One of Many Risks of the Regulatory Accountability Act: Flawed Risk Assessment Guidelines
By Genna Reed, Union of Concerned Scientists
Tomorrow, the Senate will begin marking up Senator Rob Portman’s version of the Regulatory Accountability Act (RAA), which my colleague Yogin wrote a primer about last week. This bill is an attempt to impose excessive burdens on every federal agency to freeze the regulatory process or otherwise tie up important science-based rules in years of judicial review.
One of the most egregious pieces of this bill as an affront to the expertise at federal agencies is the provision ordering the White House Office of Management and Budget’s (OMB) Office of Regulatory and Information Affairs (OIRA) to establish guidelines for “risk assessments that are relevant to rulemaking,” including criteria for how best to select studies and models, evaluate and weigh evidence, and conduct peer reviews. This requirement on its own is reason enough to reject this bill, let alone the long list of other glaring issues that together would fundamentally alter the rulemaking process.
The RAA is a backdoor attempt at giving OIRA another chance to try and prescribe standardized guidelines for risk assessment that would apply to all agencies, even though each agency conducts different types of risk assessments based on statutory requirements.
OIRA should not dole out science advice
The way in which agencies conduct their risk assessments should be left to the agencies and scientific advisory committees, whether it is to determine the risks of a pesticide to human health, the risks of a plant pest to native plant species, the risks of a chemical to factory workers, or the risks of an endangered species determination to an ecosystem. Agencies conduct risk assessments that are specific to the matter at hand; therefore an OIRA guidance prescribing a one-size-fits-all risk assessment methodology will not be helpful for agencies and could even tie up scientifically rigorous risk assessments in court if the guidelines are judicially reviewable.
OIRA already tried writing guidance a decade ago, and it was a total flop. In January 2006, OMB released its proposed Risk Assessment Bulletin which would have covered any scientific or technical document assessing human health or environmental risks. It’s worth noting that OIRA’s main responsibilities are to ensure that agency rules are not overlapping in any way before they are issued and to evaluate agency-conducted cost-benefit analyses of proposed rules. Therefore OIRA’s staff is typically made up of economists and lawyers, not individuals with scientific expertise appropriate for determining how agency scientists should conduct risk assessments.
OMB received comments from agencies and the public and asked the National Academy of Sciences’ National Research Council (NRC) to conduct an independent review of the document. That NRC study gave the OMB a failing grade, calling the guidance a “fundamentally flawed” document which, if implemented, would have a high potential for negative impacts on the practice of risk assessment in the federal government. Among the reasons for their conclusions was that the bulletin oversimplified the degree of uncertainty that agencies must factor into all of their evaluations of risk. As a result, the document that OIRA issued a year later, under Portman’s OMB, was heavily watered down. In September 2007, OIRA and the White House Office of Science and Technology Policy (OSTP) released a Memorandum on Updated Principles for Risk Analysis to “reinforce generally-accepted principles for risk analysis upon which a wide consensus now exists.”
Luckily, in this case, the OMB called upon the National Academies for an objective review of the policy, which resulted in final guidelines that were far less extreme. As the RAA is written, it does not require that same check on OIRA’s work, which means that we could end up with highly flawed guidelines with little recourse. And the Trump administration’s nominee for OIRA director is Neomi Rao, a law professor whose work at the George Mason University Law School’s Center for the Study of the Administrative State emphasizes the importance of the role of the executive branch, while describing agency policymaking authority as “excessive.” I think it’s fair to say that under her leadership, OIRA will not necessarily scale back its encroachment into what should be expert-driven policy matters.
Big business is behind the RAA push
An analysis of OpenSecrets lobbying data revealed that trade associations, PACs and individuals linked to companies that have lobbied in support of the RAA also contributed $3.3 million to Senator Rob Portman’s 2016 campaign. One of the most vocal supporters of the bill is the U.S. Chamber of Commerce, whose support for the bill rests on the assumption that we now have a “federal regulatory bureaucracy that is opaque, unaccountable, and at times overreaching in its exercise of authority.” Yet this characterization actually sounds a lot to me like OIRA itself, which tends to be fairly anti-regulatory and non-transparent, and has a history of holding up science-based rules for years without justification (like the silica rule). Senator Portman’s RAA would give OIRA even more power over agency rulemaking by tasking the agency with writing guidelines on how agencies should conduct risk assessments and conveniently not requiring corporations to be held to the same standards.
When OIRA tried to write guidelines for risk assessments in 2006, the Chamber of Commerce advocated for OIRA’s risk assessment guidelines to be judicially reviewable so they could be “adequately enforced,” claiming that agencies use “unreliable information to perform the assessments,” which can mean that business and industry are forced to spend millions of dollars to remedy those issues. It is no wonder, then, that the Chamber would be so supportive of the RAA, which would mandate OIRA guideline development for risk assessments, possibly subject to judicial review. OIRA issuing guidelines is one thing, but making those guidelines subject to judicial review ramps up the egregiousness of this bill. All sorts of troubling scenarios could be imagined.
Take, for example, the EPA’s human health risk assessment for the pesticide chlorpyrifos, which is just one study that will be used for the agency’s registration review of the chemical, which has been linked to developmental issues in children. The EPA sought advice from the FIFRA Scientific Advisory Panel on using a particular model to better determine a chemical’s effects on a person based on their age or genetics and to predict how different members of a population would be affected by exposure, called the physiologically-based pharmacokinetic and pharmacodynamic (PBPK/PD) model. The agency found that there is sufficient evidence that neurodevelopmental effects may occur at exposure levels that are well below previously measured exposure levels.
If OIRA were to produce risk assessment guidelines that were judicially reviewable, the maker of chlorpyrifos, Dow Chemical Company, could sue the agency on the grounds that it did not use an appropriate model, consider the best available studies, or that its peer review was insufficient. This would quickly become a way for industry to inject uncertainty into the agency’s process and tie up regulatory decisions about its products in court for years, delaying important public health protections. A failure to ban a pesticide like chlorpyrifos based on inane risk assessment criteria would allow more incidences of completely preventable acute and chronic exposure, like the poisoning of 50 farmworkers in California from chlorpyrifos in early May.
“Risk assessment is not a monolithic method”
A one-size fits all approach to government risk assessments is a bad idea, plain and simple. As the NRC wrote in its 2007 report:
Risk assessment is not a monolithic process or a single method. Different technical issues arise in assessing the probability of exposure to a given dose of a chemical, of a malfunction of a nuclear power plant or air-traffic control system, or of the collapse of an ecosystem or dam.
Prescriptive guidance from OIRA would serve to squash the diversity and flexibility that different agencies are able to use depending on the issue and the development of new models and technologies that best capture risks. David Michaels, head of OSHA during the Obama Administration, wrote in his book Doubt Is Their Product that regulatory reform, and in this case the RAA, offers industry a “means of challenging the supporting science ‘upstream.’” Its passage would allow industry to exert more influence in the process by potentially opening up agency science to judicial review. Ultimately, the RAA is a form of regulatory obstruction that would make it more difficult for agencies to issue evidence-based rules by blocking the use of science in some of the earliest stages of the regulatory process.
The bill will be marked up in the Senate Homeland Security and Governmental Affairs Committee tomorrow, and then will likely move onto the floor for a full senate vote in the coming months. Help us fight to stop this destructive piece of legislation by tweeting at your senators and telling them to vote no on the RAA today.