Environmental Federalism

Nearly all major federal environmental legislation divides the responsibility for controlling pollution between the federal government and the states. The exact division of responsibilities varies substantially from statute to statute.

Consider the Clean Air Act. It requires the federal government to set air quality standards for each of six pollutants and to establish performance standards for new stationary sources of pollution (such as factories and utilities) and mobile sources (such as automobiles). States are required to develop implementation plans, which must be approved by the Federal government, to reach these goals; states also share the responsibility for enforcement. In sharp contrast, under the Clean Water Act the individual states are directed to set water quality standards. We thus have uniform national standards for air quality but state-specific standards for water quality.

Which approach is better? Should we move toward greater centralization, thus giving more responsibility for controlling pollution to the federal government? Or should we encourage further decentralization and allow state and local governments a greater voice in environmental policy?

This question is part of a much broader issue. The debate over the proper division of governmental responsibilities is long-standing. It was a key issue in the framing of the Constitution in the 1780s; it lies at the heart of the debate over the Reagan administration's call for a "new federalism" in the 1980s. Many of the questions about environmental federalism should be familiar to those who have thought about other issues in federalism.

Decentralized policy making

The case in favor of decentralized decision making is simple, yet powerful and persuasive. For the moment, let's set aside the question of who should set policy and look at the principles we would like to see embedded in policy making decisions. "Optimal" environmental policy—from an economic standpoint, at least—requires us to continue to reduce pollution in all parts of the country up to the point that the cost of further reducing pollution (what is known as marginal abatement cost) equals the benefits from that reduction (that is, marginal social damage). Focusing only on the economics of the issue, if it costs society $100 to reduce emissions by one additional ton but the benefits from doing so were valued at $200, we would be wise to cut back on pollution further; if, on the other hand, benefits were valued at only $50, we have set environmental standards too high. If we continue with this line of argument, the logic behind the economic principle that the optimal standard is such that marginal abatement cost equals marginal social damage becomes clear.

While it is true that we should apply the same principle in every region, we should not set the same environmental standard everywhere unless the costs and benefits of pollution control are the same in each region. There are excellent reasons to believe that this is not the case. For example, different industries are clustered in different regions, and it is much more costly for firms in some industries to reduce emissions than others. Thus differences in regional industry mix imply differences in regional abatement costs. Differences in meteorology, topography, and land-use patterns also play important roles in abatement costs. Few would disagree with the proposition that it will be much more costly for Los Angeles to reach any given level of air quality than most other cities; by one estimate, the Los Angeles area would have to reduce gasoline consumption by 82 percent in order to meet the air quality standards set under the Clean Air Act.

Similar considerations arise on the benefit side of the equation. Benefits, after all, are to some extent subjective. It could well be the case that the residents of some states would be willing to pay $200 for a slightly cleaner environment, but that residents of others would be willing to pay only $50. Population size also plays a role. Everyone in an area suffers the damages pollution causes. Therefore, everything else being equal, the greater the population in an area, the greater the benefit from a cleaner environment.

Ongoing research sponsored by RFF and funded by the National Science Foundation points to the validity of this argument. Paul R. Portney, Albert McGartland, and Wallace E. Oates have assembled data on damages and abatement costs for total suspended particulates (TSP) for Baltimore and St. Louis. Their preliminary estimate is that the optimal standard for TSP concentration for Baltimore is nearly 50 percent less stringent than the optimal standard for St. Louis. That is, if we were to equate the costs and benefits of reducing TSP in both cities, the permissible level of pollution would need to be nearly twice as high in Baltimore.

How should we divide the responsibility for setting environmental policy in order to reach this optimal outcome (again recalling that we are only focusing on economic issues)? Federal legislation is rarely sensitive to regional differences in costs and benefits. Typically, federal legislation implies uniform standards; for example, under the Clean Air Act, the federal government sets uniform maximum allowable levels of pollution. Therefore, vesting the federal government with the authority to set environmental policy is unlikely to lead to an optimal outcome.

This line of reasoning should sound familiar to anyone who has been involved in the debate over fiscal federalism. It would make little sense to provide the same menu of public services in every community, or for the federal government to try to tailor policies to meet the local circumstances in each community. Thus we allow state and local governments to decide how much to spend on education, when trash is to be collected, and which beats police officers will patrol. The logic leading to this conclusion also suggests that state and local governments may be in a better position than the federal government to choose the correct level of environmental quality.

