MERRITT, Circuit Judge.

The appellant, Brown Williamson Tobacco Company (B W), appeals the decision of the District Court for the Western District of Kentucky dismissing its suit against the Federal Trade Commission (FTC). Brown Williamson challenges the following proposed actions and statements of the FTC which the agency planned to publish in the Federal Register:

(1) the FTC has concluded that its present testing methodology does not accurately assess the "tar" and nicotine yields of B W's Barclay cigarettes;

(2) the FTC's December, 1981 Report, which stated that the "tar" yield of Barclay cigarettes is 1 mg., is inaccurate and should be corrected;

(3) pending a revision in the test methodology, future FTC reports, if any, will not include results for Barclay cigarettes; and

(4) the FTC has concluded that "there is a significant likelihood" that the FTC method of testing does not assess accurately the "tar" and nicotine yields of B W's Kool Ultra and Kool Ultra 100's which use the same filter design as the Barclay cigarette.

This June 25, 1982 announcement concluded a year long study by the FTC into the "tar" and nicotine content of Barclay cigarettes.

I. Background

The FTC has a long history of involvement in the testing and reporting of the "tar" and nicotine content of cigarettes. In 1967, the FTC established its own testing laboratory; the agency has made periodic reports to the public since that time. In 1971, while the FTC was in the process of promulgating a rule requiring the cigarette companies to include "tar" and nicotine figures in their advertisements, the major companies voluntarily entered into an agreement among themselves to advertise the figures.

In June, 1981, the FTC received a complaint from the R.J. Reynolds Tobacco Company, a competitor of B W, alleging that the testing system presently used by the FTC to measure the "tar" content of the Barclay cigarette is inaccurate. The FTC method was never intended to provide precise measurements of "tar" and nicotine delivery to each smoker because consumers smoke cigarettes in different fashions. Instead, the tests were designed to provide consumers with figures by which to compare the many brands of cigarettes on the market.

Cigarette manufacturers have traditionally lowered the "tar" content of cigarettes by allowing air to be mixed with the smoke to dilute the intensity of the smoke. Most low "tar" cigarettes have a filter surrounded by porous paper with one or more rows of small ventilating holes encircling the filter which allow air into the smoke channel. The Barclay cigarette, however, has four lengthwise channels to conduct the air from the ventilating holes directly into the mouth. The air does not mix with the smoke until both are in the smoker's mouth. The amount of air dilution in the Barclay cigarette may be reduced if the smoker crushes or blocks the channels with his lips.

Citing its own studies, R.J. Reynolds argued to the FTC that the unique filter on the Barclay cigarette was being enclosed by smokers' lips so that the claimed reductions in "tar" and nicotine were not occurring with Barclay. R.J. Reynolds claimed that the testing machine does not collapse the filter as does a human smoker and that the machine indicates a much lower "tar" content when testing Barclay than is typically delivered by Barclay to consumers.

After receiving the R.J. Reynolds letter, the FTC solicited comments from the five major cigarette companies including Brown Williamson. The FTC also retained three independent experts who reviewed the extensive studies submitted by the cigarette companies. After a year of study, the FTC determined that the Barclay cigarette was indeed incorrectly measured by the testing machine. To prevent further dissemination to the public of what the FTC believed to be misleading information, the agency in a press release made the announcement described above which it proposed to publish in the Federal Register.

On that same day, June 25, 1982, Brown Williamson filed suit in District Court to prevent the FTC from taking the actions described in the proposed Federal Register notice. Circuit Judge Boyce F. Martin, sitting as District Judge by designation, issued a temporary restraining order so as to preserve the status quo pending a comprehensive review of the issues. The TRO prevented the FTC from:

(1) prohibiting B W from relying upon the FTC's present cigarette testing methodology and upon the figures for Barclay reported in the December, 1981 FTC Report to substantiate B W's advertising claim that Barclay is a 1 mg. "tar," 0.2 mg. nicotine cigarette;

(2) amending the December, 1981 FTC Report as to the "tar" and nicotine content of Barclay;

(3) refusing to continue to test Barclay cigarettes with the FTC's present methodology or to publish the results of these tests in future FTC Reports;

(4) publishing in the Federal Register any notice of changes or proposals to alter the FTC's present cigarette testing or reporting program.

District Judge Ballantine continued the temporary restraining order on July 27, 1982 and further ordered that the administrative record and other documents filed by the FTC be placed under seal. On September 27, 1982 the District Court dissolved the temporary restraining order and denied Brown Williamson's motion for injunctive relief after concluding that the court lacked jurisdiction to hear the case. The District Court ruled that the FTC had not taken final agency action that was reviewable in federal court at that time.

Judge Ballantine did, however, issue a stay identical to the temporary restraining order pending appeal to this Court in order to prevent immediate and irreparable injury to Brown Williamson. The FTC challenged the stay before a motions panel of this Court, but the motion was denied. After oral argument on March 22, 1983, this Court ordered that the stay be lifted, with the exception of the District Court's seal on all documents. We held that the appellant had failed to show that it was likely to succeed on the merits, one of the four requirements for a stay pending appeal.

