[An updated version of this article can be found at Pollution Controls in the 2nd edition.]
While there is general agreement that we must control pollution of our air, water, and land, various interest groups, public agencies, and experts have disputed just how we should control it. The pollution control mechanisms adopted in the United States have tended toward detailed regulation of technology. In 1970 popular concern about environmental degradation coalesced into a major political force, resulting in the creation of a federal Environmental Protection Agency (EPA) by President Nixon, and the first of the major federal attempts to regulate pollution directly—the Clean Air Act Amendments of 1970. Since then, the federal role in regulating pollution has grown immensely, unleashing a cascade of regulation upon the EPA, local governments, and the business community. But that has begun to change. Although the command-and-control approach is the norm, environmental lobbyists and legislators are beginning to consider market-based approaches to pollution control.
In virtually every antipollution law, Congress has instructed the EPA to establish and enforce specific pollution standards for individual polluters. These standards are generally based on some notion of the "best available" technology for each source of pollution in each industry. Because each pollutant has many sources, the EPA often must set literally hundreds of maximum discharge standards for any single pollutant.
Existing pollution sources (such as old factories) are generally required to meet less onerous standards than those applicable for new sources, largely because it is considered more costly to retrofit an old factory than to build pollution control devices into a new one. And even the definition of "new" requires further regulations because EPA must distinguish, for example, among rebuilding a fossil-fuel-fired boiler, replacing it, or replacing the entire facility of which the boiler is only a part. Complicating matters further, standards for existing sources and new sources are often stricter in regions with a higher-quality environment (i.e., cleaner air, cleaner water, and so on).
The Cost of Pollution Controls
The way that pollution controls are often built into the production process makes any estimation of their cost extremely difficult. In addition, pollution controls often discourage new investment and production, but no one currently calculates such indirect costs as the value of what is not produced. The federal government has, however, estimated a subset of costs, namely direct expenditures on pollution controls. These expenditures cost governments and private entities an estimated $100 billion in 1988 alone. Some $40 billion was spent on air-pollution abatement, $40 billion on water-pollution controls, and $20 billion for a variety of solid-waste, hazardous-waste, and other programs.
The most costly and complex federal pollution control policy is the motor-vehicle program. In order to enforce automobile standards set by Congress, the EPA must test each model line of new cars and test and random sample vehicles already on the road. The Clean Air Act requires that emission controls work for at least the first fifty thousand miles driven. Direct expenditures for compliance with these vehicle standards totaled an estimated $14 billion in 1988, costs shouldered primarily by consumers.
Among the programs funded by the federal government, two are especially costly. The larger of these is the Municipal Sewage Treatment Construction Grant program begun in 1973. Through this program, the federal government directly underwrote grants totaling over $43 billion by 1983 to pay for municipal sewage treatment plants.
The second program is more well-known. In 1980 Congress established Superfund to finance the cleaning of hazardous waste sites. This program required private entities responsible for hazardous dumps to clean them up. But if these parties could not be found, the cleanup would be funded by the government, through general revenues and a tax on petroleum feedstocks. In 1986 a new statute—the Superfund Amendments and Reauthorization Act—levied a federal tax on all corporations with taxable income over $2 million to help fund these remedial actions. This new tax is expected to generate about $8.5 billion over five years for waste cleanup. Thus, corporations that had nothing to do with old hazardous waste sites or that do not even generate toxic waste are required to pay for the pollution others left behind.
The Economic Effects of Pollution Controls
Pollution controls divert economic resources from other economic activities, thereby reducing the potential size of measured national output. As long as the increase in the value of the environment is at least one dollar for each additional dollar spent on controls, the total value of goods, services, and environmental amenities is not reduced. Unfortunately, that seldom happens, for at least three reasons.
First, the Congress or the EPA may decide to control the wrong substances or to control some discharges too strictly. Congress's own Office of Technology Assessment concluded, for example, that attempting to reach the EPA's goal for urban smog reduction could cost more than $13 billion per year, but result in less than $3.5 billion in improved health, agricultural, and amenity benefits.
Second, regulatory standards can result in very inefficient patterns of control. Some polluters may be forced to spend twenty-five thousand dollars per ton to control the discharge of a certain pollutant, while for others the cost is only five hundred dollars per ton. Obviously, shifting the burden away from the former polluter toward the latter would result in lower total control costs for society for any given level of pollution control.
Third, pollution controls can have deleterious effects on investment in two ways. First, by making certain goods—chemicals, paper, metals, motor vehicles—more expensive to produce in the United States, they raise the prices of these goods and thereby reduce the amount of each demanded. Second, because controls are generally more onerous for new sources than for older, existing ones, managers are more likely to keep an old plant in use rather than replace it with a new, more efficient facility, even though the new facility would produce the same goods as the old one.
