Searching over 5,500,000 cases.

Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.

McConnell v. Federal Election Commission

May 1, 2003


The opinion of the court was delivered by: Richard J. Leon, District Judge



The passage of the Bipartisan Campaign Reform Act of 2002 ("BCRA"),*fn1 as the record amply indicates, was a considerable legislative achievement that many thought would never come to pass. Indeed, the law constitutes the most comprehensive reform of our national campaign finance system in the past twenty-eight years. As such, it is the latest chapter in the history of a longstanding and recurring problem that our government has been wrestling with since the administration of Theodore Roosevelt.

Titles I and II of BCRA focus principally on a major campaign finance development that has dramatically unfolded over the past decade: the use of corporate and union treasury funds, either directly or through soft money donations to political parties, to finance electioneering communications masquerading, predominantly, as"issue ads." In an attempt to prevent actual and apparent corruption arising from the funding of such sham issue advertisements, Congress enacted a sweeping set of reforms that effectively: alters the methodology of our national, state, and local parties and transforms their relationship with each other; limits the ability of corporations, unions, individuals, and interest groups to engage in communications on public policy; and diminishes the role of federal officeholders in fundraising for political parties and nonprofit interest groups.

For the reasons set forth in the following opinion, I find that the defendants have more than adequately demonstrated the constitutionally necessary basis for Congress: (1) to restrict the use of soft money donations by national, state, and local parties to fund certain types of campaign communications (particularly candidate-advocacy"issue" advertisements) which are designed to, and which do, directly affect federal elections; and (2) to restrict the airing of corporate and union electioneering communications which promote, oppose, attack, or support specific candidates for the office which they seek.

Notwithstanding this conclusion, however, I do not find that the defendants have demonstrated a sufficient constitutional basis to support Congress's decision: (1) to ban the solicitation, receipt, and use of soft money by national parties for purposes that do not directly affect federal elections; (2) to ban state and local parties from using soft money to fund a variety of election activities that do not directly affect federal elections; (3) to ban the use of corporate and union treasury monies to fund genuine issue advertisements that are aired during a particular time period preceding election even though they do not directly advocate the election or defeat of a federal candidate; (4) to ban national parties from donating to and soliciting soft money for certain Section 501(c) and Section 527 organizations under the Internal Revenue Code; (5) to prohibit federal officeholders from raising soft money for their national parties; and (6) to require broadcast licensees to collect and disclose certain records in connection with requests to purchase broadcast time. To the contrary, I find that in trying to do so Congress has unconstitutionally infringed upon the First Amendment rights of the various political actors and their supporters.

In short, the defendants, in my judgment, have been able to establish in some respects, but not in others, a sufficient basis for Congress's intervention in dealing with these problems. As to those where they succeeded, I believe it would make a mockery of existing Supreme Court precedent and the regulatory scheme that it has heretofore blessed, to hold otherwise. As to those where they have not, the protections accorded the plaintiffs under the First Amendment more than adequately warrant their undoing.

The following is a brief outline of my opinion, which has been organized on a title by title basis. With respect to the opinion itself, to the extent I have agreed with both the judgment and reasoning of either of my colleagues (or both), I have so noted and refrained from writing. As to those sections where I have agreed only in the judgment of one or more of my colleagues, I have limited the discussion of my reasoning to that necessary to explain how I reached my holding. To the greatest extent possible I have tried to acknowledge and address any disagreements we have had factually. However, in light of the importance and enormity of the record, I have included in my opinion a complete set of my Findings of Fact which I relied upon in reaching my judgments.

Table of Contents


I. Title I: Restrictions on Nonfederal Funds

A. New FECA Section 323(a): National Party Ban

1. Standard of Review for Restrictions on

Donations to Political Parties

2. Donations Used to Directly Affect Federal Elections are Regulable by Congress

3. New FECA Section 323(a) Unconstitutionally Regulates the Use of Nonfederal Funds for Nonfederal and Mixed Purposes

4. Severability of New FECA Section 323(a)

B. New FECA Sections 323(b) and 301(20)(A): Restrictions on"Federal Election Activities"

1. New FECA Sections 323(b) and 301(20)(A)(i), (ii), and (iv)

2. New FECA Section 301(20)(A)(iii)

C. New FECA Section 323(d): Restrictions on Tax-Exempt Organizations

D. New FECA Section 323(e): Restrictions on Federal Candidates

E. New FECA Section 323(f): Restrictions on State Candidates

II. Title II: Noncandidate Campaign Expenditures

A. Section 201: The Primary Definition

B. Section 201: The Backup Definition

C. Section 204: The Wellstone Amendment

D. Section 213: The Party Choice Provision

III. Section 318: Prohibition of Donations by Minors

IV. Section 504: Public Access to Broadcasting Records

V. Findings of Fact

VI. Conclusion

I. Title I: Restrictions on Nonfederal Funds New FECA Sections 323(a), 323(b), 301(20)(a), 323(d), 323(e), and 323(f)

A. New FECA Section 323(a): Nonfederal Fund Restrictions on National Parties

I agree with Judge Henderson's conclusion, although for different reasons, that Congress, in essence, is constitutionally prohibited from regulating a national party's ability to solicit, receive, or use nonfederal funds (i.e., soft money) for nonfederal and mixed purposes. To the extent that Section 323(a) seeks to regulate donations to national parties that are used for purposes that at the most indirectly affect federal elections (i.e., nonfederal or mixed purposes), the defendants have failed to demonstrate that Section 323(a) serves an important government interest, or even if they had, that it is sufficiently tailored to serve that interest.

However, I find that Congress can restrict a national party's use of nonfederal money to directly affect federal elections through communications that support or oppose specifically identified federal candidates. Therefore, like Judge Kollar-Kotelly, I find constitutional Congress's ban on the use of nonfederal funds by national parties for Section 301(20)(A)(iii) communications. As a result, I concur in part in, and dissent in part from, Judge Henderson's judgment and reasoning regarding Section 323(a).

1. Standard of Review for Restrictions on Donations to Political Parties

The right to freedom of political association under the First Amendment is a fundamental right of donors to political parties, see Buckley v. Valeo, 424 U.S. 1, 24-25 (1976),*fn2 and arguably of the political parties themselves, see FEC v. Colorado Republican Fed. Campaign Comm. ("Colorado II"), 533 U.S. 431, 448 n.10 (2001). Our national political structure is firmly anchored by our two major parties.*fn3 The role those national parties play in defining wide-ranging political agendas and bringing together individuals (and their financial resources) on behalf of those political agendas is critical to the stability our political system has enjoyed over the past 200 years.*fn4

Upon giving money to a political party, or to any political organization for that matter, a donor hopes that the organization will amplify his political perspective or candidate preference.*fn5 Any number of activities by a political party can amplify the donor's political voice. Some activities, like direct contributions to state candidates, are for a nonfederal purpose because they have no effect on a federal election. Others, like public communications that advocate the election or defeat of a particular federal candidate, are for a federal purpose because they directly affect federal elections. And still others, like generic voter registration and genuine issue advertisements,*fn6 are for"mixed purposes" because they indirectly affect both state and federal elections. Regardless of the purpose served, all of these party activities, paid for with aggregated donations, express loudly, and often effectively, the donor's political position.*fn7 For this reason, an individual's donation to a political party is an act of political association protected by the First Amendment. But as important as this right is, it is not absolute.

