This morning, CPR Member Scholar Tom McGarity testifies at the Senate Committee on Environment & Public Works on "Strengthening Public Health Protections by Addressing Toxic Chemical Threats." His testimony can be found in full here. McGarity contributed the following blog post in advance of the hearing.
The Chemical Safety Improvement Act: The Wrong Way to Fix a Broken Federal Statute
We live in an era in which human health and the environment are threatened by toxic chemicals that have not been adequately tested and that are subject to a federal regulatory regime that is badly broken.
The fact that we do not often read about disease outbreaks caused by toxic chemicals in the newspapers probably stems from the fact that we know so very little about the risks posed by the thousands of synthetic chemicals that we encounter on a daily basis.
We are only beginning to learn of the adverse health effects of endocrine disrupting chemicals and various flame-retardants. The impacts of many ubiquitous chemicals on ecosystems are largely unknown. We know, for example, that the Potomac River contains high levels of endocrine disrupting chemicals and that there is an unusually high incidence of “intersex” (male fish exhibiting female characteristics) in smallmouth bass in the river. At the same time, people continue to be e exposed to some chemicals, like the carcinogen formaldehyde, that have well-known adverse effects on human health.
There is a federal law in place that is supposed to require manufacturers of chemical substances to conduct adequate toxicity testing and to empower the Environmental Protection Agency (EPA) to ban, phase-out, label, or otherwise regulate chemicals that pose unacceptable risks to human health or the environment. That statute is the Toxic Substances Control Act of 1976 (TSCA).
But that statute is severely broken. Congress is considering a bill, referred to as the Chemical Safety Improvement Act, that purports to fix TSCA’s problems. Unfortunately, that bill is not likely to be as effective as its supporters think, and it contains two provisions that will make matters affirmatively worse.
Why TSCA Is A Broken Statute
When Congress enacted the Toxic Substances Control Act of 1976 (TSCA), I was a young attorney in the division of the Environmental Protection Agency’s Office of General Counsel in the division that was responsible for implementing the brand new statute. Although it took five years for the final version to emerge from Congress, most observers agreed at the time that the statute would for the first time allow the federal government to protect American citizens from the serious risks posed by potentially toxic substances in the environment.
TSCA was supposed to fill in the considerable jurisdictional gaps left by the topical statutes like the Clean Air Act, the Clean Water Act, the Resource Conservation and Recovery Act, and the Federal Insecticide, Fungicide and Rodenticide Act. The hope was that EPA would in short order require companies to test the thousands of chemicals for which rudimentary toxicity studies were lacking. The Agency would then require manufacturers, processors and distributors of risky toxic chemicals to use proper warnings, limit human exposure to those chemicals, and, in some cases, to take dangerous chemicals off the market altogether.
Thirty-five years later, it has become painfully apparent that, with some modest exceptions, TSCA is a failed statute.
TSCA also empowers the EPA to promulgate rules requiring manufacturers to test an existing chemical if EPA can demonstrate that insufficient information is available to evaluate its safety and that human beings or the environment are heavily exposed to the chemical or that the chemical is likely to be toxic. The problem with this “selective interdiction” program is that it places the burden on EPA to justify a testing requirement. Given the toxic substances program’s perennial lack of resources, this requirement has effectively driven TSCA’s testing function underground as EPA and manufacturers negotiate testing agreements outside of the public rulemaking process envisioned by Section 4 of the statute. More important, only a very few chemicals to which the public and the environment are routinely exposed (sometimes at high levels) have been the object of TSCA’s testing requirements. Consequently, thousands of “grandfathered” chemicals have not undergone the full range of testing necessary to determine whether they are safe for human beings and the environment.
TSCA’s greatest disappointment, though, is EPA’s inability to take effective action under section 6 of the statute to ban, label, or otherwise limit exposure to existing toxic substances. Section 6 provides that when EPA finds that the manufacture, processing, distribution, use or disposal of a chemical substance presents an "unreasonable risk of injury to health or the environment," it must issue a rule applying "one or more" of eight requirements "to the extent necessary to protect adequately against such risk, using the least burdensome requirements." When EPA, in an early test of its rulemaking powers under section 6, promulgated a rule providing for a gradual phase-out of the manufacture, processing and distribution of asbestos for most domestic uses, the Fifth Circuit Court of Appeals set the rule aside in an opinion that made it abundantly clear that EPA would be hard-pressed to take effective action to protect the public health and the environment under section 6 in the future.
