OPINION OF THE COURT

BECKER, Chief Judge.

This proceeding is before us pursuant to 2 U.S.C. § 437h, which channels constitutional challenges to the Federal Election Campaign Act, 2 U.S.C. § 431 et seq. ("FECA"), as amended, directly to the en banc Court of Appeals. The present challenge was filed in the District Court for the Middle District of Pennsylvania by Renato P. Mariani. A criminal indictment pending in that court charges Mariani and other officers of Empire Sanitary Landfill, Inc., and Danella Environmental Technologies, Inc., with violating the FECA, 2 U.S.C. § 441b(a) and 441f, by making campaign contributions to a number of candidates for federal office through enlisting company employees and others to forward contributions to the candidates that were thereafter reimbursed by one of the companies. Mariani argues that §§ 441b(a) and 441f violate the First Amendment to the United States Constitution.

Mariani's principal argument regards "soft money," or funds lawfully raised by national and congressional political party organizations for party-building activities from corporations, labor unions, and individuals who have reached their federal direct contribution limits. Soft money is sometimes used to fund so-called "issue advocacy," advertisements that advocate a candidate's positions or criticize his opponents without specifically urging viewers to vote for or defeat the candidate. Issue ads are often only marginally distinguishable from ads directly supporting a candidate, which corporations cannot lawfully fund under the FECA.

Mariani contends that § 441b(a), which proscribes corporate contributions made directly to candidates for federal office, has been completely undermined by the staggering increase in recent years of the amount of corporate soft money donations. In Mariani's submission, this avalanche of soft money has made § 441b(a) so underinclusive, and so incapable of materially advancing the intended purpose of the federal election statute, that it must be struck down. Alternatively, because the bellwether cases in this area, including Buckley v. Valeo, 424 U.S. 1 (1976) (per curiam), validate statutes limiting campaign contributions, but not banning them outright, and recognize that corporate speech is protected under the First Amendment, see First National Bank of Boston v. Bellotti, 435 U.S. 765 (1978), Mariani challenges the total ban on direct corporate contributions as inconsistent with the First Amendment. Mariani also challenges the constitutionality of § 441f, which prohibits making campaign contributions in the name of another to a candidate for federal elective office.

The Supreme Court has construed § 437h so that, if a district court concludes that a challenge to the FECA is frivolous, the court may dismiss the case without certifying it. See California Med. Ass'n v. Federal Election Comm'n, 453 U.S. 182, 193-94 n. 14 (1981). The District Court concluded that the challenge to § 441b(a) was not frivolous, made comprehensive findings, and certified Mariani's challenge to this Court. Section 437h, as construed by the Supreme Court, required the District Court to make fact findings. Many of the District Court's findings were stipulated to by the parties and are uncontested. The government and the Federal Election Commission ("FEC"), however, assail other findings and the Court's 21 ultimate findings of fact as being excessive or beyond its powers. They also argue that a number of them, including the ultimate findings, are unsupported by the record. Our review of the District Court's findings, made in a setting outside the traditional adversary crucible, is not deferential. As we note in section II, we agree that some of the District Court's findings are unsupported by proper evidence and that some stray from appropriate fact finding into legal conclusions. But even assuming that the role of soft money is that asserted by Mariani and found by the District Court, we conclude that the record could not support a holding that § 441b(a) violates the First Amendment.

The government and the FEC not only defend the constitutionality of §§ 441b(a) and 441f, but contend that Mariani's challenges are legally frivolous and thus never should have been certified to the en banc court. They also submit that the District Court employed an insufficiently stringent standard for measuring frivolousness. We are satisfied that the District Court did not apply an incorrect standard of legal frivolousness and that it acted correctly in not dismissing the case without certifying it, at least with respect to the challenges to § 441b(a), for which it made an independent assessment of frivolousness. Though the District Court did not make an independent assessment of the frivolousness of the challenge to § 441f as it should have, the government does not challenge the lack of an independent assessment here, and because the pending criminal case awaits a determination of this action, we will reach the challenges to § 441f without remanding for such a determination.