Brown Williamson maintains in this appeal that the District Court had jurisdiction to decide this case because the actions to be announced in the Federal Register notice constituted final agency action. On the substantive issues, appellant claims first, that the FTC did not follow the procedures mandated by the Administrative Procedure Act, 5 U.S.C. § 551 et seq., governing the promulgation of a rule by an agency. Specifically, B W alleges that the FTC failed to comply with the notice and comment requirements of 5 U.S.C. § 553(b) and (c). Second, Brown Williamson argues that the actions of the FTC were arbitrary, capricious and an abuse of agency discretion. The appellant urges this Court to remand the case for review on the merits by the District Court and to reinstate the stay pending the lower court's ruling on the motion for an injunction.

The FTC argues that the District Court's dismissal of the case was appropriate because its attempt to publish the announcement did not constitute agency action but was merely a statement of future policy. Even if the announcement should fall within the APA's definition of agency action in 5 U.S.C. § 551(13), the FTC maintains that it was not final agency action under 5 U.S.C. § 704 and, therefore, not subject to judicial review. The FTC points to the lack of any enforcement proceeding pending or threatened. The agency stresses that preenforcement judicial intervention is warranted only upon a showing that the aggrieved party must incur unduly burdensome business expenses to comply or face grave risks of criminal and civil sanctions for noncompliance.

The Public Citizen Health Research Group, a non-profit consumer organization which engages in research, education and advocacy on a range of health and safety issues, has filed a brief in this case as amicus curiae. Their primary objective is to have this Court lift or substantially modify the seal placed by the District Court on the administrative record and all other documents filed by the FTC. They maintain that the public has a right of access to these papers under the Freedom of Information Act, the First Amendment, and the common law.

For the reasons discussed below, we hold that the District Court erred when it dismissed this action for lack of jurisdiction. The steps taken by the FTC on June 25, 1982 were final agency action subject to judicial review to prevent substantial hardship to Brown Williamson. We also hold, reaching the merits, that Brown Williamson's claims under the APA are lacking in substance and are unsupported by the record. Finally, we reverse the District Court's order placing the documents in the case under seal and hold that under applicable legal principles they should be released for public inspection as are other court records and documents.

II. Final Agency Action

Judicial review under the Administrative Procedure Act may be sought only by a "person suffering legal wrong because of agency action or adversely affected or aggrieved by agency action within the meaning of a relevant statute. . . ." 5 U.S.C. § 702. Agency action is defined in 5 U.S.C. § 551(13) as "the whole or a part of an agency rule, order, license, sanction, relief, or the equivalent or denial thereof, or failure to act. . . ." The FTC denies that it issued an order or promulgated a rule. The agency seeks to portray its June 25, 1982 announcement as the sole action taken. It analogizes the proposed publication to an announcement of an investigation or an enforcement policy, or the issuance of a complaint — which are generally not reviewable in court.

The FTC fails to acknowledge, however, that the announcement was only a small part of the action taken on June 25, 1982. The agency did notify the public of the Barclay investigation and of its doubts as to the accuracy of the Barclay "ultra-low tar" advertising claim. This alone might not have been subject to immediate review in court. But the FTC also halted the testing of Barclay cigarettes, stated that it would refuse to publish any "tar" or nicotine content figures for Barclay, and amended the 1981 Report which lists Barclay as a 1 mg. "tar" cigarette. As a result, it would now appear to be inappropriate for Brown Williamson to continue to cite the figures in the 1981 Report in its Barclay advertisements. Although the FTC did not order Brown Williamson to desist from making the "ultra-low tar" claim in its advertisements for Barclay, that is the effect of the action taken on June 25, 1982. The FTC may not have labelled its action on that date an order or a rule, but we find, nevertheless, that such dispositive decisionmaking falls within the definition of agency action.

The more difficult question is whether this agency action was final and otherwise ripe for judicial review. Ruling on this issue, the District Court concluded that:

[The] FTC's proposed publication in the Federal Register is not a definitive ruling or regulation. It has no legal force or practical effect on plaintiff's daily business other than the disruptions that accompany any major litigation. Immediate judicial review would serve neither efficiency nor enforcement of the Act. These pragmatic considerations counsel against the conclusion that the proposed publication is final agency action.

We reverse the District Court because we believe the court failed to apply the principles of judicial intervention enunciated by the Supreme Court in Abbott Laboratories v. Gardner, 387 U.S. 136, 87 S.Ct. 1507, 18 L.Ed.2d 681 (1967), and its two companion cases, Toilet Goods Ass'n v. Gardner, 387 U.S. 158, 87 S.Ct. 1520, 18 L.Ed.2d 697 (1967) and Gardner v. Toilet Goods Ass'n, 387 U.S. 167, 87 S.Ct. 1526, 18 L.Ed.2d 704 (1967).

In Abbott, the Supreme Court was asked to review a regulation promulgated by the Food and Drug Administration even though the agency had not initiated an enforcement action. The Food, Drug and Cosmetic Act required that the generic name of a prescription drug be printed prominently on the drug's label. Interpreting this provision, the agency issued a regulation requiring that the generic name appear every time the brand name was printed on the package. The government took the position, and the Third Circuit agreed, that the drug companies would have to wait to test the validity of the regulation until the agency initiated an enforcement proceeding for mislabelling, denied a license for a new drug, or brought a criminal proceeding against a manufacturer. Abbott Laboratories v. Celebrezze, 352 F.2d 286 (3d Cir. 1965).

