Dear Mr. Charney:

The Center for Regulatory Effectiveness would like to applaud OMB for moving forward on implementation of the data access provision of P.L. 105-277. We believe that implementation of the new law is integral to the public's continued ability to participate effectively in regulatory policy development. We are particularly pleased to see that OMB reaffirmed its commitment to issue the final rule revising Circular A-110 by the end of the current fiscal year, September 30, 1999.

CRE submits the following comments on OMB's proposed revisions to Circular A-110, in response to the Federal Register notice of August 11, 1999 (64 Fed. Reg. 43786).

Definition of "Data"

CRE recommended in its initial comments that OMB adopt a definition of data that was broad enough, at a minimum, to allow for replication of the research study results, and we are pleased that this suggestion was substantially adopted in the August reproposal. CRE also notes that OMB has clarified that the data access provision excludes "preliminary analyses, drafts of scientific papers, plans for future research, peer reviews, or communications with colleagues [and] physical objects (e.g., laboratory samples)". The reproposal therefore seems to address many commenters' concerns regarding the possible disclosure of meeting notes, laboratory samples, secondary sources, etc.

CRE is very concerned, however, that the August 11 reproposal effectively allows the recipient, not the federal awarding agency, to determine whether certain data or information qualify as "research data" for purposes of the public access provision. The proposed revision requires award recipients to make their "research data" publicly available, but the reproposal also says, in relevant part:

Research data . . . do not include: (A) Trade secrets, commercial information, materials necessary to be held confidential by a researcher until publication of their results in a peer reviewed journal, or information which may be copyrighted or patented; and (B) Personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy, such as information that could be used to identify a particular person in a research study.

Our specific concern with the above exclusionary portion of the "research data" definition is that the preamble to the reproposal states:

OMB proposes to define "research data" in a way that does not require recipients to transmit information which, in their judgment, includes "trade secrets, commercial information", or "personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy." The Federal awarding agency would retain its right to ask the recipient for additional information, if it believed the recipient's application of these principles was improper.

64 Fed. Reg. at 43789, col. 1 (emphasis added). OMB's reproposal thus effectively gives the recipient, the person in possession of the data, primary authority to determine whether the data may be, for example, or "trade secrets", and therefore not "research data" within the OMB definition. The same broad discretion would be granted to researchers to categorize their data as "preliminary analyses" or "communication with colleagues" or to otherwise conclude that their data fall outside the definition of "research data".

This procedure is an unwise delegation of the awarding agency's decision making authority to a party that may often be the least interested in sharing the data with outsiders. While we are encouraged that OMB proposes a saving provision allowing the awarding agency to request the recipient for additional information, the delay inherent in this apparent appeal process could significantly impact fair and timely access to research data. Moreover, appeals by disappointed FOIA requesters through this secondary agency review provision could easily become the rule rather than the exception. CRE believes the sounder approach is for the awarding agency to review the data in the first instance, guided by the traditional strictures of FOIA disclosure exemptions and the added protections afforded by the OMB's data access rules.

The data access statute requires that the data be made available to the public "through procedures established under [FOIA]." These procedures include federal agency regulations and established practice and judicial case law that govern determinations regarding what information is subject to the exclusions in FOIA, specifically 5 U.S.C. § 552(b). The entities with the legal responsibility to comply with FOIA and with the public access provision of P.L. 105-277 are OMB and the federal awarding agencies. OMB's proposal to empower the recipient to determine whether certain data or information is or is not "research data" that must be disclosed is an illegal delegation of FOIA responsibilities, which are obligations of the federal agencies.

If OMB chooses to give the recipient the power to determine what are or are not "research data" that are potentially subject to public disclosure, OMB at a minimum should also revise Circular A-110 to require the recipient to identify the general substance of each item of information that the recipient determines is not "research data". Absent a specific requirement that the recipient identify what data or information the recipient is not disclosing, neither the federal awarding agency nor the FOIA requester will be able to know that data or information is in fact being withheld by the recipient. The requirement to prepare a list of withheld data or information should track existing FOIA practices that require agencies to identify generally each agency record that the agency is withholding, including identification of the particular FOIA exclusion upon which the decision to withhold is based. If such a list does not exist, the requester has no way to challenge, either through the agency or in court, the propriety of the recipient's decision to withhold one or more pieces of potential "research data".