Although not legally frivolous, Mariani's challenge to § 441b(a) fails. As we explain in detail, both the underinclusiveness and outright ban challenges are interred by the Supreme Court's jurisprudence in the area. See especially Austin v. Mich. Chamber of Commerce, 494 U.S. 652 (1990), and Federal Election Comm'n v. Nat'l Right to Work Comm., 459 U.S. 197 (1982). Although Mariani's factual portrayal of the impact of soft money on contemporary elections is impressive, it falls short. Section 441b(a) is not fatally underinclusive under our precedents, because we cannot say that there is no meaningful distinction between hard and soft money. We cannot exchange our robes for togas; any reform in this area must be sought from Congress.

Finally, we conclude that the challenge to § 441f is patently without merit. Accordingly we shall enter judgment in favor of the government.

I. Procedural History

In October 1997, the United States filed an indictment charging Mariani and several other individuals with, inter alia, violating the FECA. That action, United States v. Mariani, No. 3:CR-97-225, is pending before the District Court. The indictment charges that between August 1994 and December 1996, Mariani and other officers and employees of Empire Sanitary Landfill, Inc. ("Empire") and Danella Environmental Technologies, Inc. ("Danella") solicited numerous employees of the corporations, as well as business associates, friends, and family members, to make contributions to the campaigns of designated candidates for federal election. According to the indictment, these contributions were reimbursed either directly or indirectly by Empire. The indictment also alleges that Mariani and other officers and employees at Empire and Danella made individual contributions to these federal candidates, which were also reimbursed by Empire.

More particularly, the indictment alleges that in April 1995, Mariani and other officers and employees of Empire and Danella contacted employees, associates, friends and family members in an effort to raise funds for the New Jersey Steering Committee, a state fundraising arm of the Robert Dole campaign for President. Contributors allegedly were asked to write personal checks in amounts of $1,000 (or, in the case of couples, $2,000) and were reimbursed with Empire corporate funds. It is also alleged that on April 29, 1995, Mariani and another defendant in the criminal case, Michael Serafini, attended a Steering Committee luncheon at which they handed an envelope containing the contributions to Dole campaign officials. When the Dole campaign reported the contributions to the Federal Election Commission ("FEC"), its filing allegedly attributed these $80,000 worth of contributions to the individual contributors, rather than to Empire. The Dole contributions came approximately ten days prior to a vote in the Senate on the Interstate Transportation of Municipal Waste bill, in which Empire and Danella were interested. Dole was the Senate majority leader at the time.

The indictment charges Mariani (and others) with violations of 2 U.S.C. § 441b(a) and 441f. Section 441b(a) of the FECA prohibits any corporation from making any contribution in connection with any campaign for federal office and renders it unlawful for any officer of a corporation to consent to any prohibited corporate contribution. Section 441f of the FECA, the conduit contribution ban or "anti-conduit" provision, prohibits one from making a contribution "in the name of another person" or "knowingly permit[ting] his name to be used to effect such a contribution." 2 U.S.C. § 441f. Mariani moved to dismiss the FECA charges in the indictment and simultaneously filed this action against the United States seeking declaratory relief pursuant to 2 U.S.C. § 437h. The FEC was granted leave to intervene as a defendant.

Section 437h provides that

any individual eligible to vote in any election for the office of President may institute such actions in the appropriate district court of the United States, including actions for declaratory judgment, as may be appropriate to construe the constitutionality of any provision of [FECA]. The district court immediately shall certify all questions of constitutionality of this Act to the United States court of appeals for the circuit involved, which shall hear the matter sitting en banc.

2 U.S.C. § 437h. The Supreme Court has construed § 437h so that, if a plaintiff brings a claim that is frivolous, a district court may dismiss the case without certifying it. See California Med. Ass'n v. Federal Election Comm'n, 453 U.S. 182, 193-94 n. 14 (1981). The Supreme Court also has interpreted § 437h to require the district court to develop a record and make findings of fact sufficient to allow the en banc court of appeals to decide the constitutional issues. See Bread Political Action Comm. v. Fed. Election Comm'n, 455 U.S. 577, 580 (1982) ("[T]he District Court, as required by § 437h, first made findings of fact and then certified the case. . . ."). The District Court concluded that the challenge to § 441b(a) was not frivolous, and that the interests of judicial economy "militated against" a separate determination that the challenge to § 441f was not frivolous. See Mariani v. United States, 80 F. Supp.2d 352, 355 (M.D.Pa. 1999). The District Court then made comprehensive findings and certified the challenge to this Court.

