Analysis of Changes to OMB Circular A-110: Extending FOIA to Federal Grantees

Last year, as part of the FY 1999 Omnibus Appropriations bill, legislation sponsored by Sen. Richard Shelby (R-AL) was enacted that effectively subjects federal nonprofit grantees to the Freedom of Information Act (FOIA), essentially overturning a 1980 Supreme Court decision limiting FOIA to governmental agencies. View the Original OMB Proposal(2/4/99) View OMB Watch's Analysis of the Original OMB Proposal (2/8/99) View OMB Watch's Comment On the Original OMB Proposal (3/31/99) View OMB Watch's Testimony on H.R. 88, a Bill to Strike the Changes to A-110 (7/15/99) View OMB Watch's Analysis of the Draft of the Second OMB Proposal (7/15/99) View the Second OMB Proposal (8/11/99) View OMB Watch's Analysis of the Second OMB Proposal (8/21/99) View OMB Watch's Comment On the Second OMB Proposal (9/10/99) View the Final OMB Change to Circular A-110 (11/6/99) View OMB Watch's Analysis of the Final OMB Rule (12/1/99) Background Last year, as part of the FY 1999 Omnibus Appropriations bill, legislation sponsored by Sen. Richard Shelby (R-AL) was enacted that effectively subjects federal nonprofit grantees to the Freedom of Information Act (FOIA), essentially overturning a 1980 Supreme Court decision limiting FOIA to governmental agencies. The Shelby amendment makes "all data" produced under a grant subject to the "procedures established under the FOIA" and would permit the granting federal agency to impose a user fee on the requestor equaling the incremental cost of obtaining the data. The amendment required OMB to modify Circular A-110, "Uniform Administrative Requirements for Grants and Agreements with Institutions of Higher Education, Hospitals, and Other Non-Profit Organizations." As a result, the Shelby amendment only applies to nonprofit grantees; it does not apply to federal contractors or federal funds going to state and local governments. Some believe that this amendment was a direct result of frustration over a fight about an EPA regulation on clean air. The National Institutes of Health had granted funds to Harvard University to collect and analyze data that was used by the EPA in determining how to write a rule central to the Clean Air Act. Conservatives who disagreed with the EPA rule wanted the underlying data that Harvard had generated so they could re-analyze it and criticize the EPA regulatory policy. EPA publicly provided aggregated data used in its risk assessment and cost-benefit analysis, but did not provide the underlying information which was retained by Harvard. OMB published the final changes to Circular A-110 in the October 8, 1999 Federal Register after receiving more than 12,000 comments in response to two proposed revisions to the Circular. In under-stated fashion, OMB says that "commenters offered strongly differing views on the" Shelby amendment. On one side, Shelby, along with Sens. Ben Nighthorse-Campbell (R-CO), Phil Gramm (R-TX), and Trent Lott (R-MS), sent a letter to OMB strongly opposing OMB's second proposal, claiming OMB's proposal would "render the [Shelby] provision potentially meaningless" and would be "contrary to the plain meaning of the statute and Congress's intent in passing the law." Others, such as the Chamber of Commerce, who represents business interests, took a more direct approach. Their March 23, 1999 web site posted bold headlines about the Shelby amendment: "In The Regulatory Reform Arena, There May Never Be A More Important Issue!" The Chamber notes that extending FOIA to nonprofit grantees will allow businesses to "challenge the agency" that develops regulatory protections. These challenges will mean "agencies will have a much harder time imposing regulations on the business community..." Some examples of regulations they want to target: protection against second-hand smoke; warnings about breast implants; workplace ergonomic standards; clean air rules; environmental justice claims; and EPA's right-to-know programs. Another web site, junkscience.org, supported the Shelby amendment and had prepared letters to go to OMB. Those who sent comments to OMB from the site were eligible to win one of five awards. On the other side, the academic and science community, led by groups such as the American Association of Universities and the American Association for the Advancement of Science, raised major concerns with the OMB proposals. They argued the Shelby amendment would significantly undermine quality science, could infringe on intellectual property, and may affect the privacy of human subjects. The nonprofit community, more generally, complained that the Shelby amendment was only targeted at nonprofits and that there was no ability for the nonprofit grantee to seek reimbursement of costs associated with complying with the FOIA request. The public interest community, save one or two groups, opposed the use of FOIA for this purpose. History of OMB's Proposals The Shelby amendment stated "That the Director of OMB amends 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: Provided further, That if the agency obtaining the data does solely at the request of a private party, the agency authorize a reasonable user fee equaling the incremental cost of obtaining the data." There was no public debate on the amendment, nor was there a hearing on the subject. It is clear, however, that the Congress was not united on this provision, as a bill repealing it (H.R. 88) was introduced on the first day of the next session. As of October 20, 1999, the bill had 54 co-sponsors. On February 4, 1999, OMB published its proposed changes to Circular A-110 in the Federal Register (see /files/files/npadv/1999/feb4a110.html). The proposal narrowed the Shelby amendment to apply to "data relating to published research findings" produced under the grant that were used by the government in "developing policy or rules." Unfortunately, OMB did not define any of the terms, such as "data" or "published" and did not provide any detail on whether a grantee could request reimbursement for the costs of providing the data or how reimbursement would occur. The