You are here

TSCA Reform Won't Reduce Chemical Spill Risks

The January 9 chemical spill in West Virginia, which temporarily contaminated the Charleston drinking water supply, has rekindled a debate related to federal chemical regulation. Clearly, this case -- and another spill that occurred this week in West Virginia -- demonstrates the failure to properly implement the many emergency planning and prevention laws and programs already on the books. And this is a problem that state regulators and policymakers there need to address.

But do these cases warrant expansion of the federal Toxic Substances Control Act (TSCA) to force companies to conduct more testing of chemical products?

My blog post yesterday pointed out that there was enough information about the chemical to manage the risks. Now we address the myth that thousands of potentially dangerous chemicals are simply unregulated because they are "grandfathered" from TSCA. Richard Dennison of the Environmental Defense Fund makes this claim:

The sad truth is this chemical is one of tens of thousands of chemicals on the market today with little or no safety data. MCHM is one of the 62,000 chemicals that were already in use when TSCA, our nation’s main chemical safety law, was passed in 1976. All of these chemicals were grandfathered by TSCA: That means they were simply presumed to be safe, and EPA was given no mandate to determine whether they are actually safe. Even to require testing of these chemicals under TSCA, EPA must first provide evidence that the chemical may pose a risk – a toxic Catch-22.

In reality, existing chemicals were not "grandfathered" from the act. They are simply not regulated as "new chemicals." TSCA set up two regulatory programs, one for these "existing chemicals" and another for "new chemicals." For "new chemicals," the law demands that manufacturers provide the agency with notice and data demonstrating safety before chemicals enter into commerce, and EPA can regulate or ban them based on that data. The agency can also request more data, or do its own testing.

The existing chemical products all had a long history of safe and beneficial use. Years of experience with these chemicals tells us a great deal about their risks: They are manageable. And the absence of government data or testing does not mean that the companies that use them haven't done testing to ensure safety. After all, it is the private companies that bear the cost and liability should their products do harm.

When they passed TSCA, members of Congress recognized that calling for more testing of 60,000 chemicals would be expensive, and it is unlikely to reveal much of anything new. Any investment in the research of these tried-and-true products would need to be prioritized, with the EPA focusing on those it deems to pose the highest risks.

Accordingly, the law empowered the EPA to set priorities and call for more data on any existing chemical if the agency found that the chemical “may present an unreasonable risk of injury to health or the environment” or if the public is exposed to high levels of a chemical. In either case, EPA must also show it needs more data to fill in data gaps.

As I detailed in a paper on TSCA, this is not a particularly high bar for EPA to meet. The law does not require the agency to prove risk or demonstrate excessive exposure. All it requires is that EPA show that a chemical “may” pose more than a theoretical risk. Greens say that's not good enough, complaining because EPA has not banned or heavily regulated many existing chemicals. But that's not evidence that they are risky or that EPA lacks power. It's testimony to the fact that these chemicals pose low risks in their current uses.

Companies must also get EPA approval of all significant new uses of existing chemicals, and they must submit data when they conduct new tests on these chemicals. Under Section 8 of the law, EPA can also conduct testing on its own.

EPA doesn't even need TSCA to collect data or do testing. It has a number of voluntary programs through which companies and other government agencies have created and submitted a large amount of data on thousands of chemicals. For example, the federal government has conducted 86 tests under an inter-agency program called TOX21 on the main chemical in the mixture (4-methylcyclohexanemethanol) that spilled on January 9, which I addressed in my post yesterday.

Still environmental activists want TSCA reform that is modeled after the European Union’s 2006-passed chemical law, called REACH, which stands for Registration, Evaluation, and Authorization of Chemicals. REACH is a massive bureaucratic exercise to study and collect data on more than 100,000 chemicals over several decades.

The tests employed to meet REACH regulations commonly involve injecting high doses of concentrated chemicals into the stomachs of lab animals that are bred to be highly susceptible to tumor formation. This type of animal study has little-to-no relevance to humans who are exposed to chemicals on a short term basis at trace levels. Like with REACH, we are unlikely to learn much by mandating more such tests under TSCA. Nonetheless, REACH will require 13 million to 54 million animals for tests conducted between 2009 and 2018, according to the European Coalition to End Animal Experiments, and REACH testing will continue beyond 2018.

Sometimes animal testing is necessary for scientific discovery and to ensure both safety and efficacy of drugs and consumer products. But policies that demand such excessive and unnecessary animal testing are simply inhumane. This is hardly a good model to follow.

If policymakers want to reduce the risk of chemical spills, they need to focus on the source of the problem rather than allow people to use it to advance unrelated agendas.