CRE is also deeply concerned that the August reproposal excludes from the definition of "research data" any "information which may be copyrighted or patented". (Emphasis added.) Under U.S. law, copyright is available for "original works of authorship fixed in any tangible medium of expression." 17 U.S.C. § 102(a). Complying with the "formalities" of copyright registration and notice (e.g., affixing the © copyright symbol) are not even necessary for material to "be copyrighted". The Copyright Act provides protection for compilations of data merely if they are "selected, coordinated, or arranged in such a way that the resulting work as a whole constitutes an original work of authorship." 17 U.S.C. § 101. Virtually any compilation of research data can be copyrighted, and OMB's proposed exception to the "research data" definition therefore would swallow the rule. Award recipients could routinely copyright their compilation of part or all of the data developed during each research project, thereby categorically excluding the data from public access.

Similarly, any recipient could withhold from public access all "information which may be [. . .] patented". CRE's understanding is that information or a useful article or other patentable product of the information cannot be patented as easily as it can be copyrighted. However, any recipient could take the position that the information developed in the research project "may be" patented and could on that basis decide that the information is not "research data" that must be provided to the federal awarding agency so that it can be made available to the public.

CRE suggests strongly that OMB conform wherever possible to established FOIA legal categories and procedures as OMB implements the public access provision. The responsibility under FOIA for determining whether information can be excluded from public disclosure lies with the federal agency, not with the award recipient, and OMB should make that clear in the revision to Circular A-110.

Definition of "Published"

In its February 4, 1999 proposed revisions, OMB initially limited the scope of the data disclosure rule to "published research findings". Commenters raised concern that vagueness of the term "published" could require the release of data to the public even before the grantee was able to formally present the results. It was argued that such prepublication disclosure requirements could create a considerable disincentive for conducting research, and could result in circulation of drafts and preliminary analyses before the researcher approves of or intends such dissemination.

Many groups suggested that "publication" be defined narrowly to mean only "publication in a scientific journal after formal peer review." CRE believes that this definition is too narrow. Under such a restrictive definition, award recipients would be able to avoid disclosure of data that would otherwise be available to the public merely by failing to submit the data to a peer reviewed journal.

CRE noted in previous comments that every willful release of results or conclusions by a scientist to the public domain could, in a real sense, be characterized as "publication". In some cases a legitimate public need for access to data will arise well before research papers are officially accepted for publication in a peer reviewed scientific or technical journal. CRE cited, for example, posters at scientific conferences that display recent research results. Such materials are often publicly displayed, and the conclusions they present have been relied upon in some cases by government officials in developing policy.

CRE therefore strongly supports, as a reasonable first step, OMB's proposal to consider the research "published" when it is cited by the government in support of a federal government action. CRE believes that research that is either unpublished or published in a non-peer-reviewed journal, including any results or conclusions that are made public by the researcher in the trade press or other public media or in a public forum such as an open scientific conference, should be made available for public scrutiny if they are cited in support of a federal government action. Indeed, it is precisely this type of non-peer-reviewed study that most warrants thorough examination by the public.

"Used by the Federal Government in Developing Policy or Rules"

OMB's February 4 proposed revision stated that the new data access provision would apply to research findings that the federal government used in developing "policy or rules." The proposal did not define the terms "rules" or "policy," however, and many commenters noted this omission. CRE commented on April 5, 1999 that the term should be expanded to include surveys, risk assessments, and federal reports. We recommended that Circular A-110 be clarified to make certain that data that form the bases of studies underlying these important government initiatives also must be publicly disclosed unless otherwise exempted.

In its August 11 reproposal, OMB indicated that it is considering limiting the scope of the rule to "regulations" only, and invited comment on this narrower interpretation of the Shelby Amendment. In issuing its narrowed reproposal, OMB stressed the administrative difficulties that could arise if the data access law were made applicable to other agency actions that do not rise to the level of rulemaking.

While CRE continues to believe that its original position is justified, we also understand OMB's desire for a bright line test in determining the law's application. Although OMB's August 11 position differs from CRE's April recommendation, the reproposal may be a workable first step in implementation. OMB could start with its August position and see how the system works. If there are too many risk assessments or other important activities that fall through the net and are not covered by OMB's revisions to Circular A-110, however, then OMB should, by necessity, expand the scope of government activities covered by the data access provisions of the Circular at a later time.

