Nearly all economists agree that the most efficient way to address environmental problems is to raise the cost of the pollution-generating activity. Whether one uses a tax as famously suggested by Pigou (1938) or a quantity constraint such as a cap-and-trade system as proposed by Dales (1968) and others, the point is to "internalise the externality" by raising the cost of pollution to the firm or individual so that they have the appropriate incentives to engage in the socially optimal level of this activity. Weitzman (1974) provides a framework for choosing between price and quantity approaches when benefits and costs of abatement are uncertain.
All of these approaches assume that policymakers will address externalities by making them more costly. The US energy policy has taken another approach. Rather than making energy sources that contribute to pollution more costly, US policies have typically subsidised clean-energy alternatives. Until the US enacts a national cap-and-trade system or a carbon tax, tax-based subsidies will provide some of the most substantial incentives to reduce energy-related greenhouse gas emissions.
In Metcalf (2009), I review the current set of tax-based subsidies for low-carbon energy and critique the subsidy-based approach to energy policy. The single largest subsidy is the 45¢ per gallon excise tax credit for ethanol. But in addition to this credit, there are production and investment tax credits for renewable electricity production, for clean-fuel burning vehicles, as well as credits and subsidies for energy efficiency investments in homes and workplaces. Many of these subsidies were recently renewed and extended in the American Recovery and Reinvestment Act of 2009. Even if a national cap-and-trade bill such as the American Clean Energy and Security Act bill (HR 2454), recently reported out of the House Energy and Commerce Committee, were enacted, it is likely that many of these subsidies to clean energy will remain in the tax system.
Subsidies for clean energy are appealing but they suffer from four problems:
- They lower the cost of energy,
- play favourites with technologies,
- are often inframarginal, and
- often interact in unexpected ways with other policies.
Problem 1: Subsidies lower the cost of energy consumption
Tax-based subsides achieve the important goal of adjusting relative prices of polluting and non-polluting energy sources in the right direction. If fuel source X causes pollution that is equal to 10 percent of its cost then we can provide the right incentive to fuel users choosing between fuel sources X and Y by raising the price of X by 10% or by lowering the cost of fuel source Y by 1/(1.10) or 9.1%. Either way, the relative cost of fuel source X to Y is now 10% higher than it was prior to the implementation of new energy policy. Either a tax or a subsidy can be effective on the margin of choosing among fuel sources where some sources cause pollution.
This creates a problem, however, on a different margin. Efficiency requires that consumers make decisions taking into account the full cost of using commodities – including the pollution costs associated with using energy. Raising the cost of the polluting fuel source X raises the overall cost of energy use and encourages a reduction in energy consumption. Subsidising the clean substitute undermines this consumer substitution effect as it leads to a lower cost of energy overall. Consumers do not reduce energy consumption as much as they would under a cost-raising policy.
Problem 2: Subsidies favour particular technologies
Subsidies tend to favour particular technologies. Consider the tax credit for hybrid vehicles put in place in the Energy Policy Act of 2005. The credit ranges from zero to $3,000 per vehicle depending on whether the vehicle meets the specific hybrid criteria and on how many vehicles have been sold. The credit phases out as the vehicle hits certain sales targets over time. Table 1 shows the subsidy cost per gallon of gasoline saved through this credit for a number of model 2009 vehicles.
Table 1. Hybrid vehicle tax credit, 2009 model values
Value of Credit
|Tax Credit per
|Chysler Aspen Hybrid
|Ford Escape Hybrid (2WD)
|Mazda Tribute Hybrid (2WD)
|Nissan Altima Hybrid
|Source: Metcalf (2009). Author's calculations of savings relative to a vehicle that gets 20 miles per gallon and is driven 12,485 miles per year. Vehicles are assumed to be driven for ten years and savings are annualised wtih a 10% discount rate.
The table illustrates several points. First, the tax credit per gallon of gasoline saved varies from zero to over $11 per gallon. Second, certain hybrid vehicles that get high mileage are excluded from the credit because they have been successful in the market place. Third, certain high-mileage vehicles are excluded from the subsidy because they do not use specified technology. Note that the Corolla gets nearly the same mileage as the Tribute Hybrid. This is the most egregious violation of technological neutrality. The tax credit provides no incentive to tinker with the internal combustion engine to achieve increases in vehicle efficiency despite the many opportunities that exist to make the internal combustion engine more efficient. Our tax policy should provide the same incentives to improve mileage regardless of the technology put in place.
Problem 3: Subsidies are wasteful
Ideally we want subsidies targeted to marginal investments. But often they end up supporting inframarginal investments, i.e. investments that would have taken place in the absence of the policy. A good example of this is the $0.50 per gallon alternative fuels mixture credit. This credit is intended to encourage the addition of biodiesel and other biomass-based fuels to petroleum to reduce petroleum use. Recently it has emerged that many paper firms are taking the credit for mixing diesel fuel with black liquor, a biomass by-product of paper making that historically has been used by the industry as a fuel source for their boilers. Controversy has arisen over whether paper firms are adding diesel fuel to black liquor purely for the purpose of claiming the tax credit biodiesel mixture tax credit (Mouawad and Krauss 2009).
This is troubling on two levels. First, it may be very inefficient if credits are being provided for inframarginal activities. This is a common problem with any subsidy. We want to provide the incentive to firms that would not have undertaken the desirable activity in the absence of the subsidy. But we don't want to provide the subsidy to firms that would have undertaken the activity regardless of the subsidy. But the example from the paper industry is troubling beyond the inframarginal nature of the subsidy. If the tax credit is raising the demand for diesel fuel in order to make the biofuel eligible for the credit, then it is having the perverse effect of raising rather than lowering demand for petroleum products.
Problem 4: Subsidies interact in unexpected ways with other policies
Subsidies can be rendered ineffective or more expensive through their interaction with other policies. A simple example here is the interaction of the hybrid vehicle tax credit and the Corporate Average Fuel Economy (CAFE) standards. Allowing tax credits for hybrids encourages the production and purchase of high-mileage vehicles. But CAFE sets minimum fleet mileage standards for automakers. Producing more hybrid vehicles relaxes the CAFE mileage constraint for automakers and allows them to sell more low mileage vehicles. The result is that federal tax collections are reduced with no improvement in automobile mileage.
An alternative to subsidies
Most, if not all, of these problems would disappear if we replaced the current system of tax subsidies for carbon-free technologies with a market-based system to set positive and gradually increasing carbon prices. It looks like we're on track to set the carbon price, either through a cap-and-trade system or some form of carbon charge. But insufficient attention is being paid to removing the various tax subsidies currently in place.
Dales, J.H. Pollution, Property, and Prices. Toronto: University of Toronto Press, 1968.
Metcalf, Gilbert E. "Tax Policies for Low-Carbon Technologies," Cambridge, MA: National Bureau of Economic Research Working Paper No. 15054, 2009.
Mouawad, Jad and Krauss, Clifford. "Lawmakers May Limit Paper Mills' Windfall," New York Times. New York, 2009.
Pigou, Arthur C. The Economics of Welfare. London: Weidenfeld and Nicolson, 1938.
Weitzman, Martin. "Prices Vs. Quantities." Review of Economic Studies, 41 No. 4(1974): 477-91.