Nonpoint Pollution and National Responsibility

Throughout the twenty-year history of the federal Clean Water Act, most water pollution regulations have been directed against pollutants generated by so-called point sources; factories and municipal wastewater treatment plants, for example, are more-or-less single points that are readily identifiable targets for regulations. Dispersed, area-wide (or "nonpoint") sources like farms, forests, and city streets have been all but ignored, in part because they are so much harder to control. This malign neglect is beginning to end, and none too soon.

True, for many pollutants—cadmium and other heavy metals, for instance—nonpoint sources contribute only a minor proportion of the nation's water pollution problems. For many others, however, including phosphorus, nonpoint sources account for the vast majority of the discharges to the nation's waterways. As these facts become more widely known, members of Congress and others are realizing that the goals of the Clean Water Act are not attainable as long as nonpoint sources remain uncontrolled.

The Clean Water Act establishes a complex mix of federal and nonfederal responsibilities, with the proportions changing as the act has been amended over the years. In its original 1965 version, the federal role was confined to approving state plans and water quality standards. The federal government could impose both standards and plans should the states fail to act, but the clear implication of the 1965 legislative language was that Congress did not relish the federal exercise of such authority. However, the 1972 amendments to the act required permits for point-source discharges, based on federal effluent guidelines, and a huge municipal treatment construction grants program featuring a 75 percent federal subsidy. In the face of this overwhelming combination, state responsibilities nearly disappeared. Congress's anxiety about too much centralized government was more than offset by the promise of federal dollars and the fear that states would compete for industries if the states wrote their own, presumably weak, effluent guidelines.

Under the Reagan administration, the attitude toward federal policy appears to be swinging back again. For example, recent changes in the regulations governing municipal wastewater treatment construction grants reduce the federal cost share and allow more state flexibility in the allocation of funds. But control over nonpoint sources never did change. Always considered a state and local responsibility, nonpoint control remains decentralized despite expressions of deep congressional concern about nonpoint pollution. Both the Senate and House versions of recent bills to reauthorize the Clean Water Act leave most decision making to the states, albeit with some promise of federal subsidies.

Putting theory to the test

According to environmental economic theory, control of nonpoint water pollution by individual states should not be the most efficient management approach. In theory, a governmental policy role (as opposed to a private one) is justified because of the externality features of water pollution. The term externality applies generally to situations in which one party created through its activities a cost (or a benefit) for another party but does not take this cost (or benefit) into account in making decisions about its own production or consumption. In the case of water pollution, injured parties and polluters often are not known to each other because of flowing water's ability to carry damaging effects many miles from the polluting source, and because the contribution of many pollution sources is mixed, thus obscuring that of any individual source. Because most water networks extend well beyond state boundaries, one would hardly expect externality problems to be confined to individual states. Congressional and administrative attitudes about federal-state roles doubtless are affected more by political and perceived practical considerations than by the theoretically most efficient strategy for attaining water quality goals.

Of course, the theoretical presumption that water pollution problems are more likely to exhibit interstate than intrastate externalities should be subject to test. After all, it is conceivable that most water pollution fails to cross state boundaries even though water networks usually do. Some water pollutants, like fecal coliforms, have relatively short lifetimes. Their adverse effects may be confined to very short geographical distances from the polluting sources.

Undertaking such a test is not simple. For example, pollutants with radioactive tracers could be dumped at thousands of locations around the country and followed to determine to what extent the pollutants remain confined within state boundaries, but this would be enormously costly. A more practical—and certainly less expensive—approach would be to use computer models to simulate the transport of pollutants between sources and affected locations. This is the method we have chosen at Resources for the Future.

Using the RFF Water Network Model and a new database developed for the U.S. Environmental Protection Agency, the U.S. Department of Agriculture, and the U.S. Geological Survey, we simulated the effects on water quality of reducing erosion from agricultural cropland. The model permitted us to "observe" changes in water quality caused by reductions in erosion ranging from 0 to 100 percent 1980 levels. In particular, we were interested in the concentration of nutrients like phosphorus and nitrogen—pollutants that can lead to oxygen depletion and the eutrophication, or premature aging, of lakes and reservoirs.

Considering that phosphorus decays in water at a rate of roughly 20 percent per day and that water in major rivers may flow more than thirty miles in a day's time, interstate movement of nutrients seems highly probable. At these rates, phosphorus concentrations still would be at half their initial levels more than a hundred miles downstream of the source of the phosphorus (assuming no lakes or reservoirs in between). Our simulation results confirm these expectations: in the rapidly flowing Missouri and Mississippi river systems, for example, water quality in Illinois is affected by nutrient loadings from as far away as Minnesota.

Politics of pollution

Our results suggest that it usually is impossible for a state acting alone to clean up its nonpoint nutrient pollution problems; in most cases, some degree of upstream and "upstate," effort will be required. Furthermore, in certain river systems, states will have to reduce pollutant loadings even though their own water has relatively low levels of nutrients. This almost certainly would produce a political response to nonpoint pollution controls in downstream states markedly different from that in states far upstream. If an upstream state's water already is fairly clean and water in a downstream state is affected by its upstream neighbor(s), in neither state is it likely that water quality will improve significantly through the state's own efforts. Neither improving the already acceptable nor improving the unacceptable to a point below standards would hold much appeal for state officials, regardless of location. If nonpoint controls were imposed on all states, the downstream state would see the quality of its water improved, for which local politicians would be expected to claim credit. But little or no local improvement would occur in states far upstream, and upstream politicians no doubt would be unenthusiastic at best about such a program.

If these conjectures are correct, a program of local or state nonpoint pollution control initiatives will not be effective for nutrients and other long-lived pollutants. A state will need the complementary actions of its upstream neighbors to attain major improvements. However, political support fir a mandatory interstate program—whether it be a federal program or, preferably, a regional, river-basin program—is not very likely from those states in the very upper reaches of major river systems.

Finally, our simulations establish only a crude association between local pollutant loadings and local water quality: water transport of pollution clearly plays a large role. Programs that implicitly assume a local association, as does the conservation reserve program embodied in the 1985 farm bill, are likely to have unpredictable effects on water quality, to the extent they are effective at all.