Kevin A. Hassett
Current federal energy tax policy is premised in large part on a desire to achieve energy independence by promoting domestic fossil fuel production. This, we argue, is a mistake. The policy also relies heavily on energy subsidies, most of which are socially wasteful, inefficient, and driven by political rather than energy considerations. Finally, the energy taxes that are in place could be more precisely targeted to specific market failures, and these higher taxes themselves would encourage the production of alternatives more efficiently than do current subsidies.
It is widely held that the United States must reduce its reliance on foreign oil. The concern over U.S. vulnerability to the disruption of supply by the Organization of the Petroleum Exporting Countries (OPEC) is understandable, given the fact that the United States imports over 60% of the oil it consumes each year. Of the oil that the United States imports, 40% comes from OPEC countries and nearly half of that from the Persian Gulf region. Many Americans are also concerned that oil monies help countries such as Iran pursue activities that are contrary to U.S. foreign policy.
A twenty-first-century U.S. energy tax policy would include an end to energy supply subsidies, a green tax swap, an end to the gas guzzler tax loophole, the possible use of "feebates," and conservation incentive programs.
As a response to these concerns, current tax policy promotes domestic oil and gas production in a variety of ways. The federal government provides a production tax credit for "nonconventional oil" (essentially a subsidy for coalbed methane), generous depreciation allowances for intangible expenses associated with drilling, and generous percentage depletion allowances for oil and gas. In addition, the Bush administration has consistently lobbied to allow additional drilling on the Alaskan North Slope.
This supply response ignores a fundamental fact: Oil is essentially a generic commodity priced on world markets. Even if the United States were to produce all the oil it consumes, it would still be vulnerable to oil price fluctuations. A supply reduction by any major producer would raise the price of domestic oil just as readily as it raises the price of imported oil. In addition, if the United States reduces its demand for oil from countries such as Iran, it has little effect on Iran, because that country can just sell oil to other countries at the prevailing world price. Indeed, this effect has been made abundantly clear by historical experience. The United States has cut its dependence on Iranian oil to zero, buying no oil directly from that nation since 1991. Despite the U.S. import ban, Iran was the world's fourth-largest net oil exporter in 2005.
A policy of energy independence that depends on boosting domestic oil and gas supplies through subsidies has several defects. First, subsidies reduce production costs and so do nothing to discourage oil consumption. Second, the policy encourages the consumption of high-cost domestic oil in place of low-cost foreign oil. A policy to encourage the United States to use up domestic reserves and thus become increasingly vulnerable in the future to foreign supply dislocations seems especially peculiar to us. Third, it is expensive. The five-year cost simply for the incentives mentioned above totals nearly $10 billion, according to the most recent administration budget submission.
Assuming that reliance on oil is unattractive, a clear sign that policy is headed in the wrong direction is the high and even recently increasing dependence on oil of the U.S. economy. Petroleum accounted for nearly 48% of primary energy consumption in the United States in 1977. Since this peak, it fell to a low of 38% in 1995 before inching up to just over 40% in 2005. Even going back to 1977, the 16% drop in the oil share from its peak to 2005 falls far short of the percentage reduction in oil share of other developed countries. The United Kingdom, for example, has reduced its oil share from a peak of 50% to just under 36%. France has reduced its oil share by 48% and Germany by 22%. In Asia, Japan has reduced its oil share by 39%, and even China has reduced its oil share by more than has the United States, with a 26% reduction. Current U.S. policies are leaving it increasingly vulnerable relative to other major oil-consuming nations.
One might argue that because the United States is such a large producer of petroleum products--the third-largest supplier behind Russia and Saudi Arabia-- domestic supply incentives in the United States help reduce the world price of oil. U.S. efforts, however, are but a drop in the bucket. One of us has estimated that the domestic oil production incentives in the U.S. tax code have lowered world oil prices by less than one-half of 1%.
To summarize, energy independence as popularly construed has little economic content. If reliance on oil is a problem, then supply subsidies make little sense, as they just encourage additional reliance on oil.
The single largest energy tax expenditure in the U.S. budget is the tax credit for alcohol fuels, with a five-year revenue cost of $12.7 billion. The 51-cent-per-gallon credit primarily benefits corn-based ethanol. The fundamentally political motivation for the subsidies to corn-based ethanol are apparent when one realizes that the United States levies a 54-cent-per-gallon tariff on imported ethanol. There is also debate in the scientific literature about whether ethanol takes more energy to produce than it contains. Even making an optimistic read of the literature, corn-based ethanol is expensive and provides little new energy to the economy. A study by Jason Hill and his colleagues at the University of Minnesota indicates that shifting the entire current corn crop to ethanol production would replace just 12% of U.S. gasoline consumption. This shift would reduce greenhouse gas emissions by less than 3%.