The Supreme Court reversed the Third Circuit in an opinion which remains the leading authority on pre-enforcement judicial review. The Court set out the following two-step test: (1) is the issue fit for judicial determination; and (2) how great is the hardship to the parties of refusing jurisdiction at this time. The fitness question was resolved in Abbott in favor of review because the Court characterized the case as involving a purely legal question with further factual development unnecessary. Analyzing the hardship prong of the test, the Court emphasized that:

These regulations purport to give an authoritative interpretation of a statutory provision that has a direct effect on the day-to-day business of all prescription drug companies; its promulgation puts petitioners in a dilemma that it was the very purpose of the Declaratory Judgment Act to ameliorate . . . . If petitioners wish to comply they must change all their labels, advertisements and promotional materials. . . . The alternative to compliance — continued use of material which they believe in good faith meets the statutory requirements, but which clearly does not meet the regulation of the Commissioner — may be even more costly. That course would risk serious criminal and civil penalties for the unlawful distribution of `misbranded' drugs.

Id. 387 U.S. at 152-53, 87 S.Ct. at 1517-18. In addition to the hardship factor, the Court noted that review at that time would speed the process of compliance if the regulation proved valid. Pre-enforcement review, in other words, was not calculated by the drug companies to stall or frustrate the operation of the regulation.

The Court applied the Abbott two-prong test in Gardner v. Toilet Goods Ass'n, 387 U.S. 167, 87 S.Ct. 1526, 18 L.Ed.2d 704 (1967), again allowing pre-enforcement review of a controversial FDA regulation. The Food, Drug and Cosmetic Act required the Commissioner of the FDA to approve all color additives in food, drugs and cosmetics. At issue was the Commissioner's determination that the statute applied to dilutents, additives used as diluting agents. Referring to the first part of the Abbott test, the Court determined that the question was essentially a legal one which would not necessarily be made clearer if raised in the context of a specific attempt to enforce the challenged regulation. The Court again emphasized that without immediate review, the companies were faced with an unreasonable choice. They could challenge the regulation through noncompliance, risking criminal seizure or injunctive suits filed by the government, or comply, which would entail substantial business expenses.

In the final companion case, Toilet Goods Ass'n v. Gardner, 387 U.S. 158, 87 S.Ct. 1520, 18 L.Ed.2d 697 (1967), the Supreme Court denied jurisdiction for lack of ripeness. The FDA regulation under scrutiny provided that the Commissioner could suspend certification to any manufacturer who refused to allow FDA employees access for inspections. The Court acknowledged that the regulations constituted final agency action and that the case raised a purely legal issue. Nevertheless, the Court denied review primarily because the drug manufacturers could not show that direct and immediate harm would come if they failed to comply and then challenged the regulation in an enforcement proceeding. The Court also found that it could not resolve the issue adequately without a more fully developed factual record containing information as to how and why the Commissioner was utilizing the regulation requiring access.

The Abbott two-part test remains the key to determining whether to permit pre-enforcement review of agency action. See 4 K. Davis, Administrative Law Treatise 405 (2d ed. 1983); Vining, Direct Judicial Review and the Doctrine of Ripeness in Administrative Law, 69 MICH.L.REV. 1443 (1971).

The issues raised by Brown Williamson in the court below are fit for judicial review. The company challenges the FTC action as violating the APA's rule-making process — a question of law which, under the record before us, requires no further factfinding. Brown Williamson's second contention, that the FTC acted arbitrarily and capriciously, is a mixed question of fact and law. However, the appellant urged below that this issue be resolved upon the facts submitted to the District Court. The FTC argues that the process of determining what Brown Williamson may now advertise is ongoing. The agency argues that, as in the second Toilet Goods case, the issues would be framed better in an enforcement proceeding if and when the FTC sues Brown Williamson for illegal advertising practices. While we agree that any controversy over future advertising campaigns is best left to an enforcement proceeding, the issues before us now involve only those decisions made by the FTC on June 25, 1982. The determination that the Barclay cigarette was incorrectly tested is well documented in the studies and discussions which are before this Court. Further factfinding on the basis for the FTC's actions would not clarify at this time whether the FTC acted inappropriately in the summer of 1982.

The second prong of the test, the hardship factor, clearly counsels in favor of preenforcement review. The resemblance between the Abbott facts and those before us is striking. In both situations, the parties challenging the agency action were faced with significant changes in business operations to comply with the agency's determinations. The impact of the agency action in both cases was "direct and immediate." Abbott, supra, 387 U.S. at 152, 87 S.Ct. at 1517. As Brown Williamson stresses, the key to the successful marketing of the Barclay cigarette has been its portrayal as a less carcinogenic product with the taste and flavor of a more robust cigarette. By eliminating Barclay from the testing and reporting program, the FTC has not merely issued a statement of intent to investigate; if allowed to stand, the FTC's actions will force changes in the marketing strategies of Brown Williamson. The hardship to Brown Williamson of requiring the company to wait to resolve the issues until an enforcement proceeding appears to us to be no less than the hardship faced by the drug companies in Abbott.

The FTC argues that we should look to the hardship standard which the Supreme Court relied upon in Gardner v. Toilet Goods Ass'n, supra, to deny pre-enforcement review. In that case, however, the drug companies were only required to permit access to FDA investigators. Compliance did not entail the expenditures or marketing alterations which will befall B W if it wishes to avoid an enforcement proceeding.

