The Basic Structure of Federal Climate Change Policy
(Return to Top)

The policy debate concerning U.S. action to address climate change is increasingly active and complicated, with a suite of proposals being considered at the federal level even as new initiatives emerge at the state, regional, and international levels. The issue briefs compiled in this report are intended to provide a guide through that debate and, in particular, to help policymakers at the federal level identify the various advantages and disadvantages of different legislative proposals. We take the position that while state-level and international activities are important - especially in terms of their eventual intersection with federal policy - it is necessary to start with an understanding of the basic questions that are central to designing an effective national policy.¹

Download PDF

In thinking about the architecture of a domestic policy, two important and potentially competing design criteria must be balanced. The first is the desire to establish a clear vision of the future and provide enough policy certainty going forward that key actors in the economy - especially those faced with making long-term investment decisions - can plan effectively and adjust smoothly to greenhouse gas (GHG) constraints. Second, the policy architecture must be sufficiently flexible to evolve over time. No domestic program adopted in the next five or even ten years is going to solve the climate change problem once and for all. Rather, it will be necessary to revisit U.S. policy at intervals and to respond to new information and developments in climate science, mitigation technologies, and international commitments. This places a premium on policies that are both capable of being modified over time without significant economic disruption and robust enough to drive the emissions reductions and technology innovation needed to produce environmentally significant results in a relevant timeframe. Policies that lock in particular actions, technologies, or political interests - for example, narrow subsidies that do not gradually phase out - may be difficult to adjust and ultimately more costly; therefore, the arguments for such policies, along with the potential for unintended consequences over the long-term, must be evaluated carefully.

Over time, a key factor in the evolution of U.S. climate policy will be the choices made and actions taken by other nations, particularly nations with large economies and significant emissions of greenhouse gases. Without concerted global action, U.S. policies alone cannot substantially alter climate outcomes. Further, to be cost-effective - and achieve global benefits at the lowest cost - national policies will need to be coordinated, if not connected. In this context, adopting a domestic policy architecture that offers the specific ability to coordinate and adjust both carbon dioxide (CO2) prices and technology incentives is particularly valuable.²

Underlying this report is an assumption that establishing a price path for CO2 and other GHG emissions - either through a tax or tradable permit program - will be a core element of future U.S. climate policy. A pricing strategy is appealing because it responds to the need for both policy clarity and flexibility - making it possible, on the one hand, to predict prices and emissions over reasonable timeframes with a reasonable degree of certainty while also facilitating smooth adjustments over time. There are other compelling economic arguments for taking this assumption as a starting point. As a means of addressing climate-change risks, GHG reductions are equally valuable wherever they occur - but they are not equally costly. From an economic perspective, this means that costs to society can be reduced by implementing a policy that achieves cheaper emissions reductions without any trade-off in environmental benefit. Setting a price on GHG emissions sends a transparent signal to everyone engaged in emissions-producing activities - including direct emitters as well as downstream consumers of emissions-producing products - about the value of reducing emissions. Those who can reduce emissions cheaply will do so, while those who cannot will face a common CO2 price.

The alternative to a single price policy is a more traditional approach to government regulation in which emissions abatement requirements or technology standards and incentives are applied to various GHG sources, such as power plants, factories, cars, and households. While this type of strategy is feasible, evidence suggests it could be much more expensive. Studies that compare the costs of traditional regulation to the costs of market-based approaches have typically found substantial differences: for example, a recent study by economists at RFF suggests that the costs of limiting U.S. GHG emissions through traditional regulatory approaches could be ten times higher than achieving the same result through a pricing policy.³ Given that the costs of addressing climate change are likely to be far from trivial under any circumstances - the cost of an efficient policy has been estimated at 1 percent or more of GDP over many years - maximizing economic efficiency should be a primary consideration for policymakers. (By comparison, the total cost of existing environmental regulation in the United States has been estimated at 2 percent of GDP.)

In fact, recent interest in market-based policies has probably had less to do with academic arguments about economic efficiency and more to do with the experience accumulated through real-world emissions trading programs, starting with the U.S. Acid Rain program in the 1990s and continuing with subsequent broad-based trading programs for both sulfur dioxide and nitrogen oxides in the United States and, as of 2005, for large industrial sources of CO2 in Europe. Of the climate-policy proposals debated in Congress, the great majority feature a cap-and-trade style approach to limiting GHG emissions. The alternative of a carbon tax has not featured as prominently in the current debate, but the larger trend toward establishing a single CO2 price appears well established.

Strong arguments are often made for additional public support and incentives to accelerate the transition to climate-friendly technologies, particularly in the electric power and transportation sectors where a large volume of emissions are concentrated and where market problems may exist. These arguments are well supported by economic theory in the case of research and development. In the case of technology deployment, however, policies that are not carefully designed to address a specific market problem tend to raise costs relative to a simple CO2 pricing policy. Public spending on research and development, for example, is typically viewed as necessary to capture the broad public benefits that arise from new knowledge: because these benefits cannot be fully appropriated by individual firms, private-sector incentives for basic R&D typically fail to reflect the full societal benefits generated by those investments. Uncertainty about future policy commitments may further dilute incentives for private-sector investments in developing new technology.