The command-and-control approach is flawed in other ways, too. It does little to encourage compliance beyond what is mandated. Regulations are introduced only after noticeable damage has occurred. Polluters who manage to avoid legislative scrutiny continue to pollute. And regulations may be difficult to enforce.
Market-Based Approach to Pollution Control
Problems like these have led policymakers to look for more efficient means of cleaning up the environment. As a result, the 1990 Clean Air Act Amendments look very different from their predecessors of two decades earlier because they include market-based incentives to reduce pollution.
Market incentives are generally of two forms: pollution fees and so-called "marketable permits." Pollution fees are simply taxes on polluters that penalize them in proportion to the amount they discharge into an airshed, waterway, or local landfill. Such taxes are common in Europe but have not been used in the United States. Marketable permits are essentially transferable discharge licenses that polluters can buy and sell to meet the control levels set by regulatory authorities. These permits have been used in the United States because they do not impose large taxes on a small set of polluting industries, as would be the case with pollution fees.
The 1990 Clean Air Act allows the EPA to grant "emissions permits" for certain pollutants. These are, in effect, rights to pollute that can be traded among polluters. Imagine a giant bubble that encloses all existing sources of air pollution. Within that bubble some emitters may pollute over the control level as long as other polluters compensate by polluting less. The government or some other authority decides on the desired level of pollution and the initial distribution of pollution rights within an industry or for a geographic region—the "bubble" that encloses these sources. Purchases and sales of permits within the "bubble" should reduce the total level of pollution to the allowable limit at the lowest total cost.
For example, a St. Louis study found that the cost of reducing particulate emissions for a paper-products factory was $4 per ton, while the cost to a brewery was $600 per ton. The Clean Air Act could require St. Louis to reduce its emissions by a certain amount. Under the traditional approach, the brewery and the paper factory would each be required to cut emissions by, say, ten tons. The cost to the paper factory would be only $40, while the cost to the brewery would be $6,000. But with tradable permits, the brewery could pay the paper factory to cut emissions by twenty tons so that the brewery could continue to operate without reducing emissions at all. The net result is the same emission reduction of twenty tons as under the command-and-control approach, but the total cost to society of the reduction is only $80 instead of $6,040.
All this is not just speculative. A market for trading emissions permits was allowed by the EPA under the Carter administration in 1979. Said Douglas Costle, EPA chief at the time: "The bubble means less expensive pollution control, not less pollution control."
The tradable permits work. In 1981 General Electric had three months to meet the state of Kentucky's deadline for emissions control. To do it, GE paid $60,000 to International Harvester to lease several hundred tons of emissions reductions that International Harvester had "saved." Not only did GE meet the deadline, but it also saved $1.5 million in capital and $300,000 in operating costs. Up through 1984 bubbles approved by the EPA alone saved an estimated $300 million compared to what would have been spent to comply under traditional pollution controls. State-approved bubbles, like that used by GE, have saved millions more. Environmental economist Thomas H. Tietenberg estimates that marketable permits can reduce the cost of pollution control by as much as 75 percent. University of Maryland economist Wallace Oates estimates that a complete switch from command-and-control to marketable permits would reduce pollution control costs by at least one-third.
Marketable permits have also been used to phase down the use of chlorofluorocarbons in order to preserve the stratospheric ozone layer. This policy was instituted in 1990, and a number of trades had already taken place by mid-1991. Moreover, the Clean Air Act of 1990 includes a provision for allowing trading of pollution rights for sulfur oxides as part of a policy to reduce these emissions by nearly 50 percent by 2000. Allowing trading of these rights could make the cost of reducing sulfur dioxide as much as $4 billion per year less than the cost that would be required by the traditional pollution standards approach.
Protecting our environment does not have to put an end to economic progress. Free markets in permits to pollute, like free markets for other resources, can assure that pollution is controlled at the lowest cost possible.
Robert W. Crandall is a senior fellow at the Brookings Institution in Washington, D.C. He served as acting director of the Council on Wage and Price Stability during the Carter administration and was previously an associate professor of economics at MIT.
Hahn, Robert W., and Gordon L. Hester. "Where Did All the Markets Go?" Yale Journal on Regulation 6, no. 1 (Winter 1989): 109-53.
Hahn, Robert W., and Roger G. Noll. "Environmental Markets in the Year 2000." Journal of Risk and Uncertainty 3, no. 4 (1990): 351-67.
Tietenberg, Thomas H. Emissions Trading: An Exercise in Reforming Pollution Policy. 1985.
U.S. Department of Energy. Office of Environmental Analysis, Assistant Secretary for Environmental Safety and Health. A Compendium of Options for Government Policy to Encourage Private Sector Responses to Potential Climate Change. October 1989.