The Supreme Court, in Buckley v. Valeo and ensuing campaign finance cases, has recognized Congress's power through FECA*fn8 to regulate, in effect, the source and amount of contributions to political parties and candidates that donors could make for the purpose of influencing federal elections.*fn9 In enacting such contribution limitations Congress had to demonstrate that it was doing so in furtherance of the important government interest of preventing actual or apparent corruption of either the officeholder or the federal electoral system. Buckley, 424 U.S. at 25; see, e.g., Citizens Against Rent Control/Coalition for Fair Housing v. City of Berkeley, 454 U.S. 290, 298-99 (1981); Nixon v. Shrink Missouri Gov't PAC, 528 U.S. 377, 387-88 (2000).*fn10 Indeed, in addressing contribution limitations, preventing actual or apparent corruption is the only government interest the Supreme Court has found sufficient to interfere with associational rights. FEC v. National Conservative Political Action Comm. ("NCPAC"), 470 U.S. 480, 496-97 (1985). In Buckley, the Supreme Court explained:

To the extent that large contributions are given to secure a political quid pro quo from current and potential office holders, the integrity of our system of representative democracy is undermined.... Of almost equal concern as the danger of actual quid pro quo arrangements is the impact of the appearance of corruption stemming from public awareness of the opportunities for abuse inherent in a regime of large individual financial contributions. 424 U.S. at 26-27. In each case where the Supreme Court upheld contribution limitations, see supra note 9, the Court reviewed those limits under Buckley's"closely drawn" scrutiny, a standard of review somewhat less rigorous than strict scrutiny, by which"[e]ven a significant interference with protected rights of political association may be sustained if the State demonstrates a sufficiently important interest and employs means closely drawn to avoid unnecessary abridgement of associational freedoms." Buckley, 420 U.S. at 25 (internal quotations omitted); see also California Medical Ass'n v. FEC ("California Medical"), 453 U.S. 182, 196, 196 n.16 (1981).

Nevertheless, the plaintiffs argue, and Judge Henderson agrees, that any limitation on a national party's ability to raise and use nonfederal money must pass muster under the strict-scrutiny standard of review that the Supreme Court has traditionally applied to analyze expenditures. See, e.g., Republican National Committee ("RNC") Opening Br. at 51-53. I disagree.

While I agree that Section 323(a)'s prohibitions on soliciting, receiving, and using nonfederal funds restricts a party's ability to spend nonfederal money, their principal effect is to limit the ability of future donors through their contributions to use the national parties to amplify their voices.*fn11 Therefore, in determining to what extent Congress can limit donations to national and state parties,*fn12 this Court should review the limitations using the closely-drawn standard of review applied to contribution limitations in Buckley, 420 U.S. at 25, and ensuing campaign finance cases,*fn13 in which the Court specifically acknowledged"that restrictions on contributions require less compelling justification than restrictions on independent spending." FEC v. Massachusetts Citizens for Life, Inc. ("MCFL"), 479 U.S. 238, 259-60 (1986); see also Shrink Missouri, 528 U.S. at 387. The Supreme Court lowered the hurdle for contributions, see Shrink Missouri, 528 U.S. at 387-88, because restrictions on contributions, in its judgment, impact associational rights less by leaving"the contributor free to become a member of any political association and to assist personally in the association's efforts on behalf of candidates," id. at 387 (quoting Buckley, 424 U.S. at 22) (internal quotations omitted), and by not preventing"political committees from amassing the resources necessary for effective advocacy," Buckley, 424 U.S. at 21. Surely, these reasons for applying a lower standard of scrutiny for donations to candidates and their political committees are no less persuasive for analyzing contributions to political parties.

The plaintiffs, nevertheless, maintain that in Citizens Against Rent Control, which involved limitations on contributions to political committees with the purpose of supporting or opposing ballot measures, 454 U.S. at 291, the Supreme Court settled on strict-scrutiny review for contribution limitations to political organizations.*fn14 This conclusion is not based on a close enough reading of the case. While the plurality does suggest the undefined standard of"exacting judicial scrutiny," Citizens Against Rent Control, 454 U.S. at 294, 298, the three concurring justices either specifically applied closely-drawn scrutiny, id. at 301 (Marshall, J., concurring),*fn15 or equated the plurality's"exacting scrutiny" with Buckley's closely-drawn scrutiny, id. at 302 (Blackmun, J., & O'Connor, J., concurring).*fn16 Thus, if anything, Citizens Against Rent Control suggests that closely-drawn scrutiny should apply, even if the political organization is established exclusively for a purpose unrelated to federal campaigns. Restrictions on national political parties, which engage in both candidatespecific and issue-oriented activities, do not deserve to be treated with greater vigilance. Indeed, considering the standard of review adopted by the Supreme Court in Citizens Against Rent Control and the fact that contribution regulations are less jarring of associational rights than are expenditure restrictions, restrictions on donations to political parties should be similarly regulable if they are closely drawn to serve the compelling government interest of preventing corruption and its appearance.

2. Donations Used to Directly Affect Federal Elections are Regulable by Congress

Section 323(a) of BCRA seeks to expand Congress's authority to regulate donations that by definition did not appear to be regulable under FECA because, ostensibly, they were not given for the purpose of influencing federal elections.*fn17 It does so in sweeping fashion: national parties"may not solicit, receive,... direct... transfer, or spend any funds that are not subject to the limitations, prohibitions, and reporting requirements of this Act." BCRA § 101; FECA § 323(a); 2 U.S.C. § 441i(b)(1). Congress seeks this expansion, principally, because the national parties have increasingly, over the past decade, exploited a so-called"loophole"*fn18 in FECA that fails to regulate the use of these nonfederal funds for various types of electioneering communications that advocate the election or defeat of a specifically identified candidate.*fn19

The defendants contend that Congress, in its attempt to close this"loophole," can limit any donation to a national party, regardless of the purpose for which it is used thereafter. See Intervenors Opp'n Br. at 26. I disagree. The plaintiffs, on the other hand, contend that Congress can only limit donations that are funneled thereafter through the party as coordinated expenditures, direct contributions to candidates, or uncoordinated expenditures for express advocacy as defined by the"magic words." See, e.g., McConnell Opening Br. at 36; McConnell Opp'n Br. at 25-26; see also Buckley, 424 U.S. at 44 n.52 (providing examples of"express words of advocacy"). With that I disagree, as well. Both contentions are calculatingly indifferent to what reason, and precedent, have shown to be the only constitutionally viable antidote to corruption or its appearance: restrictions on donations to political parties based upon their use to directly affect federal elections.*fn20 In this sense, from my perspective, the issue before the Court is not whether Congress can limit donations to political parties,*fn21 but to what extent it can do so.