In addition to the hardship to Brown Williamson, we must consider the impact on the FTC and the public of permitting preenforcement judicial intervention. The FTC resists judicial review because the agency would like to proceed informally with Brown Williamson to negotiate a permissible advertising strategy. We believe, however, that these negotiations could only be enhanced by the resolution of this case so that both parties may know the legal effect of the FTC's decisions of June 25, 1982.

In addition, there are clear advantages to the general public of judicial review at this time. The FTC's involvement in cigarette testing and reporting is designed to reduce the health risks of smoking by disseminating current, valid statistics on the content of the many cigarette brands. Declining jurisdiction would only delay the dissemination of the most up-to-date information on Barclay cigarettes. The reliance that smokers who are concerned about health risks place on representations of the "tar" and nicotine content of cigarettes underscores the public interest in resolving this controversy in this forum.

III. The Merits [33] A. Procedural Adequacy

The District Court never ruled on Brown Williamson's two arguments on the merits because the court below erroneously held that it lacked jurisdiction. Rather than remand the case which would entail further delay, we have decided in the interest of judicial economy to reach the merits of this case. After careful review of the entire record, we find that the appellants have failed to present supportable arguments for further disrupting the FTC's proposed program concerning the testing and reporting of Barclay cigarettes.

Brown Williamson first contends that the actions of the FTC constituted rule-making and, therefore, must be invalid because the FTC failed to follow the APA's rule-making procedures as provided in § 553:

(b) General notice of proposed rule making shall be published in the Federal Register, unless persons subject thereto are named and either personally served or otherwise have actual notice thereof in accordance with law. The notice shall include —

(1) a statement of the time, place and nature of public rule making proceedings;

(2) reference to the legal authority under which the rule is proposed; and

(3) either the terms of substance of the proposed rule or a description of the subjects and issues involved.

Except when notice or hearing is required by statute, this subsection does not apply —

(A) to interpretative rules, general statements of policy, or rules of agency organization, procedure, or practice; or

(B) when the agency for good cause finds (and incorporates the finding and a brief statement of reasons therefore in the rules issued) that notice and public procedure thereon are impracticable, unnecessary, or contrary to the public interest.

(c) After notice required by this section, the agency shall give interested persons an opportunity to participate in the rule making through submission of written data, views, or arguments with or without opportunity for oral presentation. After consideration of the relevant matter presented, the agency shall incorporate in the rules adopted a concise general statement of their basis and purpose . . . .

(d) The required publication or service of a substantive rule shall be made not less than 30 days before its effective date, except —

(1) a substantive rule which grants or recognizes an exemption or relieves a restriction;

(2) interpretative rules and statements of policy; or

(3) as otherwise provided by the agency for good cause found and published with the rule.

5 U.S.C. § 553.

The parties spend a great deal of time debating whether the actions taken by the FTC on June 25, 1982 fall within the exceptions in the statute for general statements of policy. We do not need to resolve that issue because we find that the appellant had actual notice and an adequate opportunity to comment on the proposed changes.

The statute makes an exception to the requirement that a notice of proposed rule-making be published in the Federal Register if the persons subject to the rule have actual notice. The documents submitted to the District Court leave no room for doubt that Brown Williamson was repeatedly notified both of the FTC's questions concerning the accuracy of the Barclay 1 mg. "tar" claim and of the agency's contemplation of a change in testing methodology. Congress enacted Section 553 to insure that any rules or regulations promulgated by federal agencies which affect the day-to-day activities of businesses and individuals be conceived in an atmosphere immune from prejudice. But when the purposes of the procedural requirements have been fully met, there is no need for the courts to require rigid adherence to formalistic rules. See Duke City Lumber Co. v. Butz, 382 F. Supp. 362, 372 (D.D.C. 1974).

Reviewing the record, it is difficult to comprehend how Brown Williamson can contend that they received no actual notice of the FTC's intended course of action. On June 11, 1981 — over a year before the proposed Federal Register notice — the FTC sent a letter to Brown Williamson advising the company that the agency had received a complaint from R.J. Reynolds and was studying the possibility of adopting a new testing methodology in order to measure Barclay cigarettes more accurately. The letter stated:

The purpose of this letter is to begin a dialogue between the Commission staff and the members of the cigarette manufacturing industry on the merits of the R.J. Reynolds proposal and to provide all members of the industry with an opportunity to comment informally on the proposal at the earliest possible stage.

App. at A26. On October 9, 1981, the FTC sent a follow-up letter apprising Brown Williamson of the status of the investigation. Although each company had submitted studies during the summer months, the letter stated that it was the staff opinion that additional information was needed. On December 15, 1981, the FTC published in the Federal Register one of its periodic reports on the "tar" and nicotine content of most brands of cigarettes. The report identified Barclay as a 1 mg. "tar" cigarette but it also included three paragraphs describing in detail the R.J. Reynolds complaint that Barclay was incorrectly measured by the FTC smoking machine. FTC Report of "Tar," Nicotine and Carbon Monoxide of the Smoke of 200 Varieties of Cigarettes, 46 Fed.Reg. 61828, Dec. 15, 1981.

By February 1982, the FTC had sent all of the material submitted by the cigarette companies to the three independent experts, Drs. Bock, Guerin and Kozlowski. Each expert concluded that the FTC smoking machine does not measure the Barclay cigarette accurately, and recommended that the testing methodology be changed. App. at A161, A-235 and A243. Copies of all of these reports were sent to Brown Williamson on March 17, 1982. Although there may be some dispute as to the number and dates, it is clear from a Brown Williamson letter dated June 11, 1982 that the company had several opportunities to talk with the Commissioners about the proposed changes shortly prior to the FTC's June 25, 1982 actions. See App. at A348.