Broadly speaking, technology deployment policies find support in economic arguments where there are information problems that impede the market response to price incentives, spillovers from learning associated with new technology investments, network effects for large integrated energy systems, and incomplete insurance markets for managing liabilities associated with new-technology investments. In these cases, technology deployment policies can lower the cost associated with achieving a given emissions-reduction target. Often, however, such policies are driven by other interests, including enthusiasm for particular technologies (renewables, biofuels, carbon capture and storage, hybrid vehicles, hydrogen, etc.), a desire to avoid imposing explicit costs in the form of higher energy prices, and a desire to shift costs to the general taxpayer via subsidies. Policies created largely for these reasons, rather than to address market problems, typically raise the overall cost to the economy of reaching the environmental goal compared to a simple pricing policy centered on an emissions tax or permit trading program.

For a given economic cost in aggregate, the distribution of costs across businesses and consumers - and over time - can vary considerably. Most discussions of cost focus on aggregate economic impacts, such as changes in the cost of energy and loss of GDP. However, the distribution of impacts across different industries, regions of the country, and demographic groups can vary considerably. Competitive industries with high energy costs, regions of the country that depend on more carbon-intensive fuels, and households that have higher energy expenditures and lower incomes, are all at greater risk.

There is also a temporal dimension to costs. Many policies currently under discussion propose to achieve relatively modest near-term reductions that lead gradually to significantly deeper reductions in the future, coupled with the flexibility to move emissions-reduction obligations over time. With the predictability and flexibility afforded by this type of approach, businesses can adjust their investments and households can save now to offset higher burdens in the future. In this way, costs should be smoothed out over time. Without predictability and flexibility - or if the policy generates inaccurate expectations that lead to poor investment decisions - costs are likely be higher in the future. Alternatively, unexpected, positive developments could lead to lower costs in the future.

Additional technology policies - including policies designed to overcome barriers to zero- and low-carbon energy sources - can be economically efficient (that is, they may lower the overall costs to society of achieving long-term environmental goals) only as complements to, rather than substitutes for, a pricing policy. Implementing public R&D investments, traditional performance standards for stationary sources or equipment, tradable portfolio standards for electricity generation or fuels, or subsidies alongside a broader pricing policy may be justified if the aim is to address other market problems while the CO2 price encourages emissions reductions. The same is true for policies that address natural gas supply, nuclear waste, the siting of renewable energy projects, electricity grid infrastructure, and efficiency, where the status quo may or may not achieve an adequate balancing of costs and benefits. Used in place of a CO2 price to achieve a given emissions-reduction target, such policies will almost certainly result in higher overall costs compared to a broad-based emissions tax or cap-and-trade program.

As a substitute for policies that effectively price CO2 emissions, for example, performance standards for energy-using equipment reduce the energy-related costs of using that equipment. The effect of lower energy costs may or may not be to cause consumers to increase their use of more efficient equipment, but certainly such standards don't serve to encourage less use. (To give a concrete example: fuel-economy standards reduce the per-mile cost of driving, thus they don't encourage consumers to use their vehicles less.) In effect, low-cost opportunities to reduce emissions by simply reducing equipment use are foregone, implying higher-cost mitigation somewhere else. Also, regulations specific to a single sector will not balance the cost of whatever actions they require against potentially less costly abatement opportunities elsewhere in the economy; as a result, someone will almost certainly spend more than necessary to meet the overall target. Third, traditional regulation does not offer the same incentives for continual innovation over time as do policies that put a price on GHG emissions - once firms meet a standard, there is no incentive to exceed it. Over time, this again leads to higher costs.

As a complement to CO2 pricing, on the other hand, technology policies may or may not lower costs or emissions, depending on the extent to which they address an existing market problem. In either case, it is important that policymakers understand the interaction between a broad-based pricing policy and narrower technology policies. If an emissions cap is in place, technology policies can, at best, only serve to reduce costs and will not produce additional emissions reductions. Similarly, under a tax or other price-setting mechanism, such policies can only serve to reduce emissions and will not lower the price.

Domestic climate policy should be viewed in the context of energy policy more broadly. More than 80 percent of U.S. GHG emissions come from the combustion of fossil fuels. Reducing these emissions implies changing the energy sources used to power the U.S. economy toward increased reliance on zero- or low-carbon fuels. Policies that affect the availability, cost, and usability of natural gas, renewable resources, nuclear power, carbon capture and storage, and end-use efficiency improvements will have important consequences for the cost and success of climate policy. While this report focuses exclusively on the design of climate policy, policies implemented to address broader energy objectives can act to significantly support or undermine climate policy goals and to mitigate or exacerbate the economic impacts associated with achieving GHG reductions. Conversely, policies intended to address climate risks may simultaneously support or undermine broader energy policy goals. For example, climate policies that reduce oil consumption could yield energy-security benefits. On the other hand, climate policies that create additional demand for natural gas in the power sector - absent new supply opportunities - could drive up natural gas prices for industrial users and give rise to additional competitiveness concerns.

1. There is obviously an important role for state action in policy areas that present potentially inexpensive emission-reduction opportunities and that have typically been addressed at the state level - examples would include building codes and utility demand-side management programs. A distinct and important question is now emerging about whether and how federal legislation might "pre-empt" state efforts to regulate - on the basis of climate concerns - in areas beyond their traditional role. On the one hand, a patchwork of state regulation creates additional costs. Further, state actions will not affect national emissions if there is a national cap: in that case, additional emissions reductions in one state as a result of state- or region-specific policies will lead to lower allowance prices and higher emissions in other states (assuming, as current federal proposals do, that there are no constraints on emissions trading between states). On the other hand, states may wish to pursue more aggressive targets in a way that prevents other states from benefiting and/or states may wish to exercise greater control over how and where mitigation occurs. Economic principles suggest it would not be efficient for states to pursue separate policies, but offer little guidance in terms of how policymakers might weigh efficiency considerations against other, competing interests and states' rights.

