House TSCA reform discussion draft: Major problem #2 – Preemption of State authority

Richard Denison, Ph.D., is a Senior Scientist.

The House’s discussion draft of the Chemicals in Commerce Act (CICA) issued last week was accompanied by statements from both its sponsor and the American Chemistry Council (ACC) claiming that it represents a “balanced” approached to reform of the Toxic Substances Control ACT (TSCA).

Despite the rhetoric, however, the draft is anything but balanced, and instead pegs the needle far to one side of the dial.  My earlier post describes the massive requirements EPA must meet in order to regulate a dangerous chemical and how far out of kilter those requirements are compared both to current TSCA and to the Chemical Safety Improvement Act (CSIA), especially as the latter is being revised via ongoing negotiations.

This post focuses on another area in which the CICA draft takes an extreme position:  its preemption of state authority, which is far more sweeping than under current TSCA or even CSIA as introduced.  But first let me start by arguing that any preemption needs to follow – not precede – final EPA actions that are based on robust information. 

Only final EPA actions based on robust information should have any preemptive effect

The most fundamental flaws of the CICA draft’s preemption provisions – which also apply to the Chemical Safety Improvement Act (CSIA), though to a lesser extent – are that they would take effect:

  • before EPA takes a final action to address a chemical’s risks; and
  • in situations where EPA’s decisions are based on partial information and made against a low standard for safety.

There are major unresolved issues in the TSCA reform debate relating to the pace at which chemical reviews and decisions should and can feasibly take place, including what deadlines should apply, whether specific throughput targets for decisions should be included, what levels of resources are needed and likely to be available, and what kind of fees should apply to companies.  What does this have to do with preemption?

The issues just noted, especially in an era of declining EPA budgets, mean there is real uncertainty as to the extent to which EPA will be able act expeditiously under any form of TSCA reform.  It is essential, therefore, that States be able to continue to act when EPA does not or cannot.  And hence, whatever its scope, any preemption should not apply until EPA takes final action on a chemical that is based on having full authority to require whatever information it needs to make a sound decision.

Yet the CICA draft greedily applies preemption wherever it suits the industry’s interests.  In particular:

  • Low-priority designations would be based on a low “not likely to result in an unreasonable risk” standard and EPA would be denied authority to get information it needs for making such decisions.  Yet any such decision would wholly preempt State and local governments from continuing or initiating action on any chemical deemed low-priority.
  • Reviews of new chemicals would also be based on the low “not likely to result in an unreasonable risk” standard, and here again, EPA would be denied authority to get information it needs for making such decisions.  Only 90 days would typically be allowed for EPA to make a decision.  Yet these decisions would have a full and everlasting preemptive effect.

A truly balanced approach would acknowledge the resource and evidentiary limitations to EPA’s ability to act and preserve the rights of States to do so until and unless EPA actually does take fully informed and final actions.

The authors of CICA say they acknowledge “the rights of states to act on their own in the absence of federal action.”  Nothing that even comes close to retaining that right of States, however, is reflected in the actual text of the CICA draft.  Let me describe just what it would preempt.

The details

All of CICA’s preemption provisions would apply to State and local requirements regardless of whether they were already on the books prior to enactment or to a specific EPA action, or were proposed thereafter.  As drafted, it can be read as preempting entire laws or regulations, not just requirements under those laws or regulations applicable to specific chemicals (I’m hoping that this, at least, is an inadvertent drafting error).

CICA would preempt not only State/local requirements on chemicals, but also on mixtures or articles.

But we’re just getting started.  Look what else would be preempted:

  • INFORMATION REQUIREMENTS.  CICA would preempt any State or local government from requiring any company to “submit” as well as to “develop” any information (CSIA applies only to the latter) whenever EPA has:
    • required  that information under TSCA, or
    • completed a safety determination for a chemical.

Consider the implications of the second condition. It means that if EPA has completed a safety determination for a chemical, no State could collect ANY information on that chemical, for ANY purpose.  A state could not require a company to report whether a chemical is present within that State’s borders in products, in storage tanks (think West Virginia’s Elk River), in workplaces, in air or water releases, in waste materials or at waste sites, in the environment or in people – regardless of whether EPA has required that information to be submitted or has considered it in its safety determination.  Nor could a State require a company to submit already available information on a chemical’s hazards or to assess the chemical or how it compares to available alternatives.  I could go on, but you get the idea.

