Interlocutory appeal on a certified question arising from a decision of the Court of International Trade (CIT)1 denying U.S. Steel's (USS) corporate in-house counsel access to confidential information. We vacate and return.

In Republic Steel Corp., supra, note 1, an action involving a negative preliminary injury determination by the International Trade Commission (ITC), the CIT denied a motion for access by USS' in-house counsel to certain confidential information while granting access to counsel retained by other parties. Relying on an earlier decision in U.S. Steel Corp. v. United States, 569 F. Supp. 870 (C.I.T.1983), vacated on other grounds, slip op. 84-12 (C.I.T. Feb. 24, 1984), the court reiterated its view that the possibility of inadvertent disclosure by in-house counsel warranted denial of access. 572 F. Supp. at 276. That earlier decision, specifically incorporated into the decision on appeal here, acknowledged USS's need for the information but said that the information's nature and volume required a focus on the possibility of inadvertent disclosure. Though it accepted representations that the present in-house counsel are not involved in competitive decisions, the CIT nonetheless denied access to in-house counsel because of their "general position" and "reasonable assumptions that they will move into other roles."

The CIT certified the access question in its decision. 572 F. Supp. at 277. This court granted USS's petition for review of that question on November 10, 1983, under 28 U.S.C. Sec. 1292(a)(1), as amended by Federal Courts Improvement Act of 1982, Pub.L. No. 97-164, Sec. 125(a), 96 Stat. 25, 36 (1982).

The case has proceeded with access granted to retained counsel and denied to in-house counsel.

The United States joins USS in arguing that the CIT's decision constitutes a per se ban on access by in-house counsel and should be reversed in favor of a case-by-case balancing test without regard to whether counsel are in-house or retained.

The ITC takes no position on the present court-denial of access, but seeks to preserve its right to deny access by in-house counsel at the administrative level. Intervenors Companhia Siderurgica Paulista, S.A. (COSIPA) and Usinas Siderurgicas de Minas Gerais, S.A., of Brazil and Companhia Siderurgica Nacional are exporters of steel products seeking affirmance of the present denial. European exporters filed a brief amici curiae urging affirmance. Bethlehem Steel Corporation filed a brief amicus curiae in support of reversal.

Whether the CIT erred in denying the present motion for access.

The authority of the CIT under 19 U.S.C. Sec. 1516a(b)(2)(B) to control access to confidential information in cases before it is not in dispute.2 In exercising that control in this case, the CIT carefully reviewed Atlantic Sugar, Ltd. v. United States, 85 Ct. Cust. 133, C.R.D. 80-18 (1980) and available authorities dealing with access in other fields of law, made clear that its rationale carried no reflection on the unquestioned integrity and unblemished record of USS' in-house counsel in adhering to protective orders, and indicated that retention of outside counsel was a reasonable way for USS to satisfy its recognized need for the requested information. Serving the interest of early and just resolution, the CIT certified to this court the question of whether access may be denied solely because of counsel's in-house status.

Emphasizing congressional concern for confidentiality and the statutory provision, 19 U.S.C. Sec. 1516a(b)(2)(B) for maintenance of confidentiality, the CIT denied access. It did so, however, only to in-house counsel, because of its concern, as it said, "solely with the greater risk of inadvertent disclosure within the corporate setting" (CIT's emphasis).

Because what the CIT called the "extremely potent" information in this case fills several volumes and is intermixed with nonconfidential information, the CIT said "its nature and volume place it beyond the capacity of anyone to retain in a consciously separate category" and that "it is humanly impossible to control the inadvertent disclosure of some of this information in any prolonged working relationship." The CIT recognized that those statements applied equally to retained counsel, but also recognized that applying it to both in-house and retained counsel would render adversarial proceedings impossible.

The CIT's well-taken concern for the nature and scope of the information would be eminently applicable to (and would doubtless complicate) the crafting of a suitable protective order. That concern, coupled with the CIT's emphasis on protection of confidentiality, might have justified denial of access to all and sundry. Once it became clear that access must be granted, however, it was error to deny access solely because of in-house counsel's "general position" and "reasonable assumptions" that present in-house counsel will move into other positions within USS.

