Unattainable Clean Water Act Standards
EPA has imposed radical risk policies in the calculation of certain water quality standards that are in many cases unattainable and could cost municipal and industrial dischargers billions of dollars. Many of these industrial facilities provide high-paying manufacturing jobs in rural communities that cannot afford to lose them. If EPA’s new policies are extended to other states and programs (e.g. Superfund, Clean Air Act), private parties and federal agencies would be faced with environmental requirements with costs that greatly exceed the benefits.
Under the Clean Water Act (CWA), states have the primary responsibility to develop water quality standards. States begin that process with EPA’s Human Health Water Quality Criteria (HHWQC), but can use other criteria, as long as they are adequately protective of human health. States also have the discretion to set the excess lifetime cancer risk (ELCR) level and the Fish Consumption Rate (FCR) — two of the variables used to calculate the HHWQC.
Extreme Risk Management Policy
EPA is making three radical changes to its risk policy that has existed since 2000. First, EPA is requiring states to adopt HHWQC based on a new ELCR level policy -- that one in a million (10-6) and only one in a million is adequately protective of high fish consuming subpopulations, such as Native American tribes. This overrides the long-standing policy, which recognizes that states have discretion to choose other risk levels, as long as they are adequately protective.
Second, EPA is requiring that HHWQC be derived based on protection of an extremely small subset of the population, instead of the general population, which is the established basis for most risk-based standards. This makes the standards so stringent that they are guarding against risks more remote than the risk of getting hit by an asteroid; they also are much more stringent than other health and safety standards (see Figure 1). For example, the Occupational Safety and Health Administration (OSHA) standard for ethylene oxide is based on an ELCR of 1 in 500 – over 2000 times less restrictive than EPA’s requirement for HHWQC.
Note how EPA’s new risk requirements compare with other regulatory programs and everyday risks that society internalizes in Figure 1. The incremental cancer risk to the general population of Washington State of EPA’s new rule represents a risk at least 5000 times less likely than OSHA’s standards for carcinogens and 250 times less likely than being struck by lightning.
EPA also is requiring states to adopt FCRs taking into account tribal populations’ reduced fish consumption based on their fear of contamination; and also using tribal populations’ FCR as a proxy for the general public, resulting in much more stringent and unattainable permit limits. If applied to other programs, these policies will determine “how clean is clean” for Superfund cleanups and make other standards more stringent and expensive, without a commensurate improvement in human health.
Finally, EPA is taking conservative and unscientific approaches to other values in HHWQC derivation – the Relative Source Contribution (RSC) and Bioaccumulation Factor (BAF). EPA’s mandated RSCs and BAFs may not be applicable to the waters of a particular state.
Imposing Unattainable Federal HHWQC in Washington, Maine and Idaho
EPA imposed a federal rule in Washington, partially rejecting the state’s HHWQC even though it had worked with EPA for years on the criteria. In 2013, a study found that if HHWQC consistent with EPA policies were applied to Washington, industries and municipalities would not be able to meet all the resulting CWA permit limits, and the potential compliance cost would be over a billion dollars. Yet, despite that expenditure, the theoretical cancer risk reduction would not be measurable.
EPA had previously disapproved Maine’s HHWQC and imposed a federal HHWQC rule for the state (with a FCR of 286 grams per day) based on similar policies. Maine brought a legal challenge to that rule, and EPA recently told the court that it had reconsidered the rule and decided not to change its position.
Late in 2016, Idaho submitted its new HHWQC to EPA for approval after an extensive stakeholder process. Throughout the process, EPA had pressured Idaho to adopt policies similar to those imposed in Washington and Maine. Idaho did not do so, and EPA told Idaho that it was inclined to reject the HHWQC, but has not yet taken that step.
Florida also undertook an extensive stakeholder process to develop generally reasonable and scientifically-based HHWQC. However, as a result of EPA pressure, the state revised and adopted HHWQC with flawed RSCs and BAFs. Challenges to the HHWQC are pending.
Proposed National Fish Consumption Survey Guidance
In December 2016, EPA issued national guidance for states on how to conduct fish consumption surveys. The guidance perpetuates EPA’s new policy of focusing the FCR on high-consuming subpopulations such as tribes, instead of the general population. It also provides further guidance on how to calculate an “unsuppressed” FCR.
Probabilistic Risk Assessment (PRA)
EPA’s HHWQC are based on exposures represented by one value on the high end of the range of possible values. In contrast, a PRA approach gives policymakers an understanding of the central tendency of estimated risk and the probability that actual risk will be on the high end of the range. PRA is more scientifically advanced; links risk targets with environmental concentrations; improves transparency; and makes greater use of available data. Numerous offices and bodies within EPA have endorsed or used a PRA approach for several years.
EPA should withdraw the federal rules in Washington and Maine; allow states such as Idaho and Florida to establish their own HHWQC without federal interference as long as they are consistent with the CWA and applicable regulations; and withdraw EPA’s survey guidance. EPA also should begin deriving its own HHWQC recommendations using a PRA approach based on sound science, encourage states to do so, and then approve the resulting criteria.