II. The District Court's Findings of Fact

Some of the District Court's findings are disputed, are unsupported by proper evidence, or go beyond appropriate fact finding into legal conclusion. For example, an opinion expressed by the New York Times Editorial page that one individual's experiences with the Democratic National Committee "deepen the cynicism of Americans" is not a proper evidentiary source for a finding that Americans have become more cynical about government as a result of the role of soft money in the political system. See Mariani v. United States, 80 F. Supp.2d 352, 412 (M.D.Pa. 1999). Similarly, the very title of the segment of the findings called "Due to the Effects of Soft Money on the Political System, FECA is not Serving the Goals it was Intended to Serve," id. at 418, as well as the finding that "[m]ost issue ads are financed in large part with soft money . . . from sources and in amounts that the FECA was meant to prohibit," id. at 377, demonstrate that the fact-finding effort sometimes metamorphosed into conclusions regarding the legal issues in this case. Id. at 418-19. Given the unique procedural posture of the case, we need not (and do not) defer to such findings in our analysis. Although some of the District Court's findings went beyond what was proper both as a matter of evidence and by crossing the line into forming legal conclusions, the court compiled an impressive factual showing that soft money plays an increasingly large role in federal elections.

Contributions made to or expenditures made on behalf of candidates for federal elective office are referred to as "hard money." Under § 441b(a), corporations are not permitted to make contributions of hard money to campaigns for federal office. Corporations can, however, make contributions to political parties in unlimited amounts. These contributions, which are referred to as "soft money," can be used to fund "issue advocacy." "Issue advocacy" includes advertisements or other campaign materials that advocate positions supported by a candidate, often comparing those positions with those of an opponent, without directly advocating the election of the candidate. Donors of soft money are able to avoid the FECA contribution limits and disclosure requirements applicable to hard money and direct advocacy. The amount of soft money contributed in each election cycle has grown tremendously in the last two decades, from about $19 million in 1980 to more than $260 million in 1996. Soft money donations by the 544 largest public and private companies more than tripled between 1992 and 1996.

With respect to Mariani's challenge, the parties agree on the following facts. Candidates for federal elective office help their parties raise soft money. Candidates who raise large amounts of soft money often receive more support from their party than candidates who are less effective at raising soft money. Committee officials often act as intermediaries between donors and candidates.

Soft money is used to fund (or partially fund) issue advocacy that, on occasion, is hard to distinguish from direct advocacy for a particular candidate for federal office. Campaigns sometimes coordinate with outside entities regarding these ads. These ads promote or criticize federal candidates in order to influence the outcome of elections, although avoiding words of direct advocacy such as"vote for," "elect," or "defeat."

Corporations play an important role in campaign finance. Candidates for federal elective office often know which corporations are large contributors of soft money. Because there are no limits on soft money contributions, soft money is easier to raise than hard money. Soft money contributions of corporate treasury funds can result in access (and thus a forum to express their interests) for corporate officials to high government officials, including elected officials, as well as to candidates for federal elective office. Large and repeat donors sometime get more access than other donors, and donating soft money can be a more effective means for getting access than hard money. Corporate soft money contributions enable corporations to some extent to circumvent the corporate hard money contribution ban and support (indirectly) candidates for federal elective office.

Corporations are solicited for and give large sums of soft money in federal elections; according to reports filed with the FEC, during the 1994 and 1998 election cycles, corporations donated more than 50 percent of all itemized soft money contributions. Additionally, in the 1995-95 election cycle, corporations in industries in which legislation was contemplated gave large sums of soft money.

III. The Test for Frivolousness

In California Med. Ass'n v. Fed. Election Comm'n, 453 U.S. 182, 193-94 n. 14 (1981), the Supreme Court stated that "we do not construe § 437h to require certification of constitutional claims that are frivolous." The Court cited with approval a district court decision from an in forma pauperis action that employed the standard from the in forma pauperis statute, 28 U.S.C. § 1915(e)(2)(B), to determine whether a challenge to FECA was frivolous. See id. at 193-94 n. 14 (citing Gifford v. Congress, 452 F. Supp. 802 (E.D.Cal. 1978)). The in forma pauperis statute authorizes a district court to dismiss sua sponte any action that it determines to be legally frivolous. An action is not frivolous under the statute where the complaint raises an arguable question of law that ultimately will be resolved against the plaintiff. Neitzke v. Williams, 490 U.S. 319, 328 (1989). The District Court applied the standard for frivolousness set forth in Neitzke and certified Mariani's challenge to the en banc Court of Appeals after concluding that "it cannot be said that the constitutional challenges are plainly foreclosed by existing precedent." Mariani v. United States, No. 3 CV-98-1701 (March 25, 1999).