From these numerous contacts and written communications, well-documented in the record, it is abundantly clear to us that Brown Williamson had actual notice of the FTC's actions — whether or not such notice was required by the Administrative Procedure Act.

In addition to notice, § 553 requires that interested persons be given an opportunity to comment and otherwise participate in the rule-making procedure. 5 U.S.C. § 553(c). Brown Williamson contends that it "had no advance opportunity whatever to comment on the action taken on June 25th." Appellant's Reply Brief at 16. The appellant tries to distinguish between the opportunities it had to comment on the "wide-ranging inquiry conducted beforehand" and the opportunity to comment on the June 25, 1982 actions of the FTC. We reject this distinction. The record contains multiple invitations for comments from the FTC to Brown Williamson and the other cigarette companies covering every aspect of the investigation and encompassing the actions taken on June 25, 1982.

As discussed above, Brown Williamson was fully aware of the proposals to change the testing methodology and to remove Barclay from the testing and reporting process. The studies submitted by B W throughout the year of agency review addressed in detail the FTC's proposed actions of June 25, 1982. Following the FTC's June 11, 1981 letter apprising B W of the R.J. Reynolds complaint, Brown Williamson representatives met with the FTC staff to present their view that Barclay was correctly labelled a 1 mg. "tar" cigarette. Counsel for Brown Williamson sent an 18 page letter to the FTC on July 16, 1981 describing the meeting and documenting B W's position against revising the "tar" figures for Barclay or the testing methodology. On June 20, 1981, the agency sent each company the submissions of their competitors and encouraged the companies to comment on or criticize the conclusions contained in the reports. On October 23, 1981, after several extensions from the FTC, Brown Williamson filed an extensive report totaling 185 pages, including three studies by the company's own experts. Brown Williamson was then given an opportunity to answer specific questions posed by the three independent consultants. Following the dissemination of the reports filed by the three experts, Brown Williamson was invited to comment on the statements and conclusions contained in the reports. Finally, representatives from and counsel for B W had the opportunity to discuss the company's position orally before the Commissioners prior to June 25, 1982.

The reports, studies and discussions in which Brown Williamson participated focused precisely on whether the FTC should take action regarding the advertising of the Barclay cigarette. The FTC repeatedly allowed and encouraged Brown Williamson to defend the accuracy of the present testing methodology. Brown Williamson knew from the proposals included in the R.J. Reynolds complaint that the FTC was considering revising the "tar" statistics of Barclay. We can see no benefit in requiring the FTC to follow further comment procedures. As the Court of Appeals for the D.C. Circuit wrote in a similar case:

If the purpose of notice when required in any case is to give notice, the appellants-petitioners here had it. There is no showing that they were deprived of the opportunity in any measure to take whatever steps their own situation might suggest, whether by way of counterproposal and comment or by evidence to establish their own position. That they are not happy over the result is clear. That they sustained legal injury is not.

Owensboro On the Air v. United States, 262 F.2d 702 (D.C. Cir. 1958).

B. Agency Abuse of Discretion

In addition to its procedural challenges, Brown Williamson maintains that the FTC's actions were arbitrary, capricious and an abuse of discretion in violation of 5 U.S.C. § 706. Although B W separates this argument into four claims — insufficient evidence, arbitrariness, discrimination, and vagueness — the company essentially objects to the Barclay brand being singled out by the FTC. B W asserts that there was no evidence before the FTC to support any different treatment than that given other cigarettes with filters that can be crushed in the mouth of the consumer.

The record, however, contains studies and opinions by the independent experts and the other cigarette companies which show just such a difference in filter effectiveness between Barclay and the other ultra-low "tar" brands. Dr. Lynn Kozlowski unequivocally stated that "Barclay presents greater risks of high tar and nicotine yields to smokers than do other ultra-low-tar brands." App. at A57. The expert concluded:

To summarize, I think that Barclay is not properly assayed to smokers out of proportion to its ranking on the FTC lists. It should be noted that all ventilated filter cigarettes . . . are subject to a similar violation of the integrity of the rankings, but that Barclay (as supported by the PM air-dilution studies, the Lorillard Uninhaled-Puff Study, and the above analysis of the Gori Studies) is significantly more prone to this violation than are conventional ventilated-filter cigarettes.

App. at A68. The second expert, Dr. Michael R. Guerin, also carefully analyzed the many studies submitted by the five companies and concluded that "[i]t is my carefully considered opinion that the contention of [R.J. Reynolds] and [Philip Morris] is correct in that the current FTC method does not appropriately rank the delivery of cigarettes containing the Barclay-type filter." App. at A235. Finally, Dr. Fred G. Bock reached the same conclusions that the Barclay cigarette is inaccurately tested. He emphasized that "the ventilation studies indicate that smokers handle Barclay cigarettes differently than other 1 mg. cigarettes . . . . The preponderance of data thus indicate that the present FTC tar and nicotine yield values of Barclay cigarettes are misleading." App. at A258.

