Seeking a Truce in the Environmental Information Wars: Replacing Obsolete Secrecy Conflicts With New Forms of Sharing

James T. O'Reilly

James O'Reilly is a Visiting Professor of Law at the University of Cincinnati College of Law. Prof. O'Reilly received his undergraduate degree from Boston College and his juris doctor from the University of Virginia.

[30 ELR 10203]

Military historians observe that the losing armies were the ones that were only prepared to fight the last war, failing to change their tactics and strategies to deal with current warfare. Today, environmental groups and U.S. industries continue to war over the public dissemination of process details and chemical formulations regarding the environmental consequences of industrial facilities. Information confidentiality disputes relating to the environment present a classic political science, economic, and philosophical conflict among competing values. While these "last wars" were based on the industry's fixation on the physical security of their data, the shape of information disputes in 2000 forward is likely to be dramatically different.

The development of corporate technological advancements through commercial research greatly benefits U.S. trade through licensing income and product export sales. For years, protection of that technology has been enhanced by safeguards such as physical security, cryptographic protection, and contractual constraints on employees with access to data. When regulators began to acquire large quantities of industrial secrets, industry applied similar constraints and won the courts' approval.¹ But is industry fighting the last war and missing the Internet's effects on the instant transferability of data that is shared with the U.S. Environmental Protection Agency (EPA)?

This Dialogue discusses the background of the environmental information confidentiality dispute and the obsolescence of industry's fixation on the physical security of regulatory submissions containing their chemical data. The Article posits an alternative method of public access that should meet the several competing interests in a post-2000 communitarian style. This solution addresses local citizen needs for answers about safety and health while being mindful of technology licensing interests of the developers of research-intensive technologies. A system of "public risk dialogue," structured in a sensitive manner respectful of citizen and industrial needs, is optimal for the future; obsessions with either total dissemination or total physical security of factory blueprints are the legacies of the past.

History

Knowledge is power, and knowledge that can be converted into income gives economic power to the owner of that knowledge. This proprietary power to assemble one's set of technical knowledge and exclusively deliver it to another is the basis for technology licensing agreements, also referred to as the licensing of trade secrets or "know-how."²

Theft and conversion of trade secrets by competing companies has a long and much-explored history.³ State and federal laws⁴ have traditionally protected the possessors of commercial secrets against those who would attempt an unconsented and uncompensated disclosure.⁵ Federal agencies use confidential business information submissions to make internal agency decisions, and courts have recognized that this function of federal data collection is discrete from collection for purposes of public dissemination.⁶ Where federal agencies "take" those secrets to disclose them, Fifth Amendment compensation conflicts will surely follow.⁷ Further, when agency employees consider disclosing those secrets, they are (or should be) aware of the criminal penalties that overshadow such disclosure decisions.⁸ Unconsented disclosures, therefore, have resulted in considerable litigation over the rights of the owners of intangible property.⁹

The Freedom of Information Act (FOIA),¹⁰ adopted in 1966, allows public access to federal agency records and exempts "trade secrets" and "confidential commercial or financial information" from required disclosure.¹¹ FOIA was intended to have no effect on the preexisting structure of commercial confidentiality and technology licensing.¹² FOIA presumed that agencies would customarily withhold [30 ELR 10204] from disclosure the documents that the submitting companies had kept confidential.¹³ Agency access to private commercially valuable data was intended to augment the agency role as regulator, not to create a new structure using the agency as a library of access to the private data.¹⁴ This distinction in purpose is recognized in the courts' receptivity to agency-only use situations, such as internal staff evaluations of chemical safety.¹⁵ Thus, while Congress is capable of assigning library functions to administrative agencies, and does so in technology transfer and technical data reproduction laws, it chose not to do so in enacting FOIA.

Technology Licensing Versus Public Access

Under most technology licensing agreements, the general dissemination of information "in the public domain" ends the duty to pay for the exclusive use of that information. The conditions of the typical licensing contract acknowledge the duty of the recipient of know-how to preserve it against further disclosure without the prior written consent of the data owner. Therefore, public access to a piece of data, such as a process flow diagram or a recipe for formulating a mixture, if "published" on the Internet or otherwise, extinguishes the potential licensing value of that piece of data.

This threat of income loss provides the economic incentive that motivates industry to oppose agencies' broader dissemination of industry-submitted technological and process data. One such response, offered in a dispute between EPA and pesticide manufacturers, sought compensation from the Agency under the Tucker Act.¹⁶ The U.S. Supreme Court looked for a "reasonable investment-backed expectation" by data owners that their data would not be disseminated by federal agencies and held that the right to exclude others from possession of the secret chemical technology "is central to the very definition of the property interest."¹⁷ In essence, this means that unless a statute has defined the private submissions as being publicly available, the private owner can exercise the rights of an owner against an agency decision to extinguish the owner's proprietary interest in that data.

Statutorily Mandated Disclosure

Congress has the authority to make the tough balancing decisions regarding environmental confidentiality, yet it has been a mere afterthought in the process of adopting many of the modern environmental control statutes. Even when disclosure was chosen as a regulatory vehicle in 1986 legislation, the legislative effort was quite imperfect.

The Post-FOIA Statutory Balancing

The adoption of FOIA did not produce a cosmic shift in the confidentiality of intellectual property such as trade secrets. FOIA did not terminate concepts of commercial secrecy but rather subsumed them into a protective exemption that covered both classic trade secrets and a broader category of "confidential commercial . . . information."¹⁸ In later statutes, political compromises over information provisions sometimes specified that, apart from a narrow disclosable class having a certain defined scope, industry claims of confidential status would be or could be honored. For example, the Toxic Substances Control Act (TSCA)¹⁹ mandates the withholding of information if the documents are FOIA exempt²⁰; other laws permit such withholding.

