EPA practices are hindering transparency and public confidence in TSCA’s new chemicals program

Stephanie Schwarz, J.D., is a Legal Fellow.  Richard Denison, Ph.D.is a Lead Senior Scientist.

Part 1               Part 2               Part 3               Part 4

This is our final post in a series spurred by our review of 69 public files for new chemicals we received from EPA’s Docket Center.  For most of these chemicals, EPA made a determination that they are “not likely to present unreasonable risk” under the Toxic Substances Control Act (TSCA), which greenlights their entry into commercial production.

In our previous post we demonstrated EPA is not complying with a number of provisions under TSCA that require the agency to make public the premanufacture notices (PMNs), notices of commencement (NOCs), and information that is submitted with them.  In this post we look further into how, through these failures and others, EPA has impeded meaningful transparency in the new chemicals program.

As originally enacted in 1976, TSCA recognized the value of public access to information, like health and safety information (see, e.g., TSCA § 14(b)).  Even in EPA’s original (1983) regulations establishing the new chemicals review program, EPA recognized that “[p]ublic participation cannot be effective unless meaningful information is made available to the interested persons” (see here p. 21737).  Among the many flaws of the original TSCA, however, was the law’s inability to ensure EPA delivered the promised transparency when it came to both information EPA receives and the agency’s decisions on new chemicals.

The amendments to TSCA in 2016 were meant, in part, to expand public access to information about both chemicals and agency decisions, and in doing so increase public confidence.  For instance, under § 26, EPA must now make available to the public “all notices, determinations, findings, rules, consent agreements, and orders.”  And under § 5, EPA must now make an affirmative determination on new chemicals, which under § 26 must be made public.  These changes, in addition to the original TSCA provisions, clearly envision a robust program under which the public is able to readily access non-confidential information on new chemicals and information on EPA’s decisions about them.  

Coupled with the policy changes EPA has made, the concerns we raise here make clear that EPA under this Administration intends to weaken a new chemicals program that Congress sought to strengthen through TSCA reform – and hide as much of it from public view as possible.

As implemented, however, a number of features of the new chemicals program severely hamper the ability of the public to understand EPA’s decision-making or engage in the new chemicals program.  In addition to the failings we have discussed in previous posts in this series, this post will address several others:

  • the convoluted and fragmented public information “system” EPA has created for PMNs;
  • the failure of EPA to provide access to agency-generated health and safety information on PMN substances; and
  • EPA’s failure to publish Notices of Commencement (NOCs) and EPA’s determinations on confidentiality claims for specific chemical identity in those NOCs.

1. Tracking the PMN review process is difficult at best.

Identifying the existence of a PMN, locating it and information associated with it, and tracking it through the review process require one to engage in a convoluted search of multiple disjointed websites and to make requests to the Docket Center, each of which has major limitations – and still leaves major gaps in information that should be publicly available:

Federal Register notices of PMNs received:  Lists of PMNs received by EPA are supposed to be published at least monthly in the Federal Register.  However, these lists lag many months behind (the most current one in the Federal Register is for PMNs received in October 2017).  In addition, as discussed in our last post, there are often multiple entries for the same PMN, which may or may not refer to different versions of the PMN itself; some apparently relate to EPA’s receipt of other information associated with a PMN, but there is no way for the public to tell.  Nonetheless, this source is the only place one can learn who submitted the PMN (unless that company’s name is claimed confidential, which is quite common).  Also listed is the chemical’s intended use and name – unless these are claimed confidential (also very common), in which case a generic use description and generic name are provided instead.

PMN status tracker:  This online table, which can be searched for a particular PMN number, is advertised as showing “interim recommendations and final determinations” on a PMN.  EPA began displaying the final determinations in the same table as the interim recommendations about 18 months ago, which was very helpful.  The table also has filters and sort-by-date buttons, though the latter have never worked properly.

EPA’s website (see the “After you submit” tab) directs PMN submitters to use this table to check on the status of their PMNs:

Many submitters want to follow the progress of their substances at intermediate points in the review process. Status reports on notices submitted to EPA under section 5 of TSCA are posted to the Status page within 14 days of a decision being made at the EPA Focus meeting.

