The Data Access Law: Decreasing Secret Science While Increasing Accountability
As the federal bureaucracy grows and Congress delegates more of its power to the agencies, the need for agency accountability increases. In October of last year, Congress passed legislation that requires federal agencies to publicly release the data from federally funded research. This right-to-know law provides a necessary check upon the power of agencies, and Congress to make rules and regulations based on "scientific" studies.
However, Representative George Brown (D-CA) has sponsored a bill, H.88, to repeal the provision, while Representatives David Price (D-NC) and James Walsh (R-NY) have co-sponsored an amendment to the House Treasury, Postal Service and General Government Appropriations bill that would delay implementation of the law for a year.
Background. The EPA promulgation of a clean air rule for particulate matter served as the catalyst for the data access law. The EPA justified the clean air rule by citing a study conducted by researchers at Harvard’s School of Public Health.2 Concerned about the sweeping scope of the rule and its tenuous link to health benefits due to this one study, Congress asked to see the study’s data. Both the EPA and Harvard refused to release the data, citing patient confidentiality. Despite the ensuing controversy, the EPA finalized the rule in 1997.3
To prevent future occurrences of agency rules based on "secret science," Senator Richard Shelby (R-AL) inserted language on data access into the Omnibus Spending Bill for FY 1999. The provision requires the Office of Management and Budget (OMB) to draft guidance to require "Federal awarding agencies to ensure that all data produced under an award will be made available to the public through the procedures established under the Freedom of Information Act (FOIA)."4 The OMB draft guidance narrowed the scope of the provision by permitting FOIA requests only for data from published research findings that were produced using a federal award and were used by the federal government in developing policy or rules.5
Why taxpayers need to know. One of the purposes of regulation is to save lives that would otherwise be lost if the regulation were not in place. However, regulations also cause lives to be lost. According to one study, every $3 to $8 million (1980 dollars) loss to the economy results in one premature death.6 Federal regulations costing $670 billion per year could cause 67,000 to 222,313 premature deaths each year.7 Most of these deaths occur among the poor.
Since regulations impose such sacrifices, regulators should provide evidence that their rules will save more lives than they lose. That is why it is essential for agencies to reveal the data on which they base their rules. Without such disclosure, agencies can arbitrarily set standards costing lives and expanding regulations without any accountability. Moreover, given that taxpayers funded this research, they should have a right to see the data – particularly when it has such profound impacts on their quality of life.
In addition, with many new rules on the horizon, the need for accountability – without delay – is critical. The OMB is still working on its advisory to agencies on how to implement the law. Yet the EPA has proposed new clean air regulations based on the same study that Harvard refused to release in the past and that served as the catalyst for the passage of the data access law. The researchers still haven’t released the data to the public even though the public paid for that research.8
The free exchange of data is the essence of good science. Not only is the release of underlying data good policy it is also good science. Science encourages doubt about and debate over results, because this process validates or discredits a theory. This testing of theories increases knowledge, whereas science conducted in secret does little to advance knowledge. Thus, the scientific establishment has long held the position that science requires the exchange of data so that other parties might analyze the conclusions drawn from that data.
For example, the National Academy of Sciences report notes the importance of data exchange:
…[E]xcept in situations involving national security, the protection of individual privacy, or proprietary rights, scientists have developed an ethic of full and open exchange of data, within and across national boundaries. Although infringements occasionally do occur, they typically generate community disapproval. Full and open exchange of information is a fundamental tenet of basic science that scientists regard as essential to optimizing their own work and that of their colleagues, as well as to enabling the advance of science overall… Scientists believe that restrictions on data access will slow the progress of science and significantly diminish the potential benefits that science renders to society.9
Opposition to the data access law unwarranted. Nonetheless, many in the scientific establishment insist that the data access law infringes on the protection of individual privacy and property rights. In their view, FOIA will not provide the necessary protections and they want to abandon the data access law altogether.
