regulations to collect and combust methane emissions. Other landfills are following suit under state laws or voluntary programs. 108 Hydrofluorocarbons (HFCs) are used as substitutes for several classes of ozone depleting substances (ODSs—hereafter, HFC-ODS).109 HFC-ODS account for 1.6% of total U.S. GHG emissions (Table 2). HFC-ODS emissions occur under different circumstances than the other, “smaller” sources discussed above—landfills, coal mines, etc. Most of the HFC emissions (86% in 2006) come from air conditioning and refrigeration units—located in vehicles, buildings, and homes—during production, operation, and disposal.110 Unlike the other “small” source categories, which emit GHGs onsite, HFC-ODS emissions enter the atmosphere from millions of locations, far away from and long after HFC production. Because of the different emission circumstances of HFC-ODS, some have proposed combining a tax at HFC production—five facilities in the United States—with a tax credit for HFCs properly recycled or disposed.111 For example, automobile air conditioners could be sent to designated sites in order to recycle or dispose of the HFCs. Norway has an HFC rebate program in place that may be instructive for U.S. policymakers.112 As with other carbon-intensive materials, Congress may consider applying a tax to imported HFCs as well. By applying the tax to iron/steel production and cement manufacturing facilities—approximately 250 entities—an additional 1.3% of U.S. emissions could be covered.113 A further 1.5% could be addressed by including methane from natural gas systems. Roughly 25% of this amount comes from field production, which may be impractical to monitor and measure accurately.114 The remainder involves emissions from normal operations, routine maintenance, or “system upsets.” Fugitive emissions (or accidental releases) are the primary source of emissions from these activities. 115 A tax on these emissions may impose administrative costs in terms setting up a process to measure and monitor these emissions. Methane emissions from underground coal mines—0.8% of total GHG emissions—may be easier to monitor under a tax than at aboveground mines. At all underground mines, methane is ventilated for safety reasons; methane from aboveground operations is released directly to the 108 See EPA’s Landfill Methane Outreach Program, at http://www.epa.gov/lmop/overview.htm. A small amount of perfluorocarbons (PFCs) are also used as substitutes (EPA Inventory, 2008). Under the Montreal Protocol and the Clean Air Act Amendments of 1990, ODSs are being phased out. See CRS Report RL30853, Clean Air Act: A Summary of the Act and Its Major Requirements, by (name redacted) et al. 110 EPA Inventory (2008). 111 See e.g., Metcalf and Weisbach (2008). 109 112 Intergovernmental Panel on Climate Change, Safeguarding the Ozone Layer and the Global Climate System, Issues related to Hydrofluorocarbons and Perfluorocarbons (2005). 113 As with fossil fuels, policymakers may want to consider applying the tax to imported steel and cement. This would increase the number of entities subject to the tax. 114 See e.g., Metcalf and Weisbach (2008). 115 EPA Inventory (2008). Congressional Research Service 30 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress atmosphere. The 0.8% figure includes methane only from underground coal mines, which account for about 61% of methane from coal mines.116 Table 2. Selected Sources of U.S. GHG Emissions and Potential Applications of a Carbon Tax GHG Emission Source CO2 from fossil fuel combustion: Percentage of U.S. GHG Emissions (2006 data) 79.9 Potential Carbon Tax Applications Entity Coal minesa Number 1,374 641 - electricity generation or - transportation Coal-fired power plantsb - industrial Power plantsc using imported coald 26 Petroleum refineriese 150 Petroleum importersf 220 Natural gas processorsg 530 Natural gas importersh 45 - commercial/residential N2O from agricultural soils 3.8 Farmsi CO2 from non-energy use of fuels 2.0 Covered by the tax applied to fuels (above)j CH4 from livestock (enteric fermentation) 1.8 Cattle operationsk CH4 from landfills 1.8 Landfillsl HFCs from the substitution of ozone depleting substances 1.6 HFC manufacturersm 5 CH4 from natural gas systems 1.5 Natural gas processors 530 CH4 from coal mines 0.8 Coal minesn CO2 from iron/steel production 0.7 Raw steel production facilities; 116 Integrated steel millso 18 118 >2 million 967,440 1,800 1,374 CO2 from cement manufacturing 0.6 Cement plantsp CH4 from manure management 0.6 Cattle operations; 967,440 Swine operationsq 65,640 Percentage of Total GHG Emissions 95.1 Source: Prepared by CRS; GHG emission data from EPA, EPA, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2006 (April 2008); data for number of entities from multiple sources, cited in footnotes. 