Letter by National Academy of Sciences President Bruce Alberts on OMB Circular A-110 Revision
F. James Charney
Office of Management and Budget
New Executive Office Building
Washington, D.C. 20503
Dear Mr. Charney:
These comments are being submitted on behalf of the National Academy of Sciences in response to the notice which appeared in the Federal Register for Wednesday, August 11, 1999 at pages 43786-91 offering interested parties an opportunity to comment on clarifying changes to the proposed revision to OMB Circular A-110 previously published in the Federal Register for Thursday, February 4, 1999 at pages 5684-85. The revision of OMB Circular A-110 would implement that provision of Public Law 105-277 (frequently referred to as the "Shelby amendment") which directs OMB to amend Section __.36 of OMB Circular A-110 "to require Federal awarding agencies to ensure that all data produced under an award will be made available to the public through the procedures established under the Freedom of Information Act."
As I have noted in previous correspondence, the Academy has issued a number of reports through the National Research Council supporting the concept of data sharing. Data sharing is particularly important when the data are the foundation for the formulation of public policy.
However, reliance upon the Freedom of Information Act (FOIA) as the mechanism for achieving this result -- as the Shelby amendment seemingly requires -- will likely cause confusion, encourage litigation, and produce unanticipated and unintended consequences that may surprise and disappoint both the proponents and the detractors of the Shelby amendment. I therefore believe that specific legislation will be needed to address these issues in a more effective manner.
In the meantime, within the constraints of the existing wording and limited scope of the Shelby amendment, OMB has done a masterful job of listening to public concerns and trying to craft an amendment to OMB Circular A-110 that will implement the Shelby amendment in a fair and reasonable way. We sincerely appreciate your efforts.
The August 11th Federal Register notice specifically requests public comment on the proposed clarifying definitions for "data," "published," and "used by the Federal Government in developing policy or rules," as well as the additional background information set forth in the notice regarding cost reimbursement. Comments on each of these points are set forth below.
The case for making the proposed amendments to OMB Circular A-110 applicable to "research data" is compelling. While the Shelby amendment speaks in terms of "data" produced under awards, the comments by Senators Shelby, Lott and Campbell on the Senate floor describing the Shelby amendment, 144 Cong. Rec. S12134 (October 9, 1998), make it absolutely clear that the Shelby amendment was intended to apply to "research data." Since OMB Circular A-110 does not currently define that term, a definition of "research data" for purposes of OMB Circular A-110 is clearly required.
In scientific research, "research data" are the various bits of factual information that are collected and recorded as a result of scientific observation, experiment, or similar methods of research. For purposes of implementing the Shelby amendment, we agree that OMB's proposed definition of "research data" as "the recorded factual material commonly accepted in the scientific community as necessary to validate research findings" -- subject to the specific exceptions that are set forth in the proposed definition, which are discussed separately below -- is a reasonable and workable definition of "research data."
One of the specific exceptions incorporated into the proposed definition of "research data" is "preliminary analyses, drafts of scientific papers, plans for future research, peer reviews, or communications with colleagues." While these kinds of things do not fall within the common understanding of the term "research data," the specific exception is nonetheless absolutely necessary to dispel any notion that OMB Circular A-110 is being amended to authorize wide-ranging fishing expeditions through the files of Federally-funded research grantees. The Shelby amendment simply deals with access to research data. It does not contemplate anything remotely equivalent to the subpoena process or the discovery process in litigation.
The proposed definition of "research data" would specifically exclude physical objects such as laboratory samples. This is an entirely reasonable clarification since physical objects such as laboratory samples are not really "research data" but rather sources of "research data."
The proposed definition of "research data" would specifically exclude "information which may be copyrighted or patented." While this makes excellent sense with respect to copyrights, some further clarification with respect to patents may be needed.
With respect to copyright, the U.S. Supreme Court's opinion in Feist Publications, Inc. v. Rural Telephone Service Co., Inc., 499 U.S. 340 (1991), makes it clear that "facts," which may not be copyrighted -- such as the factual information that results from scientific observation or experiment -- and compilations of facts that possess the requisite originality to qualify for copyright protection, are mutually exclusive domains. 499 U.S. at 347-51. As the Supreme Court explained, facts are not original and therefore may not be copyrighted but a factual compilation is eligible for copyright if it features an original selection or arrangement of facts, although in no event may copyright extend to the facts themselves. 499 U.S. at 350-51. Therefore, compilations of research data which qualify for copyright protection -- as opposed to the actual research data themselves -- are necessarily excluded from the definition of "research data," and the specific exclusion in your revision for "information which may be copyrighted" is therefore entirely appropriate.