The FTC did not act arbitrarily. The conclusions of the three experts are consistent and clear. All three experts reviewed the extensive materials filed by the companies and weighed the persuasiveness of the methodologies utilized in the studies. Faced with such overwhelming evidence that Barclay was inaccurately being reported as a 1 mg. "tar" cigarette, the FTC appropriately took action to halt the dissemination of misleading and incorrect statistics to the public.

Nor did the FTC act in an impermissibly discriminatory manner. The studies focused on the difference between the Barclay type of filter and the filters prevalent in the marketplace. The experts all concluded that there were significant differences between the Barclay inaccuracies and the possible errors in testing other brands. The problems of testing Barclay arise from its unique filter construction. The FTC did single out Barclay in its actions of June 25, 1982, but the agency did so for legitimate and well-documented reasons. Agencies must be able to distinguish in their orders among products on the market which subject the public to varying degrees of risk. In this case. Brown Williamson had every opportunity for over a year to show that Barclay delivers as little "tar" and nicotine to the consumer as do other ultra-low "tar" brands. They failed to do so.

IV. The Seal On Documents

The District Court's orders placed all documents filed by the FTC under seal and continued the seal pending this appeal. The Public Citizen Health Research Group has filed a comprehensive amicus brief opposing this action, and Brown Williamson has ably defended it. Because of its importance, we reach the question on our own motion. Under the First Amendment and the common law, we conclude that the District Court erred by failing to state findings or conclusions which justify nondisclosure to the public. The order of the District Court sealing the documents in the case is, therefore, vacated.

According to Brown Williamson's brief and the affidavit of Martin London, appellant's counsel, the parties agreed at the outset of the administrative proceedings that the record be sealed to protect the agency's confidentiality commitment to Brown Williamson and the other four cigarette companies which submitted information to the FTC. After the FTC subsequently objected to the in camera treatment of all of the documents in the judicial proceeding, Judge Ballantine requested that the parties submit a list of the documents which they agreed should not be disclosed, with briefs discussing the remainder. When the FTC failed to meet with counsel for Brown Williamson or to file briefs on the issue, Judge Ballantine simply placed a seal on the entire record in the case.

The Citizen Group argues that the seal was improper on three grounds — the Freedom of Information Act, the First Amendment guarantees of freedom of speech and press, and the common law presumption of access to court documents. Brown Williamson, on the other hand, argues that the FOIA has no relevance outside of the context of an administrative agency's decision whether to release documents to the public. Furthermore, the company asserts that the District Court has broad discretion to control access to documents submitted in cases on the court's docket. The company stresses that this discretion was not abused because the Court properly honored the agreement between the five tobacco companies and the FTC that all information furnished the agency during the year of investigation be kept confidential.

A. The Freedom of Information Act

We disagree with the Citizen Group that the FOIA operates as a limiting standard in this situation. Congress enacted the Freedom of Information Act in order to provide the public with access to records of federal agencies. See, e.g., S.Rep. No. 813, 89th Cong., 1st Sess. 3-6 (1965). Included in the FOIA are a list of exceptions to the general requirement that agencies furnish upon request all documents within their possession. 5 U.S.C. § 552(b). The Freedom of Information Act specifically exempts the federal courts from its disclosure requirements. 5 U.S.C. § 551(1)(B). It is clear that the Act was not intended to restrict the federal courts — either by mandating disclosure or by requiring non-disclosure under the § 552 exemptions. Crystal Grower's Corp. v. Dobbins, 616 F.2d 458 (10th Cir. 1980). Thus, we reject the argument that the District Court below should have limited its seal to those documents not available to individuals under the exemptions to the FOIA.

B. Discretion of Trial Court

Although we reject the suggestion that the FOIA places limitations on the District Court's discretion to seal court documents, the First Amendment and the common law do limit judicial discretion. Brown Williamson points out that courts on occasion have emphasized the discretion of the trial court in this area. In Nixon v. Warner Communications, Inc., 435 U.S. 589, 98 S.Ct. 1306, 55 L.Ed.2d 570 (1978), the Supreme Court stated that "every court has supervisory power over its own records and files, and access has been denied where court files might have become a vehicle for improper purposes." Id. at 598, 98 S.Ct. at 1312. In Krause v. Rhodes, 671 F.2d 212 (6th Cir. 1982), this Court reiterated the Supreme Court's position that "the decision as to access is one best left to the sound discretion of the trial court." Id. at 219. See also, United States v. Criden, 681 F.2d 919 (3rd Cir. 1982); Note, The Common Law Right to Inspect and Copy Judicial Records: In Camera or On Camera, 16 GEORGIA L.REV. 659, 666-72 (1982).

Having "supervisory power" or "discretion" to deny access to documents does not, however, imply that the District Court operates without standards. In this case, the District Court placed a seal on all of the documents filed by the FTC without discussion. The District Court's decision is not insulated from review merely because the judge has discretion in this domain. The District Court's discretion is circumscribed by a long-established legal tradition.

C. The Standards

The English common law, the American constitutional system, and the concept of the "consent of the governed" stress the "public" nature of legal principles and decisions. Throughout our history, the open courtroom has been a fundamental feature of the American judicial system. Basic principles have emerged to guide judicial discretion respecting public access to judicial proceedings. These principles apply as well to the determination of whether to permit access to information contained in court documents because court records often provide important, sometimes the only, bases or explanations for a court's decision.