The government and the FEC argue that the District Court should have used a more exacting standard for frivolousness and rejected Mariani's challenge. They submit that the correct standard is that set forth by the Ninth Circuit in Goland v. United States, 903 F.2d 1247, 1257 (9th Cir. 1990), which viewed the role of the District Court as akin to that of a single judge deciding a motion to convene a three-judge court to hear a constitutional challenge and noted that this standard is closer to the standard used to review a claim under FED. R. CIV. P. 12(b)(6) than it is to the in forma pauperis standard.

We need not decide which standard applies, because under either standard Mariani's claim is not frivolous. As the Ninth Circuit noted, a genuinely new variation on an issue raised under a particular section of the FECA that already has been challenged and upheld may give rise to a nonfrivolous challenge to that section: "[o]nce a core provision of FECA has been reviewed and approved by the courts, unanticipated variations also may deserve the full attention of the appellate court. At the same time, not every sophistic twist that arguably presents a `new' question should be certified." Goland v. United States, 903 F.2d at 1257; see also Khachaturian v. FEC, 980 F.2d 330, 331 (5th Cir. 1992). Mariani's challenge to § 441b(a) is not simply a sophistic twist, but can fairly be characterized as a new challenge based on the rise in importance in campaign finance of soft money and issue advocacy. Moreover, the facial validity of the statute never has been squarely determined by the Supreme Court.

The District Court did not make an independent assessment of the frivolousness of the challenge to S 441f. Hereafter, district courts considering challenges to separate provisions of the FECA should make the required determination regarding frivolousness for each of the challenges. However, because the government does not challenge the lack of an independent assessment here, and because the pending criminal case awaits a determination of this action, we will reach the challenges to S 441f without remanding for a determination regarding frivolousness.

IV. The Challenge to § 441b(a)

Section 441b(a) bans corporations and unions from using funds from their corporate treasuries to contribute to or make expenditures in connection with any campaign for federal office. See 2 U.S.C. § 441b(a). In Fed. Elec. Comm'n v. Nat'l Right to Work Comm., 459 U.S. 197, 208-09 (1982), the Supreme Court chronicled the history of § 441b(a):

Seventy-five years ago Congress first made financial contributions to federal candidates by corporations illegal by enacting the Tillman Act, 34 Stat. 864 (1907). Within the next few years Congress went further and required financial disclosure by federal candidates following election, Act of July 25, 1910, 36 Stat. 822, and the following year required pre-election disclosure as well. Act of August 19, 1911, 37 Stat. 25. The Federal Corrupt Practices Act, passed in 1925, extended the prohibition against corporate contributions to include "anything of value," and made acceptance of a corporate contribution as well as the giving of such a contribution a crime. 43 Stat. 1070.

The first restrictions on union contributions were contained in the second Hatch Act, 54 Stat. 767 (1940), and later, in the War Labor Disputes Act of 1943, 57 Stat. 167, union contributions in connection with federal elections were prohibited altogether. These prohibitions on union political activity were extended and strengthened in the Taft-Hartley Act, 61 Stat. 136 (1947), which broadened the earlier prohibition against contributions to "expenditures" as well. Congress codified most of these provisions in the Federal Election Campaign Act of 1971, 86 Stat. 3, and enacted later amendments in 1974, 88 Stat. 1263, and in 1976, 90 Stat. 475.