2. This report focuses on policies that aim to mitigate GHG emissions. Additional domestic policy concerns and international considerations apply in the case of policies aimed at spurring investment in adaptation and possible geo-engineering options for reducing climate-change risk.

3. Pizer, W. et al, 2006. Modeling Economywide Versus Sectoral Climate Policies Using Combined Aggregate-Sectoral Models, Energy Journal 27(3), 135-168. For further discussion of these issues, see Issue Brief #5: Taxes Versus Tradable Permits.

Key Questions for the Design of U.S. Climate Policy
(Return to Top)

As an introduction to the more detailed issue briefs that follow, it is useful to consider a series of key questions that arise in putting together a comprehensive climate-policy package. These questions can be used to build up a proposal from scratch, or to "unpack" and understand the core elements of an existing proposal. Taken together they touch on, and provide a framework for organizing, important themes from all 15 issue briefs. The first five of these questions concern the core design of a mechanism to price emissions; the remaining three explore additional policies to address specific technology opportunities (and often unique features) in key sectors.

1. What is an appropriate, overall domestic emissions objective for the United States at this time and how does that objective balance competing considerations and risks with regard to environmental protection, economic costs, technological development, institutional constraints, and global participation? A fundamental challenge in designing a domestic climate change policy is defining an initial target emissions trajectory that will, over time and in conjunction with actions by other major emitters, achieve environmental objectives at an acceptable cost to the economy. Most global analyses designed to help answer this question start with the relationship between atmospheric GHG concentrations and temperature change, work backwards from different atmospheric stabilization targets to investigate their implications for emissions, CO2 prices, and economic output, and then contemplate a reasonable balance of costs and objectives.⁴ To select an appropriate national-level, emissions-reduction goal, policymakers will need to consider the relationship between domestic action and future technology development, broader action by other major emitting nations, and longer-term emissions trends at the global level. As a starting point, it is useful to examine - as we do in Issue Brief #2 - what trajectory for U.S. emissions and carbon prices would be consistent with the results obtained from global analyses of cost-effective paths to different stabilization scenarios.⁵

Figure 1 is based on the synthesis report of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC). It summarizes current understanding of the relationship between atmospheric concentrations of GHGs and likely changes in global average surface temperature (where "likely" is defined as corresponding to a probability range of 67 percent or higher). In the context of current emissions trends, which - if continued - would result in atmospheric GHG concentrations (in CO2-equivalent terms) of 1,000 parts per million (ppm) by volume or more by the end of this century, stabilizing GHG concentrations in the 450-650 ppm CO2-equivalent range would significantly reduce the magnitude of expected warming and associated risks to human welfare and ecological integrity.⁶ Stabilization at 1,000 ppm implies long-term warming of as much as 3-8°C relative to present conditions, whereas stabilization at 450-650 ppm reduces the range of predicted temperature change to 0.5-4.5°C.

Translating stabilization targets of 450-650 ppm into emissions scenarios, CO2 prices, and economic costs requires models of economic activity and emissions. Considerable analysis has been done on 650 ppm scenarios, with results that suggest achieving stabilization at this level will require global emissions to flatten out over the next several decades and CO2 prices to reach $5-$30 per metric ton by 2030 and $20-$90 per ton by 2050 - assuming global participation and efficient policies.⁷ The model scenarios point to significant technology shifts, including increased reliance on systems that capture and store CO2 emissions from coal-fired power plants, nuclear power, renewable energy, and energy efficiency. The costs incurred to achieve stabilization (again assuming efficient policies and global participation) in these 650 ppm scenarios generally amount to a reduction of less than 1 percent of GDP compared to the business-as-usual forecast over the next decade. Less is known about the economic impact of achieving more aggressive stabilization targets, but these scenarios generally involve significantly higher costs and require global emissions to begin declining in the next decade or sooner. The IPCC estimates that global stabilization in the 450-550 ppm range could be achieved at a cost of less than 3 percent of world economic output and with CO2 prices below $100 per ton in 2030 (again, assuming global participation and efficient policies).

Detailed analyses of various scenarios for limiting U.S. GHG emissions over the next two decades produce cost and carbon-price estimates that are roughly consistent with the results obtained in global analyses.⁸ The more aggressive domestic policy scenarios assume that emissions through 2030 are limited to roughly 1990 levels; the less aggressive scenarios assume emissions roughly stabilize at current levels or even increase slightly over the same timeframe. Modeled costs in all scenarios are less than 1 percent of forecast GDP for 2015. Costs reach almost 2 percent of GDP in the more aggressive scenarios by 2030, but remain below one-half of 1 percent of GDP under the less stringent scenarios.⁹ Carbon prices range from $15 to $100 per ton CO2 and are therefore comparable to the carbon-price estimates obtained when modeling cost-effective global stabilization scenarios in the 450-650 ppm range. Importantly, all of these estimates are for the near- to medium-term timeframe. Considerable uncertainty exists about the longer-term costs of achieving these targets. On the one hand, deeper reductions will be required as we approach the mid-century mark; on the other, capital stocks will have had more time to adjust and promising technologies may have emerged in that timeframe.