The denial of access here rested on the court's stated general assumption that there is "a greater likelihood of inadvertent disclosure by lawyers who are employees committed to remain in the environment of a single company". Denial or grant of access, however, cannot rest on a general assumption that one group of lawyers are more likely or less likely inadvertently to breach their duty under a protective order. Indeed, it is common knowledge that some retained counsel enjoy long and intimate relationships and activities with one or more clients, activities on occasion including retained counsel's service on a corporate board of directors. Exchange of employees between a client and a retained law firm is not uncommon. Thus the factual circumstances surrounding each individual counsel's activities, association, and relationship with a party, whether counsel be in-house or retained, must govern any concern for inadvertent or accidental disclosure.

The CIT distinguished in-house from retained counsel because, as it said, "a clear and more sustained relationship can be presumed as an outgrowth of the employer-employee relationship". It therefore saw exclusion of in-house counsel as providing "a meaningful increment of protection". Like retained counsel, however, in-house counsel are officers of the court, are bound by the same Code of Professional Responsibility, and are subject to the same sanctions. In-house counsel provide the same services and are subject to the same types of pressures as retained counsel. The problem and importance of avoiding inadvertent disclosure is the same for both. Inadvertence, like the thief-in-the-night, is no respecter of its victims. Inadvertent or accidental disclosure may or may not be predictable. To the extent that it may be predicted, and cannot be adequately forestalled in the design of a protective order, it may be a factor in the access decision. Whether an unacceptable opportunity for inadvertent disclosure exists, however, must be determined, as above indicated, by the facts on a counsel-by-counsel basis, and cannot be determined solely by giving controlling weight to the classification of counsel as in-house rather than retained.3

Meaningful increments of protection are achievable in the design of a protective order. It may be that particular circumstances may require specific provisions in such orders. In such cases, the order would be developed in light of the particular counsel's relationship and activities, not solely on a counsel's status as in-house or retained.

In a particular case, e.g., where in-house counsel are involved in competitive decisionmaking, it may well be that a party seeking access should be forced to retain outside counsel or be denied the access recognized as needed. Because the present litigation is extremely complex and at an advanced stage, and because present in-house counsel's divorcement from competitive decisionmaking has been accepted by the CIT, forcing USS to rely on newly retained counsel would create an extreme and unnecessary hardship.

Our decision here bears no relation to, and can have no effect on, ITC's rule establishing a per se ban on disclosure to in-house counsel in its administrative proceedings. That rule is not before the court. The policy of an administrative agency faced with specific tasks and deadlines cannot of course control a trial court's discretion in managing the litigation before it. Congress has granted discretion to control access to confidential information, in cases like the present, to the CIT. Whether the exercise of that discretion in the course of litigation would unacceptably "chill" the willingness to disclose such information at the administrative level is a matter for the Congress. On the other hand, our holding here, that access by retained as well as in-house counsel should be governed by the facts, may serve to reassure disclosers of confidential information.

It is unnecessary for us to resolve the parties' dispute over whether the apparent emphasis on confidentiality in 19 U.S.C. Sec. 1516a(b)(2)(B), or the asserted emphasis on discovery in Rule 26, Fed.R.Civ.P. should control in this case. Though the requirement to consider the facts rather than status of counsel sounds in Rule 26 terms, it relates here only to cases in which the court has decided to grant access in accord with the authorization in the second sentence in 19 U.S.C. Sec. 1516a(b)(2)(B), supra, note 2. Nothing here said diminishes the clear authority of the CIT to deny access to all where the specific facts indicate a probability that confidentiality, under any form of protective order, would be seriously at risk. We do not here reverse the denial of access from which the certified question arose. Nor do we order a grant of access in the case listed in note 1, supra. We hold only that status as in-house counsel cannot alone create that probability of serious risk to confidentiality and cannot therefore serve as the sole basis for denial of access.

We have considered and find it unnecessary to discuss the arguments: that the CIT was here creating a per se rule requiring denial to all in-house counsel of access to any confidential information in all future cases; that the denial of access here constituted a violation of USS' right to choice of counsel or a disenfranchising of counsel without due process; that Rule 26, Fed.R.Civ.P., rather than 19 U.S.C. Sec. 1516a(b)(2)(B), should have been applied; and that the "staleness" of the information sought should dictate access.