In the leading case of Richmond Newspapers, Inc. v. Virginia, 448 U.S. 555, 100 S.Ct. 2814, 65 L.Ed.2d 973 (1980), the Supreme Court elaborated on the historical and philosophical underpinnings of the right of access. There the trial court had closed the proceedings to the press and public on the motion of the defendant and without objection from the prosecution on the ground that jurors would improperly obtain information through the media. Reversing the trial court, the Supreme Court in Richmond Newspapers found that the public right of access applies to criminal trials primarily because of the long history of open courtrooms. In England the practice developed from an obligation to attend into a right of access to be enjoyed at will by members of the community. Id. at 565, 100 S.Ct. at 2821. The Supreme Court concluded:

[T]he historical evidence demonstrates conclusively that at the time when our organic laws were adopted, criminal trials both here and in England had long been presumptively open. This is no quirk of history; rather, it has long been recognized as an indispensible attribute of an Anglo-American trial.

Id. at 569, 100 S.Ct. at 2823. See also Globe Newspaper Co. v. Superior Court, etc., 457 U.S. 596, 605-07, 102 S.Ct. 2613, 2619-20, 73 L.Ed.2d 248 (1982).

The Supreme Court's historical argument is based on policy considerations developed in the past that remain valid today. First, public trials play an important role as outlets for "community concern, hostility, and emotions." Richmond Newspapers, supra, 448 U.S. at 571, 100 S.Ct. at 2824. When judicial decisions are known to be just and when the legal system is moving to vindicate societal wrongs, members of the community are less likely to act as self-appointed law enforcers or vigilantes. "The crucial prophylactic aspects of the administration of justice cannot function in the dark; no community catharsis can occur if justice is `done in a corner [or] in any covert manner.'" Id. at 571, 100 S.Ct. at 2824.

Second, public access provides a check on courts. Judges know that they will continue to be held responsible by the public for their rulings. Without access to the proceedings, the public cannot analyze and critique the reasoning of the court. The remedies or penalties imposed by the court will be more readily accepted, or corrected if erroneous, if the public has an opportunity to review the facts presented to the court. In his concurrence, Justice Brennan emphasized this link between access to the courtroom and the popular control necessary in our representative form of government. Id. at 592, 100 S.Ct. at 2835. Although the federal judiciary is not a majoritarian institution, public access provides an element of accountability. One of the ways we minimize judicial error and misconduct is through public scrutiny and discussion.

Finally, Justice Brennan points out that open trials promote "true and accurate fact finding." Id. at 596, 100 S.Ct. at 2838. See also, Globe Newspaper Co. v. Superior Court, etc., 457 U.S. 596, 607, 102 S.Ct. 2613, 2620, 73 L.Ed.2d 248 (1982). When information is disseminated to the public through the media, previously unidentified witnesses may come forward with evidence. See In Re Oliver, supra. Witnesses in an open trial may be less inclined to perjure themselves. Public access creates a critical audience and hence encourages truthful exposition of facts, an essential function of a trial.

The Supreme Court's analysis of the justifications for access to the criminal courtroom apply as well to the civil trial. The Supreme Court has acknowledged the broad application of these principles. Justice Burger's plurality opinion notes that "whether the public has a right to attend trials in civil cases is a question not raised by this case, but we note that historically both civil and criminal trials have been presumptively open." Richmond Newspapers, supra, 448 U.S. at 580 n. 17, 100 S.Ct. at 2829 n. 17. Justice Stewart, concurring, states emphatically that "the First and Fourteenth Amendments clearly gives the press and the public a right of access to trials themselves, civil as well as criminal." Id. at 599, 100 S.Ct. at 2839. The historical support for access to criminal trials applies in equal measure to civil trials. See Gannett Co. v. DePasquale, 443 U.S. 368, 386 n. 15, 99 S.Ct. 2898, 2908-09 n. 15, 61 L.Ed.2d 608 (1979) ("For many centuries, both civil and criminal trials have traditionally been open to the public.") See also Fenner Koley, Access to Judicial Proceedings: To Richmond Newspapers and Beyond, 16 Harv.C.R. — C.L.L.Rev. 430-31 (1981); Cox, Foreword: Freedom of Expression in the Burger Court, 94 Harv.L.Rev. 1, 156 n. 42 (1980); Note, Trial Secrecy and the First Amendment Right of Public Access to Judicial Proceedings, 91 Harv.L.Rev. 1899, 1921-23 (1978).

The policy considerations discussed in Richmond Newspapers apply to civil as well as criminal cases. The resolution of private disputes frequently involves issues and remedies affecting third parties or the general public. The community catharsis, which can only occur if the public can watch and participate, is also necessary in civil cases. Civil cases frequently involve issues crucial to the public — for example, discrimination, voting rights, antitrust issues, government regulation, bankruptcy, etc.

The concern of Justice Brennan that secrecy eliminates one of the important checks on the integrity of the system applies no differently in a civil setting. In either the civil or the criminal courtroom, secrecy insulates the participants, masking impropriety, obscuring incompetence, and concealing corruption.

Finally, the fact-finding considerations relied upon by Justice Brennan obviously apply to civil cases. Openness in the courtroom discourages perjury and may result in witnesses coming forward with new information regardless of the type of the proceeding.