The question frequently arises, of course, how any domestic emissions target can be environmentally meaningful - or indeed worth incurring costs to achieve - absent full international participation in emissions-reduction efforts. More to the point, how can U.S. policymakers choose a domestic target while consensus on an appropriate global objective is still lacking and while other major emitting countries have yet to adopt their own emissions-reduction commitments? While acknowledging that broader international participation and a clear roadmap to achieving global reductions will eventually be necessary to address climate risks, at least four different kinds of considerations can provide justification for near-term U.S. action and can help inform the selection of appropriate domestic targets.

First, a serious policy commitment by the United States is likely to have a significant effect on the actions of other major emitting nations (and may even represent a necessary precondition for establishing broader participation, especially on the part of some developing countries). The European Union, for example, has announced a 2020 target of reducing emissions 20 percent below 1990 levels - and has indicated it will increase the reduction target to 30 percent below 1990 levels if other countries take comparable action. One way to approach the design of domestic policy is by asking what scale of reduction commitments we seek in other countries to achieve a globally meaningful result. Second, domestic policy can stimulate international actions more directly by recognizing offset credits associated with verifiable emissions-reduction projects undertaken in developing countries.¹⁰ In fact, it is likely that international offsets will play a role in meeting domestic targets, especially if those targets are relatively aggressive. Thus, one could imagine defining the objectives for a U.S. program in terms of the emissions reductions we seek to achieve domestically plus some volume we expect to pursue in poorer countries.

A third option is to consider what emissions price we want to encourage in other countries, rather than what level of emissions reductions. That is, we can design a domestic policy to produce an emissions price consistent with a particular stabilization target and encourage other countries to seek a similar price. It will arguably be easier to converge toward a globally harmonized carbon price, and to develop some confidence in the feasibility of attaining a given stabilization target, once the United States has achieved political consensus on a pricing policy. Finally, a fourth argument can be made on the basis of technology considerations. Significant emissions reductions in the future - in the United States and in other countries - will depend on significant technology developments. Absent effective market incentives in the world's most advanced economies, the technology developments needed to achieve substantial global reductions and lower abatement costs to globally affordable levels will be unlikely to materialize. Accordingly, an important consideration for domestic policy is how effectively it will encourage necessary long-term technology advances.

In the end, the choice of an appropriate, initial goal for U.S. climate policy must balance multiple considerations and objectives, including the need to achieve meaningful environmental benefits, motivate broader international participation, minimize costs to the domestic economy, address competitiveness concerns, and promote low-carbon technology development and deployment (including technology transfer to developing countries).

2. What sectors of the economy should be covered by a single carbon price and where in the energy supply chain should energy-related CO2 emissions be regulated? From the standpoint of maximizing economic efficiency, policymakers should try to cover as many emissions sources as possible with a single policy and a single emissions price. This approach expands the pool of low-cost emission reduction opportunities that can be exploited to achieve a given policy target, thereby minimizing overall costs to the economy. Equally important, this approach addresses the risk that a market-based policy will create incentives to shift emissions to sources that are not covered under the policy. For example, if households and small businesses are excluded from an emissions cap imposed on the electric power sector, these end-users may shift some of their energy consumption away from electricity and toward increased use of primary fuels (such as oil and natural gas) for space conditioning and water heating. While such shifts may, in some cases, produce gains in energy efficiency they represent a form of emissions "leakage" and will undermine the achievement of environmental objectives since emissions, rather than being reduced, merely shift from regulated to unregulated sources. Nonetheless, arguments are often made for treating some sectors differently - either by regulating them via a separate policy mechanism or by excluding them completely. In some sectors, the economic impacts of GHG regulation may be more severe or there may be a desire to create regulations tailored to specific sector needs. The latter reflects the view that different sectors and sources face different hurdles that may be best addressed through different policies, with the government choosing technologies or performance improvements rather than firms doing so in response to market signals. All of these arguments tend to run counter to conventional economic thinking and to more than a decade of research that suggests broad, market-based policies can substantially reduce costs relative to targeted regulatory approaches.

Tied up with the question of program coverage is the question of where to regulate energy-related CO2 emissions. In contrast to most conventional air-pollutant emissions, energy-related CO2 emissions can be effectively regulated anywhere in the fossil-fuel supply chain based on a simple calculation involving fuel carbon content and through-put. It should be noted that the same is not true for non-energy-related CO2 emissions and for other (non-CO2) GHGs; however, these collectively constitute a much smaller share of overall emissions.

Two additional observations concerning the choice of where to regulate are important. First, while past tradable permit programs have typically allocated free permits or allowances to regulated sources, there is no reason why CO2 permits or allowances cannot be allocated to other entities in the fossil-fuel supply chain that are directly or indirectly affected by regulation. In other words, decisions about how to allocate permits or allowances need not be tied to decisions about which entities will be required to submit permits or allowances under a trading program. This distinction is important because stakeholders, if they fail to understand it, will tend to assume that decisions about where to regulate also constitute de facto decisions about how to distribute permits with a likely asset value, in aggregate, on the order of tens of billions of dollars per year (we return to this point below).