trigger for allowing a FOIA request - "used by the Federal Government in developing policy or rules" - was extremely broad and ill-defined, and no exemptions were provided for sensitive personal or commercial data, although the standard FOIA exemptions would be applied by the granting agency. (For a full analysis of the February 4th proposal, see /files/files/info/1999/foia-rider.html.) After receiving over 9,000 comments on the original proposal, OMB decided to make a second proposal available for public comment. OMB's second proposal -- published in the August 11, 1999 Federal Register (see /files/files/npadv/1999/aug11a110.html) -- made several positive changes to the February proposal, but did not address all of the concerns. Definitions were provided for key terms such as "data" and "published." The FOIA trigger was further clarified, and an outline for grantee reimbursement was included. (For a full analysis, see /files/files/npadv/1999/a110rev2.html.) Public interest groups and organizations representing higher education and scientists remained opposed to using FOIA as the vehicle for making scientific information held by non-governmental entities available to the public. Additionally, public interest groups complained that the OMB proposal was too broad and still lacked information about cost reimbursement. At the same time, Shelby, along with business interests, such as the Chamber of Commerce, voiced significant concern with the second OMB proposal. OMB received roughly 3,000 comments on its second proposal and then issued a final rule on October 8, 1999 which was very similar to the August 11th proposal. Summary of OMB's Final Rule The final rule modifies Circular A-110 in several ways:
  • It makes "research data relating to published research findings produced under an award that were used by the Federal Government in developing an agency action that has the force and effect of law" publicly accessible under the procedures of the Freedom of Information Act. It is important to note that only data produced by nonprofit grantees, and not data from contractors or state and local governments, is covered.
  • It states that in order for a grantee's data to be subject to FOIA an agency must officially cite the "research findings in support of an agency action that has the force and effect of law," or the data must be published in a peer-reviewed journal and used in the development of an agency action that has the force and effect of law.
  • It allows the agency to charge the FOIA requestor a "reasonable fee equaling the full incremental cost of the agency, the recipient, and applicable subrecipients." It does not establish a procedure for grantees to request or collect this fee from the agency.
  • It exempts certain data from disclosure under the definition of "research data," such as personnel and medical files, as well as "trade secrets, commercial information, materials necessary to be held confidential by a researcher until they are published, or similar information which is protected under law."
Analysis of OMB's Final Rule This section covers information about: Who's Affected by the Rule; The FOIA Trigger; Cost Reimbursement; Definitions Used in the Rule; Confidentiality and Quality Research Issues; and Public/Private Research Partnerships. The modification to Circular A-110 is effective on November 6th, 1999, 30 days after it was published in the Federal Register. OMB states that the revised circular does not apply retroactively. Yet, it also indicates that the changes affect "those continuing awards which are renewed after the effective date." Depending how this is implemented, this may cause some retroactivity problems for ongoing research which has been annually renewed. Who will be Affected? Circular A-110 establishes administrative requirements for federal grants given to nonprofit organizations, including hospitals and institutions of higher education. The impact of the changes on the nonprofit community will not be as obvious as the impact on universities and nonprofit research institutions, but it will be equally as important since many types of nonprofits that receive federal grants also publish research findings. As the Shelby amendment passed Congress, Senator Campbell stated "the amended Circular shall apply to all Federally funded research." 144 Congressional Record S121234 (October 9, 1998). This is not accurate, as Circular A-110 only applies to nonprofit grantees. Thus, research funded through contracts or through grants to states and local governments will not be covered. In other words, research conducted by a nonprofit, such as a YMCA, would be covered by FOIA requests, but the same research conducted under a contract to a company like Boeing would not. Moreover, research conducted by a counseling center operated as a nonprofit federal grantee is covered, but a counseling center that it operated by the local government that receives federal grants would not be covered. There is nothing preventing OMB from correcting this by extending the revisions of Circular A-110 to the Federal Acquisition Regulations (covering contracts) and to Circular A-102 (covering state and local governments). The OMB Circular also applies to subawards, such as those from state and local governments or those from a federal grantee to a commercial entity. This raises a question as to whether FOIA requests can be targeted to subgrantees and whether the subgrantee would even be aware that they are covered by the Shelby amendment. The FOIA Trigger FOIA requests can only be filed for "published research findings" that have been "used by the Federal Government in developing an agency action that has the force and effect of law." OMB provides two situations that satisfy the condition for a FOIA request. First, when the research is published in a peer-reviewed scientific or technical journal and the agency officially cites the research in support of a rulemaking. Second, if the federal agency officially cites the research in support of a rulemaking. Thus, the mere citation of the research, regardless of whether it has been