Perhaps the closest debate between disclosure and nondisclosure occurred in the final days before the adoption of TSCA. Unlike the permissive opportunity to claim exemption that EPA was given by FOIA, TSCA parsed the reporting obligations of chemical makers and required "health or safety studies" to be disclosed,²¹ while data that was FOLA exempt (apart from such studies) was required to be withheld.²² The clause resulted from legislative tradeoffs that cabined EPA's discretion and segregated such classes of private records into withheld or disclosed categories. The advocacy organizations effectively convinced Congress to include the provision in the new TSCA as a counterpart to the disclosure provisions in the Clean Air Act (CAA) and Federal Water Pollution Control Act (FWPCA) that had required certain other sets of data to be disclosed. By doing so, Congress selected an endpoint that would justify a disclosure norm, overriding private interests. Under the air and water laws, courts would equate such studies under TSCA as a disclosure endpoint akin to the previously adopted laws compelling disclosure of effluent and emission data.

During the early years of federal environmental legislation, the disclosure of industry-submitted data was not seen as an absolute benefit. The environmental control legislation of the 1970s recognized that the most crucial endpoints for the public measurement of performance endpoints, data on specific emissions²³ and effluents²⁴ of a facility, should be made public, but apart from those endpoints, the commercial secrecy model should be honored. During some of these debates on confidentiality issues, nongovernmental organization advocates' zeal for empowering regulators to control was always greater than their desire for regulators to inform; when trade offs inevitably occurred, the legislative staff members who favored tighter controls usually tried to appease industrial advocates by accepting their textual proposals for the nondisclosure provisions of the environmental laws.

A philosophical reliance on accountability underscores both the control approach to environmental lawmaking and the informational approach. The control approach imposed reporting commands and time schedules so that regulators could hold companies accountable. FOIA imposed access norms so that the regulators themselves could be held more [30 ELR 10205] accountable. The more recent informational laws, such as the Emergency Planning and Community Right-To-Know Act (EPCRA),²⁵ have put aside regulators and used the agency as a "pass through" for companies' data to be interpreted by any member of the public, a form of do-it-yourself accountability to those who choose to use their time reading and assessing facility reporting data. One's view of this public policy choice will depend on one's perspective about the balance of proprietary and community interests.

Statutory Mandates for Disclosure

In 1986, Congress responded to a New Jersey state law that required chemical facility disclosures²⁶ by adopting a similar but not identical federal requirement in EPCRA²⁷ as part of the Superfund reauthorization. A hot summer Sunday afternoon session of the conferees produced some unusual compromises.²⁸ Section 313 of the Act²⁹ established a list of chemicals³⁰ that Congress intended to be subject to a special quantitative release reporting. The Act decreed that for this set of chemicals, statistics on amounts released from particular sets of factories would be reported annually.³¹ Each facility owner must determine if they are covered and then submit the forms, from which a database of information called the Toxic Release Inventory (TRI) is compiled and publicly disclosed.

Statistical information about chemical releases, published by the press from EPA's TRI databases, has been widely viewed by the public and studied by environmental groups. As a result of the TRI disclosures, more companies have paid more attention to reducing their toxic releases, so this publicity-related response program was deemed a success.

The TRI essentially promotes a "wholesale" disclosure, an aggregated comparison of chemical release volumes from a plant, compared across multiple sites and locations, for the chemicals deemed by Congress to be worthy of analysis. Prior disputes tended to seek what could be labeled "retail" disclosures—what the particular plant was doing caused environmental opponents to seek the site's specific data. Attention to the "wholesale" effect, however, was most likely the more relevant comparison. As the TRI has been expanded, conflicts over the confidentiality impacts of the changes have been debated.³²

Indirectly, the wider use of company-specific data in the debates that have followed the TRI has had a moderating effect on earlier disputes about the evils of industry trade secrets. Routine annual reports about chemical releases appear to be less newsworthy with each passing year. This moderating effect has dissipated the level of fear concerning industry secrecy because the select group of residents or neighbors who care about industrial emissions now seem to perceive that the "secret" of pollution has been uncovered by the TRI. Less fear from more disclosure is an ideal result for industry and regulators. The comparisons of the specific trade secrets and the TRI-disclosed data could show qualitative contrasts, but the TRI's reassuring presence has disarmed many critics of industry who previously could have attacked a facility's nondisclosure of technology as being the concealment of risks.

The 1986 adoption of the TRI was followed by Congress' enactment of risk management plan (RMP) requirements under § 112(r) of the 1990 CAA.³³ Each RMP includes an isopleth, a geographic prediction of where a chemical cloud would go if all the plant's safety measures failed at once.³⁴ EPA struggled with the confidentiality issues surrounding RMPs because its advisors warned of serious sabotage risks.³⁵ EPA was trying to deal with the inherently disparate incentives of the public-warning aspects of the statute as well as proprietary confidentiality concerns.³⁶ In late 1999, Congress passed special legislation to allow portions of the RMP to be withheld from the Internet, but this was a tailored statute rather than an indicator of future trends.

A contrast between past secrecy and present routine disseminations is significant. Now that the public "knows" about facility releases, in volume and in potential spread if disastrous failures occurred, the news value of the story has diminished and the amount of public fear will diminish over time. Perhaps as a result of demystifying risk fears, the advocacy group pressures on manufacturers to also disclose recipes and concentrations of chemicals are much less than before.

Protection of Private Information

The level of trust between the regulator and the regulated is especially strained when the regulators don't seem to care about the private proprietary research of regulated companies. Beyond the legislative framework, the evolution of EPA practices has affected the willingness of research-based industrial companies to cooperate fully with regulators' desires.