A second important point is that the decision about where to regulate - whether upstream or downstream - generally does not change the economic burden imposed on different entities in the fossil-fuel supply chain.¹⁴ The price signal generated by an emissions tax or trading program is passed forward and backward between upstream and downstream entities and achieves the same ends regardless of where it is actually imposed. Important caveats may apply in situations where products are not competitively priced (as, for example, in regulated utility markets). Finally, the point of regulation does affect which entities bear the administrative burden of demonstrating compliance under a tradable permits program.¹⁵

3. How much emphasis should be placed on providing certainty about future GHG emissions versus certainty regarding the future cost of the policy? A particularly contentious issue in the debate over the design of a federal cap-and-trade program for U.S. GHG emissions is whether total emissions should be strictly capped (that is, limited), as has traditionally been the case in existing programs of this type. The alternative is to make additional allowances available when the market price of allowances reaches a predetermined maximum. This mechanism, which is frequently termed a "safety valve," trades emissions certainty in favor of cost certainty - effectively, it means that the level of the emissions cap is not fixed but rather becomes contingent on a maximum price. Coupled with a mechanism to create a price floor - which could involve the government either (a) re-purchasing allowances if the price reaches a specified minimum, (b) specifying a minimum price in allowance auctions, or (c) tightening future emissions caps in response to persistently low prices - trading programs can, to a large extent, mimic the price certainty of a tax.¹⁶ Disagreements about whether a cap-and-trade policy should include a safety valve are often intense because they pit two fundamental concerns - protecting the environment and protecting the economy - against each other. Because climate impacts ultimately hinge on the long-term accumulation of global emissions, the case for choosing price certainty over emissions certainty is strongest in the early years of a U.S.-only policy. Over longer horizons and with broader global efforts, fixed emissions targets can be increasingly advantageous as they are more closely tied to actual environmental outcomes (for example, stabilizing atmospheric GHG concentrations at a particular level). This suggests that if a safety valve is used, it may be more valuable in the short run.

A number of other cost-containment mechanisms have been proposed as alternatives to a safety valve; in most cases these aim to provide similar benefits (in terms of limiting economic impacts and allowance-price volatility), even as they shift the balance back toward greater environmental certainty. Many of these proposals involve allowance banking and borrowing: for example, businesses could borrow allowances from the government in one year and pay them back in a future year, with interest. This would tend to stabilize allowance prices in response to short-term fluctuations in demand and supply, but would not affect long-term drivers of CO2 price such as expectations about future targets, technologies, and energy demand. Of course, such expectations might still be subject to substantial uncertainty given the potential for politically motivated adjustments to longer-term term targets and other program parameters. (For example, if borrowing resulted in an acute shortfall of allowances in some future year, the political pressure to increase allowance budgets - at least temporarily - could be intense.)

A more recent proposal for reducing economic risk in connection with a domestic GHG cap-and-trade program involves a distinct government agency charged with balancing environmental and economic objectives and given the authority to intervene in markets by buying and selling allowances (and possibly in other ways). In principle, this concept could represent an attractive compromise, one that reassures private industry while promising greater environmental integrity. In practice, however, neither objective would be well served if such an agency is poorly designed, if its interventions are badly executed, or if statutory constraints tie its hands. It is worth noting that the Federal Reserve Board, which provides something of a model for this idea, was established only in response to a financial crisis nearly 100 years ago. Moreover, its performance was widely criticized during many decades of its existence. Notwithstanding these pitfalls, the concept of a "carbon Fed" is sufficiently promising that it merits further exploration.

In the often heated debate about cost-containment mechanisms, it is also important that policymakers not lose sight of the larger objective: to implement a well-designed program with broad coverage of emissions sources and clear rules concerning targets, trading, compliance, and flexibility. Such a program will deliver the most environmental benefit at the lowest cost and should serve as the starting point for any discussion of additional mechanisms for enhancing cost certainty.

4. How should the distributional consequences of an emissions pricing policy be addressed? Specifically, how should revenues (in a tax system) or the asset value represented by emissions allowances (in a cap-and-trade program) be distributed back to society? Under the original U.S. Acid Rain trading program, as under most of the trading programs that have followed since, the great majority of allowances has been distributed gratis (at no cost) to directly regulated entities. This need not be the case, however: permits can be given to entities other than those that are directly regulated under the program (including, for example, households or state governments). Regulated firms then buy allowances from allowance recipients. Moreover, allowances need not be given away at all: they can be sold or auctioned by the government, which can then retain and re-distribute resulting revenues for other purposes.¹⁷

The likely market value of emissions allowances in many proposals is not trivial, amounting to tens if not hundreds of billions of dollars annually (with similar revenue arising from a comparable carbon tax). This overall allowance value is not an indication of the overall cost of the program to the economy; rather, it represents a transfer from those who directly or indirectly pay for allowances in the form of higher fossil energy prices to those who hold allowances (whether those holders are taxpayers, in the case where government auctions allowances, or private-sector entities, in the case where government distributes allowances for free to selected firms).

Economic efficiency argues for the government selling allowances and using the revenue to cut other taxes. By some estimates, this approach could produce net economic gains as lower labor and capital taxes will encourage more employment and investment; in any case, it reduces the net burden imposed on the economy. Indeed, even if allowance revenues are not used to cut other taxes, they can fund valuable government expenditures that otherwise require an increase in taxes.¹⁸ At the same time, this economically efficient solution could have undesirable distributional properties, imposing very different cost burdens on different sectors of the economy and different regions of the country depending on the fuels they use and their ability to pass through costs. By contrast, arguments for a free allocation are typically premised on the need to address distributional concerns by targeting free allowances to those sectors, firms, and regions that would otherwise be most adversely affected by the policy.

Any free allocation that changes in response to future business developments - such as one that continually updates firm-level shares of the total allowance pool based on production output - must be carefully scrutinized in terms of its incentive properties. Updating allocation methodologies can produce inefficient outcomes by creating incentives that promote excess production, discourage the retirement of inefficient facilities, or - depending on the specifics of the methodology - encourage continued investment in high-emitting technologies. While these incentive properties might be desirable in some cases - for example, to promote continued domestic production in industries that might otherwise be motivated to move their operations overseas - they might produce perverse outcomes in other instances (for example, a new entrant allocation that would unnecessarily encourage coal-fired power plants over lower-emitting alternatives).

