TSCA reform legislation: Enhancing EPA testing authority

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

Part 1              Part 2              Part 3              Part 4             Part 5

[UPDATE 5-17-15:  On April 28, 2015, the Senate Environment and Public Works Committee passed a revised version of the Lautenberg Act out of the committee on a bipartisan 15-5 vote.  On May 14, 2015, the House Subcommittee on Environment and the Economy passed a revised version of the TSCA Modernization Act of 2015 out of the subcommittee on a bipartisan 21-0 vote.  UPDATE 5-28-15:  The legislation was formally introduced as H.R. 2576 on May 26, 2015.  The new versions made no significant changes to the testing provisions discussed below.]

While most of the attention around legislation to reform the Toxic Substances Control Act (TSCA) has focused on the issue of preemption, it’s important not to lose sight of how new legislation would address fundamental problems in the current law.  This post will be the first in a series examining flaws in TSCA and how recent bipartisan reform proposals would address them.

The Lautenberg Act, S. 697, is the bipartisan TSCA reform legislation introduced in the Senate in March.  A bipartisan process has also begun in the House, leading to last week’s release of a discussion draft of “The TSCA Modernization Act of 2015.”  In this series of posts, I’ll describe how each of these legislative vehicles would address the specific problematic area of the current law I’m discussing.

First up, EPA testing authority.  

Under current TSCA, for chemicals already on the market, EPA must generally go through notice-and-comment rulemaking (which is usually a multiyear process) before it can require testing. EPA must also first make certain risk or exposure findings in order to require testing, namely that a chemical:

  • may present an unreasonable risk; or
  • is produced in substantial quantities and either enters the environment in substantial quantities or there is or may be significant/substantial human exposure.

The first of these requirements is a Catch-22:  showing potential risk is not an easy task without the data that required testing would provide.  The second requirement is also circular in nature:  for example, a highly potent carcinogen could be harmful at much lower exposures than a weak one, yet this statutory requirement has forced EPA to define “substantial” and “significant” on a generic basis, in the absence of the hazard information that testing would provide and would be used to determine what actually are substantial or significant levels of production, release or exposure of a given chemical.

As a result of these constraints, EPA has managed to require testing on fewer than 300 of the 62,000 chemicals that were on the market and grandfathered in when TSCA was first enacted.

EPA fares a little better with respect to new chemicals, where it can use consent orders to require testing.  EPA staff estimate the agency imposes some type of testing requirement on about 4% of the Premanufacture Notices (PMNs) it reviews.  With EPA having reviewed nearly 40,000 PMNs over the course of TSCA, that’s about 1,600 new chemicals for which any testing was required.

Let’s generously assume all of those chemicals went on to enter commerce (a clear overestimate, as only about half of PMN’d chemicals go on to enter commerce).  This would mean that on the order of 2,000 of the 85,000 chemicals – 2.4% – on the TSCA Inventory have had some testing required under TSCA.  Put another way, EPA has not required any testing of 97.6% of the chemicals that have been on the market at some point since TSCA passed in 1976.

How would TSCA reform legislation address this problem?

The Lautenberg Act allows EPA simply to issue orders to require testing instead of going through rulemaking (though it must justify why it is using an order rather than a rule).  It also strikes the requirement that EPA first show potential risk or high release or exposure in order to require testing.  EPA can require testing to inform all actions it must take, including new chemical reviews, prioritization, safety assessments/determinations and development of risk management rules.  The bill generally requires EPA first to request submission of the needed information before mandating testing; and EPA cannot require testing as a means to establish minimum information sets for chemicals generally.

So, while the bill does not provide EPA with unfettered authority to require testing, it provides a major expansion in such authority, a clear improvement over current TSCA.

The House discussion draft also allows EPA to require testing via issuing an order rather than through a rulemaking, and does not include a requirement that EPA justify why it is using an order rather than a rule.  Except for testing necessary to conduct a risk evaluation, however, the draft retains TSCA’s current Catch-22 requirement that, before it can require testing, EPA must first show potential risk or high release or exposure.

Next up:  New chemicals.

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