The Legacy of Polaroid

It was a shocking moment for those few audience members present at an informal hearing at EPA headquarters. Polaroid Corporation's general counsel stated that prior to the 1976 adoption of TSCA, only three employees had knowledge [30 ELR 10206] of the company's secret film-chemical recipes and inventories. When TSCA was adopted, a fourth person was given the same access so that reports could be prepared. But after responding to a FOIA request, Polaroid learned that approximately 150 people in EPA and its contractors had been allowed access to these same formulas. When EPA and Polaroid could not agree on the protection of its secrets, Polaroid won a preliminary injunction in federal court,³⁷ and the case ultimately settled with EPA agreeing to upgrade its security of commercially valuable information.³⁸

At the post-settlement public hearing, EPA staff agreed to create a confidential business information (CBI) document protection system that would approximate the military's classified document safeguards. EPA adopted a very significant process for the physical protection of documents with a security manual, locked rooms, controlled access, and passwords. This satisfied virtually all of the concerns of the innovative developers of chemicals such as Polaroid. The protective system dissuaded other companies from resisting EPA in court, and it laid the foundation for two decades of relative peace between industry and EPA's chemical regulatory information staff. Both EPA employee access and EPA contractor access to formula and process data was sharply curtailed, and the system's cumbersome operation provided frequent Federal Register notices when documents were shared with EPA contractors.³⁹

Due Process for Data Submitters

The 1979 Supreme Court decision in Chrysler v. Brown⁴⁰ established the due process rights of data owners to sue agencies in order to prevent the disclosure of their business confidential data. The Court treated agency decisions to disclose as forms of "informal adjudication" for which a reasoned statement by the agency would be reviewable under the Administrative Procedure Act (APA).

During the 1980s, reform of FOIA was frequently discussed.⁴¹ Although the Bush Administration did not change FOIA, it responded to the narrow procedural question left open by the Chrysler decision. Under Chrysler, the owner of a piece of commercial data who had submitted the data to a federal agency would have an APA right to notice before the agency disclosed the data. Implementing that right of notice, and the opportunity to comment before an agency's final disclosure decision, was left to the agencies. Ultimately, the centralized process for notice and comments was adopted in the form of Exec. Order No. 12600, for all Executive Branch agencies.⁴² EPA follows that mandatory requirement in its rules.⁴³

The effect of Exec. Order No. 12600 on the dissemination of industrial submissions to EPA is threefold. First, it imposes a duty on agencies to notify industry prior to the dissemination of a document that has been properly marked as confidential. Some reassurance comes from procedural norms that limit arbitrariness; even a rabidly hostile staff employee who despised the views of the submitting company would be bound to deliver such notice and an opportunity to object to disclosure. Second, it requires the agency to allow an opportunity for disagreement to be aired before the actual release of the records. Under the APA informal adjudication called for in Chrysler, there must be an exchange of ideas both for and against the dissemination of records.⁴⁴ Third, the Executive Order imposes a concomitant duty on the owner of records to inform the federal agency that the records should be withheld as commercially confidential. Thus, one who ignores the opportunity to mark the submitted documents as being confidential may be waiving their ability to complain subsequently that the agency had disclosed the corporate treasures. In a legal environment of openness, the person seeking confidential handling may suffer for their silence.

Why Critical Mass Hasn't Been Critical

The expectations for business data submitted to federal agencies have shifted with the case law of FOIA in 1974 and 1992. The dominant standard had asked whether the agency's release of privately submitted data would cause a data owner "substantial competitive harm." The substantial competitive harm standard was borrowed from the National Parks & Conservation Ass'n v. Morton⁴⁵ decision of the D.C. Circuit. A 1974 panel of the circuit, construing the exemption for "confidential commercial information," added to the then-8-year-old law a norm of "substantial competitive harm" as a prerequisite for records to have "confidential" status.⁴⁶ This raised the threshold that had been lower in prior applications of FOIA to business data.⁴⁷

The great majority of record submissions to environmental agencies are compelled by regulation or statute. The D.C. Circuit's decision in Critical Mass Energy Project v. Nuclear Regulatory Commission⁴⁸ was an interesting reexamination of the basis for the commercial confidential treatment of submitted data.

Nuclear plant operators had voluntarily provided the Nuclear Regulatory Commission (NRC) safety reports they had performed without governmental command. As the NRC had not required them to be submitted, the agency opposed a FOIA request for the reports' disclosure with considerable help from the affected facility owners. The D.C. Circuit en banc upheld the NRC's decision to withhold the [30 ELR 10207] voluntarily submitted records, establishing a norm that "customarily confidential" business records that had voluntarily been provided to agencies could be withheld. But the decision has had virtually no impact on the landscape of environmental data disclosure because the court's decision retained the "substantial competitive harm" test for commercial records that were required to be submitted, such as under the FWPCA monitoring rules.⁴⁹

The Critical Mass decision theoretically could have an impact on an actual environmental dispute if (1) EPA chose to withhold (2) a voluntarily given business record (3) that the submitting facility could show was routinely kept from public knowledge. In practice, EPAis institutionally eager to disclose data; most records it receives are arguably required to be submitted, and so FOIA-exempt status usually requires a showing that there would be substantial harm to the competitive position of the affected facility. Bureaucracies expand their lust for more data as they sense opportunities to refine their mission, so it is abnormal for both submitter and regulator to consider a document submission to be truly "voluntary." If one or the other does not consider the record "voluntarily submitted," then the Critical Mass decision holds that the agency must continue to determine exempt status based on "substantial competitive harm" criteria.

