By Andrew Rosenberg, Union of Concerned Scientists
A leaked draft of the Environmental Protection Agency’s soon-to-be-released supplemental proposal to their so-called Transparency Proposed Rule from last year shows that the administration is making a bad proposal worse.
The supplement is, in theory, intended to address issues raised in the more than 600,000 comments the agency received last year. All this supplement does, however is underline just how bad this whole effort is. If the proposed rule change succeeds, it will undermine the scientific process at the agency, hinder and make more political than ever policy decisions that are supposed to protect public health and safety, and ensure that the entire process of making decisions slows to crawl, with ever increasing costs to both the public and the agency.
When reading the supplemental proposal I was struck by five points the EPA decided to “clarify”:
It applies to almost all science used by the agency
The proposed rule on transparency that the EPA released last year stated that it would only consider as part of its policy process studies on “dose-response” where all of the underlying raw data, all models, computer code, and any other information pertaining to the study was fully publicly available. The agency’s leaders advanced this idea despite concerns that had been raised many times before in Congress, and at the agency, that it would preclude consideration of critical scientific evidence. Now, in the supplemental proposal, the EPA states that this requirement would be applied not just dose-response studies but to all science the agency relies on for policymaking. The only exception is when a statute specifies the types of studies to be used in implementation.
Apparently, that is intended to clarify that even survey studies, environmental assessments, modeling studies, studies on the impact of pollution in drinking water, and other efforts will only be considered if the data is public. If the data is confidential, no matter how legitimate the reason, the EPA would be required to ignore the science or at least give it less weight.
It is easy to see where this goes. Much of the data on public health impacts cannot and should not be released, because it’s based on private medical records, so this rule would disqualify it from consideration in making regulations to protect public health. On the other hand, industry or other researchers could make public only the data they want the agency to rely on, and keep everything else under wraps. That’s not the way to make informed policy.
It also clarifies that the requirement to make all raw data public doesn’t just apply going forward, but to all studies done in the past that the agency might wish to rely on. Galileo, where is your data? And Sir Isaac, can you show me the raw numbers on gravity?
The supplement also requires that for every study the agency relies on that it should be reanalyzed and that all assumptions should be tested with sensitivity analysis. Their definition of “reanalyze” is “to analyze exactly the same data to see if the same result emerges from the analysis by using the same programs and statistical methodologies that were originally used to analyze the data.”
In other words, the EPA will now check the math and computer coding for every study it intends to rely on. Why? Well, there you have me. All of that error checking is routinely done by competent researchers and their colleagues, and fundamental errors are picked up usually in peer review when the results seem off to expert reviewers. That has certainly been the case throughout my career with every paper I have published or reviewed in more than 30 years.
And sensitivity analysis for all parameters, not just those that are critical? That requires redoing all calculations many many times as each model parameter varies. Usually that is also left to expert reviewers to identify key parameters. Now the rule says just do all parameters, without noting there may be hundreds or even thousands.
That should keep the EPA technical staff re-running analysis pretty much forever, without ever meeting the requirement.
Forget the best science
Almost invariably, scientific recommendations on whether and what policy action is needed rely on the weight of the evidence. That means that experts review all of the studies available, consider how strong the evidence is in each study based on design, sample sizes and comprehensiveness, methods and the clarity of the results, and then consider the overall weight of the evidence collectively. Makes sense doesn’t it?
But the EPA is proposing to jettison all that, and now weigh the evidence primarily on the criterion of the public availability of the raw data (something that is rarely if ever needed to evaluate the scientific merit of a study). In other words, let’s have a bureaucratic standard for how to consider the science, rather than a scientific standard.
Authority? Who needs authority?
For an agency to implement a new regulation, and for it to survive a legal challenge, there must be a clear authority for the action given to the agency by law. In other words, if Congress didn’t say you can do it, by and large, you can’t.
In the proposed rule, for authority the EPA basically just listed a bunch of the statutes it works under; Clean Water Act, Safe Drinking Water Act, Clean Air Act and so forth. The problem is that none of those laws give the agency clear authority to exclude consideration of science by an arbitrary criterion such as public availability of data. So, the supplemental proposal says the real reason they are doing this is because of the authority of the Federal Housekeeping Act of 1966, which says: “the head of an Executive department or military department may prescribe regulations for the government of his department ….”
So, the EPA seems to think it is just tidying up a few things by restricting its staff from considering a vast body of evidence concerning public health, safety, and the environment by an arbitrary standard. Basing a decision like this on the whim of political appointees is a far cry from the sensible, transparent process the administrator claims to be pursuing.
Political influence über alles
Finally, the agency doesn’t clarify one point that has troubled many, many scientists. The proposed rule from last year says that the administrator of the EPA will retain the ultimate authority to waive these requirements for data to be public if a study is to be considered. That means that if the administrator wants a study in, it’s in. And there are no criteria for that decision—it is wholly political. It makes a mockery of the purported “transparency” rationale to be able to suspend it with no explanation.
So what’s the real point of this entire rule-making exercise? It is clearly intended to slow the work of making regulations, and turn it from a science-based process aimed at the broad public interest to a purely political process. Whoever has the most political influence—like the well-funded industries who have been pushing for this change—will win out. And that means the rest of us lose.