In Shrink Missouri, the Supreme Court made it clear that the amount of evidence needed to satisfy judicial scrutiny of restrictions on associational rights depends on the"novelty and plausibility of the justification raised." 528 U.S. at 391. Here, Congress relies upon the government interest of preventing actual and apparent corruption to justify the restrictions on nonfederal funds. When nonfederal funds are being used by national parties for nonfederal or mixed purposes, the government's interest in preventing corruption or its appearance to justify this restraint is so novel, and implausible, that it requires a substantial amount of evidence to withstand constitutional scrutiny. However, when nonfederal funds are being used by national parties for the federal purpose of directly benefitting the election of candidates*fn22 through either express advocacy or"issue" advocacy of the type defined in Section 301(20)(A)(iii),*fn23 the government's use of that interest to justify congressional intervention is neither novel, nor implausible, because the risk of corruption, see infra Part I.B.2, naturally flows from circumstances where a donor's contribution to a party is used thereafter to directly benefit a candidate's campaign. Indeed, as I discuss at length in the later in relation to Section 301(20)(A)(iii), id., the record overwhelmingly demonstrates that candidates are aware of who makes the large soft money donations, and in many instances, participate in raising money from them. Furthermore, the record clearly establishes that the public perceives that those large soft money donors receive special access to the legislators and have special influence on the legislative process. Id.

The notion that using donations for a federal purpose can implicate corruption is consistent with Congress's definition of"contribution" in FECA:"any gift, subscription, loan, advance, or deposit of money or anything of value made by any person for the purpose of influencing any election for Federal office." 2 U.S.C. § 431(8)(A)(i) (emphasis added). This definition of contribution was part of FECA when the Supreme Court upheld contribution limitations in Buckley, stating that a donation is regulable (that is, a"contribution") because"it is connected with a candidate or his campaign," thus having"a sufficiently close relationship" to the government interest in preventing actual or apparent corruption. 424 U.S. at 78. Additionally, the contention that a party's use of a donation to influence a federal election is conducive to corruption, or its appearance, is also supported by the widely accepted premise that Congress can restrict donations used for party express advocacy as defined by the so-called"magic words" requirement of Buckley. Indeed, the plaintiffs concede as much,*fn24 and this is perfectly consistent with the regulatory scheme that was propagated by the FEC.*fn25 Surely, if donations used for express advocacy can be limited to prevent corruption, then donations used for candidate advocacy that is tantamount to express advocacy -- assuming some minimal"quantum of empirical evidence" of corruption or its appearance, see Shrink Missouri, 528 U.S. at 391 -- should be regulable for the same reason.

The notion that Congress may limit donations based on their use for certain purposes is also consistent with Supreme Court precedent which intimates that donations closely connected to a candidate's campaign -- even if they are not direct contributions or coordinated expenditures -- raise, at a minimum, the specter of corruption. In First Bank of Boston v. Bellotti, the Court rejected a Massachusetts statute prohibiting corporations from making contributions or expenditures to influence the vote on referendum proposals. 435 U.S. 765, 787-95 (1978). In rejecting the statute, the Court explained that the interests in preventing corruption and thus preserving the integrity of the electoral process were not served by limiting contributions and expenditures that affected referendum discussion. Id. at 789-92. The Court stated:"Referenda are held on issues, not candidates for public office. The risk of corruption perceived in cases involving candidate elections simply is not present in a popular vote on a public issue." Id. at 790. Alternatively, one can infer that perceived corruption is likely to be present in cases involving candidate elections.

The extent to which certain uses of donations create the risk of corruption was also at issue in both California Medical, 453 U.S. at 193-201, and Citizens Against Rent Control, 454 U.S. at 292-300, where the Court considered donations to organizations, not candidates. It is difficult to reconcile these two cases without drawing the conclusion that the Court was primarily concerned with the purpose for which the organizations were using the donations.*fn26 In California Medical, the Court held that Congress can restrict the amount of donations to multicandidate political committees"which advocate[] the views and candidacies of a number of candidates." 453 U.S. at 197. Multicandidate political committees assuredly spend some of their funds on"independent expenditures," as the Court in California Medical concedes, id. at 195-96, but the Court seemingly concluded that those uncoordinated expenditures, by a"multicandidate" political committee, are made on behalf of candidates, whether direct contributions or uncoordinated expenditures. See 2 U.S.C. § 441a(a)(4) (defining multicandidate political committee as a political committee"which has received contributions from more than 50 persons, and... has made contributions to 5 or more candidates for Federal office."). The Court said as much when it dismissed the ACLU's concerns that donation restrictions would hinder the PAC's efforts to collectively express political views. California Medical, 455 U.S. at 197 n.17. Restricting contributions to committees like the one at issue in California Medical, the Court maintained, is different than efforts to regulate groups expressing common political views. Id. In this sense, the nature of the organization -- that it is established solely to benefit federal candidates -- was enough to conclude that most, if not all, of its contributions and expenditures were for the purpose, and had the effect, of benefitting a federal candidate. Conversely, in Citizens Against Rent Control, where the Court found that the City of Berkeley could not restrict donations to political committees that support or oppose ballot propositions, 454 U.S. at 295-300, the organization was established exclusively to advocate on behalf of a public issue, id. at 291 (explaining that the issue before the Court was whether donations to associations"formed to support or oppose ballot measures" could be regulated). That the association was formed only to oppose a public issue and that its speech was unrelated to candidates in any way, id. at 296-98, led the Court to find that the restriction"does not advance a legitimate governmental interest significant enough to justify its infringement of First Amendment rights," id. at 299.

Of course, political parties are unique; they are neither super multicandidate political committees formed entirely to support candidates for federal office nor political associations completely uninvolved in candidate advocacy. Justice Kennedy described political parties this way:

Political parties have a unique role in serving [the principle of open, robust debate on public issues]; they exist to advance their members' shared political beliefs.... A political party has its own traditions and principles that transcend the interests of individual candidates and campaigns; but in the context of particular elections, candidates are necessary to make the party's message known and effective, and vice versa. Colorado I, 518 U.S. at 628 (Kennedy, J., concurring in the judgment and dissenting in part) (citations omitted).*fn27

With such varying purposes, national political parties merit a hybrid treatment in regulating the funds donated to them.