Rethinking EPA's CBI Rules

EPA's 1976 regulations on FOIA implementation⁵⁰ had been in place for almost two decades when the Agency, in light of Clinton Administration policy changes, attempted to completely revise the rules.⁵¹ Burdens on the submitter would increase, and presumptions would favor disclosure. Strong industry opposition to the changes resulted in a decision to postpone action on the proposal, and a regulatory negotiation was considered but never adopted for budgetary reasons.⁵² According to the Agency's 1999 Regulatory Agenda,⁵³ the proposed rule remains in limbo and is awaiting reproposal.

Separate EPA rules within the CBI system apply to TRI reports⁵⁴ because of the specificity of the 1986 legislation regarding the protection of facilities' particular processes and details.⁵⁵ EPA has adopted several statute-specific rules in its generic CBI rules to account for peculiarities of the several statutes.⁵⁶ In the absence of a new norm, current submissions of CBI remain subject to the well-accepted norms that require marking and provide for analysis under the substantial competitive harm approach.⁵⁷

Technology's Impact on Public Disclosure

In past decades, career bureaucrats in hierarchical agencies strained to produce lengthy and complex written directives that gave definitive commands, after months of internal deliberation. When the letters arrived by mail, expert interpreters could be hired to spend weeks planning a strategy of response. The instant arrival of the kilobyte has replaced the biting repartee of those lengthy debates.

Physical Forms of Information

The information revolution that has resulted in massive sets of digitized industrial data poses a significant challenge to the classic tools for preserving trade secrets, the locked cabinet andthe signout sheets used for decades. The challenge is in the very elusive transferability of technical data. The transferability of data across borders is virtually instantaneous. Physical document theft is no longer the vehicle of choice for the industrial spy. As the Los Alamos spy case suggests, the ready downloading by agency staffs from secure to non-secure computer systems may provide eager customers a form of remote access to data that leaves little or no trail to be traced. Yet much of the industry effort to seal up the abundant leaks from EPA and other agency offices was premised on locking up the sheets and papers on which industry reported its information to the agency staffs. Despite EPA's similar policies for handling "records" in paper or electronic forms, the ease of computer file access will ultimately supplant "hard copy" safeguards that are rapidly obsolescing.

"It Must Be True, It's on the 'Net"

Etymologists of environmental terms who are overwhelmed by NIMBY, BACT, and NAAQS may be less familiar with GIGO, an expression used by computer professionals: Garbage In, Garbage Out. Inaccurate data recorded in federal databases will produce inaccurate reports, misleading those who believe that if it's on the Internet, it must be true.

The government's ability to enjoy an aura of factual accuracy has never been greater than it is with government Internet websites. Yet the appearance of data on an EPA website does not mean that the data is accurate or that its premises and assumptions are correct. Some GIGO is inevitable in an agency website containing hundreds of thousands of pages; but to which documents can the public ascribe a presumption of EPA endorsement of accuracy? Long debates over the quality of data have called into question EPA's ability to inform the public accurately about environmental controversies. The loading of data onto a website gives even flawed data an imprimatur of accuracy, but disclaimers either do not exist or are functionally invisible to data users. The problem is exacerbated by cross-linked websites where the transferred searcher finds data that is even less assured of accuracy than data at EPA's own site.

In the ideal future world, data quality assurance will be recognized as a government obligation. Regulators who dump thousands of pages of data into a database will have a duty to certify that the records are either accurate or that they are encoded so that every download and printout bears a [30 ELR 10208] user caution or disclaimer. Unfortunately, we are not there yet.

Significance of the Technological Change

Once disseminated by computer, a private person's valuable technological data is irretrievable. The significance of the technological change is very familiar to publishers, information technology specialists, and workers at remote sites, all of whom are watching their workloads shift as a result of the rapid dissemination of the work via Internet-based systems. The physical security of a document might be well protected, but the remotely accessible computer format of the same information may now be accessed by adversaries, including industrial spies and litigation adversaries.

A possible trend in administrative agency data collection may be the conscious decision to avoid collecting CBI where the access to such information ties up the agency in disputes over the post-collection disclosure of the CBI. In a pair of 1994 litigated cases, EPA was enjoined from gathering formulation data that the Agency had intended to use in a way that would have revealed industrial secrets.⁵⁸ An agency's conscious choice not to collect CBI in certain situations could prevent it from being subject to, and losing, future suits.

The format in which records are being delivered to regulators may be the most significant change. The electronic data interchange is becoming so central to the reporting of data to regulatory agencies that the old-fashioned fear of "an inspector at every smokestack" has been replaced by fears that continuous emissions monitoring remote sensors will be placed on stacks, with their readouts and exceedance alarms ringing inside the distant office of the environmental regulator. In a real-time access world, adversaries who link up with technology gurus may develop a series of online real-time citizen suits. "Emit today, sue today" speed is possible. The prospect of automated detection of violations, tied to automatic electronic filing of citizen suits, results from the use of technology to remove barriers to data-related litigation. It may seem far-fetched to conventionally minded counsel, but the enforcement mechanism has reached such a level of fixed response criteria that anything is now possible. Imagine a recorded voice message: "Welcome to 1-800-Exceedance-Penalties.com; press 1 to e-mail your Clean Air Act citizen suit complaint to the district court covered by that zip code . . . ."

Disclosure's Unintended Consequences

Watching the plume of smoke from a chemical plant explosion and fire has a sobering effect on the viewer. One easily recognizes that the prospect of a serious adverse event is always present with volatile materials. In a world filled with guns, bombs, and grudges, the avoidance of chemical facility sabotage is of national security importance. So when the RMPs required under CAA § 112(r)⁵⁹ were developed, federal and state-law enforcement agencies encouraged Congress to pass legislation that would withhold from nonofficial recipients the "worst-case scenario" results from projections of fires or explosions. That legislation was adopted on August 5, 1999,⁶⁰ but the majority of the data has already been disseminated by activists opposed to industry's security concerns.

