COMMENT BY: Tom Bartman

I am an environmental lawyer very concerned with overreaching in some regulatory areas by the Environmental Protection Agency at great economic cost to businesses and American society in general. From this vantage, I applaud the CRE's emphasis on the quality of scientific data generated by government agencies and the CRE's watch list of particular regulations that pose data quality concerns. While I can accept policy differences over the expense an agency might seek to impose on the regulated community to protect public health and welfare, any debate over such costs absolutely must occur on the basis of accurate data and scientifically legitimate and defensible interpretations of such data.

A huge volume of scientific data constitutes an integral component of the workings of our modern regulatory society, and independent peer review is a crucial guarantor of quality of this data. The importance of independent peer review is widely recognized. E.g., National Envt'l Policy Institute, Enhancing the Integrity and Transparency of Science in the Regulatory Process 9 & 15-16 (Draft discussion report Fall 1996). Where, however, an agency tampers with the independence of peer review, potentially to skew the interpretation of data, there can be no assurance of the trustworthiness of data underlying policy debates.

Concretely at issue here is the fact that EPA, an agency which promulgates rules that potentially impose significant and expensive regulatory compliance burdens across large swaths of the U.S. economy, has established a new means of compromising the independence of peer reviews and thereby potentially fudging the production of risk assessments underlying its major rules.

In 1997, EPA entered into an umbrella contract with a consultant to provide miscellaneous services, and then issued a specific work assignment to the contractor to "provide logistical support" for a peer review panel reviewing EPA's update of its risk assessment for benzene. 62 Fed. Reg. 35172 (June 30, 1997). The benzene risk assessment will be the basis of a major Clean Air Act rulemaking for regulation of pollutant emissions from mobile sources (i.e., motor vehicles).

The contractor's primarily logistical duties under the work assignment included arranging for conference facilities; providing panel members with travel, per diem, and associated expenses; resolving issues concerning meeting format, public attendance, and logistics; handling pre-meeting mailings and preparing nametags and placards for panel members; and maintaining an address list and handling copying. More significantly, the contractor was to select peer review panelists from both a list provided by EPA and from its own list and to submit its panel choices for EPA approval; designate a panel chair, "upon consultation with EPA;" and draft and send a summary report of the committee meeting to EPA prior to finalizing the report.

To itself, EPA reserved the following responsibilities and veto authorities:

• provide the contractor with a list of potential panel members;
• approve the contractor's proposed panel choices (a right reserved in bold typeface in EPA's work assignment);
• approve the contractor's proposed choice of panel chair;
• review the committee members, written comments prior to the panel's public meeting;
• review and provide written comments on draft report of the panel's meeting, prior to finalization of report.

One possible result of this arrangement was that the selected panel included a panelist who had a scientific conflict of interest that favored EPA.

EPA has issued a schedule of peer reviews that makes clear that contractor-led peer reviews will constitute a significant proportion of approvals of its risk assessments for pollutants it is required to assess under the Clean Air Act. The primary reason for EPA's use of this device may be its interest in avoiding the open government provisions of the Federal Advisory Committee Act ("FACA"), but it may well also be interested in ensuring the presence of favored and favorable scientists on peer review panels that review its science. Meanwhile, it is not clear that EPA saves any money in performing peer reviews by means of this device instead of through its established, FACA-compliant Science Advisory Board or Council of Scientific Advisors.

EPA's proxy-peer-review device is a terrible example for other agencies potentially interested in avoiding FACA's openness provisions. (So far, EPA's use of this device has been sanctioned under current law by the Court of Appeals for the D.C. Circuit.) Following this model, why would an agency requiring outside advice or judgment needlessly saddle itself with FACA compliance requirements? Particularly an agency that might prefer to avoid public scrutiny would not bother with such requirements, nor would agencies that find FACA burdensome. (The District Court for the District of Columbia has cited "attempt[s] by the Executive Branch to escape the toils of FACA".) All this at serious cost to the independence of peer reviews.

It may require action by Congress to stop EPA's compromising the independence of scientific peer review in its apparent attempts to circumvent FACA. Certainly, an agency that seeks advice through channels subject to FACA should not simply be able to hire a contractor as a proxy and thereby evade FACA. The broader question, however is the necessary independence of peer review and its role in protecting the integrity of scientific data. The CRE's focus on the quality of data generated by federal agencies and Congress' growing interest in scientific accuracy and the value of independent peer review will be crucial in this effort.