Further, Colorado I and Colorado II do not preclude Congress from regulating donations that directly affect federal elections, even if the parties use those funds independently of the candidates*fn28 and even if there is no coordination between the donor and the candidate. In Colorado I, the Supreme Court merely found that Congress could not limit uncoordinated party expenditures. 518 U.S. 617. Indeed, the Court even reiterated that Congress has every power to limit the amount of donations to parties that are"used for independent party expenditures for the benefit of a particular candidate." 518 U.S. at 617. Thus, if anything, Colorado I serves to bolster the proposition that Congress can regulate donations used"for the benefit of a particular candidate" because that is where"the greatest danger of corruption" arises. Id. *fn29 Reading Colorado I together with Buckley, Bellotti, Citizens Against Rent Control, and California Medical leaves one with a clear impression: donations used directly for the purpose of uncoordinated federal activity, like express advocacy, can engender corruption, or the appearance thereof, and are therefore regulable.*fn30 Finally, Colorado II, in which the Supreme Court determined that Congress could limit the amount of coordinated party expenditures, 533 U.S. at 440-65, is relevant because it stands for the proposition that a contribution by the party to the candidate, even absent coordination between the donor and candidate, can be regulated. Thus, donations to political parties used thereafter for purposes that directly affect federal elections, such as candidate"issue ads," even if there is no coordination between the donor and the candidates in advance of the donations to the party, should likewise be regulable. Common sense and the evidence introduced by the defendants support that conclusion. See infra Part I.B.2. And, until such time as Buckley and its progeny are overruled, allowing such donations to occur without regulation is an affront to a regulatory system that has been blessed by the Supreme Court and in place since the adoption of the 1974 amendments to FECA.

3. New FECA Section 323(a) Unconstitutionally Regulates the Use of Nonfederal Funds for Nonfederal and Mixed Purposes

As sure as the evidence, legal precedent, and common sense support Congress's power to regulate the use of nonfederal funds for federal purposes, they do not support Congress's effort to regulate nonfederal funds used for nonfederal and mixed purposes.

National parties need to raise and use nonfederal funds for a variety of purposes. Sometimes they raise and use nonfederal funds for the nonfederal purpose of contributing to state and local candidates in"off-year" elections when there are no federal candidates on the ballot.*fn31 Other times they need to raise and use funds for mixed purposes that only indirectly affect the election of federal candidates, such as generic voter mobilization efforts and genuine issue advertisements.*fn32 The defendants do not deny that the national parties use nonfederal funds for both nonfederal and mixed purposes that at the most indirectly affect federal elections. They contend, nonetheless, that nonfederal donations to national parties -- regardless of their use -- create actual or apparent corruption. See Intervenors Opp'n Br. at 26. To support that expansion of Congress's power in contravention of the First Amendment rights of the donors and national parties, the defendants would have to demonstrate that using nonfederal funds for either nonfederal or mixed purposes gives rise to either corruption or an appearance of corruption, such that the blanket restriction on nonfederal funds is not overbroad. For the following reasons, they have not done so.

First, the suggestion that the appearance of corruption, let alone actual corruption, exists regardless of any perceived, or actual, benefit to a federal candidate does not comport with the conventional legal understanding of corruption and apparent corruption. The Supreme Court has defined corruption as something more than a quid pro quo arrangement in which a legislator sells his vote for one or more contributions to his campaign, see, e.g., Colorado II, 533 U.S. at 440-41, as well as"improper influence" or conduct by a donor that results in a legislator who is"too compliant" with the donor, Shrink Missouri, 528 U.S. at 389. Of course, the Supreme Court has also recognized that Congress has an equally compelling government interest in preventing the appearance of corruption in the public's mind. Id. at 390; Buckley, 424 U.S. at 27. The reason for this is simple: like corruption itself, the appearance of corruption undermines the public's confidence in our system of government and frustrates participation in the political process by causing the public to believe elected representatives are not acting independently of the individuals, corporations, and unions who contribute to representatives' campaigns and parties. Shrink Missouri, 528 U.S. at 390. Indeed, contribution limits to federal candidates and parties were enacted, and have been upheld by the Supreme Court, to prevent this very perception in the mind of the public. Id.; Buckley, 424 U.S. at 23-35. And it has been Congress's province to set the dollar limit above which this perception starts to ferment. See Buckley, 424 U.S. at 30; Shrink Missouri, 528 U.S. at 397. Thus, whether the corruption is actual or perceived, every traditional and accepted definition to date depends on the donor conferring, or being perceived as having conferred, a benefit on the candidate in return for something.*fn33 In NCPAC, for example, the Supreme Court defined corruption in the following way:"Corruption is a subversion of the political process. Elected officials are influenced to act contrary to their obligations of office by the prospect of financial gain to themselves or infusions of money into their campaigns." 470 U.S. at 497. Without financial gain to themselves or money into their campaigns, why would candidates elect to act contrary to their obligations? In short, since donations cannot logically foster corruption, or its appearance, unless the candidate benefits or appears to benefit in some way, donations to a party with no prospect that they will be used to directly affect the candidate's election, cannot, absent substantial evidence to the contrary, give rise to either actual or apparent corruption.

Second, the defendants' contention, in essence, that Congress can regulate the use of soft money donations by national parties for either nonfederal, or mixed, purposes is equally unsupportable by the record and common sense. If a national party uses nonfederal funds to support generic voter registration, or to conduct training seminars for state parties on getout-the vote activities, the benefit to the federal candidate, assuming his election is even being contested, is attenuated at best, Colorado I, 518 U.S. at 616, because it is generic in nature and diluted among a far greater number of state and local candidates. See infra Part I.B.1. No credible evidence has been submitted by the defendants that demonstrates that federal candidates either are, or are perceived to be, indebted to donors as a result of such mixed-purpose party activities.*fn34 Moreover, donations used for generic issue advertisements that may be helpful to both state and federal candidates, another example of a mixedpurpose activity by a party that indirectly affects federal elections in a way unlinked to any particular candidate's election or re-election, also do not foster actual or apparent corruption. Political parties, like many other political organizations, engage in noncandidate-related speech (i.e., genuine or pure issue ads) to influence public opinion on issues of the day.*fn35 Just recently, for example, the RNC funded a generic issue advertisement on the radio that touted the Republican's education proposal. It broadcasted the following:

Male: Every child can learn...

Female:... and deserves a quality education in a safe school.

Male: But some people say some children can't learn...

Female:... so just shuffle them through.

Male: That's not fair.

Female: That's not right.

Male: Things are changing. A new federal law says every child deserves to learn.

Female: It says test every child to make sure they're learning and give them extra help if they're not.

Male: Hold schools accountable. Because no child should be in a school that will not teach and will not change.

Female: The law says every child must be taught to read by the 3rd grade. Because reading is a new civil right.

Male: President Bush's No Child Left Behind Law.

Female: The biggest education reform and biggest increase in education funding in 25 years.

Male: Republicans are working for better, safer schools...

Female:... so no child is left behind.

Male: That's right... Republicans.

Anncr: Learn how Republican education reforms can help your children. Call.... Help President Bush and leave No Child Behind. Josefiak Decl. ¶ 91(e) & Exhibit X.