Time will tell whether the improved targeting of plants for sabotage results. An interesting future torts exam may ask whether the groups that used the Internet to make these targets more attractive to potential assailants are liable for damages to the victims of some future attack for facilitating the commission of a predictable crime.

Is There Room for Owners to Preserve the Value of Their Technologies?

Is there any room for trade secrecy in a public law system dominated by ideals of transparency? Yes, there still are social policy rationales for innovation-preservation short of patenting.⁶¹ Aspects of the production process that are either pure "trade secrets"⁶² or fit the lesser status of "circumstantially relevant business information"⁶³ still merit the possibility of protection when the owner cares enough and is astute enough to satisfy the criteria for the withholding of its regulatory data submissions.⁶⁴ But the continued social acceptance of the protection efforts will require more education, more shared responsibility with regulators, and recognition of several principles:

1. Companies must narrowly define what data warrants protection, including what data truly aids the competitors, as compared to what is merely desired to be kept away from critics and potential litigation adversaries.⁶⁵

2. Advocacy organizations should reexamine demands that privately generated data submitted to EPA should be routinely placed on EPA's website. Accurate, relevant, and timely data supporting public health protection should be the public agency's goal, and reaching that goal does not require or need a total dissemination of private-sector data.

3. The debates about the role of disclosure versus confidentiality should be tempered with a greater understanding of each position's views. Traditionally, "public interest" groups were arrayed against private owners' advocates, and the tone was absolutist on both sides. Perhaps the first step toward moderating this dichotomy was marked by the antidisclosure stance of the largest Washington-based public interest groups in 1999 when they argued against the Shelby Amendment, which mandates more public access to research data created by nonprofits such as colleges and universities.⁶⁶ [30 ELR 10209] After decades of strident insistence on access as an unfettered value, one would have expected groups such as OMB Watch to favor public access and disclosure of data produced with federal grant funding, but their posture was against the disclosure mandate of that 1998 statute.⁶⁷ If this policy approach foreshadows a rethinking of the debate over access and secrecy, the Shelby Amendment controversy may prove useful in the larger debates.

4. The accessibility of large volumes of documents on the Internet does not help a local resident or group without a valid and credible translation of the data that is understandable to the local neighbors. Apart from the comparative emissions work done by some researchers or epidemiology studies done on a larger geographical scale, it is better to use the expertise of a translator to address specific safety concerns in a succinct report rather than to bear the economic displacement of losing technology licensing value for the total content of the regulated company's environmental filings.

It is ultimately in the best interests of companies engaged in research and manufacturing to temper the public/private "trade secret" conflicts so that real assets can be preserved, while public acceptance and support will be facilitated. Attorneys aren't trained to think enough about accommodation and sharing because they are schooled for battle over the single-minded preservation of every tangible and intangible property right of our clients. There is ample space for protection of commercial core technology assets, while assuaging the fear and suspicion that media hits such as A Civil Action have dramatically portrayed. It may be a generational attribute that the Generation Y group raised on the Internet access phenomenon will view data secrecy as an alien concept.

Public Risk Dialogue: A Workable Solution

There may be a better way to handle information disputes. The massive "data dump" from the private firm's files to an agency, and from the agency's website to those sufficiently interested and sophisticated members of the public does not answer the question most likely asked by residents: "Am I at risk?" The alternative remedy, therefore, will need to offer some translation of the data into lay terms through some intermediate mechanism.

A Possible Alternative: Structuring the Task

The critical elements of a successful and balanced approach to disclosure are public legitimacy of the process, accuracy of the conclusions drawn, and effectiveness of the credible translator or interpreter.⁶⁸ The best public legitimacy would be drawn by declaring as a precondition of access that at least one elected official of the county or city where the industrial pollution is alleged to be occurring be involved in any request for this alternative form of report. The experience of seeking election has a marvelous cleansing effect on one's zealotry for a cause; those readers who have stood for election and survived will understand the moderating value of "representing the public" better than those who espouse some esoteric notion of local needs in some perfect democracy. Abstractions fall away after the 100th doorbell is rung or the 200th handshake; environmental safety information is not an abstract issue but a matter of pragmatically addressing peoples' needs.

Requiring involvement in this alternative communication process by an elected official also reflects the reality that local government "owns" the problem of local pollution to a degree that distant theorists or centralized planners cannot. Alternatives have costs, and the costly disclosure process could easily be abused. The litigation tests of voluntary organizations' "standing" in the "zone of interest"⁶⁹ are too loosely defined to serve as a gateway for the expense involved here. Yet moderating the degree and frequency of such demands for disclosure is a constant function of the elected official's role.

Accuracy of the conclusions drawn is assured by giving an outside evaluator access to the full set of statistical and chemical data under a standard confidential disclosure agreement that would be prescribed in regulations. Risk assessment is built on a series of technically valid assumptions; the quality of data may be broad or narrow depending on the assignment, but it is essential that evaluators get the data they will need to be effective and comprehensive in their review of risks.

The public's acceptance of the intermediary translator is important. Citizens generally do not trust government agencies or industry. They tend to trust their local elected officials more than distant bureaucrats. Acceptability can be aided by separating the evaluator qualification process from the particular dispute at any given time. EPA should hold an annual open competition for the listing of technically qualified risk assessment companies or groups that aspire to provide this risk evaluation service when called upon. EPA, the Food and Drug Administration, the Center for Disease Control, the National Institute of Health, and other agencies have had long experience with some of the risk assessment processes and with the contractors who are capable of performing the tasks.