In the end, decisions about how to allocate allowances or tax revenues, both within and between sectors, are deeply political in nature as they involve the re-distribution of significant wealth and require a careful balancing of competing claims. Policymakers will need to weigh a wide range of concerns and objectives: the desire to reward leaders versus help laggards, for example, or to accommodate new entrants without over-accommodating them in ways that creates perverse incentives to continue investing in higher-emitting sources. At a macro-economic level, additional trade-offs exist between equity and efficiency.¹⁹ That is, policymakers must weigh the merits of using free allowances to compensate entities that will otherwise bear a disproportionate share of the economic burden of the policy against the overall efficiency benefits that could be realized by using allowance revenues to reduce other taxes. At the same time, policymakers will have to address the concern that an overly generous free allocation could result in unjustified windfall gains for some firms and industries.

5. How should the policy address international competitiveness concerns? A chief concern surrounding most proposals for a mandatory GHG reduction policy is that pricing emissions will adversely affect the competitiveness of U.S. businesses and may encourage businesses to move their operations overseas. This would obviously undermine the ability to achieve stated environmental goals, along with public support for the policy. A variety of strategies have been proposed to address these concerns.

The simplest involves starting with a modest "first step" domestically while linking more aggressive future targets to international progress. This approach recognizes that the potential for competitive distortions depends on the degree of disparity that exists between the scope and stringency of climate policies in the United States and the policies that exist in other nations. The idea would be to limit economic costs until similar efforts are underway among key trading partners. The argument against this approach is an obvious environmental one: it delivers less environmental benefit, weaker incentives for technology development, and less pressure for international participation.

Other strategies involve singling out especially vulnerable industries for special treatment. Identifying these sectors can be challenging, however: evidence suggests a need to focus on both the energy intensity of domestic producers and the level of international competition they face. Typically, industries that make primary, bulk-produced products (iron and steel, aluminum, cement, and glass) are the most vulnerable to competition from overseas suppliers. Once vulnerable industries have been identified, at least four options exist for addressing competitiveness concerns related to a GHG policy. The simplest is to exclude those operations from the policy altogether; however, this is also the most inefficient response since completely excluding an activity means forgoing possibly inexpensive mitigation opportunities that would not drive production overseas. A second option is to limit emissions from these sources using traditional, tailored forms of regulation that might be less likely to create similar competitiveness concerns. Again, however, this solution is likely to be inefficient and potentially costly. A third approach would be to use free allowances to compensate industries for higher energy-related costs (in this case, the free allocation would need to be tied to continued domestic production). The challenge would be to specify an allocation formula that adequately offsets regulatory costs without over-subsidizing production and without unfairly advantaging some firms relative to others, or U.S. firms in general relative to their foreign competitors.

The last option would be to implement additional policies that directly target imports (and/or exports) of goods to the United States, rather than attempting to adjust the impacts of the GHG policy on domestic producers. The idea would be to regulate energy-intensive, bulk commodity goods imported from countries that lack comparable CO2 policies on the basis of embedded CO2 content and in a manner that parallels the price impacts of domestic regulations. This approach would have the dual advantage of directly addressing competitiveness concerns while also creating incentives for other countries to adopt comparable policies. Its key downside is the potential to provide cover for unwarranted and inefficient forms of protectionism that, in addition to their immediate costs, would hinder the long-term economic development and technology transfer needed to achieve global progress in addressing climate change. In the end, a combination of strategies may be necessary to address the competitiveness concerns of different stakeholders. Among these, excluding a sector completely or using alternate, traditional regulation tend to have the most costly consequences for the rest of the economy (assuming the overall emissions goal is held constant).

6. To what extent should a domestic climate policy create additional requirements and/or incentives (beyond the GHG price signal) for accelerated technology development and deployment? Alongside the debate about how to price emissions, policymakers must confront an additional set of questions concerning the appropriate government role in technology development. At one end of the spectrum are those who believe the private sector - once motivated by a price on GHG emissions - is best positioned to make R&D and technology investments. At the other end of the spectrum are those who see a much greater role for government.²⁰ As noted at the outset, a relatively clear economic case can be made for government involvement in research and development, both basic and applied. That said, the best approach to managing such investments is far from clear, especially in the case of applied research. U.S. Department of Energy program offices, a new public agency, a new quasi-public corporation, and/or private research consortia could all be used to manage an increased public budget for applied energy research - each has advantages and disadvantages in terms of fostering effective management and performance, providing stable funding, degree of insulation from politics, and public accountability.

The case for public investment becomes less clear moving from research to technology deployment, a frequent target of additional policies and regulation - particularly in the electricity and transportation sectors, as discussed below. On one hand, legitimate market imperfections can justify public support for technology deployment. On the other hand, technology deployment policies often go well beyond this initial motivation in practice (if such a motivation existed in the first place) - in ways that imply substantial costs and efficiency losses beyond those incurred by the CO2 pricing policy alone.

Nevertheless, such policies continue to have strong political appeal, perhaps in large part because their costs tend to be less explicit and/or fall more heavily on the general taxpayer, and because they provide a more visible means to promote popular technologies.²¹

Different technology deployment policies make different tradeoffs with respect to risk, cost burden, and relative efficiency. They can be used to guarantee quantitative outcomes (via standards) or fix the price of new technologies (via subsidies). Subsidies can be structured as fee-bates to shift the financing burden from the taxpayer to lower-performing technologies. Policies that allow greater flexibility, typically in the form of crediting, banking, and trading, will tend to lower costs.