Even though the table still claims to provide “interim recommendations,” EPA abruptly stopped doing this in August 2017.  In December, EPA started adding entries again, but for virtually all of the entries added to the table since last summer, the table now merely indicates that a “Focus Meeting Occurred” for the chemical.  This denies the public any knowledge of whether EPA professional staff identified initial concerns or a lack of sufficient information on a new chemical. There are now more than 200 such nearly useless entries.

Once a final determination is made, it shows up here relatively quickly, and 2-3 weeks later EPA inserts a link to the relevant decision document (a “determination document” for “not likely” findings, or a “consent order” for other determinations) housed in its ChemView database, which is helpful.  These documents have to be retrieved one at a time, however, making it very difficult to conduct broader analyses across multiple PMNs.

The table does not provide the company or chemical name (whether specific or generic) or the use (whether specific or generic).  That requires going back to the Federal Register notices or extracting the information from the decision documents.

Getting the actual PMN itself:  None of what we’ve described so far gets you access to the PMN itself (not even a redacted version “sanitized” of claimed CBI), nor does it provide access to any information associated with the PMN.  As we blogged about before, to get the PMN one has to submit a request to EPA’s Docket Center and wait a couple weeks until a CD-Rom arrives.  Earlier episodes in our blog series have noted the major gaps and deficiencies in such information.

Notices of commencement (NOCs):  The only way to know whether a new chemical has actually entered commercial production is to determine whether a NOC has been filed for it.  The only way to do that is to go back to the Federal Register notices, where EPA also publishes a list of the NOCs it has received in a given month, identified by PMN number.  As noted earlier, the publication of these notices is lagging months behind.

This step does not get one to the NOC itself, however.  Despite the fact that EPA’s own regulations (40 C.F.R. §§ 700.17(b)(1), 720.95) require NOCs to be made electronically accessible to the public, EPA has failed to do so and we have yet to identify a means by which the public can obtain them.  We discuss this issue further below.

We hope the above description gives you a feel for what it takes to track even a single PMN through the process.  Imagine now trying to make sense of patterns of EPA decisions by looking across multiple or all PMNs in a given time period.  We’ve been doing just that, but it has required painstaking and tedious work to extract data from multiple sources and integrate it into linked spreadsheets.  Even with this effort, we are increasingly stymied by information EPA is intentionally now withholding from the public.

So much for transparency.

2. No EPA-produced health and safety information is present in the public files.

During EPA’s review of a new chemical under § 5, the agency typically generates health and safety information about the PMN substance, and EPA’s final determinations generally indicate when EPA has done so (see, P-16-0580).  For example, EPA uses EPI (Estimation Programs Interface) Suite to estimate a number of physical-chemical and fate properties, including the potential of a substance to bioaccumulate and to biodegrade in the environment.  EPA also uses the Ecological Structure Activity Relationships (ECOSAR) Predictive Model to estimate certain environmental hazards of a new chemical.  Both of these programs generate information that constitutes “health and safety studies” as defined under both TSCA (§ 3(8)) and EPA’s implementing regulations, which is broadly defined to include “any data that bear[s] on the effects of a chemical substance on health or the environment.” This and other information EPA generates are summarized in various standard reports for each new chemical.  Such information clearly should be included in the public files for a PMN.

We reviewed the 69 determination documents associated with the PMNs for which we received public files. This review revealed that EPA referred to information it generated using EPI Suite or ECOSAR at least 35 times.  Yet none of these data are included in the public files we received, even though EPA occasionally provides this type of agency-generated information in other contexts, such as in the dockets it is required to establish when it proposes a Significant New Use Rule (SNUR) (see, e.g., here, S-17-0004).

The EPA-generated health and safety information is critical in EPA new chemical reviews, especially because few PMN submitters provide any health and safety information (see here p. 8, and here Q118-5).  EPA’s failure to disclose the health and safety information it develops in its review of a new chemical leaves the public in the dark about how a decision was made, and without access to any health and safety information on the PMN substance.  How can the public have any confidence in new chemicals that enter the market when EPA is failing to make public the only health and safety information available on those chemicals?