Defining the Alternative

An alternate means of evaluating risk, in the form of a "public risk dialogue report," could ensure a win-win solution to information conflicts. The mechanics of the public risk dialogue system can be simply proffered: a public agency responsible for the regulation of this type of danger, such as a state environmental agency, would establish a list of regional or local contractors capable of doing environmental engineering risk assessments. Environmental planning data, pre-construction designs, product formulations, etc., would flow into the agency from regulated firms under existing rules. No change in quantity of reports would be [30 ELR 10210] made, although fewer document demands might be made if regulators had such a public process "in reserve" if it were needed. The submitting firms can choose to mark the confidential portions as they now may do.⁷⁰ But a firm that marked a confidential piece of data would have a corresponding obligation to engage in a public risk dialogue about the site's safety, when and if the community asked for this type of alternative dialogue through its elected official. The initiation of the public risk dialogue by a listed, qualified organization is an event that the elected official will request. EPA and the company will then negotiate with the requesting organizations. Presumably, EPA and local officials would then subpoena records or take discovery in an enforcement or permit denial case where warranted.

If and when a significant community concern arose, such as a plant's expansion of a tank farm, the relocation of a refining process, new construction at the edge of the perimeter, an environmental permit renewal, a zoning change, or other significant occasion, the neighborhood would have an opportunity to ask the elected official to call on the permit-issuing agency to host a public risk dialogue. Then, one of the pre-certified contractors would be jointly selected by the agency and the company to perform thereview of the confidential data. The report of the contractor would be sent to the elected official and to a named representative of the community group that sought the report, such as the mayor or the chairman of a citizens' advisory board.

Understandable end products would be produced rather than lengthy essays that are unnecessary and costly. The report would be limited to the risk that has been identified for evaluation. To the extent possible, no chemical or process data that would disclose proprietary advantages would be included in the dialogue report.

Because this is an expensive alternative, it is important that the triggering event be more than just a neighbor's passive dislike for a particular site. The threshold of request should be more than nominal. The persons able to request the public risk dialogue, such as for a significant alteration in the scale of releases shown by annual TRI reports for the facility, the first commercial manufacture of a significant amount of a new chemical substance,⁷¹ or a citizen petition for environmental enforcement action, should be elected officials of the county or city. An analogy might be drawn to the mechanisms for site evaluations by the Agency for Toxic Substances and Disease Registry (ATSDR),⁷² but that system is federally funded and centrally managed, whereas this public risk dialogue system relies on subsidiarity, performing the dialogue at the local level to resolve localized concerns.

The cost could be shared two-thirds by the company and one-third by the elected official's entity, such as a county or city. State environmental agencies can decide whether to underwrite the one-third in their states. The public risk dialogue, however, should be a voluntary alternative. Allocating such costs to a private firm without its consent raises "takings" and warrantless search concerns, so cooperation and agreement is essential. Thus, if the company declines to participate, the alternative will not operate. Nevertheless, for industrial managers who consider the options, a mechanism such as public risk dialogues will be ultimately less expensive than traditional disclosure disputes and less likely to jeopardize trade secrets.

Purpose and Benefit of Public Risk Dialogue

The trade off between public health and commercial secrecy is most crucial when the risk question relates to safety of a process, such as a chemical newly brought on site under a premanufacture notification for its first commercial use.⁷³ For example, process controls for handling catalysts are among the know-how that make refineries more efficient. The ability to license that know-how to customers in other nations is an asset that the refiner can convert into a stream of income. In lieu of licensing know-how, the refiner can preserve the asset by declining to license the technology and instead continue to sell its end-products. This potential cash value, which warrants commercial privacy, should be balanced as a matter of public policy against the public need for assurance that the neighborhood facility will operate safely.

The value to having public access to risk information, e.g., "Will the plant explode? Will cancer rates be increased?", is greater than the value of having minutely detailed access to the particular data points such as the recipe. In some contexts of potential public safety concern, the process data could help the neighbors of a refinery to better understand the risks of explosion. A site's explosion potential is not obvious or readily apparent from the raw data, but rather may relate to a particular modified aspect of the proprietary process. An analogy could be drawn to a food regulatory device referred to as a "hazard analysis of critical control points."⁷⁴ The chemical handling process for volatile oxidizer is the potentially hazardous "control point." Failures of safety systems at the control point are the focus of protection when the failure scenario would adversely affect neighbors' health. To win acceptance of zoning changes, new construction permits, air quality variances, etc., it is important to assure neighbors and community leaders that the safety systems will operate successfully.

Defining risk results rather than recipes or diagrams as the core item of public benefit does not mean that particular data points would be uninteresting or have no value. The point is that they do have value for economically interested parties. The voluminous raw data may be useful to a competing firm that is designing its own competitive factory. A process diagram of reverse osmosis filtration with catalyst recycling would advance country X's ability to build its own competing factory without paying royalties to the U.S. innovator. The process flow data, however, means very little to the resident next door. The community wants access to answers, not to a pile of printouts. An advocate who demands "data dumps" for local public consumption is probably neither a representative consumer of such data nor a representative [30 ELR 10211] of the needs of the local public. Here again, the involvement of the elected official is a justified precondition.

Conclusion

The "last war" between disclosure advocates and industry involved hard issues and hard copies. The 21st century framework of debates will be less linear, more digital, and somewhat more similar to the chaos theory. The new availability of evaluation tools will facilitate the translation of complex sets of data into understandable safety and health data. The use of an alternative that is accurate, credible, and efficient makes sense as a replacement for the past modes of conduct. The public risk dialogue process begins with sharing more useful data with an increasingly data-savvy public. Replacing the information wars with a better-directed alternative will help the goal of environmental peacemaking between the regulated community and the general public.