  • LOW-PRIORITY CHEMICALS.  CICA would preempt any State or local government from enforcing any existing requirement or establishing a new requirement that restricts in any manner any chemical EPA has deemed low-priority.  (This is much broader even than CSIA as introduced, which only applied such preemption to new State or local requirements.)  That means any existing chemical EPA finds is “not likely to result in an unreasonable risk” is off-limits – forever.   (I’ll put aside for now the many problems with the draft’s continued unqualified reliance on TSCA’s fatally flawed “unreasonable risk” standard.)
    It’s one thing to trigger preemption once EPA has conducted a thorough risk assessment, with full authority to obtain whatever information it needs and concludes a chemical “will not result in an unreasonable risk.”  It’s quite another to do so based on a much lower “not likely” standard – especially when CICA would deny EPA any authority to require development of information prior to making its prioritization decision.
         NOW, HERE’S THE ULTIMATE IRONY:  Under CICA a chemical can be designated low-priority because of reductions in exposure that are due to existing State (and Federal) regulations.  Yet such a designation would then preempt those very State regulations!!
  • CHEMICALS SUBJECT TO EPA REQUIREMENTS BEFORE ENACTMENT.  Astonishingly, the CICA draft would preempt any State or local government from enforcing any existing requirement or establishing a new requirement that applies to ANY new or existing chemical that was, prior to enactment, the subject of ANY rule, consent agreement or order issued by EPA.
    Now it’s true that under TSCA EPA has only restricted a handful of chemicals using rules issued under its Section 6 authority.  But this provision is not limited to those rules.  It would appear to apply to every chemical EPA has ever subjected to:

    • a test rule under section 4 – several hundred chemicals;
    • a consent agreement or order for a new chemical under section 5, whether or not that agreement or order imposed any risk management conditions – several thousand chemicals;
    • a significant new use rule under section 5, which does not restrict but only requires notification of a chemical’s manufacture under certain circumstances – several thousand chemicals; or
    • a reporting rule under section 8(a) or 8(d), which merely require the submission of already existing information – several thousand chemicals.

(A charitable reading of the language is that the authors only intended this provision to apply to actions EPA has taken under sections 5 and 6 of TSCA, in which case this preemption would not apply to the chemicals described in the first and last of the sub-bullets above.)

  • NEW CHEMICALS.  Finally, and shockingly, the CICA draft would apply its sweeping preemption to every new chemical that enters commerce, including any such chemical:
    • EPA finds is “not likely to result in an unreasonable risk,”
    • that is subject to any rule, consent order or order, or
    • for which the review period (typically limited to 90 days) has expired.

That is, State and local governments would forever be preempted from acting on any new chemical that comes into commerce.  EPA reviews an estimated 1,000-1,500 new chemicals every year, a large fraction of which go on to commence manufacture.  As with low-priority existing chemicals, this preemption would be triggered based on a new chemical merely meeting the lower “not likely to result in an unreasonable risk” standard.  And here again, the CICA draft denies EPA any authority to require development of information on new chemicals before making that decision.

Last but certainly not least, the CICA draft provides no leeway for States to act in ways that differ from the actions EPA takes, for purposes that differ from those of TSCA, or on uses EPA fails to consider in its safety assessments:

  • No exceptions are provided for State or local governments’ information-related requirements for reporting, disclosure, monitoring, or providing warnings or labels.
  • No exceptions are provided for requirements adopted under State laws addressing wholly different objectives such as air or water quality, waste treatment or disposal, climate change or ozone depletion.
  • No exceptions are provided for State or local governments’ to address uses of a chemical EPA does not include in its assessment and determination.
  • No ability for States to seek waivers is provided.

(TSCA and CSIA both provide most or all of these.)

As a consolation prize, I do want to note one way in which the CICA draft actually improves upon CSIA:  CICA’s preemption of existing and new State or local requirements for high-priority chemicals would apply only when a final safety determination is made.  In contrast, CSIA applies preemption to new State or local requirements much earlier – upon publication of an assessment/determination schedule for high-priority chemicals; and revisions negotiated to CSIA to date would do so only slightly later, when EPA commences a safety assessment.

Even here, however, a change is needed:  Where a safety determination finds risk, any preemption should not apply until the requisite risk management rule takes effect.


The breath and scope of CICA’s preemption provisions are truly stunning and deeply disturbing.  None of them are even remotely consistent with protecting States’ rights or the public’s health.

The only fair and balanced approach is one that preserves the rights of States to act until and unless EPA actually does take fully informed and final action on a chemical.


This entry was posted in Health policy, TSCA reform and tagged , , , . Authors: . Bookmark the permalink. Both comments and trackbacks are currently closed.