Under Buckley v. Valeo, 424 U.S. 1, 19, 22 (1976) (per curiam), it is clear that spending for political campaigns is protected speech that implicates both the right to free expression and the right of free association. Moreover, because there is "no support in the First or Fourteenth Amendment, or in the decisions of this Court, for the proposition that speech that otherwise would be within the protection of the First Amendment loses that protection simply because its source is a corporation," First Nat'l Bank of Boston v. Bellotti, 435 U.S. 765, 784 (1977), the ban on corporate contributions under § 441b(a) is subject to the same level of scrutiny as other regulations limiting spending for political campaigns. In Buckley, 424 U.S. at 16, the Court held that limitations on spending for campaigns should be subjected to "exacting scrutiny": "this Court has never suggested that the dependence of a communication on the expenditure of money operates itself to introduce a nonspeech element or to reduce the exacting scrutiny required by the First Amendment." The Court added that the First Amendment guarantee "has its fullest and most urgent application precisely to the conduct of campaigns for political office." Id. at 15 (citing Monitor Patriot v. Roy, 401 U.S. 265, 272 (1971)).

Buckley, of course, distinguished campaign contributions from direct expenditures, striking down a limit on expenditures while upholding a limit on campaign contributions. As the Court's recent decision in Nixon v. Shrink Missouri Gov't PAC, 120 S.Ct. 897, 904 (2000) (citing Buckley, 424 U.S. at 20-21), explains, in the area of contributions, even under the exacting scrutiny standard, "limiting contributions [leaves] communication significantly unimpaired." "[U]nder Buckley's standard of scrutiny, a contribution limit involving `significant interference' with associational rights could survive if the government demonstrated that contribution regulation was `closely drawn' to match a `sufficiently important interest.'" Shrink Missouri, 120 S.Ct. at 904 (citation omitted). Accordingly, in considering Mariani's challenge to § 441b(a), while we treat campaign contributions from the corporate treasury as speech and subject the ban on them in § 441b(a) to exacting scrutiny, we do so against a background principle that limits on contributions — though not necessarily bans on contributions — can withstand this scrutiny if they are "`closely drawn' to match a `sufficiently important interest.'"

The District Court certified two issues regarding S 441b(a) to this Court. The first is whether the prohibition in § 441b(a) on contributions by corporations from corporate treasuries to candidates for federal elective office is unconstitutional on its face. The second is whether the prohibition in § 441b(a) on contributions by corporations from corporate treasuries to candidates for federal office, in the context of the presently existing law that otherwise permits corporations to expend unlimited amounts of corporate funds to influence the outcome of federal elections (via soft money contributions), violates the First Amendment.

A. The Constitutionality of § 441b(a) on its Face

In considering the $1,000 contribution limit at issue in Buckley, the Supreme Court stressed the importance of the right to association through support of the candidate of one's choice:

[T]he primary first amendment problem raised by the Act's contribution limitations is their restriction of one aspect of the contributor's freedom of political association . . . [T]he right of association is a `basic constitutional freedom,' Kusper v. Pontikes, 414 U.S. 51, that is "closely allied to freedom of speech and a right which, like free speech, lies at the foundation of a free society." Shelton v. Tucker, 364 U.S. 479, 486 (1960). In view of the fundamental nature of the right to associate, governmental "action which may have the effect of curtailing the freedom to associate is subject to the closest scrutiny." NAACP v. Alabama, [357 U.S.] at 460-461.

Buckley, 424 U.S. at 24-25 (internal citations partially omitted).

Nevertheless, the Court concluded that the $1,000 limit was constitutional. The Court identified two principal reasons for upholding the limit. First, the Court recognized a strong governmental interest in deterring corruption and the appearance of corruption in campaign finance, particularly from large contributions. Id. at 28; see also id. at 30 ("Congress was justified in concluding that the interest in safeguarding against the appearance of impropriety requires that the opportunity for abuse inherent in the process of raising large monetary contributions be eliminated."). Second, the Court concluded that the $1,000 limit was narrowly tailored insofar as it still permitted individual donors to register their political preferences in a substantial way, reasoning that the expressive value of the contribution lies in the act of contributing rather than the amount given. See id. at 21. Accordingly, Buckley seems to leave open the question whether an outright ban on campaign contributions — such as that found in § 441b(a) — would pass constitutional muster.

The government and the FEC argue that, even if Buckley left the door open for a constitutional challenge to an outright ban, Federal Election Comm'n v. National Right to Work Comm, 459 U.S. 197 (1982) (hereinafter NRWC), slammed the door shut. In NRWC, the Supreme Court addressed indirectly the issue of limiting direct corporate contributions to candidates. There, the Court upheld federal restrictions upon corporate solicitation of campaign funds from individuals found in a subsection of S 441b441b(b)(4)(c) — that prohibits nonstock corporations from soliciting funds to be used for political purposes (through a separate segregated fund) from people who are not members of the corporation. See id. at 198 n. 1, 205-11. 103 S.Ct. 552.

