The government collects an enormous amount of information from companies that it stores, analyzes, and disseminates to government agencies, other companies, and the public. This practice increases the chances that information disclosed to the government that should remain secret does not. Accidental disclosures of confidential and trade-secret information do occur. Over the last few years, several government agencies have inappropriately handled or inadvertently disclosed company trade secrets. In one case, for example, the EPA disclosed one organization’s trade secrets to an environmental organization. In another case, the FDA has been accused of inappropriately disclosing to a prisoner the secret formula to a drug, and posting on its website another company’s trade secrets contained in a New Drug Application.

Disclosure of company trade secrets by the government to the public is already addressed in the elaborate regulatory scheme of agency rules and regulations, as well as in the FOIA case law. However, there is a paucity of case law and other guidance specifically relevant to cases where a company refuses to submit trade-secret information to the government (“refusal-to-submit” cases) and when the government is entitled to a company’s trade-secret information. I propose a “shield or disclose” model that, among other things, makes clear the roles and burdens the various players must assume. It requires a threshold determination that the information in question qualifies for trade-secret protection under the common law. It also requires evidence of need, relevance, and potential harm before a court could order disclosure.


The unique nature of trade secrets—that they exist only so long as they are not disclosed or disclosed in confidence—requires an arrangement that ensures against accidental, unauthorized, or other improper disclosure. The owner of trade-secret information should never make a disclosure, either voluntary or involuntary, without enforceable restrictions against general disclosure. However, when it is determined that it is in the public’s interest that the information be disclosed to the government, a delicate balance must be observed.

Accordingly, a clear model is needed to determine when trade-secret information should be submitted to the government in the first place. The model suggested here addresses disclosure to the government, not the subsequent and separate step of disclosure by the government to the public. The latter is already addressed, albeit not perfectly, in the elaborate regulatory scheme of agency rules and regulations, as well as in the reverse-FOIA case law. Thus, once the government has the information in its possession, whether received voluntarily or through compliance mandates, the current regulations are applicable to protecting them.


The body of law that would allow corporations to refuse to submit proprietary information to the government, when they are not required to do so under a regulatory scheme, is trade-secret law. Because of the dearth of case law and other guidance specifically relevant to refusal-to-submit cases, I have considered a wider body of cases that implicate disclosures of trade secret by the government as well as disclosures made in the context of pending litigation. The end result is what I will refer to as the “shield or disclose” model.

Ultimately, the proposal creates a procedural and substantive path to identify the circumstances under which a court should compel a trade-secret holder to produce its trade-secret to the government. As a general policy matter, when the public interest in the disclosure outweighs the harm to the company from disclosure, the court may justifiably compel disclosure. What does that mean, however? How is it determined? These are the questions for which this framework aims to provide guidance. It is, admittedly, not the sole answer, but rather a modest step in the direction of achieving a more principled approach to refusal-to-submit cases, a step that is grounded in trade-secret law and consistent with the policy considerations that underlie governmental access and disclosure.

1. Company Establishes Trade-Secret Status and Harm

The first step of the process, having both procedural and substantive significance, requires that the trade-secret owner establish that the requested information qualifies for trade-secret protection and that harm will result from disclosure of the trade secret to the government. Whether the information in question meets the status of a “trade secret” should always be the threshold question, and it is the trade-secret owner’s burden to make that showing. While companies often try to claim protection for confidential and proprietary business information, trade-secret protection applies only to the smaller subset of information that qualifies as an actual trade secret. Therefore, in most cases, this first question could be determinative of the entire issue of disclosure because if the information is not a trade secret, then that significantly weakens the argument against disclosure. If the trade-secret owner is unable to establish trade-secret status, then the inquiry likely ends in favor of the government.

The trade-secret owner must then articulate the competitive harm that would be caused from disclosure of the information to the government. This is in keeping with trade-secret law’s focus on protecting against unfair competition and the existing articulation of harm in some of the regulations as competitive harm. For example, SEC regulations require that a business provide information regarding the adverse consequences that could result from disclosure of confidential information, including any adverse effect on its competitive position. Similarly, on the likelihood that the disclosure would result in harm, EPA regulations require a showing of “substantial harmful effects to the business’ competitive position” and “an explanation of the causal relationship between disclosure” and the harm. Moreover, in the FOIA context, competitive harm has been interpreted to mean that the harm flows directly from a competitor rather than from a customer or employee or other source. Thus, a trade-secret owner would need to establish the likelihood that such harm would occur if the information it produced to the government were to be obtained by its competitors.

2. Government Establishes Relevance and Need

Once the trade-secret owner has established the trade-secret status of the information and the harm that is likely to result from its disclosure, the burden then shifts to the party requesting the information (i.e., the government) to prove relevance and need for the information. This is similar to the good-cause burden under the discovery rules. Given the unique nature of a protectable trade secret and the devastating harm that could result from its disclosure, the better policy is that a trade secret should not be ordered produced unless the actual trade secret (as opposed to some other information related to the trade secret) is directly relevant to the inquiry for which it is sought.

Relevance, for the purposes of this proposal, is similar to the standard that has been used under Rule 26 of the Federal Rules of Civil Procedure. However, relevance should probably not be interpreted as broadly for trade-secret purposes as it is under the discovery rules. Whereas the underlying rules and policies in the discovery context favor greater disclosure between the parties, trade-secret law, on the other hand, is grounded in secrecy and the requirement that trade secrets should not be disclosed without appropriate assurances of confidentiality by the receiver. Accordingly, this might suggest that the relevance standard should be interpreted narrower than under the discovery rules. The current FOIA rules and cases do not require that the party requesting information from the governmental agency establish relevance. However, given the higher scrutiny that should be given to the disclosure of trade secrets, it makes sense to require a showing of relevance.