While pure issue advocacy, like the above advertisement, can indirectly affect a federal election,*fn36 it is unlikely, and there is no evidence to the contrary, that candidates will feel indebted to those who helped fund such advertisements. Moreover, the fact that state and local parties would still be able to use nonfederal money to engage in genuine issue advocacy*fn37 serves to undermine Section 323(a)'s complete ban on national parties being able to do the same.*fn38 Lastly, if the above analysis is true with regard to the inability to demonstrate even an appearance of corruption when nonfederal funds are being used for mixed purposes that indirectly affect a federal election, it is even more true when the purpose is nonfederal and has no effect on any federal candidate's election or re-election. In short, the defendants have provided no legal basis to restrict a national party's use of nonfederal funds for nonfederal purposes.

Third and finally, the defendants' contention that candidates, who raise soft money donations for their national parties, regardless of their subsequent use, are indebted to the donors due to"internal party benefits" they subsequently receive for raising the nonfederal donations,*fn39 is equally tenuous from both a theoretical and an evidentiary standpoint, and, in any event, Section 323(a) remains insufficiently tailored based on that justification to pass constitutional muster. The defendants' contention is theoretically flawed because it proceeds from the premise that the corruption, or appearance of corruption, necessary to warrant congressional intervention can be satisfied by a federal candidate receiving a benefit other than personal financial gain or direct assistance, monetary or otherwise, to his election effort. The Supreme Court has never defined corruption, or its appearance, in those terms. As stated previously, the only benefit the Supreme Court has based a finding on is"the prospect of financial gain to themselves or infusions of money into their campaigns." NCPAC, 470 U.S. at 497.*fn40 Even if the Court did, however, bless the notion that a non-monetary benefit from someone other than the donor could, under the right circumstances, give rise to an indebtedness that the public could perceive as"corrupting" its legislator's independence, there is no evidence on this record that such corruption has either occurred or is perceived by the public to exist. The evidence relating to the appearance of corruption, such as it is, only establishes that the public believes there is a connection between large donations to national parties and the influence and access the pubic believes the donors receive. See infra Part I.B. It does not, however, establish that this connection exists independent of how the parties use the funds. If anything, the public's regular exposure to so-called"issue ads" sponsored by the parties and crafted to help their candidates,*fn41 combined with its lack of knowledge of the difference between soft and hard money and its lack of knowledge of the campaign finance regulations (both of which have been demonstrated),*fn42 should lead this Court to reasonably infer that the public believes that these donations are used in whole, on in part, by the parties to directly help their candidates. It is that perceived benefit, in my judgment, that gives rise to the public's view that officeholders, either out of gratitude, or in hope of similar future contributions, provide increased access and influence to those donors. And in light of the utter absence of evidence on the record establishing any other reason that the public believes accounts for federal officeholders granting increased influence and access to those who give large donations to their party, this Court should infer the same. In any event, the defendants' argument is flawed because their supposed internal-party-benefit rationale was not even relied upon by Congress to limit the national parties use of nonfederal funds. If it had been, Congress would have only restricted nonfederal funds that federal candidates themselves solicited. Thus, Section 323(a)'s sweeping restriction, even if acceptable theoretically and factually based on the internal-party-benefit rationale, is not sufficiently tailored to"alleviate [the] harm in a direct and material way." Turner Broadcasting Sys., Inc. v. FCC ("Turner I"), 512 U.S. 622, 664 (1994).*fn43

In sum, conduct which only indirectly affects a federal election requires a greater degree of evidence of corruption, or appearance thereof, to warrant congressional regulation. Thus, in the absence of sufficient proof to warrant expanding FECA in this direction, Congress may only prohibit the national parties from using nonfederal money for federal purposes such as those defined in Section 301(20)(A)(iii), which are clearly designed to directly affect federal elections. The use of nonfederal funds for nonfederal or mixed purposes, which at the most indirectly affect federal elections, is simply not regulable by Congress because it does not give rise to corruption or the appearance of corruption. Thus Section 323(a)'s complete ban on the use of nonfederal funds is not closely drawn to serve the designated government interest.

4. Severability of New FECA Section 323(a)

Because Section 323(a) prohibits all uses of nonfederal funds by national parties, and because I only uphold Congress's power to prohibit the use of nonfederal funds for federal purposes (as defined in Section 301(20)(A)(iii)), see infra Part I.B.2, a considerable issue is presented as to whether we can isolate and uphold that prohibition from the remaining undefined, unconstitutional prohibitions in Section 323(a) in a manner consistent with both the severance clause and Supreme Court precedent. For the following reasons, I believe we can and should.

It is a"cardinal principle" of statutory construction to save as much of a statute as possible. See NLRB v. Jones & Laughlin Steel Corp., 301 U.S. 1, 30 (1937) (Hughes, C.J.).*fn44 Indeed, Congress itself in Section 401 of BCRA provided us with a severability clause which directed us to do as much.*fn45 In attempting to save a statute, however, the Supreme Court has made it clear that a court must take great pains to avoid"rewriting" the statute.*fn46 Thus, a severability clause, due in part to separation of powers concerns, is merely an"aid," not a"command," to the judiciary. Nonetheless, the Supreme Court has frequently found statutory provisions unconstitutional (or constitutional) as to particular applications without invalidating (or validating) the entire provision.*fn47 In the campaign finance arena, the Supreme Court in Colorado I and Colorado II did not object to severing applications of Section 441a(d), a FECA provision that caps how much political parties can spend.*fn48 In Colorado I, the Supreme Court invalidated Section 441a(d) as applied to uncoordinated, or independent, expenditures. 518 U.S. at 613-20. It then remanded the question of whether Section 441a(d)'s application to coordinated expenditures was constitutional, directing the lower courts to determine"whether or not Congress would have wanted [Section 441a(d)'s] limitations to stand were they to apply only to coordinated, and not to independent, expenditures." Id. at 625-26. On remand the district court found that Section 441a(d)'s application to coordinated expenditures was severable from its application to uncoordinated expenditures. 41 F. Supp. 2d 1197, 1206-07 (D. Co. 1999). The district court found the two applications severable because of the"strong" severability provision, which is almost identical to the provision at issue here, and because there was"no evidence" that Congress would not have rejected Section 441a(d) as applied to coordinated party expenditures. Id. at 1207. Both the Tenth Circuit and the Supreme Court, noting that the severability argument was not renewed upon appeal, let the district court's decision stand. Colorado II, 533 U.S. at 440 n.5; FEC v. Colorado Republican Fed. Campaign Comm., 213 F.3d 1221, 1225 n.3 (10th Cir. 2000). In the end, though the language in Section 441a(d) makes no reference to coordinated or uncoordinated expenditures, the Supreme Court upheld certain applications of the provision (coordinated party expenditures) but invalidated other applications (uncoordinated party expenditures).

Typically the Supreme Court invalidates specific applications, while letting others stand, when it can confidently discern congressional intent. In one recent case where the Supreme Court refused to limit the application of a statute which banned all honoraria to government employees, United States v. National Treasury Employees Union ("NTEU"), 513 U.S. 454, 479 (1995), it did so because it was faced with considerable uncertainty as to how Congress would have defined the honoraria restriction if it had known the complete ban on honoraria would have been rejected as unconstitutional. Id. *fn49 For a number of reasons, we do not have that problem here.