In addition to the ethanol subsidy, the federal tax code provides investment tax credits for solar and geothermal power production and advanced coal-burning power plants under section 48 of the tax code. Our recent research shows that the 20% investment tax credit for new integrated gasification-combined-cycle coal plants makes this technology cost-competitive with new pulverized coal plants. The subsidy for solar-generated electricity, however, is not large enough to make solar energy cost-competitive with natural gas or with other shoulder or peaking power plants.
Section 45 of the tax code provides production tax credits for wind power, biomass, and other renewable power sources. The tax credit is currently 1.9 cents per kilowatt hour (kWh). The section 45 and 48 tax credits are the second-largest energy tax expenditure, with a five-year cost of over $4 billion. The production tax credit for wind and biomass makes these two power sources cost-competitive with natural gas. The problem with production tax credits is that they must be financed somehow, either with reduced federal spending elsewhere in the budget or with higher taxes. Presumably, the credits are in place to encourage non-fossil fuel electricity production. The credit, however, distorts behaviors among non-fossil fuel power sources.
A better approach on both of these counts would be to levy a tax on the power sources that one wishes to discourage. If, for example, the concern is carbon emissions, then a carbon tax is an appropriate response. A tax of $12 per metric ton of carbon dioxide in lieu of production tax credits for wind and biomass would make these renewable sources competitive with natural gas. Unlike the subsidies, however, the tax would raise revenue, which could finance reductions in other distortionary taxes. Additionally, whereas subsidies lower the costs of electricity for consumers, increasing the quantity of energy consumed, taxes lead to decreased consumption. In units perhaps more familiar to most readers, a carbon tax of this magnitude would raise the price of gasoline by 10 cents if it were fully passed on to consumers.
Other production tax credits in the tax code include a production tax credit for electricity produced at nuclear power plants (section 45J). Qualifying plants are eligible for a 1.8cent-per-kWh production tax credit, up to an annual limit of $125 million per 1,000 megawatts of installed capacity for eight years. This limit will be binding for a nuclear power plant with a capacity factor of 80% or higher, thereby converting this into a lump-sum subsidy for new nuclear power plant construction.
U.S. subsidies discourage conservation and promote the consumption of inefficient sources of energy, a result that is irreconcilable with the goals of any rational energy policy. Alternative energy subsidies that are currently in place play political favorites and would be unnecessary if the types of energy that policymakers view as undesirable were taxed at an efficient rate. With undesirable forms of energy more costly, the market, rather than government officials, would determine which alternatives are best.
Redesigning Energy Taxes
First, we note that the literature suggests that U.S. energy tax rates may well be too low. Taking into account accident externalities, congestion, and unpriced pollution, one recent paper by Ian Parry and Kenneth Small finds that the optimal gasoline tax in the United States is $1.00 per gallon, over twice the current rate, taking into account federal and state motor vehicle fuel taxes.
Second, the sole tax policy to discourage low-mileage automobiles, the gas guzzler tax, contains a loophole large enough to drive a sport utility vehicle (SUV) through. The gas guzzler tax is levied on automobiles that obtain fewer than 22 miles per gallon and explicitly excludes SUVs, minivans, and pickup trucks. This excluded class of vehicles represented 54% of new vehicle sales in 2004. The light truck category (comprising SUVs, minivans, and pickup trucks) is the fastest-growing segment of the new vehicle market, growing at an annual rate of 5.5% between 1990 and 2004. In contrast, new car sales are falling at an annual rate of 1.6%. Unofficial congressional estimates suggest that phasing out this loophole over four years would raise roughly $700 million annually once the phaseout was complete. Optimal tax policy does not support treating similar assets differently, and current policy introduces a significant distortion that could easily be fixed.
A 21st-century U.S. energy tax policy would include an end to energy supply subsidies, a green tax swap, an end to the gas guzzler tax loophole, the possible use of "feebates," and conservation incentive programs. Ending subsidies to fossil fuel production would level the playing field among energy sources and shift the country from a policy of promoting fossil fuel supply to encouraging a reduction in fossil fuel consumption. In addition, it would move the United States away from the reliance on inefficient corn-based ethanol.
The United States should also implement a green tax swap. A green tax swap is the implementation of environmentally motivated taxes, with the revenues used to lower other taxes in a revenue-neutral reform. For example, Congress could reduce reliance on oil and other polluting sources of energy through the implementation of a carbon tax. The revenues could be used to finance corporate tax reform or to finance reductions in the payroll tax. Consider a tax of $15 per metric ton of carbon dioxide. Focusing only on carbon and assuming a short-term reduction in carbon emissions of 10% in response to the tax, a $15-per-ton tax rate would collect nearly $80 billion a year, a number that represents 28% of all corporate taxes collected in the United States in 2005. If carbon tax were fully passed on into consumer prices, it would raise the price of gasoline by 13 cents per gallon, the cost of electricity generated by natural gas by 0.6 cents per kWh, and the cost of electricity generated by coal by 1.4 cents per kWh.
