Four New (Old) Ways the White House is Trying to Restrict Science for Policymaking

April 25, 2019
Andrew Rosenberg
Director, Center for Science and Democracy

Update 4/25/19 3:30 pm: This post previously erroneously indicated a study of costs was conducted by GAO. That study was carried out by CBO, the Congressional Budget Office

Yesterday, the Office of Management and Budget (OMB) in the White House issued new “guidance” for the Administration to “Improve Implementation of the Information Quality Act”. Unfortunately, it reads like a re-hashing of some of the worst ideas for restricting the use of science in policymaking from the last five years or so. Way back in 2015, when some members of Congress were trying some of these same tricks to tip the scales in favor of regulated industry we summarized them in a Policy Forum article in Science. Here we go again—but this time, the Trump administration is trying to push these changes through unilaterally, the latest round in a long list of efforts to push science to the sidelines.

False transparency

For literally years now, my colleagues and I have been writing about proposals that supposedly are designed to increase the transparency of the regulatory process by requiring that all data and other information be made public. It sounds like a good idea, but isn’t because there are many justifiable reasons that some information used in a scientific study, such as personal medical information, must be kept private. The effect of the requirements to make all data and information public, actually greatly restricts the scientific studies that agencies can rely on, because they can’t and shouldn’t make confidential information public. For example, if a study of health impacts of a pollutant relies on health outcomes in a certain city or town, that health information for individuals may not be released publicly but the study might be critical for understand the population level effects of the pollutant.

OMB takes another shot at this issue but the results are murky. To be fair, they go to great lengths to assert the importance of privacy and to require agencies to protect privacy. But here’s the rub. If an agency does as required to protect privacy, can they use the analysis based on that private data in making decisions? The directive is unclear on this point, but it looks like they can’t or are strictly limited in doing so. And if that is the case, then for example, the EPA which is charged with protecting public health, may be restricted from using public health studies in implementing rules – and that makes no sense at all.

Déjà vu all over again

The directive has a long section on reproducibility of scientific results. And once again this idea is based on an oversimplification of how science works. Maybe we all learned that doing an experiment in a lab many times over can give you confidence in the results and that is the “scientific method.” Made sense in grade school. But lots and lots of critical scientific information and even analyses are not “reproducible” in this sense. Take, for example, the impact of a toxic pollutant on a local community. Should we release it again to see if it is really harmful? Or the study of a natural disaster? Should we wait for it to happen again to reproduce the results? The Environmental Data and Governance Initiative illustrated the many real-world examples of scientific studies that are neither feasible nor ethical to reproduce.

The directive requires that important studies include sufficient descriptions of the data and analyses to allow them to be “reproduced by qualified third parties” to test them. Now, I am all for detailed descriptions of data and methods being made clear and public, but for most studies, that wouldn’t allow the reproduction of the analyses except by statistical simulation. And who are the third parties? Regulated industry. And the directive requires that all computer code be released, even though it may be proprietary. So if you can’t release the data because of privacy concerns and you can’t “reproduce the study by a third party” because they need the raw data, what happens? Does that mean you can use a critical analysis of health impacts because of these barriers? The directive doesn’t say.

Correction or delay?

Finally, the directive requires that agencies allow the “public” to challenge data and technical information. Agencies then must respond in writing and have independent experts review their response before it is finalized. And then the “public” can challenge the response, and around the circle we go again.

Why did I put “public” in quotes? Because essentially this is designed as an avenue for regulated industry to challenge and delay regulatory actions once again. Don’t get me wrong, the public should have the right to comment and raise issues on agency proposals including the science basis for a decision. And an agency should be required to address those concerns in writing as they are currently required to do under the Administrative Procedures Act and other legal requirements. That has been the case for a long time. But this directive takes the process to a whole new level of obfuscation. Essentially it allows a near endless and expensive round of challenges to every detail of the information an agency uses, including on how well the agency complied with this directive itself, until everyone is satisfied – i.e. never. And who has the resources to make all these challenges? Really only the regulated industry. This bad old new idea is an echo of the unlamented “Regulatory Accountability Act” that was proposed before that put a huge number of barriers in the way of agency rulemaking, stuffing the process with ever more bureaucracy for agencies decision-making.

No consultation no evaluation

This memo was issued yesterday without warning, even to those of us who follow science policy closely. It broadly affects the way federal agencies will use science in the regulatory process, changes the interpretation of existing guidance, and inserts a whole new set of requirements. So, you would think a memo of such importance would be accompanied by a careful rationale for the what problems are being addressed and how it should be implemented including the costs of doing so. You would be wrong. There is no such analysis. Other than saying decision-making depends on high-quality information (I agree), the directive doesn’t say how the requirements it contains will address current problems that agencies or the public face with the information that is the basis of regulation. Unfortunately, we have seen that time and again in the Trump Administration – new rules proposed with virtually no coherent justification or analysis of their impacts.

In the case of this directive, even though it doesn’t estimate the costs of implementation as it should, we do have some idea from a CBO report on one of the proposals it contains on making data publicly available. For one agency alone (EPA) the CBO estimated that the cost would be in the hundreds of millions. Where is that money and the time of agency staff to come from? You guessed it, it will be diverted from doing the science that is the basis for public health, safety and environmental protections.

Overall, OMB’s “new” directive on information quality isn’t new, but it is directive. It resurfaces bad ideas that were stopped because – you guessed it- they were bad ideas. But now, even though this memo was issued without any public input or analyses, it can direct agency action across the government. The Trump Administration has gone to great lengths to sideline science in decision making. Congress needs to step in and tell the administration this is not the intent of the law and doesn’t serve the public. It is only of benefit to big regulated industries and the detriment to the rest of us.