This is the eighth in a series of questions that highlights RFF’s Expert Forum on EPA’s Clean Power Plan.
RFF asks the experts: Should EPA allow states to use a carbon tax to comply with the Clean Power Plan, and if so, how could a state demonstrate the associated carbon emissions reductions?
EPA's proposed Clean Power Plan sets carbon emissions targets for each state, denoted in pounds of carbon dioxide per megawatt-hour of electricity, but does not explicitly mention a carbon tax as an option for states to achieve their targets. Some argue that a carbon tax represents a simple and effective way to reduce emissions. A state that wishes to use a carbon tax to comply with the Clean Power Plan would likely have to convince EPA that this approach would allow the state to achieve its emissions target. Should EPA allow states to use a carbon tax to comply with the Clean Power Plan and, if so, how could a state demonstrate the associated carbon emissions reductions?
“We believe EPA should revise its proposal to be consistent with a state carbon tax as a primary means of achieving the emissions guidelines. . . . Just as for other compliance strategies, a state can demonstrate through economic modeling that its carbon tax program is likely to achieve the emissions standard set by EPA.”
—Michael Wara, Associate Professor and Justin M. Roach, Jr. Faculty Scholar, Stanford Law School (See full response.)
—Adele Morris, Fellow and Policy Director for the Climate and Energy Economics Project, The Brookings Institution (See full response.)
“EPA should allow states to achieve their carbon intensity targets through an upstream cap-and-permit system . . . [such a system] is preferable to a carbon tax or fee without an upstream cap—even if the fee is accompanied by dividends—because it sets the quantity of carbon that can be burned rather than the price.”
—Peter Barnes, Entrepreneur and Author of With Liberty and Dividends For All (See full response.)
“EPA’s Clean Power Plan allows cap and trade as a compliance option, so it is natural to allow carbon taxes as well. … However, EPA needs to be careful about how that option is implemented.”
—Roberton Williams III, Senior Fellow and Director of Academic Programs, Resources for the Future (See full response.)
Michael Wara, Associate Professor and Justin M. Roach, Jr. Faculty Scholar, Stanford Law School
Adele Morris, Fellow and Policy Director for the Climate and Energy Economics Projects, The Brookings Institution
A carbon tax is a simple, effective, and efficient means for reducing greenhouse gas emissions. Despite the proposed Clean Power Plan’s flexibility, it does not acknowledge or explicitly allow for the use by states of carbon excise taxes. As we explain in our comments to the agency, we believe EPA should revise its proposal to be consistent with a state carbon tax as a primary means of achieving the emissions guidelines. A carbon tax as a state implementation strategy makes environmental, economic, and administrative sense because it does the following:
Discourages each fuel’s use in exact proportion to its damage to the climate.
Incentivizes changes at power plants (for example, more efficient boilers and lower-carbon fuels) and greater energy efficiency by consumers.
Is market-based, flexible, compatible with existing fuel mixes, and accommodates the “remaining useful life” of equipment.
Encourages abatement in ways EPA and states can’t predict; for example, by helping drive a market for new technologies.
Is easy to implement: some states already have excise taxes on fuels, and they already monitor greenhouse emissions from regulated sources.
Gives states the choice of how to use new revenue—they could use it to lower inefficient taxes, potentially providing pro-growth state tax reform along with environmental benefits; or to offset some of a carbon tax’s impacts on low-income residents.
Just as for other compliance strategies, a state can demonstrate through economic modeling that its carbon tax program is likely to achieve the emissions standard set by EPA. In fact, compliance using a carbon tax will most likely be easier to demonstrate than for many of the policies EPA suggests in developing its proposed emissions guidelines. And just as for other approaches, if the carbon tax for some reason failed to produce the forecast emissions rates, a state can increase the stringency of its policies, for example, by increasing the tax trajectory.
A carbon tax is a simple approach to using market forces to reduce emissions. In contrast to cap-and-trade, under a tax approach states would not have to allocate allowances, administer auctions, create an allowance registry, monitor trades and positions, or enforce a price floor. A carbon tax can also be implemented by a single state agency, rather than a complex amalgam of air, energy, and other regulatory bodies, as envisioned in the EPA proposal. To ensure compliance with a rate-based standard, all a state needs to do is monitor fossil fuel use and collect the money. In addition, states can easily expand their tax base when and if EPA regulates additional source categories under section 111(d).
Consistent with the language of the Clean Air Act, with EPA’s public commitments and with Supreme Court precedent, states should be free to comply with section 111(d) rules using policies of their own choosing, so long as the policies achieve the goals set by EPA. Taxing carbon is an effective, simple, pragmatic, and cost-effective approach to reducing CO2 emissions. EPA should allow states to use carbon taxes as a compliance strategy. We explain in our comments to EPA that although the EPA proposal as written will allow states that wish to adopt a carbon excise tax to do so, it inadvertently precludes the simplest option of imposing the tax liability on electric generating units themselves. Instead, the proposal’s expanded definition of "emission standard" would require states to place the statutory incidence of the tax only on entities upstream or downstream from regulated sources.
EPA could remove this barrier in two ways:
EPA could change the section 111(d) implementing regulations to clarify that any measure that is enforceable and reduces emissions qualifies as an “emission standard.” Current regulations allow only for rate based standards, allowance systems, or equipment specifications; or
EPA could revise the expanded definition of "emission standard" in the proposed rule to include reductions in emissions caused by regulation of any affected entity, including electricity generating units themselves. The current proposal somewhat paradoxically credits any effective policy that applies to entities other than power plants but only a limited set of options that apply directly to the plants themselves.