Possible problems

Many people would accept the basic argument in favor of decentralization, but would continue to urge a strong role for the federal government in environmental matters nonetheless. They would contend that while the outcome under federal uniform standards is unlikely to be optimal, it is better than the outcome that would emerge if we were to allow state and local governments to set environmental policy.

Though it might not be immediately obvious, most of the arguments against decentralization turn on a single principle that has a strong basis in economic logic. If there are costs or benefits associated with the environmental policies adopted by one region that are borne by the residents of other regions, then it is unlikely that decentralized decision making will lead to an optimal outcome. We can apply this principle in a range of circumstances.

The clearest cases are those where pollution generated in one state crosses political boundaries into another. For instance, if Illinois firms are forced to reduce pollution, we would expect the citizens of Illinois (and, hopefully, Illinois policy makers) to take into account the benefits from a cleaner environment that Illinois citizens would realize. Thus, Illinois would require its firms to reduce pollution to the point that marginal abatement cost equals marginal damage suffered by Illinois residents. Under such a decentralized approach, we would not expect Illinois citizens to necessarily take into account the benefits Indiana residents receive. An optimal policy would require Illinoisans to go further and continue to reduce pollution until marginal abatement cost equaled the sum of damages to both Illinois and Indiana residents.

Clearly, in many cases this argument against decentralization is quite persuasive. The problems in coordinating efforts by Delaware, Virginia, Maryland, the District of Columbia, and Pennsylvania to clean up the Chesapeake Bay are well documented (and the Chesapeake is often offered as one of the more successful cooperative undertakings). In some cases, whether or not a pollutant is local is a policy issue; damages from sulfur dioxide emissions can be contained in a small area or dispersed over a much larger area if firms are required to use tall stacks. Some pollutants are international, acid rain being a good example, as the strained relations between Canada and the U.S. can attest. Some are truly global, carbon dioxide's impact on the climate (often called the greenhouse effect) and the impact of chlorofluorocarbons on the ozone shield being extreme examples of environmental issues that cannot be addressed by states or even nations. In such cases, the benefits from controlling pollution extend well beyond traditional political borders.

Many pollutants, however, cause damages over a much more limited area. Lead, carbon monoxide, and TSP (total suspended particulates) are good examples. In many cases, the control of hazardous waste sites and the establishment of drinking water standards also fall into this category.

Similar issues arise repeatedly when assessing fiscal federalism. Suppose the residents of Community A build parks which are used not only by residents of A but by those of Community B as well (an example of a "fiscal spillover"). Under these conditions, we would expect that Community A had built too few parks. The standard policy solution in this case would be a grant from the federal government to Community A to encourage it to build additional parks.

We can analyze a second argument against decentralization in this framework. People throughout the country derive benefits from the quality of the environment in unique national sites such as national parks and wilderness areas; the residents of New York care about clean air in Yellowstone, as do the residents of Wyoming and Idaho. In many ways this issue is identical to the interjurisdictional pollution problem discussed above, though here the damages from pollution spill over state boundaries even though the pollution itself does not. Again, optimal policy almost certainly requires some federal role.

As a third argument, there are sometimes gains from uniformity and coordination. (Anyone who doubts this should try playing a tape made on a Beta video-cassette recorder on a VHS machine, or using a piece of software written for an Apple computer on an IBM.) In the context of environmental policy, it would be very expensive to demand that Detroit make cars that meet different standards in Montana, New Jersey, and so on. Clearly there are some benefits from uniform standards; automobiles are probably the most important example. But even in the case of automobiles, the benefits to be gained from a policy on uniformity do not rule out the possibility that the "correct" number of standards may be a happy medium between the current number—two (the Clean Air Act allows California to set its own standards for auto emissions)—and fifty, that is, one that applies to each state.

A fourth argument against allowing lower levels of government to set environmental policy focuses on the effects of interjurisdictional competition. The fear is that in their eagerness to attract new business and jobs, local officials would set environmental standards that are too lax. Thus, so this argument goes, the federal government must set environmental policy in order to "save the states from themselves." Such a stance is consistent with a more general view that nearly all forms of competition among jurisdictions are destructive. The Advisory Commission on Intergovernmental Relations (the focal point for interest in federalism issues within the federal government) for years argued that state and local governments should be encouraged to cooperate with one another.