1. Polaroid Corp. v. Costle, 11 Env't Rep. Cas. (BNA) 2134 (D. Mass. June 23, 1978); Proposal Announced to List Dyes, Pigments; Confidential Data Omitted After Suits Filed, 25 Env't Rep. (BNA) 1554 (Dec. 9, 1994).

2. ROGER M. MILGRIM, TRADE SECRETS (1967).

3. Id. The trade secret definitions most often used are found in RESTATEMENT OF TORTS § 757 cmt. b (1938) and the UNIF. TRADE SECRETS ACT § 1(4), 14 U.L.A. 438 (1990)

4. See, e.g., Trade Secrets Act, 18 U.S.C. § 1905.

5. Ruckelshaus v. Monsanto Co., 467 U.S. 986, 14 ELR 20539 (1984).

6. The Second Circuit drew the distinction between EPA access for regulatory decisionmaking, which was legitimate, and access to redistribution to the public, which would raise more serious legal concerns. Union Carbide Agric. Prods. Co. v. Costle, 632 F.2d 1014, 10 ELR 20846 (2d Cir. 1980), rev'd on other grounds sub nom. Thomas v. Union Carbide Agric. Prods. Co., 473 U.S. 568, 15 ELR 20698 (1985).

7. U.S. CONST. amend. V; see Union Carbide, 632 F.2d at 1014, 10 ELR at 20846; see also Ruckelshaus, 467 U.S. at 986, 14 ELR at 20539.

8. See Trade Secrets Act, 18 U.S.C. § 1905; Clean Air Act (CAA), 42 U.S.C. § 7414(c), ELR STAT. CAA § 114(c); Toxics Substances Control Act (TSCA), 15 U.S.C. § 2613, ELR STAT. TSCA § 14.

9. See, e.g, Kewanee Oil Co. v. Bicron Corp., 416 U.S. 470 (1974).

10. 5 U.S.C. § 552.

11. Id. § 552(b)(4). The voluminous case law is covered in 1 JAMES O'REILLY, FEDERAL INFORMATION DISCLOSURE ch. 14 (2d ed. Supp. 1999) [hereinafter FEDERAL INFORMATION DISCLOSURE].

12. The extensive history of the debate over FOIA exemptions is addressed at length in FEDERAL INFORMATION DISCLOSURE, supra note 11, §§ 2.03-2.04.

13. Critical Mass Energy Project v. NRC, 975 F.2d 871, 22 ELR 21373 (D.C. Cir. 1992), cert. denied, 113 S. Ct. 147 (1993).

14. Senate Comm. on Judiciary, Hearings, Freedom of Information, 88th Cong. 198 (1963).

15. Union Carbide Agric. Prods. Co. v. Costle, 632 F.2d 1014, 10 ELR 20846 (2d Cir. 1980), rev'd on other grounds sub nom. Thomas v. Union Carbide Agric. Prods. Co., 473 U.S. 568, 15 ELR 20698 (1985).

16. 28 U.S.C. § 1491.

17. Ruckelshaus v. Monsanto Co., 467 U.S. 986, 1011, 14 ELR 20539, 20545 (1984).

18. 5 U.S.C. § 552(b)(4).

19. 15 U.S.C. §§ 2601-2692, ELR STAT. TSCA §§ 2-412.

20. Id. § 2613, ELR STAT. TSCA § 14.

21. Id. § 2613(b), ELR STAT. TSCA § 14(b).

22. Id. § 2613, ELR STAT. TSCA § 14.

23. CAA, 42 U.S.C. § 7414(c), ELR STAT. CAA § 114(c).

24. FWPCA, 33 U.S.C. § 1318(b), ELR STAT. FWPCA § 308(b). See also RSR Corp. v. Browner, 924 F. Supp. 504, 26 ELR 21353 (S.D.N.Y. 1996).

25. 42 U.S.C. §§ 11001-11050, ELR STAT. EPCRA §§ 301-330.

26. N.J. STAT. ANN. § 34:5A-14 (West 1999).

27. 42 U.S.C. §§ 11001-11050, ELR STAT. EPCRA §§ 301-330. Because it was placed in Title III of the Superfund Amendments and Reauthorization Act (SARA) of 1986, Pub. L. No. 99-499, § 300, 100 Stat. 1613, 1728-58 (1986), EPCRA is also known as SARA Title III.

28. H.R. REP. NO. 99-962, at 292 (1986).

29. 42 U.S.C. § 11023, ELR STAT. EPCRA § 313.

30. H.R. REP. No. 99-962, at 295. New Jersey's list, premised on the degree of a chemical's hazard in case of a release, was combined with Maryland's list, which was premised on total volumes present in the state. The insistence of a Maryland senator at the conference subgroup led to this "shotgun wedding" of dissimilar chemical lists, and in turn led to months of EPA effort to prune the inert or low-risk materials from the statutorily adopted list.

31. See Toxic Chemical Release Reporting: Community-Right-To-Know, 40 C.F.R. § 372 (1999).

32. Including Chemical Use Data in TRI Would Encourage Spying, Industry Says, 28 Env't Rep. (BNA) 1000 (Sept. 26, 1997).

33. 42 U.S.C. § 7412(r), ELR STAT. CAA § 112(r).

34. 40 C.F.R. § 68.150 (1997).

35. Some Advisors Support Internet Caution, Fear Attacks From Issuing Hazard Data, 29 Env't Rep. (BNA) 991 (Sept. 18, 1998).

36. Risk Management Plans Under CAA § 112(r)(7) Amendments, 64 Fed. Reg. 964, 968 (Jan. 6, 1999); James Kennedy, Agency Details Restrictions on CBI Claims, 29 Env't Rep. (BNA) 1744 (Jan. 8, 1999).