We note that a carbon tax is preferable to a carbon cap-and-trade system, as is currently implemented in Europe. Although a carbon charge and a cap-and-trade system could be designed to bring about the same reduction in carbon emissions in a world with no uncertainty over marginal abatement costs, the instruments are not equivalent in a world with uncertainty.
Given the uncertainties with respect to the introduction of new technologies to reduce carbon emissions, tax and permit systems can have very different efficiency costs. A number of researchers, including Richard G. Newell of Resources for the Future and Richard Pizer of Duke University, note that quantity restrictions such as cap-and-trade are appropriate only when either atmospheric pollutants are short-lived or the marginal costs of additional pollution above a threshold are extremely steep. Otherwise, price controls, like carbon taxes, are likely to be more efficient.
Because global warming depends on the stock of carbon in the atmosphere rather than on emissions in any one year, the expected efficiency costs of a carbon charge policy are likely to be much lower than the costs of a carbon cap-and-trade system. Essentially, the marginal damages from emissions in any given year have a roughly constant marginal damage so long as we are not at or near a threshold. Setting a price through a tax ensures that we avoid the risk of permit prices diverging dramatically from the marginal damages and thereby creating a large economic loss to society.
Moreover, although a cap-and-trade system could be designed in which the carbon permits are sold rather than given away, experience to date suggests that they will be given away. In that case, governments give up substantial revenue from cap-and-trade systems with which they could lower other distortionary taxes and limit the economic harm from the environmental policy. In a related vein, cap-and-trade systems generate substantial rent-seeking behavior, as firms lobby for grandfathering and generous allowances of permits once a program is put in place. Although firms are likely to lobby over the specific carbon charge rate and possibly coverage of the tax, a carbon charge is not conducive to lobbying over allocations, unlike permit systems.
A common criticism of carbon taxes is that they do not provide any certainty of emission reductions. A carbon tax provides certainty over the price of emissions but no certainty over emissions; a cap-and-trade system provides certainty over emissions but no certainty over the marginal cost of those emissions. (Note that this certainty over emissions is lost if a safety valve is incorporated in the cap-and-trade system.)
What we ultimately care about, however, are the economic and ecological consequences of higher concentrations of greenhouse gases in the atmosphere resulting from global emissions. Global climate models are impressively sophisticated, reflecting the enormous complexity of the climate system. Our understanding of the climate system is improving with ongoing research, and one result is that our sense of the emission reductions that will be required to stabilize the planet's temperature and prevent large economic and ecological losses is also evolving. To give primacy to specific emission reductions regardless of the cost is to suggest a greater certainty in the climate science than currently exists, and implicitly but implausibly makes controlling emissions the top policy priority, trumping all others.
This is not an argument for policy delay. Given the long lags between emissions and climatic response, it would be imprudent to wait for greater precision in the climate science before taking action. But we should not act as if we know the precise level of emissions reductions to undertake. Instead, we should balance reductions against the economic cost of achieving those reductions as represented by the marginal cost of abatement. A tax does this automatically because profit-maximizing firms will operate at the point where marginal abatement costs equal the tax rate. With a clear and unambiguous schedule of carbon tax rates over time, businesses and households can rationally plan to reduce their carbon footprint through their capital purchase decisions as well as through their use of current capital.
An additional concern magnifies the advantage of carbon taxes. Carbon emissions are a global problem; the externalities from Chinese emissions hurt the United States just as much as emissions from the United States. Curbing carbon emissions requires an international solution. Cap-and-trade policies provide a serious moral hazard problem for governments of the developing world. Can we really expect developing countries to honestly police themselves, especially when quota violations would boost local economies? On the other hand, a carbon tax provides its own incentive for a government to closely police polluters. Governments, after all, are committed and competent revenue collectors.
Next, Congress should eliminate the gas guzzler tax loophole for light trucks. Congress might also consider strengthening the gas guzzler tax by shifting to a "feebate" approach, whereby low-mileage vehicles are taxed at increasing rates, as under the current gas guzzler tax, and fuel-efficient vehicles receive a tax subsidy. This could be structured to be revenue-neutral if desired.
Our final energy tax proposal is to increase the conservation investment incentives that were recently introduced in the Energy Policy Act of 2005. In a study of energy conservation incentives contained in the Energy Tax Act of 1978, we found that the tax credit was much more successful at raising investment levels than was a comparable energy price increase. We speculated that the credit program may have publicity effects that spur investment that the energy price increase does not have. In addition, uncertainty over the permanence of future energy price increases makes the certainty of the tax credit at purchase more valuable. A conservation credit that is technologically neutral would be a worthy accompaniment of a higher tax on carbon-based fuels if reducing reliance on these forms of energy is a policy objective.
The policies we advocate shift the United States away from fossil fuels and toward renewable energy. They also reduce the cost to federal taxpayers, while aligning private and social interests. This is the making of a 21st-century energy policy.
Kevin A. Hassett is a senior fellow and director of economic policy studies at AEI. Gilbert E. Metcalf is a professor of economics at Tufts University and a research associate at the National Bureau of Economic Research.