First, with regard to Title I, Congress's intent, based on both the severability clause and the text of the statute, is unambiguous. The clause itself, of course, directs that the Court save"the application of the provisions... to any person or circumstance." BCRA § 401. However, as stated previously, such a clause only creates a presumption of severability.*fn50 It does not relieve this Court of its obligation to determine if the limiting construction of Section 323(a) can stand alone, and if Congress would have enacted such a construction knowing that its broader position would be held unconstitutional.*fn51 With regard to both the former and the latter, Congress, by defining"federal election activity" in Section 301(20)(A)(iii) to include certain communications which directly affect federal elections and by constitutionally prohibiting state parties from engaging in such activity in Section 323(b), has unequivocally indicated its intent that such activity -- as defined -- would be among the undefined uses of nonfederal funds that national parties were similarly being prohibited from engaging in under Section 323(a). To conclude otherwise would be to turn a blind eye to an obvious reason why Sections 323(b) and 301(20)(A) were written in the first place: to prohibit donors (especially corporations and unions) and the national parties from circumventing Section 323(a) by funneling soft money through state and local parties for Section 301(20)(A) purposes.*fn52 By limiting the prohibited uses of nonfederal funds by national parties in Section 323(a) to communications of the kind defined by Congress in its own words in 301(20)(A)(iii), I am neither employing a saving construction that"rewrites" Section 323(a), nor ignoring Congress's clear intention to save an implicit feature of that section consistent with BCRA's severance clause admonition.

Finally, by applying Section 301(20)(A)(iii), which I hold to be constitutionally acceptable for Section 323(b), see infra Part I.B.2, to define the prohibited conduct in Section 323(a), I avoid the Supreme Court's additional concern of creating a definition the constitutionality of which has not been decided.*fn53 Accordingly, for all of the above reasons, I find that Section 323(a)'s implicit prohibition on national parties to use nonfederal money to fund communications of the kind defined in Section 301(20)(A)(iii) is constitutionally severable from its remaining unconstitutional applications.

B. New FECA Sections 323(b) and 301(20)(A): Restrictions on Nonfederal Funds for"Federal Election Activities"

Unlike Section 323(a)'s total ban on the use of nonfederal funds by national parties, Section 323(b) only prohibits state parties from using nonfederal funds for certain"federal election activities," BCRA §101; FECA § 323(b); 2 U.S.C. § 441i(b)(1), which it defines in Section 301(20)(A). BCRA § 101; FECA § 301(20)(A); 2 U.S.C. § 431(20)(A). Thus, in order to assess the constitutionality of the restraint on the state parties in Section 323(b), we have to simultaneously assess the constitutionality of Section 301(20)(A)'s definition of federal election activity.

Section 301(20)(A) defines federal election activity to include: (1) voter registration activity during the period 120 days before a regularly scheduled federal election; (2)"voter identification, get-out-the-vote activity, or generic campaign activity conducted in connection with an election in which a candidate for Federal office appears on the ballot"; (3)"a public communication that refers to a clearly identified candidate for Federal office... and that promotes or supports a candidate for that office, or attacks or opposes a candidate for that office (regardless of whether the communication expressly advocates a vote for or against a candidate)"; and (4) services provided by a state or local party committee employee who spends more than twenty-five percent of that individual's compensated time"on activities in connection with a Federal election." BCRA § 101; FECA § 301(20)(A); 2 U.S.C. § 431(20)(A).

Only Section 301(20)(A)(iii), however, describes conduct which is targeted exclusively at federal elections and which directly affects federal elections. Accordingly, for the reasons set forth below, I find that Section 323(b) and Sections 301(20)(A)(i), (ii), and (iv) are substantially overbroad in that they seek to restrain state parties from using nonfederal funds for election activities, which only indirectly affect federal elections, and thus do not give rise to the appearance of corruption necessary to warrant congressional intervention. As to Section 301(20)(A)(iii), however, I find that it is constitutionally permissible because, by contrast, it focuses on election activities that directly affect federal elections and as such, give rise to the appearance of corruption necessary to warrant Congress's restraint on the First Amendment rights of the donors.

1. New FECA Sections 323(b) and 301(20)(A)(i), (ii), and (iv)

Section 323(b) is premised, in part, on the congressional belief that certain mixedpurpose activities by state parties, when funded with nonfederal funds, sufficiently affect federal elections that they give rise to an appearance of corruption between the donors and the candidates whose campaign receives the benefit of these activities. I disagree. Setting aside the considerable issue of whether the Elections Clause, U.S. Const. art. I, § 4; Buckley, 424 U.S. at 13 & n.16,*fn54 can be fairly read to allow Congress to regulate state party activities such as those defined in Sections 301(20)(A)(i), (ii), and (iv), see CDP/CRP Opening Br. at 20-27, the justification and evidence submitted here fail to establish that those provisions serve a sufficient government interest to justify an infringement on First Amendment rights.

The Supreme Court pointed out in Colorado I that"the opportunity for corruption posed by" nonfederal funds for mixed-purpose activities like voter registration and get-outthe-vote"is, at best, attenuated." 518 U.S. at 616. Though dictum it may be, it is particularly telling. See Central Green Co. v. United States, 531 U.S. 425, 431 (2001) ("dicta'may be followed if sufficiently persuasive' but are not binding" (quoting Humphrey's Executor v. United States, 295 U.S. 602, 627 (1935))). One can infer from the Supreme Court's plain statement that the opportunity for corruption is less because there is no clear link between donations used for these predominantly generic activities and whatever benefit accrues to the candidate.

When nonfederal funds go to political parties, not candidates, and are spent for purposes that do not directly affect federal elections, there is less concern about donors having quid pro quo arrangements with candidates. Indeed, given that Section 301(20)(A)(iii) prevents communications that promote federal candidates, the activities defined in Sections 301(20)(A)(i) and (ii) are necessarily"generic" or specific only to a nonfederal candidate: that is, these activities are not directed at a specific federal candidate.*fn55 Conversely, such activities potentially assist a considerably larger number of state and local candidates that greatly outnumber the one, or possibly two, federal candidates on the ballot,*fn56 many of whom invariably run unopposed or in clearly noncompetitive races. There is also evidence that state and local parties undertake voter mobilization efforts principally for state and local candidates,*fn57 mostly from nonfederal money they raised on their own,*fn58 and in elections where the federal candidates are practically uncontested.*fn59