The scientific community is right that we should ensure the protection of individual privacy and of proprietary interests. But they are wrong in maintaining that such protections do not exist under the data access law. The Office of Management and Budget (OMB) built protections into the law, and the law uses FOIA, not only as the vehicle for the dissemination of information, but also as a filter.
Privacy concerns seem a particularly curious concern to raise in relation to FOIA. After all, scientific studies conducted directly by the agencies are already subject to FOIA. Former NIH Director of Extramural Programs Management Office, William H. Goldwater notes that NIH has dealt with FOIA for years and protections for individuals exist both in FOIA and in court decisions that interpret FOIA. FOIA provides explicit protection for "personnel and medical files and similar files the disclosure of which would constitute a clearly unwarranted invasion of personal privacy."10 Furthermore, scientists retain the ability to mask data so as to protect the privacy of individuals, and the agencies themselves have the ability to narrow the reporting rules in order to protect individual privacy as NIH has done.11
Goldwater also points to the property protections that exist under FOIA and subsequent court cases. FOIA exempts from disclosure "trade secrets and commercial or financial information obtained from a person and privileged or confidential."12 Court decisions have denied access to data when the access would harm the competitive position of the supplier of the data.13 And again, the agency can do what NIH has done to protect data by narrowing the release by requiring specific requests for data and information so as to avoid "fishing expeditions."14 Plus, under the rule, FOIA will grant access to published information only. By the time researchers publish a study, they should have protected all intellectual property rights that might be involved in the study.
Conclusion. The increasing amount of regulation based on scientific premise demands the existence of a data access rule. The rule will not stifle science. The discipline of science has always held the position that good science requires the free exchange of information: good science is open science. Science conducted in secret or by acclaim furthers neither the cause of science or responsible regulation. Secret science does not increase knowledge; it only increases confusion and permits itself to be manipulated for political ends. In order to prevent such manipulation, taxpayers should have access to the data that supports the regulations that have such impact on their lives.
1 Jennifer Zambone (email@example.com) is a Research Assistant at the Competitive Enterprise Institute
2 James Freeman, "New Rule Forces Release of Taxpayer-Funded Studies," The Boston Globe, Nov. 9, 1998.
3 www.epa.gov/oms/regs/Id-hwy/tier-2/nprm/ria/ch-vii.pdf , Tier2/Sulfur Draft Regulatory Analysis, Chapter VII, 43.
4 Treasury and General Government Appropriations Act, 144 Cong. Rec. S12134 , October 9, 1998. (Statement of Sen. Lott).
5 64 Fed. Reg. 5684, 1999.
6 Ralph Keeney, "Mortality Risks Induced by Economic Expenditures," Risk Analysis, Vol. 10, No. 1, 1990, pp. 147-159, 154. $10 million per premature death is an estimate. The study actually states that the risk of a premature death can occur when there is a loss to the economy of $3 to $8 million dollars in 1980 dollars.
7 Thomas D. Hopkins, Prepared Statement to the House Government Reform and Oversight Committee, National Economic Growth, Natural Resources and Regulatory Affairs Subcommittee, Committee on Government Reform, U.S. House of Representatives, May 16, 1996.
8 Eventually, they did agree to release it to Health Effects Institute, an organization jointly funded by EPA and the auto industry, for analysis. The organization plans to complete analysis of the data by June of this year.
9 "Bits of Power, Issues in Global Access to Scientific Data," National Research Council, 1997, p. 47.
10 5 U.S.C. § 552(b)(6) (1994), as amended by Electronic Freedom of Information Act Amendments of 1996, 5 U.S.C.A. § 552 (West Supp. 1997).
11 William A. Goldwater, "Letter to the Editor," 282 Science 1823, 1998.
12 5 U.S.C. § 552(b)(4) (1994), as amended by Electronic Freedom of Information Act Amendments of 1996, 5 U.S.C.A. § 552,West Supp. 1997.
13 See. National Parks and Conservation Ass'n v. Morton, 498 F.2d 765, (D.C. Cir., 1974.)
14 William A. Goldwater, "Letter to the Editor," 282 Science 1823, 1998.