This lecture will discuss policies for controlling pollution of surface water and groundwater. A general review of federal pollution control policies illustrates the old adage "The road to hell is paved with good intentions." In general, these policies have targeted overly ambitious environmental quality standards; they have relied almost exclusively on complex and impractical regulations rather than economic incentives (emissions charges or marketable permits); they have largely ignored the economic costs of these regulations; and they have mostly fallen short of achieving their stated environmental quality objectives.
Water contamination is caused by pollutant discharges from point sources into streams (e.g., factories and sewage-treatment plants) and surface runoff into streams or leaching into groundwater from non-point sources (e.g., cropland and urban storm-water). Point sources are generally far easier to identify, monitor and regulate than non-point sources. Water contamination may occur in surface waters, or groundwater or both, depending on water solubility of the pollutant, soil permeability, etc.
Many fund pollutants are bio-degradable, but excessive quantities deplete dissolved oxygen (DO) levels and disrupt aquatic ecosystems. Thus DO provides a useful index of in-stream water quality. Other fund pollutants include thermal pollution from industrial cooling systems; excess plant nutrients (nitrates and phosphates) which generate excess aquatic weeds and turn waterways eutrophic; persistent chemical pollutants such as DDT which bioaccumulate in species which are high in the food chain; and toxic organisms such as Pfiesteris piscicida which may thrive in polluted waters.
Stock pollutants include heavy metals which may bioaccumulate in the food chain. Surface waters can assimilate various pollutants which groundwater cannot. Thus some contaminants which are fund pollutants in surface waters are essentially stock pollutants in aquifers.
Standards
US water pollution policy is almost entirely based on "command-and-control" standards. There are three basic types of standard. An ambient standard specifies minimum DO levels or maximum pollutant concentrations at receptor points. The ambient standard states the pollution control objective pretty directly, but it generally doesn't provide much practical guidance on how it should be achieved.
An effluent standard specifies maximum pollutant loadings from individual sources; this typically requires monitoring of loadings from all sources. As pollutants mix and diffuse in water, the connection between loadings from point sources and ambient quality may be uncertain.
A technology standard specifies the method of pollution control each polluter must use to reduce pollutant loadings. Monitoring compliance with technology standards is typically easier than monitoring compliance with effluent standards, but the connection between technology standards and ambient quality is even more uncertain.
We have based the logic of Pigovian pollution taxes and Coasian bargaining to correct a pollution externality upon highly simplified externality models. More realistically, the environment commingles pollution from heterogeneous sources with differing marginal abatement cost schedules. The contributions of individual sources to aggregate pollution levels are often difficult to trace. And the environment distributes pollutants in varying concentrations to different receptor points that have differing marginal damage cost schedules. So the economically efficient pollution standard or tax may be different for each source, balancing the individual source's marginal abatement cost schedule against the summation of all receptors' marginal damage costs that are attributable to that source. The accounting required to determine the optimal set of standards or taxes is messy, albeit not particularly sophisticated. Various intermediate-level environmental economics textbooks include mathematical formalizations of the multi-source multi-receptor problem.
Control of surface water quality in the US was traditionally a concern of individual states. The Consitution's interstate commerce clause gives Congress has a clear mandate to regulate interstate wateways, and Federal authority has gradually been extended to all navigable waters.
The principal water quality law in the US is the Federal Water Pollution Control Act, as amended in 1972 and the 1977 Clean Water Act. The 1972 amendments focused on technology standards, requiring all point sources to obtain discharge permits, and industrial point sources to implement "best practicable technologies" (BPT's) by 1977, "best available technologies" (BAT's) by 1983, and eliminate all discharges by 1985. The 1972 FWPCA was a total failure (and control of point sources was supposed to be the easy part!). The EPA quickly got bogged down in the complexities of defining effluent standards, BPT's and BAT's; the states were supposed to enforce whatever standards the EPA did promulgate, but generally didn't; and the implementation deadlines were never met.
The 1977 Clean Water Act (FWPCA amendments) postponed these deadlines, and focused more effort on toxics. Reauthorized in1987, the CWA again focused on toxics, authorized citizen lawsuits against polluters, and included funding for sewage treatment plants under the Construction Grants Program. The CWA allows EPA to delegate many permitting, administrative, and enforcement functions to state governments.
The original zero-discharge goals for all pollutants sounds noble, but is obviously unattainable. First, a zero discharge objective is economically inefficient for more benign pollutants. Second, as harsher and harsher penalties are established for non-compliance, the public and the courts view the law as increasingly unreasonable. The old FWPCA's very tough language actually impaired its effectiveness!
Focusing on BPT's tends to divert focus from actual pollution abatement. A firm may comply with the law by simply installing the specified abatement technology, but has no incentive to improve on it, run it efficiently, or even maintain it propertly. Not surprisingly, technology standards often yield much smaller gains in environmental quality than regulators had envisioned.