The right of access is not absolute, however, despite these justifications for the open courtroom. Courts have carved out several distinct but limited common law exceptions to the strong presumption in favor of openness. The exceptions to the practice of maintaining openness in the courtroom fall into two broad categories: those based on the need to keep order and dignity in the courtroom and those which center on the content of the information to be disclosed to the public. The first type of access restriction resembles the traditional time, place and manner restrictions on speech. In Richmond Newspapers, supra, the Supreme Court noted that "a trial judge, in the interest of the fair administration of justice, [may] impose reasonable limitations on access to a trial. . . . It is far more important that trials be conducted in a quiet and orderly setting than it is to preserve that atmosphere on city streets." Id. 448 U.S. at 581 n. 18, 100 S.Ct. at 2830 n. 18. Any such regulations must pass the following three-part test: that the regulation serve an important governmental interest; that this interest be unrelated to the content of the information to be disclosed in the proceeding; and that there be no less restrictive way to meet that goal. See United States v. O'Brien, 391 U.S. 367, 377, 88 S.Ct. 1673, 1679, 20 L.Ed.2d 672 (1968). These limitations on access, such as regulating the number of spectators or the use of flashbulbs or cameras, have been accepted in many instances as based on the legitimate societal interest in protecting the adjudicatory process from disruption. See, e.g., Illinois v. Allen, 397 U.S. 337, 343, 90 S.Ct. 1057, 1060, 25 L.Ed.2d 353 (1970); Sheppard v. Maxwell, 384 U.S. 333, 354, 86 S.Ct. 1507, 1517, 16 L.Ed.2d 600 (1966); Fenner Koley, Access to Judicial Proceedings, To Richmond Newspapers and Beyond, 16 Harv.C.R. — C.L.L.Rev. 444-46 (1981).

It is the second group of limitations on access to court proceedings that concerns us here. Under the common law, content-based exceptions to the right of access have been developed to protect competing interests. In addition to the defendant's right to a fair trial, these interests include certain privacy rights of participants or third parties, trade secrets and national security. See Nixon v. Warner Communications, Inc., 435 U.S. 589, 598, 98 S.Ct. 1306, 1312, 55 L.Ed.2d 570 (1978); Note, Trial Secrecy and the First Amendment Right of Public Access to Judicial Proceedings, 91 Harv.L.Rev. 1899 (1978).

Simply showing that the information would harm the company's reputation is not sufficient to overcome the strong common law presumption in favor of public access to court proceedings and records. In Joy v. North, 692 F.2d 880 (2d Cir. 1982), for example, the special litigation committee of a bank sought to have their report placed under court seal. The report recommended to the corporation that a stockholder derivative suit be terminated. The committee argued in part that the report contained a candid review of internal business operations of the bank which, if made public, would adversely affect the bank and the local community. Lifting the District Court's seal, the Second Circuit said:

[A] naked conclusory statement that publication of the Report will injure the bank in the industry and local community falls woefully short of the kind of showing which raises even an arguable issue as to whether it may be kept under seal. The Report is no longer a private document. It is part of a court record. Since it is the basis for the adjudication, only the most compelling reasons can justify the total foreclosure of public and professional scrutiny. The potential harm asserted by the corporate defendants is in disclosure of poor management in the past. This is hardly a trade secret.

Id. at 894. The Second Circuit was responding in the case above to the natural desire of parties to shield prejudicial information contained in judicial records from competitors and the public. This desire, however, cannot be accommodated by courts without seriously undermining the tradition of an open judicial system. Indeed, common sense tells us that the greater the motivation a corporation has to shield its operations, the greater the public's need to know. In such cases, a court should not seal records unless public access would reveal legitimate trade secrets, a recognized exception to the right of public access to judicial records.

Brown Williamson seeks to justify the imposition of the court-ordered seal on the FTC documents in this case by pointing to the confidentiality agreement between the five cigarette companies and the FTC. There is no trade secret issue. All of their major competitors have had complete access to the documents. The cigarette companies voluntarily submitted their reports to the FTC under the Federal Trade Commission Act, 15 U.S.C. § 57b-2(c) which provides for confidential treatment of certain information as follows:

(c)(1) All information reported to . . . the Commission . . . shall be considered confidential when so marked by the person supplying the information and shall not be disclosed. . . .

Brown Williamson maintains that the understanding between the agency and the companies entered into at the time when the documents were submitted and which is sanctioned by the FTC Act should override the presumption of public access. In other words, B W seeks to have an exception created to maintain in a court proceeding the confidentiality promised during an agency investigation.

The Federal Trade Commission Act, however, specifically limits the confidentiality provisions so that they apply only to the agency:

(2) Any disclosure of relevant and material information in . . . judicial proceedings to which the Commission is a party shall be governed by . . . court rules or orders. . . .

15 U.S.C. § 57b-2(d)(2). Thus, the court must apply the rules and principles governing the right of access to court documents as in any other civil case. The confidentiality agreement between the parties does not bind the court in any way.

We decline to carve out an exception to the right of access in order to protect the secrecy of an administrative record. The public has a strong interest in obtaining the information contained in the court record. The subject of this litigation potentially involves the health of citizens who have an interest in knowing the accurate "tar" and nicotine content of the various brands of cigarettes on the market. The public has an interest in knowing how the government agency has responded to allegations of error in the testing program. The public has an interest in ascertaining what evidence and records the District Court and this Court have relied upon in reaching our decisions.

Accordingly, the orders of the District Court sealing documents in the case are vacated. The order of the District Court dismissing the case on jurisdictional grounds is reversed. Judgment will be entered for the defendant on the merits for the reasons stated.