While the FOIA rules do not require a showing of need either, evidence of need is required under my proposal. The government requestor should demonstrate need for the information separate and apart from relevance. This inquiry would focus on such considerations as: (1) whether the information sought is available elsewhere; (2) whether acceptable substitutes for the information can be found from other sources; (3) whether the public interest in receiving the information can only be protected via receipt of the trade-secret information; and (4) whether the public could suffer injury to their health or safety if the information is not released to the government.

3. Court Balances Need Versus Potential Injury

Satisfied that the requested information qualifies for trade-secret protection, the court must ultimately balance the government’s showing of relevance and need against the trade-secret owner’s claim of injury that could result if disclosure is compelled. In considering the government’s need for the information, the court could factor in who the requestor is and the purpose for which the information is sought. None of the existing approaches pay particular attention to these questions, but they could add value when dealing with trade-secret cases. A further consideration may be whether the trade-secret information is critical to the government or the public. If not, perhaps the company can comply without disclosing the specific trade secret. If yes, then disclosure with greater assurance of protection might be advisable, possibly ordering, for instance, that the information is outside the agency’s discretion to disclose or that it be “sealed.”

Moreover, the court could also consider the nature of the trade secret in evaluating need and risk of injury. Because trade secrets can be virtually any kind of business information, it may matter whether the information sought is the secret formula to the company’s core product, or a list of the company’s customers; the encryption code to a black box or the record of drivers’ braking patterns during an unintended-acceleration incident.

The shelf life of the information could also be considered to determine whether the nature of the information is such that it will no longer be secret after a short period of time. It could, for instance, be a marketing-related secret that will be divulged or reverse-engineered after a product is released. In that situation, the court may lean toward not ordering disclosure, since the requestor will likely have access to the information by legitimate means in a relatively short period of time. This assumes, however, that there is no immediate critical need for the information.

Whether the trade-secret owner can persuade the court of the harm that could result if the trade secret is ordered disclosed is a very important part of the balance that the court must aim to achieve. The scope of harm, whether limited to that by a competitor, for instance, compared to a more widespread public harm, may matter. However, defining harm can be difficult. The risk of harm is an important component of this evaluation and could be influenced by the government’s assurances of safeguarding the trade secret.

Besides harm to the company, the court may also consider harm to the public if the information is not produced to the government. This would therefore allow for those circumstances where the health and safety of the public are so threatened that disclosure should be compelled. In other words, the harm to the public would outweigh any competitive harm that the proprietor of the trade secret may suffer. As recognized under well-settled takings principles, the government’s use of information for the public good (with adequate compensation) would not likely violate constitutional norms. Accordingly, a court in its discretion could order disclosure in those circumstances.

4. Court Determines Scope of Order

After weighing the various considerations, a court could find that production of the trade-secret should not be ordered. This will end the inquiry and the trade-secret owner prevails. On the other hand, if the court determines that production should be compelled, then the delicate task of crafting an appropriate protective order will remain. A court should not compel production of a trade secret without a protective order and appropriate safeguards for protection of the secret.

The court could choose from a range of options, depending on the particular case, to determine the appropriate scope of the order. For instance, limited disclosure could be ordered, such that the government may not receive the entire trade secret, but part of it. This does not appear to be the current scheme under FOIA, which is an all-or-nothing approach. Thus, there could be a middle-ground approach, one that would meet the requestor’s need for the information while still protecting the trade-secret owner’s interests. A mosaic approach might also work, where the trade-secret information is disaggregated such that the disaggregated form does not reveal the trade secret, yet it remains valuable information to the requestor. The FTC rules, for instance, provide for the disclosure of disaggregated information to other agencies, and to the extent the secret information can be segregated from the nonsecret information, a portion of the record may be disclosed to a FOIA requestor. In some circumstances, a court could order disclosure contingent upon some payment to the trade-secret owner. This would be akin to a compulsory license where, for instance, the court has deemed that withholding the information from the public will have an injurious effect on the public welfare.


Refusal-to-submit cases raise some delicate issues on both sides, since the interests of all involved parties must be given very serious consideration. While these cases will necessarily be decided on a case-by-case basis, an approach that takes into account trade-secrecy principles, in addition to a more structured approach to government-disclosure policies, will better achieve the balance between secrecy and access. It will also allow for the crafting of more creative solutions that are better able to serve the needs of the respective parties. The shield-or-disclose model presented here helps meet those ideals by making clear the parties’ burdens of proof on showing harm, relevance, and need before trade secrets are to be handed over to the government. Ultimately, it provides a more balanced, more specifically tailored approach that offers more of a middle-ground solution than currently exists.

Professor Rowe’s piece is derived from a law review article which first appeared in the Iowa Law Review. The article more thoroughly presents and explores the background research leading to the development of the shield-or-disclose model, including a discussion of existing agency regulations governing trade secrets. See Elizabeth A. Rowe, Striking a Balance: When Should Trade-Secret Law Shield Disclosures to the Government? 96 Iowa Law Review 791 (2011). Professor Rowe thanks Lamar Miller for providing research assistance on this piece.