A well-known simulation study of the Delaware Estuary compared the economic efficiencies of achieving target DO levels via (1) uniform effluent treatment standards; (2) uniform effluent charges; (3) zoned effluent charges reflecting ambient quality in different zones; and (4) a least cost system of variable effluent charges based on ambient quality at each discharge point. The effluent charges were much more cost-efficient than the uniform treatment standards, and zoned charges were close to least-cost. These results are consistent with the experiences of various European countries which use effluent charges. Nevertheless, the US has generally stuck with inefficient standards.
Non-point sources
Does the fact that non-point sources are harder to control, justify even more intensive controls on point sources? Only if the marginal damages from non-point sources are lower, or the marginal costs of non-point controls are higher.
Until recently, control of non-point water pollution sources was largely left to the states. Under Section 303 of the CWA, the EPA's TMDL (Total Maximum Dailing Load) program will have the states list and prioritize all impaired or threatened water bodies, implement ambient quality standards for multiple pollutants in them, and establish permitting processes for all sources of these pollutants. TMDL's can be established on a segment-specific basis or on a watershed basis. (It makes more sense to manage non-point sources by watershed than arbitrary political boundaries.) A TMDL specifies the maximum amount of a pollutant that a waterbody can receive, as determined by the states, based on each waterway's uses (drinking water, swimming, fishing, etc.). The TMDL program is (eventually) also supposed to allocate allowable amounts by individual pollutant source. TMDL's are supposed to include a "margin of safety" to insure these uses remain viable.
1997 TMDL rule changes have given EPA stricter control over state TMDL programs. Under current rules, states have sole responsibility for implementing TMDLs by requiring clean ups and source control for waterbodies under its jurisdiction. EPA basically approves states' lists of impaired water bodies (submitted every two years) and their TMDL plans. In recent years, EPA has lost numerous citizen lawsuits filed under the CWA to force EPA to take a more active role in setting the requirements for listings and approving TMDLs. So EPA is revising its TMDL rules to address non-point sources such as agriculture and forestry directly.
One proposal is to require new or expanding sources of water pollution to obtain "offsets" if they discharge into impaired waters for which TMDLs have not yet been established. EPA is proposing that new discharges should provide 1.5-to-1 to 2-to-1 offsets from existing sources (most likely non-point sources).
The USDA's Conservation Reserve Program is another non-point source program at the federal level (as well as being an important agricultural price support program): farmers retire environmentally sensitive cropland in exchange for annual rental payments.
In Delaware, agriculture appears to be a major contributor of phosphorous (P) runoff into streams, ponds and bays. The major aggregate source of agricultural P runoff is land-applied poultry manure, which has very high concentrations of P. Runoff from lawns is another source. In 1999 Delaware established the Delaware Nutrient Management Commission to oversee regulation of nutrient runoff. Largely dominated by farmers and industry representatives, this commission has been slow to implement substantial remediation programs.
A few years ago I proposed an alternative runoff control policy for Delaware's Inland Bays watershed in Eastern Sussex County, DE. This research is summarized on the Spatial Analysis Lab website. My proposal was to manage runoff by sub-basin, charging a tax per pound on poultry grown in each sub-basin based on phosphorous monitoring at each sub-basin outlet. The tax revenues from each sub-basin would be returned to a runoff management cooperative comprised of the poultry growers and other phosphorous sources in the sub-basin, to be spent on whatever runoff abatement strategy they choose. As its runoff is reduced, the sub-basin's poultry tax would be reduced accordingly. A supporting GIS analysis showed that a poultry tax of less than 0.2 cents per pound would be sufficient to fund conversion of all cropland riparian buffers to runoff-control vegetation at prevailing cropland rental rates.
Groundwater
Groundwater pollution problems reflect diverse land uses and complex physical properties of soils and aquifers. The EPA's federal groundwater quality standards are easily met in some areas and impossible to meet in others. Toxics (mainly from point sources such as industrial waste sites and landfills) are controlled under CERCLA (see toxic waste notes); controlling other groundwater contaminants (mainly from non-point agricultural sources) requires changes in agricultural practices and land uses hich state and local governments have been slow to implement.
The Safe Drinking Water Act (1974) is supposed to protect actual and potential drinking water sources, both above-ground and underground. It charges EPA with establishing safe standards of purity, and requires operators of public water systems to comply with primary health-based standards. State governments, which assume this power from EPA, also encourage attainment of secondary nuisance-related standards.
Groundwater in much of southern Delaware contains nitrates in excess of the EPA's maximum safe drinking-water standard of 10ppm. A Spatial Analysis Lab study conducted in the early 1990's by John Sparco summarized water test results for Sussex County and showed about 30 percent of all wells with nitrate concentrations above this level. Nitrate concentrations are typically higher in well-drained soils, where highly-soluble nitrates can leach quickly to the water table. The principal nitrate source appears to be agriculture--land-applied poultry manure plus commercial N fertilizer. There is little prospect for reversing the long-term buildup: since lateral movement of groundwater is slow, even with a total cessationn of N applications, Sparco's simulation model predicted it would be 50+ years before there are any significant improvements in contaminated aquifers.