116 a. This figure accounts for mines in operation in 2007. EIA, Coal Production and Number of Mines by State and Mine Type, (2008). b. In 2006, there were 641 plants with at least 1 coal-fired generator unit (EIA, 860 Database). EPA Inventory (2008). Congressional Research Service 31 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress c. Number of plants comes from EIA database, Monthly Nonutility Fuel Receipts and Fuel Quality Data (Database 423). CRS was unable to determine the number of companies that act as coal importers, analogous to petroleum importers. d. In 2006, the United States imported approximately 36 millions short tons of coal (EIA, Quarterly Coal Report (2008), table 4)—3.5% of the amount to coal consumed domestically in that year (EIA, Annual Coal Report 2006 (2007), table 26). Coal imports have increased by more than 200% since 2002. e. This figure represents the number of “operable” refineries. EIA, Refinery Capacity Report (2008). f. EIA, Company Level Imports (as of November 2008). Note that some of these companies may import only crude oil, whose emissions would be covered by the tax at the domestic refineries. Thus, this figure represents an upper bound of petroleum product importers potentially subject to a carbon tax. g. EIA, Natural Gas Processing: The Crucial Link Between Natural Gas Production and Its Transportation to Market (2006). h. This includes pipelines and liquefied natural gas facilities. EIA, About U.S. Natural Gas Pipelines (as of September 2008). i. U.S. Department of Agriculture, Farms, Land in Farms, and Livestock Operations: 2007 Summary (2008). The resource defines a farm as “any place from which $1,000 or more of agricultural products were produced and sold, or normally would have been sold, during the year.” j. Fossil fuels are used for a wide range of non-energy purposes. EPA estimates that of the total carbon consumed for non-energy purposes, approximately 62% is stored in products, and not released to the atmosphere (EPA, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2006 (April 2008), tables 3-14 and 3-15). The 2% value in Table 4 represents the emissions. In an upstream carbon tax system, fuels would be taxed before they are used. Congress may want to consider providing tax credits for the amount of carbon stored in products. k. U.S. Department of Agriculture, Farms, Land in Farms, and Livestock Operations: 2007 Summary (2008). l. EPA, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2006 (April 2008), citing BioCycle, 15th Annual BioCycle Nationwide Survey: The State of Garbage in America (2006). m. Intergovernmental Panel on Climate Change, Safeguarding the Ozone Layer and the Global Climate System, Issues related to Hydrofluorocarbons and Perfluorocarbons (2005), Figure 11.1. n. Methane from underground mines, which accounts for about 61% of coal mine methane, is removed through ventilation systems for safety reasons. These emissions would be easier to monitor under a carbon tax than aboveground coal mine methane emissions. o. Data from U.S. Geological Survey, Mineral Commodity Summary, Iron and Steel Production (2008), at http://minerals.usgs.gov/minerals/pubs/commodity/iron_&_steel/. p. Cement manufacturing information from Portland Cement Association, at http://www.cement.org/basics/cementindustry.asp. q. U.S. Department of Agriculture, Farms, Land in Farms, and Livestock Operations: 2007 Summary (2008). Other animals—chickens, horses, and sheep—contribute approximately 10% of the total emissions from manure (EPA, Inventory of U.S. Greenhouse Gas Emissions and Sinks: 1990-2006 (April 2008), table 6-6). Level of Taxation If Congress were to use a carbon tax to control GHG emissions, setting the level of the carbon tax would be a subject of intense debate. Regardless of the chosen tax schedule, some parties would argue that the selected rate was either too low or too high. Although the decision would likely involve political calculations, there are economic principles that could be used to inform the debate. This section identifies and discusses two approaches that policymakers could take to set the level of a carbon tax. The first approach is based on the estimated benefits associated with avoiding climate change impacts. The second approach is based on a GHG emissions target and the tax rate needed to achieve that target. The first approach is arguably stronger from a Congressional Research Service 32 