In the last several years, however, there has emerged a strong sense that competition among governments can be beneficial, just as competition among firms can be beneficial. According to this view, competition forces local government to make decisions that are in the best interest of their constituents. Thus competition stops local officials from overspending; if they were to try to set public budgets that were too high, firms would vote with their feet and move to competing communities.

The following simple example captures the spirit of this argument in the context of environmental policy. Suppose Illinois were about to set a standard for some pollutant. Who would bear the costs associated with this new policy? Since Illinois competes against other states to attract and hold firms, clearly the burden cannot fall on Illinois firms; they have the option to move to another state. Therefore the cost of reducing pollution must fall on Illinois residents in the form of fewer jobs and lower wages.

Under these circumstances, what level of environmental quality should Illinois policymakers set? They might reason as follows. Illinois residents will pay the costs of reducing pollution. Therefore if Illinois policy makers act in the best interest of their constituents, they should set a standard such that the costs of reducing pollution further are just balanced by the benefits. If they set a less stringent standard, Illinois residents would be willing to sacrifice additional jobs and wages in return for a cleaner environment; if they set a more stringent standard, the benefits from less pollution would not be worth the cost. This is exactly what optimal environmental policy requires: marginal abatement cost should equal marginal social damage.

Undoubtedly the real world is much more complicated than the above situation. Firms cannot instantaneously move from Illinois to Hawaii; Illinois firms may have quite a bit of influence with Illinois lawmakers; the costs of pollution control will be concentrated among workers in certain industries while the benefits of a cleaner environment are diffuse; residents of other states will realize some of the benefits if Illinois firms generate less pollution. But this illustration does make an important point. Competition among jurisdictions implies that the costs of government would be borne largely by those who realize the benefits from those policies and will therefore encourage efficient public decisions.

Children and grandchildren

Finally, as some people have asked, if we vest state and local governments with the authority to set environmental policy, will they properly take into account the interests of future generations? Will communities be inclined to follow policies that yield benefits now (such as more jobs) but would result in substantial damages in the future (such as increased exposure to long-lived pollutants like hazardous wastes)? Problems of this kind would seem to be more acute at the state and local levels than at the federal level. We would presumably all want the federal government to take into account the interests of our children since it is likely that they will live somewhere in this country. But at least some of our children will live in communities elsewhere, and therefore their environmental heritage will depend on the decisions of others. Geographic mobility, so this argument goes, could result in myopic local decisions and suboptimal environmental quality for future generations.

This line of reasoning is not altogether persuasive. There is substantial evidence that many of the factors that determine the quality of life in a community are reflected in community property values—that is, people are willing to pay more to live in a community with good schools, a low crime rate, and a clean environment. Suppose a community were considering allowing a new firm to locate in that community, and suppose further that this firm would cause $100 worth of environmental damages in the future. If environmental quality is reflected in property values, the value of homes in this community would fall by (the present value of) $100 if the firm were allowed to enter. At least in this simple case, current generations could not escape the future cost and benefits of their environmental decisions. Here again, the example ignores some important issues; in particular, it assumes that future residents are aware of and can evaluate the damages from long-lived pollution. But it does suggest that some of the concerns about decentralized policymaking need to be reevalutated.

Sensible solutions

These considerations are not meant to suggest that we should abolish the U.S. Environmental Protection Agency and ask state and local governments alone to protect the environment. But I would argue that we should seriously consider moving some policymaking responsibility to lower levels of government, while still leaving the federal government with an important role in environmental issues. There are some strong arguments in favor of decentralization. An environmental policy that is correct in one region is unlikely to be correct in all. Federal regulation is rarely sensitive to these differences; instead, it often implies a single uniform policy in all regions.

It is important not to minimize the problems associated with this proposal. In particular, the spillover problem is difficult; the damages from some pollutants do not respect state boundaries. But sensible solutions may be available. Federal matching grants to state and local governments are often used to deal with the problem of fiscal spillovers. Perhaps similar policies (for instance, grants from the federal government to a state in return for an agreement to control pollution more tightly) would be effective in the interjurisdictional pollution problem. Further, we might envision a regulatory structure where states (and possibly lower levels of government) assume the responsibility for controlling some pollutants while the federal government retains the responsibility for others. It is not clear that we can solve all of the problems in environmental federalism, but the issues are sufficiently important that they deserve careful attention.