One reason for the court’s willingness to substitute its judgment for that of the agency was the odd standard of review that TSCA provides for rulemaking. Although the standard of review in the Administrative Procedure Act (APA) for informal rulemaking is the familiar “arbitrary and capricious” test under which the courts are supposed to defer to the agencies’ exercise of expert judgment, section 19 of TSCA provides that the reviewing court shall set aside a rule promulgated under that statute if it is not supported by “substantial evidence in the rulemaking record.” Some courts have construed this test, which the APA reserves for formal adjudications and formal rulemakings, to call p for more stringent judicial review than the arbitrary and capricious test.
One provision in the original statute that is clearly not broken is section 18, the statute’s preemption clause. That section provides that, with certain exceptions, the courts are not to interpret the statute to “affect the authority of any State to establish or continue in effect regulation of any chemical substance, mixture, or article containing a chemical substance or mixture.” The exceptions are for state testing requirements for a chemical after EPA has promulgated a testing rule for the same chemical. In addition, state regulations other than outright bans that differ from rules promulgated under sections 5 and 6 could be restricted. Under this relatively straightforward preemption provision, the states and the federal government have effectively stayed out of each other’s way for thirty-five years with very little, if any, controversy.
The Chemical Safety Improvement Act
The Chemical Safety Improvement Act (S. 1009) is an attempt to fix some of the problems that have plagued TSCA implementation for the past three decades. Senators Frank Lautenberg (D- New Jersey) and David Vitter (R-Louisiana) introduced it as a “bipartisan” bill aimed at getting TSCA back on track. Senator Lautenberg had originally introduced a far more comprehensive bill called the “Safe Chemicals Act” (S. 696) that attracted 28 sponsors from both parties, including Senators Barbara Boxer, Al Franken and Elizabeth Warren. Many of the most protective aspects of that bill wound up on the cutting room floor, and a much smaller number of Democratic senators (not including the three above) signed onto the “compromise” Lautenberg-Vitter Bill, ) and many other Republicans. It also garnered the support of the chemical industry.
In this day and age of extreme partisanship, a “bipartisan” bill offers a very attractive vehicle for moderate Democrats and Republicans to demonstrate to their constituents that Congress can enact socially beneficial legislation. S. 1009 is also attractive to Democratic senators because it is the “Lautenberg Bill.” Not long after he introduced the bill with Senator Vitter, Senator Lautenberg died. One of the last of the New Deal liberals, his death represented a terrible loss to the Democratic Party. Many supporters of S. 1009 have urged Congress to enact it into law as a tribute to the late great senator. Sadly, if Congress does pass the severely diminished bill in its present form, it may well do more to tarnish that great senator’s legacy than to burnish it.
The Lautenberg- Vitter bill, S. 1009 appears to provide a systematic mechanism for prioritizing and evaluating the tens of thousands of grandfathered chemicals that have not been adequately tested to determine the risks, if any, that they pose to human health and the environment. But appearances can be deceiving.
S. 1009 recognizes that thousands of chemicals will never receive adequate toxicity testing if EPA must meet the burden of demonstrating the need for testing in advance. The best way to ensure that new and existing chemicals get tested is to put the burden on the manufacturer to test before placing chemicals on the market and to require manufacturers of existing chemicals to test their products by a statutorily predetermined deadline.
The Bill does place the burden on manufacturers to test “high priority” chemicals for which EPA, by rule or order, would require additional testing. But it accomplishes this improvement through an extremely convoluted process that, among other things, requires the agency: to develop a “chemical assessment framework” for collecting and analyzing existing information on chemicals; to promulgate criteria for evaluating the quality of individual studies; to develop a “structured evaluative framework” for deciding what action to take with respect to chemicals; to come up with a risk-based screening process (within one year of the date of enactment) for identifying “high priority” and “low priority” “active” chemicals; to prioritize existing chemicals under this screening process (making “every effort” to complete the prioritization of all “active” substances in “a timely manner”); to determine an order for performing safety assessments on all high priority chemicals; and to publish and maintain a list of high priority and low priority chemicals.