One way to evaluate technology policies as potential complements to a common CO2 price is to consider the following questions: Does the policy address a market problem distinct from reducing CO2 emissions and thereby provide additional, otherwise un-priced benefits? Or, are there other aspects of the policy that make it more appealing and therefore worth incurring higher costs? Are the higher costs reasonable for the volume of emissions reductions and other public benefits achieved? Is the policy as flexible and cost-effective as possible, given other key features and constraints? If the answer to these questions is yes, the benefits of the additional policy under consideration are more likely to outweigh its costs.

7. How can climate policies be designed to address equity concerns and confront multiple technology challenges in the electricity sector, given the considerable variation in resource portfolios and regulatory structures that characterizes this industry? The electric power industry represents one of the largest and most concentrated sectors for GHG emissions, and one where international competitiveness concerns generally do not apply.²² With only 3,000 facilities that together account for 33 percent of U.S. GHG emissions, any long-term climate solution must deal with the challenge of transforming the nation's power sector. Complicating this challenge is the fact that the electricity industry is enormously diverse, with different regions of the country relying on a different combination of fuels and generation technologies - and hence characterized by different CO2-emissions profiles - and being governed by different regulatory structures. This variation has important implications for the use of complementary policies and for the design of a CO2 pricing policy itself.²³

Given the importance of the electric power industry, it is perhaps not surprising that numerous complementary policies - in addition to a CO2 pricing policy - have been proposed for this sector. These range from increased public support for basic research and development to additional technology policies, including direct subsidies, performance standards for new facilities, portfolio standards for existing facilities, and energy efficiency programs. These policies need to be evaluated carefully because while some may resolve various market problems others can lead to more costly solutions than are otherwise necessary. For example, carbon capture and storage may be a critical technology that does not benefit from adequate incentives under a broad-based CO2 price because of long-term policy uncertainty and the additional liabilities associated with underground storage and other unfamiliar aspects of the technology. While additional incentives may be justified by these and other considerations, it remains the case that additional incentives for carbon capture and storage and other climate-friendly technologies in the electric sector must be carefully monitored to avoid creating imbalances with other abatement opportunities.

In the context of an emissions trading program, special complexities arise with respect to allocating allowances within the power generation sector. In general, the electricity industry as a whole should be able to pass through a large fraction of the cost associated with GHG regulation (including the cost of both mitigating emissions and purchasing necessary allowances) as electricity prices rise to reflect emissions costs. This means that if a free allocation is meant to offset regulatory cost burdens, the bulk of any free allocation should go to electricity users and only a relatively small share of the allowances associated with electricity-related CO2 emissions needs to be given to electric power generators. In different regions of the country, however, the way in which emissions costs are passed through and the consequences of free allocation to generators will be very different depending on whether electricity markets are competitive or regulated. In regulated regions, generators are traditionally protected by rules that guarantee a rate of return on investments and could expect to be allowed to pass through all costs. At the same time, regulators are likely to ensure that the value of any free allowances allocated to generators in regulated regions will be passed on to consumers by way of reducing the price impact of the CO2 policy. In competitive regions, the price of electricity is set by the marginal generator - which will rise to reflect the opportunity cost of CO2 allowances regardless of any free allocation generators receive. Here, the degree to which individual generators can pass through emissions costs depends on their emissions profile compared to the marginal generator. Thus, in competitive regions, free allocation can be used to offset those emissions costs that are not passed through and are borne by generators. The possibility also exists, however, that free allocation to some generators in competitive regions will more than offset costs and result in increased profits.

These and other considerations have led to a variety of proposals for handling allocation in the electric sector, including proposals that establish a greater role for auctions.²⁴ At a minimum, many current proposals now envision any free allocation that exists in the early years of program implementation will be phased out over time. Such a phase-out aligns with the underlying notion that free allocation is supposed to compensate for unequal regulatory burdens - burdens that eventually become more evenly distributed as existing capital depreciates and new investments are made. Some have proposed that free allowances be allocated to load-serving entities instead of generators, with the idea that this could lead to more consistent outcomes - in terms of the impact on retail electricity prices - across regulated and competitive regions alike. Allocation to load-serving entities could be based on a variety of measures including electricity consumption, population, or emissions by generators in a state or region. Still other proposals include allocations to end-users, including both large industrial users as well as state governments ho could then use allowance assets to address local issues.

8. What are possible approaches to address emissions in the transportation sector, where achieving reductions comparable to those in other sectors might otherwise require significantly higher carbon prices and longer lead times? How do available policy options compare in terms of the emissions reductions they achieve, the costs they impose, the distribution of those costs, and their short- and long-term effects on different drivers of transport-sector emissions, including vehicle fuel economy, fuel carbon content, and vehicle miles traveled? As in the electric power sector, numerous additional policies have been proposed to reduce GHG emissions and advance other policy objectives in the transportation sector. It is unclear whether this interest in additional sector-specific policies stems from transportation's large share of overall emissions (28 percent of the U.S. total, including 16 percent of the U.S. total from light-duty vehicles alone), from the historic regulation of light-duty vehicle fuel economy, or from the observation that under a typical carbon pricing policy, transport-sector emissions are unlikely to decline very much. Regardless, various policies have been proposed to directly address two of the three factors that drive overall GHG emissions from this sector: the fuel-economy of new vehicles and net GHG emissions from the production and use of different transportation fuels. The remaining factor is vehicle miles traveled, which has increased by 25 percent over the last decade for light-duty as well as larger vehicles. There are few policy alternatives to a carbon price for delivering incentives to reduce travel demand.²⁵