Subsection 441b(b)(4)(c) permits corporations to make limited campaign contributions from separate segregated funds solicited explicitly for that purpose. See id. at 201-02. In upholding the statute, the Court suggested that Congress could prohibit direct contributions by corporations to candidates for public office, stating that

The first purpose of § 441b, the government states, is to ensure that substantial aggregations of wealth amassed by the special advantages which go with the corporate form of organization should not be converted into political "war chests" which could be used to incur political debts from legislators who are aided by the contributions. See United States v. United Automobile Workers, 352 U.S. 567, 579, 77 S.Ct. 529, 535, 1 L.Ed.2d 563 (1957). The second purpose of the provisions, the government argues, is to protect the individuals who have paid money into a corporation or union for purposes other than the support of candidates from having that money used to support political candidates to whom they may be opposed. See United States v. CIO, 335 U.S. 106, 113, 68 S.Ct. 1349, 1353, 92 L.Ed. 1849 (1948). We agree with the government that these purposes are sufficient to justify the regulation at issue.

Id. at 207-08. See also Fed. Election Comm'n v. Nat'l Conservative PAC, 470 U.S. 480, 495 (1985) (stating that NRWC upheld "the prohibition of corporate campaign contributions to political candidates").

Although § 441b(a) was not directly at issue in NRWC, the Eleventh and Sixth Circuits have read NRWC to uphold the constitutionality of its ban on contributions from corporate treasuries. See Kentucky Right to Life, Inc. v. Terry, 108 F.3d 637, 645-46 (6th Cir. 1997); Athens Lumber Co., Inc. v. FEC, 718 F.2d 363, 363 (11th Cir. 1983) (en banc). There is some room for doubt as to whether the Court can be said to have held squarely that the ban in § 441b(a) is constitutional. NRWC stated that "We are also convinced that the statutory prohibitions and exceptions we have considered are sufficiently tailored to these purposes to avoid undue restriction on the associational interests asserted by respondent." Id. at 208 (emphasis added). Moreover, the first purpose identified by the Court — limiting the effect of the advantage flowing from the corporate form — could be met by a limit on contributions from corporate treasuries instead of a ban; and the second purpose could perhaps be addressed in corporate charters and state laws regulating corporations. Nevertheless, we feel constrained to read NRWC, and the Court's statements on NRWC in Nat'l Conservative PAC, as at least strong suggestions that § 441b(a) is constitutional.

Austin v. Michigan Chamber of Commerce, 494 U.S. 652 (1990), which upheld a Michigan statute that prohibited corporations from using corporate funds for independent expenditures in support of or in opposition to candidates for state office, also implies that the flat ban in § 441b(a) is constitutional. The analysis proceeds from Buckley, which distinguished independent expenditures from contributions: "[A]lthough the Act's contribution and expenditure limitations both implicate fundamental First Amendment interests, its expenditure ceilings impose significantly more severe restrictions on protected freedoms of political expression and association than do its limitations on financial contributions." Buckley, 424 U.S. at 23. Austin upheld a ban on independent expenditures from the corporate treasury because it found the ban sufficiently narrowly tailored to the purpose of limiting the influence of the unique state-conferred benefit of the corporate structure, which allows corporations to amass large treasuries. See Austin, 494 U.S. at 660-61. Because Buckley treats limits on independent expenditures as more severe than limits on contributions, Austin suggests that a ban on contributions from the corporate treasury also would be constitutional if sufficiently narrowly tailored to achieve the goal.

Austin also counsels that the ban on contributions from the corporate treasury here is sufficiently narrowly tailored to the interest of limiting the influence of corporate treasuries amassed under the state-conferred corporate structure. Austin reasoned that the Michigan statute prohibiting independent expenditures by corporations was sufficiently narrowly tailored to its purpose because, by permitting corporations to make independent political expenditures from separate segregated funds, it avoided an absolute ban on all forms of corporate political spending. See 494 U.S. at 660-61. The FECA also permits such indirect corporate political expenditures (via soft money), and under the teachings of Austin would thus seem to be sufficiently narrowly tailored to pass constitutional muster.