In light of this focus on state and local candidates, there is every reason for this Court to doubt whether the federal candidates themselves would view such generic activities by state and local parties as sufficiently helpful to their campaigns as to warrant even a token sense of indebtedness to the soft money donors to the political parties. The evidence, such as it is, regarding Sections 301(20)(A)(i) and (ii) activities fails to demonstrate either the degree of effect such activities have on the federal candidate's re-election, or the existence of a public perception that donations used to fund such efforts create a sense of indebtedness between the federal candidate and those who make large donations to the party. That there is no evidence that the public perceives corruption in these circumstances is not surprising. Since the public would expect state parties to engage in such voter mobilization efforts for the benefit of all of its candidates, we can, and should, reasonably infer that the public would correspondingly view a federal candidate's sense of indebtedness, if any, to be diluted among the numerous state and local candidates who equally benefit from these activities. See Feingold Dep. at 126-27 (acknowledging that soft money being used for generic campaign activity is less likely to create an appearance of corruption). Such uncertainties are hardly a sufficient basis from which to allege that precluding state and local parties from using nonfederal funds for mixed-purpose activities serves the government interest in preventing corruption, or its appearance. Thus, in the absence of a substantial evidentiary showing to the contrary, it is"mere conjecture," Shrink Missouri, 528 U.S. at 392, by the defendants that an appearance of corruption arises from donations to state parties, or transfers from national parties, that are used for these generic or noncandidate-specific activities set forth in Sections 301(20)(A)(i) and (ii).

Finally, even if the defendants had demonstrated that some mixed-purpose activities do somehow create an appearance of corruption, Sections 301(20)(A)(i), (ii), and (iv) are not sufficiently tailored to be constitutionally acceptable. By limiting donations for so many unmistakably noncorrupting activities, like donations for voter registration on behalf of state candidates, these sections extend too far. It is simply not enough to claim that just because the use of a donation has some effect on a federal election, it must be completely funded with federal funds.*fn60 To reach such a conclusion would inevitably sweep in too many activities that deserve First Amendment protection. Section 301(20)(A)(iv)'s percentage based definition of the amount of time a party official must spend in connection with a federal election, BCRA § 101; FECA § 301(20)(A)(iv); 2 U.S.C. § 431(20)(A)(iv), provides the starkest example of this insufficient tailoring. If a state party employee spends 26 percent of his compensated time"in connection with a Federal election," then his entire salary must be paid using federal funds. Clearly, 74 percent of the employee's compensated time, which has no relation whatsoever to a federal election, is being regulated. There is no adequate basis in the record before us to determine whether the nature of his conduct sufficiently impacts the election to give rise to an appearance of corruption between the donors who may be funding his efforts through the state party and the candidate receiving the benefit of his services. Thus, even if Sections 301(20)(A)(i), (ii), and (iv) served to prevent some appearance of corruption, they are not drawn closely enough to survive scrutiny.

2. New FECA Section 301(20)(A)(iii)

As I indicated previously, Congress has the power to require both national and state parties to use only federal money for election activities that directly affect federal elections.*fn61 Section 301(20)(A)(iii) focuses on one such type of activity:"a public communication that refers to a clearly identified candidate for Federal office... and that promotes or supports a candidate for that office, or attacks or opposes a candidate for that office." BCRA § 101; FECA § 301(20)(A)(iii); 2 U.S.C. § 431(20)(A)(iii) (emphasis added). Judge Henderson concludes in her opinion that the Supreme Court in Buckley required certain"magic words" that specifically advocate the election or defeat of a candidate in order for a communication to be the type of advocacy regulable by Congress.

All other communications, even if they directly affect a federal election, from her perspective, are not regulable.*fn62 I disagree. For the same reasons, in part, set forth by Judge Kollar-Kotelly in her opinion on Title II, I believe that Buckley did not set forth a"brightline test." See J. Kollar-Kotelly Op. at Part III.I.C.1. But, even if the Supreme Court did set forth a bright-line test for Congress's regulation of expenditures by nonparties in Buckley, 424 U.S. at 39-51, that rationale is not necessarily applicable to, and binding on, Congress's power to regulate donations to political parties under BCRA.

Therefore, the question before this Court is not so much whether Congress can regulate the use of nonfederal funds for uncoordinated, nonexpress advocacy that directly affects federal elections, but whether Congress has defined such nonexpress advocacy in Section 301(20)(A)(iii) in a way that will withstand constitutional scrutiny. For the following reasons, I have concluded that Congress not only can regulate uncoordinated, nonexpress advocacy which directly affects federal elections, but has defined that nonexpress advocacy in Section 301(20)(A)(iii) in a way that is sufficiently tailored to serve the government interest of preventing actual or apparent corruption, and in a way that is not unconstitutionally vague.*fn63

In defining"candidate advocacy" as it did, Congress chose to couple two concepts together in order for communications to qualify as regulable: (1) the identification of a candidate for a federal office; and (2) words that promote, oppose, attack, or support that candidate for that office. BCRA § 101; FECA § 301(20)(A)(iii); 2 U.S.C. § 431(20)(A)(iii). Therefore, unlike genuine issue advocacy, which both national and state parties have every right to participate in with nonfederal funds, and which help all party candidates in a generic sense, Congress in this definition was seeking to focus on parties using nonfederal funds for communications intended to directly help a specific federal candidate. Because such assistance is focused on a specific candidate,*fn64 it is natural for that candidate to feel indebted towards those whose donations funded the communication, even if he does not know exactly which soft money donors' funds actually made it possible. Concomitantly, it is natural for the public to perceive that those whose large soft money donations funded the national and state parties' communications are not only known by the parties' staffs, but by the federal candidates who directly benefitted from the donations.

The evidence submitted by the defendants overwhelmingly corroborates these common sense conclusions. The record is clear that many, if not most, of the party so-called"issue ads" refer to a specific federal candidate.*fn65 And the evidence also demonstrates that the advertisements are designed to, and do, support or oppose those candidates for that office.*fn66 To illustrate one extreme of this genre: in 1996, the Republican National Committee ("RNC") ran a supposed"issue ad" called"The Story" which requires only a quick reading to discern its true purpose and effect.

Audio of Bob Dole: We have a moral obligation to give our children an America with the opportunity and values of the nation we grew up in.

Voice Over: Bob Dole grew up in Russell, Kansas. From his parents he learned the value of hard work, honesty and responsibility. So when his country called... he answered. He was seriously wounded in combat. Paralyzed, he underwent nine operations.

Audio of Bob Dole: I went around looking for a miracle that would make me whole again.

Voice Over: The doctors said he'd never walk again. But after 39 months, he proved them wrong.

Audio of Elizabeth Dole: He persevered, he never gave up. He fought his way back from total paralysis.

Voice Over: Like many Americans, his life experience and values serve as a strong moral compass. The principle of work to replace welfare. The principle of accountability to strengthen our criminal justice system. The principle of discipline to end wasteful Washington spending.

Voice of Bob Dole: It all comes down to values. What you believe in. What you sacrifice for. And what you stand for. Fabrizio Dep. Exhibit 2; McCain Decl. ¶ 15; Huyck Decl. ¶ 3; Fabrizio Dep. at 49-55.