As in the electric power sector, the use of additional policies (beyond a GHG price) will tend to raise the overall cost of reducing emissions unless those policies are addressing additional market problems. In the case of fuel economy standards, the concern is frequently voiced that consumers do not adequately value fuel economy, thereby justifying the existing CAFE (Corporate Average Fuel Economy) program and creating momentum to strengthen current standards.²⁶ Recent changes in the CAFE standard for light trucks offer some guidance for making the overall program more cost-effective and could be applied to cars. Meanwhile, additional program reforms (such as trading across fleets and manufacturers, a safety valve mechanism, and/or shifting to a feebate program) could improve efficiency even further.

Fuel requirements, such as a renewable fuels standard or low-carbon fuel standard, by contrast, represent a relatively new policy approach for addressing transport-sector GHG emissions.²⁷ In their most flexible form, fuel standards specify an average life-cycle emissions rate per gallon that must be met in aggregate, and are designed to achieve that rate as cost-effectively as possible. Nonetheless, both fuel standards and vehicle efficiency standards should be evaluated carefully to ensure that they do not go too far in creating higher mitigation costs in a narrow area of activity when cheaper emission-abatement opportunities exist elsewhere.²⁸

In contrast to the electric power sector, where additional policies (such as a renewable portfolio standard) are typically viewed as complementary to a carbon pricing policy, fuel and vehicle performance standards in the transport sector are sometimes viewed as potential substitutes for including the sector in a unified GHG pricing policy, particularly since any policy that can be portrayed as raising the price of gasoline tends to be politically unpopular. The argument is also often made that demand for transportation fuel is relatively inelastic at the level of price signal contemplated in most current GHG cap-and-trade proposals; therefore, excluding the transportation sector from an economy-wide CO2 price would not be expected to have the effect of foregoing a significant quantity of emissions abatement. Nevertheless, over time excluding transport sector emissions from a broader pricing policy and relying instead on fuel and vehicle standards is likely to be increasingly inefficient, as CO2 prices rise and the potential impact of higher fuel prices on vehicle miles traveled could become more important. Equally important, distinct transportation policies such as low-carbon fuel requirements and vehicle fuel economy standards do not trade-off CO2 mitigation opportunities across sectors.

4. An entirely different approach is to assign economic value to climate change impacts and then compute the value of mitigation. Such an approach can generate a wide range of results depending not only on the valuation, but also rather critically on how effects over time are discounted. For example, William Nordhaus estimates mitigation benefits of $6.40 per ton CO2 while Nicholas Stern estimates benefits at $85 per ton. Some of the findings from this literature are summarized in Issue Brief #3 on costs.

5. Issue Brief #2 on U.S. Climate Mitigation in the Context of Global Stabilization examines the first part of this question; Issue Brief #3 on Assessing the Costs of Domestic Regulatory Proposals examines the second part, as well as the benefit estimates in the preceding note.

6. Here and throughout, we discuss GHGs and GHG concentrations in carbon-dioxide equivalent (CO2e) terms. This means that when discussing various stabilization targets we include the atmospheric concentration of carbon dioxide plus the effect of other greenhouse gases that enter the atmosphere as a result of human activities, where those other gases are converted to a global warming-equivalent volume of carbon dioxide. Expressed in terms of carbon dioxide concentrations only, the same stabilization targets are typically 50 to 100 ppm less.

7. Here and elsewhere, prices are in current (2005-2006) dollars unless specifically indicated otherwise.

8. It is important to recognize that available models provide only a stylized representation of the likely economic impacts of different policies. As policies diverge from a simple economy-wide CO2 pricing regime (implemented through either a cap-and-trade system or an emissions tax), overall costs could rise significantly.

9. The total cost of all existing U.S. environmental regulation is around 2 percent of GDP, according to estimates by the Bureau of Economic Analysis and the U.S. Environmental Protection Agency (EPA). The sulfur dioxide (acid rain) trading program is estimated to cost around 0.01 percent of GDP. Over the time period that most of these regulations have been introduced and implemented (that is, over the past 40 years), average household income has grown at an average annual rate of 1.3 percent while median household income has grown at an annual average rate of 0.7 percent.

10. Issue Brief #15 discusses international offsets in some detail.

11. We devote an entire issue brief (Issue Brief #14) to the topic of regulating "nontraditional" emissions - both process CO2 emissions and other gases. Much of this category of emissions consists of fugitive emissions from land-use changes that would be difficult to capture outside of an offset program (see also Issue Brief #13 on agriculture).

12. The distribution of source size and potential for regulation is discussed in Issue Brief #1.

13. Under an upstream program, adjustments would have to be made for imported and exported fuels, sequestered emissions such as carbon capture and storage, and uses of fossil fuels that do not result in emissions. Note that the European Union Emissions Trading Scheme currently accounts for emissions on the basis of fuel volume and carbon content and not through direct emissions monitoring, even though emissions are regulated at the combustion source rather than upstream.

14. This is analogous to the observation that it does not generally matter whether employer or employees pay income and payroll taxes - the effect on the eventual wage paid by the employer and received by the employee end up being the same.

15. In a well-designed program administrative costs are likely to be small in a relative sense, compared to allowance prices and the cost of reducing emissions. However, these administrative costs - which include the tran