We are mindful that the flat ban on corporate contributions has never been directly addressed by a holding of the Supreme Court, and that this issue involves important First Amendment values. Because of the strong implication we draw from NRCW, Nat'l Conservative PAC, and Austin, however, we feel compelled to reject Mariani's facial challenge to § 441b(a). It will be for the Supreme Court itself to decide otherwise.

B. Section 441b(a) and Soft Money

The second challenge Mariani raises with respect to § 441b(a) is that the development of issue advocacy and the prevalence of soft money in campaigns for federal office has so eroded the theoretical distinction between hard and soft money that any justification for the ban on contributions from corporate treasuries has been vitiated. Mariani argues that under present conditions the ban cannot advance a compelling state interest and therefore must be invalidated. Significantly, Mariani does not complain that S 441b(a) itself fails to ban contributions from corporate treasuries. Rather, he argues that under the FECA — as interpreted by the Supreme Court and FEC regulations — it is possible for corporations to accomplish through other means that which they cannot accomplish through direct contributions from corporate treasuries. Mariani contends that, by funding soft money issue advocacy, contributors come so close to accomplishing what they would accomplish by hard money campaign contributions that the two are basically indistinguishable in terms of the danger they pose of corrupting the political process.

This contention amounts to an argument that § 441b(a) does too little by way of banning corporate political spending and is thereby fatally underinclusive. The Supreme Court has made clear, however, that Congress can act incrementally in this and other areas. See Buckley, 424 U.S. at 105 ("[A] statute is not invalid under the constitution because it might have gone farther than it did.") (citations omitted). As we have explained in a case regarding solicitation of campaign funds by a candidate for judicial office, the government may "take steps, albeit tiny ones, that only partially solve a problem without totally eradicating it." Stretton v. Disciplinary Bd. of the Supreme Court of Penn., 944 F.2d 137, 146 (3d Cir. 1991).

The underinclusiveness analysis employed for First Amendment questions does not change this principle. The First Amendment requires that the rule chosen must"fit" the asserted goals, City of Cincinnati v. Discovery Network, Inc., 507 U.S. 410, 428 (1993), and it must also strike an appropriate balance between achieving those goals and protecting constitutional rights. Underinclusiveness analysis serves to "ensure that the proffered state interest actually underlies the law," Austin, 494 U.S. at 677 (Brennan, J., concurring). But a rule fails the test only if it cannot "fairly be said to advance any genuinely substantial governmental interest," Federal Communication Comm'n v. League of Women Voters, 468 U.S. 364, 396 (1984), because it provides only "ineffective or remote" support for the asserted goals, id. (citing Central Hudson Gas Elec. Corp. v. Public Serv. Comm'n, 447 U.S. 557, 564 (1980)), or "the most limited incremental" support, Bolger v. Youngs Drug Prods. Corp., 463 U.S. 60, 73 (1983).

Thus, First Amendment underinclusiveness analysis requires neither a perfect nor even the best available fit between means and ends. See City of Renton v. Playtime Theaters, Inc., 475 U.S. 41, 52-53 (1986) (zoning ordinance regulating adult theaters was not constitutionally underinclusive "in that it fail[ed] to regulate other kinds of adult businesses . . . We simply have no basis on this record for assuming that Renton will not, in the future, amend its ordinance to include other kinds of adult businesses."). See also Blount v. SEC, 61 F.3d 938, 946 (D.C. Cir. 1995) ("[A] regulation is not fatally underinclusive simply because an alternative regulation, which would restrict more or the speech of more people could be more effective. The First Amendment does not require the government to curtail as much speech as may conceivably serve its goals.").

Applying this standard, section 441b(a) is not fatally underinclusive. The regulation in Fed. Communications Comm'n v. League of Women Voters, 468 U.S. at 397, which banned editorial speech by station management, but not editorial control over the content of programs and guests on news programs, was struck down because it did "virtually nothing" to prevent noncommercial stations from serving as outlets for expression of narrow partisan views. In contrast, § 441b(a) prevents corporations from donating hard money entirely. The important theoretical differences between hard and soft money, which include that a candidate cannot directly control how to spend soft money, are intended to avoid the corrupting influence of large contributors supporting a particular candidate. The practical distinctions between hard and soft money may have diminished in the past decade with the rise of issue advocacy, but not to such an extent that we can say that there is no benefit from distinguishing between the two. If hard and soft money were equivalent, it would be hard to imagine why Mariani would have gone to the lengths he allegedly went to in order to give hard money instead of soft.