peer-reviewed, will be the trigger for FOIA requests. In one sense, this is inducement for federal agencies not to cite literature that supports its actions. In another, it misses the point. The trigger should not be based solely on whether the agency simply cites the research in support of the action. Rather, it should be based on whether data from the cited research was part of the underlying assumptions or assessments used in developing the regulatory action. The main vehicles for incorporating such research data would be cost-benefit analyses, which are only required of major rules, and risk assessments, which are only conducted on certain major rules where the tool makes sense to use. Even though it makes most sense to limit the FOIA trigger to major rules (rules that have an annual economic impact of $100 million or more) in which the research was used as part of the underlying assumptions or assessments, it will apply to all regulations, big and small -- roughly 2,000 each year. OMB specifically rejected limiting the FOIA requests to major rules, but added "we may revisit the issue." Another concern about the FOIA trigger is the definition of "an agency action that has the force and effect of law" is not completely clear. Federal agencies take thousands of actions every year that have the "force and effect of law," including regulations and administrative orders. A significant portion of these rules likely cite federally funded research, thereby making a broad body of research subject to FOIA requests. Such broadness in Circular A-110 will create opportunities to harass federal grantees and misuse the FOIA requirement. OMB agreed to retain the "force and effect of law" clause rather than use the term "regulation" because "we think that agencies rarely rely on Federally-funded research in the context of their administrative orders." But OMB leaves open what is covered by the clause. Agency guidance, for example, often has enormous impact on the regulated community. OMB simply states that "agency guidance documents and other issuances that do not have the force and effect of law" are not covered. Thus, we can expect some debate over whether certain agency documents have the "force and effect of law." Cost Reimbursement Issues The Shelby amendment states that "If the Federal awarding agency obtains the research data solely in response to a FOIA request, the agency may charge the requestor a reasonable fee equaling the full incremental cost of obtaining the research data." OMB provides modest guidance on how to establish the user fee. OMB suggests that a grantee would assess the costs of complying with a FOIA request, and the agency would then request "appropriate prepayments" from the requestor. No mention is made of prepayment to the grantee, even if it is required to comply with the FOIA request. OMB only states that a grantee will be reimbursed after it "transmits the responsive research data to the agency." The revised Circular contains no clarification of what exactly is covered by this fee, or exactly who is reimbursed. Surprisingly, there is no mention of the rights of a grantee to request reimbursement. The absence of this authority raises questions about burdens placed on the grantee. It is likely that an agency will not know the exact cost of obtaining data from a grantee. The reimbursement plan provided by OMB does not seem to allow a grantee to request reimbursement from the agency in order to comply with a FOIA request. Further, it is not clear if expenditure of grant funds to comply with a FOIA request is an allowable expense under OMB cost principles. It is especially important that the grantee be able to request reimbursement if the agency is seeking "data" from the grantee or that the cost of such a transfer be considered when preparing the cost of the grant. If the grantee cannot be reimbursed quickly (or at all), it may be impossible to comply with a FOIA request, which could subject the grantee to legal or other penalties. Without clearly defining a fee and penalty structure, it is impossible for a grantee to know the consequences of non-compliance with a FOIA request, as this is the first time data that is not a direct agency record is covered under FOIA. Definitions Research Data: The OMB definition of "research data" is helpful in clarifying the limits on what an agency can request from a nonprofit grantee. By exempting personnel and medical files and information that could identify individuals, OMB has addressed some of the concerns of researchers. OMB also exempted "trade secrets, commercial information, materials necessary to be held confidential by a researcher until they are published, or similar information which is protected under law." Nonetheless, the definition leaves several uncertainties. For example, while the rule does exclude "physical objects (e.g. laboratory samples)" from being considered research data, it is unclear if things such as video or audio tapes of a family interaction session used in sociological research are "physical objects." This type of detail is important, but is missing. It is also very likely that the judgement of researchers in applying these exemptions will be questioned by those seeking information under FOIA, and these disputes will likely result in litigation. The definition of "research data" also leaves some uncertainty as to required data formats. It is possible that some data may be completely indecipherable to anyone but the researcher, either due to its complexity or due to format. For example, some data may be stored on a proprietary computer program. Is the researcher obligated to provide the data in a format that is commonly useable, or can she simply hand over disks that contain the data, but are useless with any other program? Most software licenses would not allow the researcher to include a copy of the program with the data. Published: Data produced under an award is subject to FOIA only if it is "published," therefore making the definition of "published" a key component of the FOIA trigger. There are still several questions surrounding the final definition. First, the