On the other hand, the patent process deals with invention rather than information per se, so the intent of your phrase "information which may be ... patented" may need some clarification. The impact of the implementation of the Shelby amendment on the patent rights of grantees is clearly highly relevant and important. For example, premature disclosure of information related to a claim to be made for a patent could have the effect in some cases of precluding the issuance of a patent. If the intent of the OMB proposal is that information which may be the basis for a patent claim would be excluded from the definition of "research data," that should be made clear.
The proposed definition of "research data" would also specifically exclude "materials necessary to be held confidential by a researcher until publication of their results in a peer-reviewed journal." In light of the proposed definition of "published" discussed below, this exception seems perfectly reasonable and indeed unavoidable. However, it may be appropriate to revise the exception to read "materials necessary to be held confidential by a researcher until the results are published," thereby fully incorporating the special definition of the term "published." Otherwise, where research findings that have not yet been published in a peer-reviewed scientific or technical journal are nonetheless deemed to have been "published" because a Federal agency has publicly and officially cited the research findings in support of a regulation, the public might still be denied access to those research data because there has been no journal publication. That would appear to be inconsistent with OMB's intent.
Finally, as explained in the August 11th notice, OMB is proposing "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.'" However, a 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."
We feel that this concept is extremely important. We recognize that "trade secrets" and "commercial information" are generally covered by Exemption 4 of the Freedom of Information Act (FOIA) and that "personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy" are covered by Exemption 6 of FOIA. But for Federal agencies the FOIA exemptions are generally discretionary, not mandatory, in nature.
OMB's proposal would convert these exemptions into true safeguards for the protection of "trade secrets" and "commercial information" and "personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy," without any need for reliance upon other Federal statutes which may or may not be applicable in particular cases. OMB's proposal would enable Federally-funded research grantees to tell their research partners and participants that the grantee is not obligated to turn over "trade secrets" and "commercial information" and "personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy," rather than having to tell their research partners and participants that the grantee must turn everything over to the Government and the Government will decide. OMB's proposal would also more appropriately recognize Federally-funded research grantees as independent legal entities with legal and ethical obligations to both the Federal Government and the grantee's research partners and participants.
For purposes of implementing the Shelby amendment, we agree that OMB's proposed definition of "published" as "either when: (A) Research findings are published in a peer-reviewed scientific or technical journal; or (B) A Federal agency publicly and officially cites to the research findings in support of a regulation" is reasonable and workable. This directly addresses the critical problem that we have addressed in earlier correspondence of the potential impact of the Shelby amendment on the publication rights of researchers. It reasonably protects those publication rights and at the same time is consistent with the underlying intent of the Shelby amendment.
We would suggest, however, that (B) be modified to read: "A Federal agency publicly and officially cites in the Federal Register to the research findings in support of a regulation." The Federal Register is official. It is readily available to the public. It ties official action to a specific date -- i.e., the date of publication. It is readily researchable. Specific reliance on the Federal Register would greatly reduce ambiguity and uncertainty with respect to the exact scope and timing of official agency action.
Used by the Federal Government in Developing a Regulation
The argument made by OMB for narrowing the phrase "used by the Federal Government in developing policy or rules" to "used by the Federal Government in developing a regulation" is highly persuasive. The substantive and procedural vagaries and complexities associated with the development of Government "policy" make the argument for excluding "policy" from the scope of the amendment to OMB Circular A-110 compelling.
The definition of "used by the Federal Government in developing a regulation" as "when an agency publicly and officially cites to the research findings in support of a regulation (for which notice and comment is required under 5 U.S.C. 553)" is reasonable and, as the August 11th notice points out, incorporates a well-established and long-standing definition of "regulation." Once again, however, for the reasons previously set forth above, we would suggest that the definition be modified to read "when an agency publicly and officially cites to the research findings in the Federal Register in support of a regulation (for which notice and comment is required under 5 U.S.C. 553)."
Further limiting the scope of the amendment to OMB Circular A-110 to regulations that meet a $100 million impact threshold, as described in the August 11th notice, would be wise public policy. It would focus the effect of the amendment on Federal grantee research results that are cited in support of those regulations which have the greatest economic impact, narrow the overall effect of the amendment on Federal grantee research as a whole, and reduce the administrative burden on the scientific enterprise.
There are still far too many uncertainties about how the proposed amendment to OMB Circular A-110 will actually be applied in practice to offer any meaningful estimates of cost impact. With respect to cost accounting, since costs of compliance will continue to be incurred after the expiration of awards -- and also since research results and data may frequently be the result of multiple awards -- indirect cost treatment would appear to be appropriate. This may create a significant problem with respect to administrative caps on indirect costs.
National Academy of Sciences