Except for the one-year deadline for coming up with the screening process, the statute does not impose any definite deadlines for accomplishing these tasks. Instead, the agency is to “make every effort” to complete the prioritization of all active chemicals “in a timely manner.”
Much of this work will require the agency to engage in notice-and-comment rulemaking, a process that has become laden with resource-draining and time-consuming analytical and procedural hurdles.
S. 1009 requires EPA to conduct a safety assessment for every high priority substance, to make a safety determination based on the safety assessment, and establish appropriate risk management requirements for those high priority substances that do not meet the safety test. The agency must develop an “appropriate science-based methodology” for conducting safety assessments that meets several statutory specifications.
For chemicals that do not meet the safety standard, the agency must promulgate a rule establishing the “necessary restrictions,” choosing from among a list that fairly closely tracks the list of regulatory alternatives in the current statute. All of the analytical operations that the original TSCA statute requires in making the safety determination and risk management prescriptions are subject to judicial review under the statute’s “substantial evidence” test.
There are two fundamental problems with this highly complex testing and evaluation regime.
One fundamental problem with the Bill is that imposes a large number of very burdensome new obligations on an agency that is struggling to keep up with its existing statutory duties. In deciding whether to force the agency through more procedural and analytical hoops, Congress should bear in mind the limited resources that are likely to be available to the agency, at least in the near term.
A second fundamental problem is the lack of judicially enforceable deadlines. Nor does the Bill address the consequences, if any, of failure to act in a timely manner or to complete studies within time limits of reasonable duration.
Take it from someone who has been studying bureaucracies for more than 35 years: If the statue does not subject EPA to judicially enforceable deadlines, the agency will still be searching for just the right chemical assessment framework and promulgating criteria for evaluating all if the possible varieties of health and environmental studies a decade from now.
Overly Risky Safety Test
The Bill does not change the test for taking regulatory action to protect the public and the environment in any significant way.
The term “safety standard” is defined as “a standard that ensures that no unreasonable risk of harm to human health or the environment will result from exposure to a chemical substance. This is virtually indistinguishable from the “unreasonable risk of injury to health or the environment” standard that the current statute employs.
The most debilitating aspect of the current statute, as interpreted by the court in Corrosion Proof Fittings, is the requirement that EPA engage in an extensive analysis of the costs and benefits of the regulatory alternative that the agency selects as well as all of the other alternative regulatory approaches indentified in the statute. The Bill clearly requires EPA to engage in the very same debilitating analysis of the costs and benefits of a set of alternatives that closely resembles the alternatives identified in the current law.
Beyond the analytical difficulties of applying the risk-benefit standard in the real world, the standard itself is insufficiently protective of human health and the environment. Even if EPA had infinite resources to devote to the analytical exercise, the risk-benefit test is inherently biased against protective regulatory action. The benefits of a chemical that is already in use are typically obvious and easily exaggerated, while the risks that the chemical poses to health and the environment are often clouded by uncertainty and easily belittled or ignored (especially in the case of environmental risks). Thus, bans and phase-outs of existing chemicals are very difficult to justify under a risk-benefit test.
Yet, history teaches that sometimes a ban or phase-out is by far the most effective way to reduce the risks that toxic chemicals pose to human health and the environment. The phase-down of lead in gasoline and the phase-out of the pesticide mirex are two examples of situations in which society benefited greatly from forceful action that, in retrospect, had very few, if any, negative effects on the economy.
The Bill also fails to take into account the special sensitivities of fetuses, infants and children to chemicals in the environment.
A better test would be the “reasonable certainty of no harm” test employed by the Food, Drug and Cosmetics Act and the Food Quality Protection Act. The latter act requires EPA to give special attention to the risks that pesticides pose to fetuses, infants and children.