definition states that when research findings are published in a peer-reviewed scientific or technical journal, they are considered published. OMB does not address the issue of multiple papers being published from one data set. Presumably, once the first paper is published, or the first citation is made by an agency, the data set is open to the public. This is especially problematic for large bodies of research, which may not be fully analyzed by researchers for years. Second, OMB also defines published if the agency "publicly and officially cites" the research in support of an "agency action that has the force and effect of law." OMB has indicated the importance of this provision by quoting a commenter on the first proposal: "award recipients would be able to avoid disclosure of data otherwise available to the public merely by failing to submit the data to a formal peer review publication." The same concerns can be raised from the other perspective. If an agency wanted to make a nonprofit grantee's data subject to FOIA, it would only need to cite the research in supporting its action-- even if the grantee has not formally published the data yet. (See discussion above.) Confidentiality and Impact on Quality of Research One of the greatest areas of concern with the changes to Circular A-110 surrounds patient confidentiality. In its comment letter on the changes to A-110, the American Heart Association stated that the "most significant area of concern for the American Heart Association [is] the issue of participant confidentiality." Although the definition of "research data" exempts information which can be used to identify subjects, there is no guarantee that this exemption will be applied perfectly. Most human subject review panels would require researchers to notify subjects of the possibility that personal information about them might become public. Potential subjects may be reluctant to give personal information if it is to become the property of the federal government or the public at large. For example, it may become harder to conduct clinical studies of those who use illegal substances. A subject may be less inclined to participate in studies on drug use, alcoholism, or similar studies if he or she knows that it is possible a government agency, or the public, may obtain records with their personal information. There are also forms of data that make it extremely difficult to protect privacy, such as videotaped interview or interaction sessions used in social science, and the definition of "research data" does not clearly address this issue. It is also possible that the changes to Circular A-110 would make it more difficult for nonprofit organizations to provide service under a federal grant. The possibility of the public release of records may frighten people from AIDS clinics, shelters for abused women, and similar services provided under a federal grant. Inevitably, the fear of possible breaches in privacy will have a chilling impact on subject participation and research quality. Even though OMB has exempted information with personal identifiers from the definition of "research data," there are likely to be privacy concerns that could affect research quality, which researchers will try to address by removing as many personal identifiers as possible from data requested under the FOIA provision. The rule defines "research data" in such a way that nonprofit grantees would not be bound to transmit information which, in their judgement, includes "trade secrets, commercial information," or "personnel and medical information and similar information the disclosure of which would constitute a clearly unwarranted invasion of personal privacy." While it is still unclear what legal recourse a FOIA requestor has under Circular A-110, to the extent there is legal recourse, the judgment of the researcher will likely be questioned. This would all occur before legal questions arise about privacy exemptions under FOIA. Impact on Public/Private Partnerships The final rule does not directly address problems that may arise from research conducted with both federal and private money. An earlier OMB draft of the proposal, as well as floor statements made by Senator Campbell, state that the amended circular will apply to all federally funded research, regardless of level of funding, or involvement of private funding. It is likely that private companies will shy away from research partnerships with grantees under the new rule, as they may feel that the research is not adequately protected, while research under a federal contract would be. Even with the exemption of trade secrets and commercial information under the definition of "data," there can be no guarantee that the research may be used by the private firm before anyone else has a chance to profit from it. Conclusion While OMB's final rule addresses some of the problems in the original proposal, it does not address all of the concerns, the most fundamental being the unfairness of subjecting nonprofit grantees to FOIA without applying a similar rule to contractors. While OMB could have improved its drafting of the revisions to Circular A-110, the real problem lies in the Shelby amendment itself. FOIA is a politically blunt instrument; it is not designed for strengthening the quality of science that is funded by the federal government. We remain concerned that the language of the Shelby amendment -- "that all data produced under an award will be made available to the public through the procedures established under" FOIA -- will invite litigation over the fact that OMB's rule is not broad enough. The Shelby amendment remains a poorly drafted piece of legislation that ultimately must be repealed. While a bipartisan bill to repeal the Shelby amendment has been introduced in the House, its future is unclear. The bill's primary sponsor and champion, Rep. George Brown (D-CA), passed away in July. Further action on the bill is also unclear, as hearings were held on July 15, 1999 without further committee action since. (Read the testimony of OMB Watch's Executive Director, Gary Bass.)
back to Blog