These changes are straightforward, but the first is more broadly applicable. It would apply not only to power plants and carbon dioxide but also to other source categories and pollutants, thus assuring states that the flexibility EPA is offering for power plants will extend to other source categories in future emissions guidelines. EPA has said that it wants to give states maximum flexibility in meeting targets in the Clean Power Plan. We suggest simple ways for EPA to provide states full flexibility, including the option of using a carbon tax to achieve required reductions if their local circumstances and politics support that approach.
Entrepreneur and Author of With Liberty and Dividends for All
EPA should allow states to achieve their carbon intensity targets through an upstream cap-and-permit system, with all first sellers of carbon or carbon-based electricity required to buy permits, all permits auctioned and all auction revenue used for per capita dividends to the state’s legal residents (people, including children, who are legal US residents, have Social Security accounts, and have lived in the state for a year or more). The states ideally could coordinate their systems with other states in their region, as in the northeast Regional Greenhouse Gas Initiative.
Each year the number of permits would decline until the state’s or region’s target is met. Only actual fossil fuel or electricity sellers could acquire permits (that is, no traders or speculators), and no offsets or other forms of leakage would be allowed. The efficacy of such a cap on first sellers lies in the fact that, if carbon doesn’t come into a region, it can’t go out. EPA would be able to track the effect of the cap by the number of permits sold.
Distribution of the dividends should not be made by utilities as a credit against energy bills, as is currently the practice in California. While a utility credit system is better than no rebates at all, a direct payment system would be much better. For example, equal dividends could be wired electronically to every resident’s bank account or debit card. This could be done through the Social Security Administration or a state agency using the Internet and/or banks to enroll eligible recipients. This is the approach that Alaska has successfully used to distribute its Permanent Fund dividends.
The reason for distributing dividends via direct payment rather than via utilities is that credits against utility bills mask the higher cost of energy and thus reduce the signal to conserve. Utility credits also allocate payments per meter rather than per capita, and don’t effectively remind state residents that they are receiving dividends that will rise along with energy prices. Such public awareness is essential for maintaining political support for phasing out carbon over the decades it will take to do so.
An upstream cap-and-permit system as described above is preferable to a carbon tax or fee without an upstream cap—even if the fee is accompanied by dividends—because it sets the quantity of carbon that can be burned rather than the price. There is no way to know in advance the effect of a carbon price on the quantity of carbon burned; innumerable other variables are involved. On the other hand, with a leak-proof upstream cap, it is possible to foresee and control the future quantities of carbon burned, regardless of rises and dips in carbon prices.
And, an upstream carbon cap would not be difficult to administer. It would apply not to emitters but to suppliers—the same companies a carbon tax would apply to, the difference being that the prices paid by the companies would be set by auctions rather than government. California and the Regional Greenhouse Gas Initiative have shown that such auctions can be conducted fairly and efficiently. If some predictability in prices is desired, a floor price on permits could be included.
As for the argument that permit revenue could be more effectively spent by state governments on conservation programs than by individuals who receive dividends, the fact is that it is the cap itself that reduces emissions, not the spending of revenue on particular programs. Such spending may shift reductions under the cap from one sector of the economy to another, but they are unlikely to increase reductions above the level set by the cap.
Roberton Williams III, Senior Fellow and Director of Academic Programs, Resources for the Future
EPA should provide states with the option of using a carbon tax to comply with the Clean Power Plan, but it needs to be careful about how that option is implemented. Allowing a carbon tax fits well with the general approach of giving states a variety of compliance options; each state can pick the approach that works best for its individual needs. And carbon taxes have several important advantages over other compliance options. But the translation from emissions rate–based goals to tax rates is important and needs to be structured to limit states’ incentives to pass other policies that undermine the effectiveness of the carbon tax.
Economists have long recognized the close similarities between a carbon tax and a cap-and-trade system—both put a price on carbon while offering flexibility in how to reduce emissions. EPA’s Clean Power Plan allows cap and trade as a compliance option, so it is natural to allow carbon taxes as well.
Moreover, a carbon tax has a number of substantial advantages. It provides a stable and predictable carbon price, which reduces uncertainty, making decisions about emissions-reducing investments simpler. That stable price also avoids any potential for market manipulation. Carbon taxes are also easier to administer than an emissions trading system and they could accomplish the added benefit of providing revenue for state budgets. Coordinating carbon tax rates across states is also simpler than setting up a multi-state trading system. Of course, carbon taxes have disadvantages too, and won’t be right for every state—but the advantages are sufficient to justify making the option available.
However, EPA needs to be careful about how that option is implemented. A key question is how to translate emissions rate–based state goals into a tax rate. In most respects, this is relatively straightforward. It requires careful modeling to estimate what tax rate is equivalent to any given emissions rate, but the same is true for converting between rate-based and mass-based goals. The models won’t always be right, of course, but they don’t need to be; as long as they’re correct on average (that is, unbiased), the carbon tax will have the same long-run effect on emissions as the equivalent rate-based approach. To guarantee unbiased modeling, EPA will either need to do that modeling itself or impose clear standards for how it needs to be done—but again, the same issue arises for converting to mass-based goals.
An important difference, though, is that a state that doesn’t want to regulate carbon could take actions to undermine the effectiveness of a carbon tax—something that would be far more difficult under a cap-and-trade or traditional regulatory system. For example, suppose such a state chose to impose a carbon tax, but then used the tax revenue to subsidize fossil fuels (or a less-obvious policy with a similar effect), thus undoing the effect of the carbon tax. Economists refer to this as "fiscal cushioning." To prevent this, modeling to determine the tax-rate target will need to take into account any other state policies that have a substantial effect on carbon emissions—and might need to be updated if those other policies change.