37. Polaroid Corp. v. Costle, 11 Env't Rep. Cas. (BNA) 2134 (D. Mass. June 23, 1978).

38. EPA adopted a CBI manual and process safeguards by guidance document rather than as an amendment to its 40 C.F.R. § 2.203 regulations, which set forth the notice to be included in EPA requests, demands, and forms.

39. See, e.g., Notice on Contractor Access to Confidential Business Information Under the Clean Air Act, 63 Fed. Reg. 54476 (Oct. 9, 1998).

40. Chrysler Corp. v. Brown, 441 U.S. 281 (1979).

41. See, e.g., S. REP. No. 98-221, Freedom of Information Reform Act (1983).

42. Exec. Order No. 12600, Commercial Information, 52 Fed. Reg. 23781 (June 25, 1987).

43. 40 C.F.R. § 2.203 (1998).

44. Dozens of disputed commercial confidentiality claims have been resolved and only a few have gone on to the stage of litigation. See cases reported in FEDERAL INFORMATION DISCLOSURE, supra note 11, at ch. 10.

45. 498 F.2d 765, 4 ELR 20385 (D.C. Cir. 1974).

46. Id. at 770, 4 ELR at 20388 (the newly minted term did not appear in the statute or the legislative history of FOIA).

47. See Consumers Union v. Veterans Admin., 301 F. Supp. 796, 801 (S.D.N.Y. 1969), appeal dismissed, 436 F.2d 1363 (2d Cir. 1971); Barceloneta Shoe Corp. v. Compton, 271 F. Supp. 591 (D.P.R. 1967).

48. 975 F.2d 871, 22 ELR 21373 (D.C. Cir. 1992), cert. denied, 113 S. Ct. 147 (1993).

49. A mandatory semiannual discharge report was augmented by more frequent reports to EPA; these more frequent reports were optional but not "voluntary" for FOIA exemption purposes. RSR Corp. v. Browner, 924 F. Supp. 504, 26 ELR 21353 (S.D.N.Y. 1996).

50. EPA Requests for Information, 41 Fed. Reg. 36903 (Sept. 1, 1976); Public Information and Confidentiality Regulations, 50 Fed. Reg. 51654 (Dec. 18, 1985).

51. Proposed Rule on Information and Confidentiality Regulations, 59 Fed. Reg. 60446 (Nov. 23, 1994).

52. CBI Rule on Hold as Regulatory Negotiation Eyed, 26 Env't Rep. (BNA) 17 (May 5, 1995).

53. EPA Regulatory Agenda, 64 Fed. Reg. 20941 (Apr. 26, 1999).

54. Trade Secrecy Claims for Emergency Planning and Community Right-To-Know Information, 40 C.F.R. pt. 350 (1999).

55. 42 U.S.C. § 11023, ELR STAT. EPCRA § 313.

56. Public Information, 40 C.F.R. pt. 2 (1998).

57. Id. § 2.203.

58. Proposal Announced to List Dyes, Pigments; Confidential Data Omitted After Suits Filed, 25 Env't Rep. (BNA) 1554 (Dec. 9, 1994).

59. 42 U.S.C. § 7412(r), ELR STAT. CAA § 112(r).

60. Pub. L. No. 106-40, 113 Stat. 207 (1999).

61. Many trade secrets ripen into patents after the laborious and expensive process is completed; this Dialogue does not relate to licensing of issued patents.

62. RESTATEMENT OF TORTS § 757 cmt. b (1938).

63. WILLIAM CASEY ET AL., ENTREPRENEURSHIP, PRODUCTIVITY AND THE FREEDOM OF INFORMATION ACT (1981).

64. The criteria most relevant to this discussion are EPA's CBI criteria set forth at 40 C.F.R. § 2.204(e)(4)(i-ix) (1998).

65. But some courts recognize a protection against disclosures where reputational loss or harassment is threatened, not economic harm; see Nadler v. Federal Deposit Ins. Corp., 92 F.3d 93, 97 (2d Cir. 1996).

66. Omnibus Consolidated and Emergency Supplemental Appropriations Act of 1999, Pub. L. No. 105-277, 112 Stat. 2681 (1998).

67. Hearings on H.R. 88 Before the House Comm. on Government Reform, 106th Cong. (1999) (testimony of G. Bass, Executive Director, OMB Watch, at 9: disclosure of grantee files "is unwarranted, unnecessary, and unwise").

68. These parallel Dean Cramton's triad of the values of accuracy, efficiency, and acceptability that have been widely cited by administrative law scholars for years. Roger Cramton, A Comment on Trial-Type Hearings in Nuclear Power Plant Siting, 58 VA. L. REV. 585, 592 (1972).

69. See, e.g., Lujan v. Defenders of Wildlife, 504 U.S. 555, 22 ELR 20913 (1992).

70. 40 C.F.R. § 2.203 (1998).

71. 15 U.S.C. § 2604(a).

72. This branch of the Centers for Disease Control and Prevention (CDC) is empowered to study health conditions that may be related to local exposures to toxic chemicals. The ATSDR can select which local conditions it wishes to study. See CDC, Agency for Toxic Substances and Disease Registry (visited Nov. 23, 1999) www.cdc.gov/maso/atsdrfs.htm.

73. Premanufacture Notification, 40 C.F.R. § 720 (1998).

74. See, e.g., Procedures for the Safe and Sanitary Processing and Importing of Fish and Fishery Products, 60 Fed. Reg. 65095 (Dec. 18, 1995) (mandating the application of hazard analysis critical control point principles to the processing of seafood).