concurring:
Our dissenting colleague’s charges notwithstanding, the majority opinion is both faithful to the Supreme Court’s holdings in Metro Broadcasting, Inc. v. FCC and respectful of Congress’s factfinding and poli-cymaking roles.
The question we are asked to decide is whether the gender preference mandated by Congress is substantially related to the achievement of greater variety in broadcasting. See Metro, 497 U.S. 547, 110 S.Ct. 2997, 3008-09, 111 L.Ed.2d 445 (1990). The Court noted, in Metro, that it “must pay close attention to ... the factfinding of Congress when establishing the nexus between minority ownership and programming diversity.” Id. 110 S.Ct. at 3011. It concluded that that nexus had been “corroborated by a host of empirical evidence,” citing in particular the Congressional Research Service analysis that is examined in detail in the majority opinion. See id. at 3017 & n. 31; maj. op. at 396-398.
*403As noted in the. majority opinion, the CRS study reveals an impressive relationship between minority ownership and program diversity. For example, 79% of stations owned primarily by blacks will broadcast black programming as contrasted with only 20% of those owned by non-blacks. This is a ratio of four to one. In the case of Hispanics, the figures are 74% and 10%, for a multiple of more than seven to one; and in the case of Indians and Alaskans, the figures are 46% and 4% — more than a ten-fold increase. No one would question, least of all the majority, that these statistics demonstrate a striking relationship between minority ownership and programming.
By contrast, the differences in programming between stations owned primarily by women and those owned by men are modest: an increase of from 28% to 35% in the case of women’s programming and, in the case of non-black owners, an increase of from 20% to 26% in black programming. These seven and six percentage point increases pale in comparison with the fifty-nine and sixty-four percentage point differentials to be found in the case of black- and Hispanic-owned stations.
The dissent asserts that this disparity is understandable because in the case of gender, the effect of ownership on broadcasting is more likely to be manifested in a diversity of viewpoints than in differences in programming. That may well be the case, but it has not been documented in the record before us. Further research may vindicate this intuition; but until the evidence is marshaled, it is hard to see how the preference accorded women in the FCC’s licensing process can pass constitutional muster.
At base, the quarrel between the majority and the dissent revolves around how substantial the relationship must be between ownership and program diversity in order to justify a gender-based distinction that would otherwise offend the Constitution’s equal protection principles. Our dissenting colleague disagrees with our conclusion that the indicated relationship between the gender of station owners and diversity in programming is not substantial enough to pass constitutional scrutiny. Fair enough. But lines must be drawn; and, with great respect, I do not believe that a disagreement over where it should be drawn warrants the charge that the majority is engaged in judicial activism.
This litigation deals with a sensitive subject, and it is not surprising that it should have aroused some passions. Unfortunately, this case has also proven the occasion for a most serious breach of trust. I refer to an article that appeared on September 30, 1991, in The Legal Times, which purported to report in some detail on the contents of preliminary drafts of the majority and dissenting opinions. The issuance today of those opinions in their final form will demonstrate the general accuracy of the information divulged to The Legal Times.
The seriousness of this violation cannot be overstated. Each member of this panel has been aggrieved by it, as have the parties who brought this case to us for adjudication. Moreover, because one or more of their number has been guilty of a willful breach of trust, this incident must cast a shadow over the dozen or more able young law clerks who had become privy to the preliminary drafts. I say “willful” because the information in the published reports was too detailed to have been the product of inadvertent disclosures.
We cannot, of course, repair the damage that may already have been done to one or more of the parties as a result of this premature disclosure. But we can and must take steps not only to ensure against a repetition in the future, but to demonstrate the seriousness with which we take this violation. I believe the appropriate measure is for this court to initiate a formal investigation in an effort to identify the source or sources of this disclosure, and I urge my colleagues to do so.
The hemorrhaging of confidential information has become endemic in the legislative and executive branches of our government, with untold cost to their ability to function. It is essential that we prevent *404this disease from invading the judiciary, as this would inevitably undermine the public confidence that is one of the major strengths of our legal system.