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress theoretical standpoint, but the second approach is arguably (considering the uncertainty associated with the first approach) a more practical option. To provide some context for potential carbon tax rates, Table A-1 of this report (located in the Appendix) includes estimates of price increases to fossil fuels and motor gasoline based on different carbon tax rates. Tax Based on Estimates of Costs and Benefits117 As noted previously, economic principles involving pollution taxes prescribe that the optimal tax should be at the intersection of the marginal costs of emission abatement with the marginal benefits of abatement. In terms of climate change, the intersection of costs and benefits is often described as the Social Cost of Carbon (SCC). The Intergovernmental Panel on Climate Change (IPCC) defines the SCC as the following: The discounted monetized sum (e.g. expressed as a price of carbon in $/tCO2) of the annual net losses from impacts triggered by an additional ton of carbon emitted today. According to usage in economic theory, the social cost of carbon establishes an economically optimal price of carbon at which the associated marginal costs of mitigation would equal the marginal benefits of mitigation.118 For multiple reasons, applying this economic principle to set a carbon tax currently presents monumental, if not insurmountable, challenges. These are highlighted below. Social Cost of Carbon Estimates Estimates of the SCC are fraught with uncertainty, thus the wide range of estimates produced should not come as a surprise. According to a 2008 meta-analysis, there are over 200 SCC estimates from 47 studies. The estimates range across several orders of magnitude, from zero to over $500 per ton of CO2-equivalent (tCO2-e) emissions.119 The uncertainty and controversy of choosing a value within this range would be formidable. The definition of SCC indicates the complexity involved in its estimation. The SCC can be defined as a monetary measure of the marginal damages (or costs)—on a global scale—of an incremental unit (e.g., one metric ton) of anthropogenic GHGs emitted to the earth’s atmosphere at some point in time (e.g., now). Estimating the SCC requires analysts to place values on goods and services that may be difficult (or controversial) to precisely measure: e.g., human health/life, water supplies, agricultural production, recreational activities. The value of these goods/services will vary from one location 117 A comprehensive discussion and analysis of climate change costs and benefits is beyond the scope of this report. This section highlights issues associated with estimating and applying climate change cost and benefit estimates for carbon tax implementation. 118 IPCC, Climate Change 2007: Mitigation, (Annex I—Glossary)Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (2007). 119 Richard Tol, “The Social Cost of Carbon: Trends, Outliers, and Catastrophes,” Economics E-Journal (August 2008). This updates a previous study: Richard Tol, “The Marginal Damage Costs of Carbon Dioxide Emissions: An Assessment of the Uncertainties,” Energy Policy 33:2064-2074 (2005) Congressional Research Service 33 Carbon Tax and Greenhouse Gas Control: Options and Considerations for Congress to the next, complicating the estimate. Moreover, the climate change processes that may impact these goods and services are not completely understood.120 The element of time particularly complicates an SCC estimation and application in several ways. An SCC estimate signals—per economic theory—what global society should be willing to pay now to avoid future damages due to additional emissions generated today. As discussed above, CO2 and other GHGs remain in the atmosphere for decades to centuries, thus current emissions could generate damages for years to come. To take time into account—that is, determine the cost of future damages in today’s dollars (referred to as net present value)—economists employ a discount rate.121 Discount rates are typically at the crux of cost-benefit analyses and their assignment often generates controversy. As with other public policy decisions that employ costbenefit analysis, the chosen discount rate is a critical factor when estimating the SCC. In terms of uncertainty in evaluating damages from climate change, the Intergovernmental Panel on Climate Change (IPCC) estimates that the discount rate is the second most important factor (clim [Text truncated at 120,000 characters. The full text is on the page linked above.] --- Source: Frix Law Library, https://www.frixlaw.com/law-library/documents/crs%3AR40242. Public record. Not legal advice.