How EPA Sets Maximum Contaminant Levels by Balancing Health and Filtration Costs

The binding constraint of public drinking water is not the absence of toxins, but the economic capacity of a municipality to filter them out. For a water system to function, the cost of purification cannot exceed what ratepayers can afford to pay. Currently, that constraint holds, but it forces a structural compromise: tap water is not engineered to be perfectly pure, but to be precisely, legally acceptable.[1][3]

This compromise is codified in the Maximum Contaminant Level (MCL), the enforceable standard that dictates exactly how much of a hazardous substance is allowed to flow through a residential faucet. Established by the U.S. Environmental Protection Agency (EPA) under the Safe Drinking Water Act, the MCL is the dividing line between safe and non-compliant water.[1][7]

But the MCL is not a pure health metric. It is the product of a high-stakes negotiation between human biology and municipal budgets. Before the EPA sets an MCL, it first determines the Maximum Contaminant Level Goal (MCLG)—the concentration at which no known or anticipated adverse health effects occur.[3][4]

For known carcinogens, the MCLG is always set at zero micrograms per liter (0 µg/L). The biological reality is that any exposure carries some risk. However, the MCLG is entirely non-enforceable. It ignores the limits of analytical detection and the astronomical cost of absolute purification.[6]

To bridge the gap between the biological ideal and the engineering reality, regulators conduct a rigorous cost-benefit analysis. The legally enforced MCL is set as close to the MCLG as is technologically and economically feasible.[1][7]

"The MCLG is set at close to zero as possible and is a non-enforceable public health goal," notes the Alaska Department of Environmental Conservation. "It does not consider analytical detection limits or treatment feasibility."[6]

This means that for many contaminants, the legal limit is intentionally set higher than the health-based goal. The EPA explicitly permits this if the costs of advanced treatment would outweigh the quantified public health benefits of a lower limit.[7]

The tension between cost and purity became highly visible during the 2001 revision of the federal arsenic standard. Arsenic is a naturally occurring carcinogen, and lowering its allowable limit required municipalities to install expensive new filtration infrastructure.[5]

A 2001 analysis published by researchers at Carnegie Mellon University highlighted the friction. "The current effort to revise the arsenic drinking water standard is one of the first times that the promulgation of a Maximum Contaminant Level (MCL) for drinking water has been influenced explicitly by benefit-cost considerations," the researchers noted.[5]

The study revealed that "large discrepancies in the aggregate national cost estimates are shown to result largely from differences in the engineering cost estimates for arsenic treatment processes." If the engineering is too expensive, the standard cannot be tightened without bankrupting small water systems.[5]

As of 2019, the EPA had issued 88 national primary drinking water standards, including 78 MCLs and 10 Treatment Techniques, covering microorganisms, chemicals, and radionuclides. Each of these 88 standards represents a specific mathematical compromise.[7]

When a contaminant is too difficult or expensive to measure accurately at trace levels, the EPA abandons the MCL entirely and issues a Treatment Technique (TT). A TT is an enforceable procedure—such as a specific filtration method—that water systems must follow to ensure control of the contaminant, rather than a strict numerical limit.[7]

The economic balancing act is currently being tested by the regulation of per- and polyfluoroalkyl substances (PFAS), often called "forever chemicals." The EPA has developed a Lifetime Health Advisory (LHA) level of 0.07 micrograms per liter (µg/L) for two common PFAS compounds, PFOS and PFOA.[6]

Developing these standards requires massive state or federal investment. California currently spends approximately $26 million a year on its own MCL program, a budget most states cannot match, forcing them to rely entirely on federal EPA guidelines.[6]

The cost of compliance is not borne by the federal government; it is passed down to local ratepayers. For large cities serving over 1 million residents, the cost of upgrading a treatment plant is distributed across millions of households, making the per-capita increase negligible.[1]

AI

For small, rural water systems, however, the math breaks down. Upgrading a facility to remove trace levels of a newly regulated chemical can double or triple the monthly water bill for a community of fewer than 3,300 people.[1]

Once an MCL is finalized by the EPA, states have exactly two years to adopt the new rule into their own legal frameworks. To prevent localized economic collapse, the Safe Drinking Water Act allows states to grant temporary exemptions lasting up to three years to public water systems.[6]

Beyond primary health standards, the EPA also maintains National Secondary Drinking Water Regulations. These cover 15 contaminants that cause aesthetic issues—such as the staining of fixtures by manganese or the metallic taste of iron—but do not pose a direct threat to human health.[2][3]

Secondary standards are non-mandatory guidelines. Because they do not carry the weight of a primary MCL, water systems are not legally required to spend millions of dollars to remove these nuisance chemicals, further illustrating the regulatory prioritization of resources.[2][3]

The architecture of the Maximum Contaminant Level reveals a profound truth about public infrastructure. Clean water is not a natural state; it is a manufactured product. The next phase of this regulatory balancing act will center on how the EPA enforces its proposed PFAS limits, and the degree to which the federal government subsidizes the multi-billion-dollar treatment facilities required to meet those new thresholds.[8]