Unchanged Standard of Review
The Bill makes no attempt to address an anomaly that lies at the heart of the current law’s ineffectiveness -- the standard for judicial review.
The curious specification of the “substantial evidence” standard for judicial review of TSCA rules has invited the courts to review those rules less deferentially than they would under the “arbitrary and capricious” test that normally applies to judicial review of informal rulemaking.
If Congress amended the statute to provide for arbitrary and capricious review, it would send a clear signal to the reviewing courts that it disapproved of the overly intrusive review demonstrated in the Corrosion Proof Fittings case.
Excessively Powerful Preemption
S. 1009 would work a fundamental change in the relationship between EPA and the states in the area of toxic substances regulation, without justification.
Under the current law, courts are not supposed to interpret the statute to “affect the authority of any State to establish or continue in effect regulation of any chemical substance, mixture, or article containing a chemical substance or mixture.”
S. 1009 provides that “no State or political subdivision may establish or continue to enforce”: (1) a data production requirement for a chemical or category of chemicals that “is reasonably likely to produce the same data and information required by an EPA data requirement; (2) a prohibition or restriction on the manufacture, processing, or distribution” of a chemical after EPA has completed a safety determination; or (3) a significant new use notification requirement for a chemical for which EPA has required such notification.
S. 1009 goes even further to prohibit states from establishing a new prohibition or restriction on a chemical that EPA has identified as a high priority substance or a low priority substance. The preemption of regulatory action against low priority substances is apparently based on the assumptions that: (1) EPA will always assess the risks posed by chemical substances accurately in the first instance; and (2) EPA will continuously update its assessments and priority lists in light of new scientific evidence. Both of these assumptions are misplaced.
EPA does not have sufficient resources to make accurate assessments of substances that are likely to fall within the low priority category. Once EPA makes the assignment, however, the states will be powerless to regulate the chemicals, even if the scientific evidence demonstrates that the chemical is far more dangerous that EPA originally thought.
The prohibition on state regulation of chemicals assigned high priority status is even more difficult to comprehend. EPA’s assignment of high priority status to a chemical represents the agency’s determination that the chemical has the potential for high hazard and/or high exposure. High hazard chemicals are likely to be the very chemicals that states are most likely to identify as in need of further regulation. But once EPA made the high hazard determination, S. 1009 would preclude any state regulatory action.
Given the absence of statutory deadlines, discussed above, it may take EPA years (or even decades) to get around to determining the most effective risk management approach for the chemical. In the interim, no governmental entity will have the authority to protect public health and the environment from the risks posed by the chemical.
Limitations on the Right to a Jury Trial
Finally, S. 1009 contains a highly unusual constraint on the discretion of local judges and juries to consider and evaluate evidence that chemical substances or products containing chemical substances are unsafe, dangerous or defective. The Bill provides that once EPA has completed a safety determination for a high-priority chemical, that determination is automatically admissible in any public or private litigation. Moreover, the safety determination is “determinative of whether the substance meets the safety standard under the conditions of use addressed in the safety determination.
In my thirty-six years of teaching and research in the area of torts, I have never seen a proposal for such an intrusive interjection of federal law into the day-to-day administration of justice at the state level.
Not only does the Bill change the state judiciaries’ procedural rules governing the admissibility of evidence, but it also takes away from state court judges and juries the ability to determine whether a chemical is abnormally dangerous or otherwise unsafe in both common law litigation and criminal trials. I have seen no evidence whatsoever that the substantive and procedural rules promulgated by the courts of the 50 states are causing any problems for the manufacturers, processors and distributors of chemical substances.
This provision is nothing less than a gift of partial immunity to manufactures fortunate enough to have their chemicals declared safe by EPA.
The Toxic Substances Control Act is badly in need of repair, but S. 1009 is not the way to fix that broken statute. The committee should reject this Bill or amend it in ways that make it more protective of human beings and the environment and less protective of the chemical industry. The Center for Progressive Reform has just published an Issue Brief on Toxic Substances Control Act reform. Congress can consult that publication for useful and innovative suggestions for turning TSCA into a robust protective statute.