We also suggest that OMB clarify that the public access provision is triggered when the research study is cited in a publicly available federal government document that clearly is destined to be part of the administrative record for a rulemaking, including at points in time before the agency issues a Notice of Proposed Rulemaking (NPRM). For example, when EPA is conducting its periodic review of an existing National Ambient Air Quality Standard (NAAQS) as required by the Clean Air Act, EPA's scientific staff prepares a "Criteria Document" which is essentially a survey and discussion of recent research involving the health and ecological effects of that particular NAAQS substance. The "Criteria Document" becomes the basis for EPA's legal, policy, and regulatory deliberations regarding whether and to what extent the current NAAQS should be amended.

CRE believes that national policy favoring informed public debate on the development of individual regulations requires allowing public access to the data in federally-funded research that EPA identifies in the "Criteria Document" as a study relevant to the agency's impending reassessment of the NAAQS. There is no valid policy reason for denying the public access to data from an A-110-funded study until after the NPRM is issued, where EPA has specifically identified the study in the "Criteria Document" -- perhaps years prior to the publication of the NPRM -- as a study that is relevant to the agency's reassessment of the NAAQS.

With regard to limiting the public access provision to research supporting regulations that have a "$100 million impact threshold" (64 Fed. Reg. at 43791, col. 1), CRE believes that the public would be ill served by such a sweeping limitation. The vast majority of federal regulations, many of which impose considerable burdens on small business and smaller industrial sectors, do not rise to the $100 million level. In addition, OMB would be acting contrary to the spirit of P.L. 105-277 and would be virtually inviting substantial legal and political controversy if it imposed the proposed "$100 million threshold" in the final revision to Circular A-110.

Cost Reimbursement

CRE believes that OMB correctly interprets the statutory user fee language in the data access provision in P.L. 105-277 as providing authority for a federal awarding agency to retain the user fees that the agency collects, in order to reimburse recipients and applicable subrecipients for reasonable incremental costs incurred in responding to a FOIA request for research data. CRE also agrees with the statement in the proposed last sentence of § __.36(d)(1) that the fee charged for the purpose of reimbursing the recipient or subrecipient "is in addition to any fees the agency may assess under the FOIA (5 U.S.C. 552(a)(4)(A))."

OMB's authority to require agencies to implement this cost reimbursement mechanism is well within OMB's discretion as the agency charged with implementing the public data access provision. Under Chevron USA v. Natural Resources Defense Council, Inc, 467 U.S. 837 (1984), federal agencies are accorded considerable deference in their interpretations of statutes they are entrusted to administer. Here, Congress clearly intended that researchers should be compensated for the time and expenses they incur in retrieving data as a result of FOIA requests. This view is supported both by language of the statute and by public statements and formal comments of Senator Shelby himself. OMB is therefore correct in adopting an approach that will secure agencies' payment of funds, generated through collection of reasonable FOIA user fees, to the responding researchers.

CRE does suggest that OMB issue, as part of its revision to Circular A-110 or as binding guidance through another appropriate mechanism, a standardized schedule of reasonable fees that recipients and subrecipients may receive for their time and expenses. OMB issued such a uniform schedule of fees applicable to expenses incurred by federal agencies responding to FOIA requests, as required by 5 U.S.C. § 552(a)(4)(A)(i). CRE believes that OMB would have relatively little difficulty in developing and issuing a similar uniform schedule that federal awarding agencies would incorporate into their agency regulations implementing FOIA, for purposes of collecting fees to reimburse recipients and applicable subrecipients pursuant to the public access provision in P.L. 105-277.

CRE Interactive Public Docket on Data Access

On its Internet website, www.TheCRE.com, CRE has opened an Interactive Public Docket ("IPD") on the data access issue. As part of the IPD, we have prepared an analysis of the key issues in the Shelby Amendment discussion, with links to comments that other prominent stakeholders have filed with OMB. CRE has also received a number of comments from the public and posted them on the IPD as well.

CRE intends to maintain the Data Access portion of its website as an active section during the coming year as we track the agencies' implementation of conforming regulations. We invite you to visit and make use of this resource.

Sincerely,
Charles J. Fromm
Executive Director