3. Notices of Commencement and Evidence of EPA CBI determinations are unavailable.

We also discovered that 22 of the new chemicals we requested public files for had notices of receipt by EPA of Notices of Commencement (NOC) published in the Federal Register (as of the latest published list for October 2017).  We reviewed these notices of receipt of NOCs in light of mandates that EPA:

  • place Notices of Commencement (NOCs) themselves in the public files (40 C.F.R. § 720.80(b)(2)(ii));
  • require substantiation for and review all confidentiality claims for specific chemical identity when a NOC is received (40 C.F.R. § 720.102(c)(2));
  • publish its determinations on those claims (TSCA § 26(j)(1)); and
  • apply a unique identifier to each specific chemical identity for which EPA has approved a confidentiality claim (TSCA § 14(g)(4)).

Our review revealed a host of concerns:

  1. The NOCs are not in any of the public files we received;
  2. The Federal Register notices suggest that EPA has received multiple NOCs for some PMNs;
  3. Eighteen of these chemicals are still identified by their generic names, even though EPA has published no determinations or provided any other indication that it has reviewed and approved the associated CBI claims for specific chemical identity; and
  4. None of the chemicals with generic names have been assigned a unique identifier.

Below is a bit more detail:

First, EPA’s regulations state that when companies submit information to EPA, if any information is claimed confidential, the submitter must provide EPA with a sanitized copy, which EPA “will place” in the public file.  These requirements apply to NOCs:

  • 40 C.F.R. § 720.80 discusses notices containing CBI claims submitted under Part 720.
  • Part 720 includes § 720.102 that is specific to NOCs and hence makes clear NOCs are subject to § 720.80.
  • Section 720.80(b)(2)(ii) states sanitized notices are to be placed in the public file.

Yet EPA is failing to place sanitized copies of the NOCs it receives in the public files.

Second, it is unclear why, in some cases, the same PMN number appears in the NOC lists in multiple Federal Register notices (e.g., search the Federal Register for, P-16-0281 and P-17-0158).  While this likely does not actually mean multiple NOCs were received (similar to what we highlighted for PMNs in our previous post), it is yet another example of EPA’s information management system failing to provide even a modicum of transparency on new chemicals.

Third, and more importantly, a number of the “NOC’d” chemicals we reviewed are still identified only by their generic names (i.e., the chemical identities in the NOCs presumably have been claimed CBI).  When submitters file NOCs, they are required to re-assert and substantiate any claim that a chemical identity should remain CBI.  EPA must then review the claim and publish its determination as to whether or not the claim is warranted.  Yet EPA has provided no means for the public to know whether the required substantiations were submitted and whether EPA has conducted the required reviews of the claims.  If it has conducted those reviews, EPA has not published any of its determinations on these claims.  These violations are not only troubling, they are further examples of EPA denying the public any transparency as to how the new chemical’s program is operating.

Fourth, even assuming EPA has approved these claims, EPA has not applied a unique identifier to any of the chemicals with specific chemical identities that are being protected as CBI, as required by amended TSCA.  A unique identifier is exactly what it sounds like: where EPA approves a claim to keep a chemical substance’s specific chemical identity confidential, EPA must assign the chemical substance a “unique identifier,” which is then to be applied to all information relevant to that chemical substance.  Once EPA applies a unique identifier to a chemical substance, the public can then identify all non-confidential information on the chemical that is linked using that unique identifier, which would allow access to information that would otherwise not be able to be linked to the substance.  EPA’s failure to implement this provision compounds the lack of transparency in the new chemicals program.

* * *

This series of blog posts has identified a large number of pervasive and systematic flaws and information gaps across all stages of EPA’s new chemicals review process that deny the public access to information on new chemicals and agency decisions about them to which the public is entitled.  Coupled with the policy changes EPA has made that we have also blogged about and commented on, the concerns we raise here make clear that EPA under this Administration intends to weaken a new chemicals program that Congress sought to strengthen through TSCA reform – and hide as much of it from public view as possible.

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