Mariani attempts to counter this analysis by citing to United States v. Nat'l Treasury Employees Union, 513 U.S. 454 (1995):

[w]hen the Government defends a regulation on speech as a means to . . . prevent anticipated harms, it must do more than simply `posit the existence of the disease sought to be cured.' . . . It must demonstrate that the recited harms are real, not merely conjectural, and that the regulation will in fact alleviate these harms in a direct and material way.

Id. at 475 (quoting Turner Broadcasting System v. Fed. Communications Comm'n, 512 U.S. 622, 664 (1994)). The underinclusiveness analysis explicated above is not inconsistent with National Treasury Employees Union. Congress may regulate speech so long as it demonstrates that the recited harms are real, and it may, consistent with that principle, choose to regulate just some part of that speech. The requirement that the regulation alleviate the harm in a direct and material way is not a requirement that it redress the harm completely. And in light of the broad language in NRWC regarding the legitimacy of Congress's purpose in enacting § 441b(a), it is simply too late in the day to argue that Congress has failed to demonstrate that the recited harms are real.

Congress might well have concluded that direct contributions from corporate treasuries were more important to regulate than expenditures or contributions made through committees, because hard money can be used by a candidate in more and different ways than soft money. We note that no party to this case has argued that there is no compelling government interest in banning contributions from corporations. Indeed, Mariani's argument that the rise of soft money fatally undermines the purpose of § 441b(a) seems to depend on the assumption that limiting corporate contributions — if done effectively — would be constitutionally valid.

V. The Challenge to § 441f

Section 441f provides that "[n]o person shall make a contribution in the name of another person or knowingly permit his name to be used to effect such a contribution, and no person shall knowingly accept a contribution made by one person in the name of another person." 2 U.S.C. § 441f. Mariani argues that the prohibition in S 441f on contributions in the name of another to candidates for federal elective office violates the First Amendment because it fails to advance any compelling state interest and because it is underinclusive since it only applies to contributions of hard money (and can be circumvented by donating soft money).

The Buckley Court accorded broad acceptance to the FECA's reporting and disclosure requirements, explaining that they impose "only a marginal restriction upon the contributor's ability to engage in free communication." Buckley v. Valeo, 424 U.S. 1, 21-22 (1976). Although acknowledging the dangers of compelled disclosure of political activity, the Court found that the governmental interests in disclosure were of such magnitude that the requirements passed the strict test established by NAACP v. Alabama, 357 U.S. 449 (1958). The Court accepted as compelling three purposes behind the disclosure requirement: to provide the electorate with information as to where political campaign money comes from and how it is spent by the candidate in order to aid the voters in evaluating those who seek federal office; to deter actual or apparent corruption; and to gather the data necessary to detect violations of the contribution limits. Buckley, 424 U.S. at 66-68.

Buckley carefully considered the danger posed by compelled disclosure. It held that the state interests promoted by the FECA's reporting and disclosure requirements justified the indirect burden imposed on First Amendment interests, and that the compelled disclosure requirements were constitutional in the absence of a "reasonable probability" that disclosures would subject their contributors to "threats, harassment, or reprisals." Id. at 74. Proscription of conduit contributions (with the concomitant requirement that the true source of contributions be disclosed) would seem to be at the very core of the Court's analysis. In light of Buckley, we reject Mariani's argument that § 441f fails to advance a compelling state interest.

We also conclude that Congress's decision to limit the disclosure requirement to contributions of hard money does not make the requirement fatally underinclusive. Mariani's argument that the disclosure requirement is fatally underinclusive is similar to his argument that § 441b(a) has been undermined by the rise of soft money. As with that challenge, however, we conclude that Congress was free to determine that disclosure of hard money donations was the most important form of disclosure, and to limit the regulation to that area.

VI. Conclusion

For the foregoing reasons, we reject Mariani's challenges to §§ 441b(a) and 441f. Judgment will be entered in favor of the government.