The certified question (whether access may be granted to retained and denied to in-house counsel solely on a presumption that inadvertent disclosure by the latter is more likely) is answered in the negative, i.e., a denial of access sought by in-house counsel on the sole ground of their status as in-house counsel is error. In further proceedings, access should be denied or granted on the basis of each individual counsel's actual activity and relationship with the party represented, without regard to whether a particular counsel is in-house or retained.

In light of the foregoing, the order denying access to in-house counsel in the case listed in note 1, supra, must be vacated, and the question returned.

VACATED and RETURNED.

NICHOLS, Senior Circuit Judge, dissenting.

I would affirm because I am not persuaded that CIT Judge Watson abused his discretion. His decision has two things going for it this court does not mention. First, he conforms practice in his court to that of the ITC. We may say the ITC rule is not before us, yet we cannot overlook the anomaly that will exist if the court and the ITC enforce conflicting rules respecting the same documents. Second, the intervenors, original sources of the information in question, are willing for the court to allow disclosure to retained but not to in-house counsel. What they think is important because, if they consider the litigation is conducted in a manner unfair to them and in effect a nontariff barrier to their trade, they could withdraw their marbles from our game and invite their own government to take retaliatory action against United States trade.

Under all the circumstances, Judge Watson well may have thought whatever faults his disposition might suffer from--and hardly could he have imagined it was faultless--alternatives were worse. Factual inquiry into the relationship of in-house counsel with the makers of business policy in their companies, has an appearance, it cannot be denied, of greater fairness. One hopes, but does not much believe, it will not degenerate in practice into an invidious effort to throw doubt on the ability--if not the willingness--of certain members in good standing of the CIT bar, who happen to be currently employed as in-house counsel, to resist pressures to violate protective orders or not to yield "inadvertently." Not in this case, perhaps, but in cases for which this will be a precedent. At best a way is found to prolong the litigation and make it more costly. The CIT judge will have to lay out a pretty rigid method of trial of this issue, one that will keep things within seemly limits and not take forever to implement, thus limiting the damage to what is endurable.

I would be, on remand, inclined to consider seriously adoption of a simple alternative rule which our court majority also seems not to exclude, i.e., if a document is too sensitive to disclose to any counsel of record, in good standing as a member of the CIT bar, it is too sensitive to disclose to any or all other such counsel. This is, I suppose, rejected by the CIT on its theory, as explained by Judge Watson, that the second sentence of 19 U.S.C. Sec. 1516a(b)(2)(B) nullifies the first once the court has examined the material in camera. Apparently the effect of the two sentences is believed to be to achieve practically nothing different from what Fed.R.Civ.P. 26 would effectuate if the Trade Agreements Act of 1979 had said nothing. The court majority here implies something different possibly to be the rule inasmuch as nothing in the second sentence requires grant of access to anyone. The supposed necessity of discriminating between retained and in-house counsel is, or may be, somewhat of a self-created dilemma. While the general rule is that sufficient necessity on the part of the discovering litigant will override any degree of sensitivity, this may not be so where Sec. 1516a(b)(2)(B) is applicable. Such an interpretation would recognize the differences in litigation where foreign traders and governments are so strongly interested in the procedure as well as the outcome, and relieves Congress of the imputation of having enacted futile "weasel" words. The matter has not been briefed and I do not wish to seem to rule upon it, even if, writing as a minority, I could. It seems to me that, without discriminating among counsel or having to decide who is trustworthy, a court might find some other way of dealing with the problem. For example, a court appointed expert, acceptable to both sides for expertise and impartiality, might examine the documents and advise the court as to what they reveal, in sanitized terms sufficient to support a legal conclusion, yet not divulging business or trade secrets.

At any rate, the effect of the decision below, if it had stood, and if United States Steel had still refused to retain outside counsel as the CIT judge hoped it would, is not necessarily denial of justice to United States Steel, but a different thing, denial of the benefit of house counsel's advocacy. If United States Steel's counsel cannot examine these papers, it becomes incumbent on the court to examine them itself, in camera, and arrive at a just and lawful decision using its own very considerable intellectual powers. If this were the result, justice might possibly gain instead of losing, and I say this not meaning to dinegrate the benefit to the court of adversary counsel's advocacy. This is a benefit, a great one, but one the court, if it must, can do without.