Though this advertisement transparently was intended to assist Senator Dole's campaign, it was funded in part with nonfederal funds. It is hard to disagree with Senator Levin, who described the advertisement this way:"It's not an ad about welfare or wasteful spending; it is an ad about why should we elect that particular nominee."*fn67

Another and more typical form of sham issue advertisement run by parties is a candidate-centered ad that focuses on the positions, past actions, or general character traits of a given federal candidate; contrasts them to the party's view of the proper outlook on those issues; and encourages the viewers to contact the candidate to ostensibly inquire why he/she is taking those positions or actions. See Findings 45 & 46. In the 2000 election, for example, the National Republican Congressional Committee and the Florida Republican Party ran television advertisements criticizing Linda Chapin, the Democratic candidate for Congress:

Announcer: Linda Chapin. Hard on taxpayers. Soft on convicts. Chapin raised taxes on your utilities, pushed to raise the county sales tax and even tried raising your property tax. Meanwhile, hard time in the county jail turned into"Chapin time." Where convicts received cable tv and lounged on padded furniture in carpeted cells. Chapin's County Commission ran this soft jail... a jail she called a"national model." Ask Chapin why she's hard on taxpayers and soft on convicts. Chapin Decl. Exhibit 2; Chapin Decl. ¶ 10; Beckett Decl. ¶ 10; Pennington Decl. ¶ 14.*fn68

This type of electioneering advertisement, and many others like it, were run by the state party, which used mostly nonfederal funds to pay for it. It takes little convincing to find that these advertisements can, and do, directly influence the outcome of a federal election*fn69 and that parties engaging in them are"electioneering in the guise of issue advocacy." Mann Expert Report at 26. Moreover, because the amount of money used for candidate advocacy is substantial,*fn70 the candidates are likely to feel even more indebted to the donors whose contributions to the political parties made possible this form of campaign assistance.*fn71 Not surprisingly, the use of nonfederal funds for this type of candidate advocacy was, to all appearances, Congress's primary concern in deciding to enact BCRA Sections 323(a) and 323(b).*fn72

The record further demonstrates that congressional candidates know who the major soft money donors are.*fn73 In some cases, they actively assist in helping the party to raise such contributions.*fn74 And as to those candidates who do not, the parties and the donors keep many of them apprised of who made the large donations. For example, Robert Hickmott, lobbyist and former DNC official, advised donors in the following manner:

[W]hen one of my clients is going to make a donation to a federal candidate or party, hard or soft money, I advise them on the manner in which they should do that. I tell them not to just send the check to the party committee, for example, to the young staff member who is collecting the checks. Instead I tell my clients that they should personally give the money to a Member of Congress who then can give the money to the Chair of the party committee, who will in turn make sure that the check reaches the young staff member. That way the donor, with one check, gets"chits" with multiple Members of Congress. Hickmott Decl. ¶ 9.*fn75

Finally, while there is no evidence in the record of actual quid pro quo corruption,*fn76 the record does establish that the public not only appreciates that there are many donors giving large sums of money (mostly corporations and unions) to the political parties,*fn77 but believes and expects that the donors -- in return -- receive privileged access to the legislators and special influence in the legislative process.*fn78 It is of little surprise that Congress was particularly concerned with the consequences of the public's perception of a correlation between large donations to parties and the special access and influence that the public believes are accorded to these donors.*fn79 There is ample evidence, including polls*fn80 and press reports,*fn81 to support Congress's judgment that the special access and perceived special influence accorded to those large donors have undermined the public's confidence in the independence of its elected representatives from those donors, and thereby giving rise to an appearance of corruption. The record clearly establishes that those large donations are used in large part by the parties to bombard the public with candidate advocacy of the type defined by Section 301(20)(A)(iii). See Findings 36-52, 138-47. Because the advertisements are recognizably sponsored by the parties, see supra note 41, it takes little for the public to conclude that candidates are directly benefitting from large donations to the parties. Moreover, the fact that the public neither distinguishes between"soft money" and"hard money," nor is knowledgeable of the restrictions on contributions,*fn82 does nothing to deter the inference that the special access and influence it perceives are accorded to the donors is in return for the plainly visible assistance the parties provide to candidates' campaigns. Given the high plausibility that federal candidates feel indebted to donors who fund, through the parties, Section 301(20)(A)(iii) communications of which they are the beneficiary, and the equally high plausibility that the public perceives that the special access and influence it believes are accorded to these donors is in gratitude for that assistance, the evidentiary documentation more than adequately convinces me that an appearance of corruption has arisen from such arrangements. Simply stated, Congress has drawn a reasonable inference of the same based on substantial evidence.*fn83

Moreover, any fear that Section 301(20)(A)(iii) is overbroad is cured by the accumulated effects of three circumstances. First, Section 301(20)(A)(iii) only regulates communications that are candidate specific, not issue specific, such that most genuine issue communications should not be subject to regulation. See Hearing Tr. (Dec. 4, 2002) at 17. Second, though there are some genuine issue advertisements that do specifically mention candidates,*fn84 the fact that the Section 301(20)(A)(iii) definition requires the communication to have language that"promotes or supports a candidate for that office, or attacks or opposes a candidate for that office" eviscerates any remaining overbreadth by focusing on advertisements that directly influence"a candidate for that office." Finally, the fact that the"speaker" running this type of advertisement is a national or state party also supports the defendants' contention that most of the advertisements, referring to a candidate in this manner, will be intended to promote or attack that particular candidate.*fn85 A political party, as opposed to corporations and other nonparties, exists to a large extent for the purpose of electing candidates of its party to office.*fn86 In Buckley, the Supreme Court observed that expenditures by political parties, and other political committees with"the major purpose of which is the nomination and election of a candidate... can be assumed to fall within the core area sought to be addressed by Congress. They are, by definition, campaign related." 424 U.S. at 79. In interpreting this language by the Supreme Court, the D.C. Circuit observed that"when an organization controlled by a candidate or the major purpose of which is election-related makes disbursements, those disbursements will presumptively be expenditures within the statutory definition." Akins v. FEC, 101 F.3d 731, 742 (en banc) (D.C. Cir. 1996).*fn87 With this understanding of parties and other political committees, it is not surprising that the Supreme Court limited disclosure of expenditures by all other groups to express advocacy, but sanctioned disclosure of all expenditures by"political committees." Buckley, 424 U.S. at 49. While not all party advertisements are intended to directly influence a federal election (i.e., genuine issue advertisements), the presumptively electoral-focus of parties suggests that party communications that do mention candidates are, more likely than not, designed to have some impact on a federal election.*fn88 For these reasons, Section 301(20)(A)(iii) is closely drawn to serve the government's interest.

Even so, plaintiffs contend that Section 301(20)(A)(iii) is still defective because its definition of communications that must be funded with federal money is too vague to withstand constitutional scrutiny. They claim that words such as"promote,""oppose,""attack," and"support" are too vague to enable party officials to determine where the line between noncandidate advocacy (i.e., pure issue advertisements) and candidate advocacy lies. Like Section 201's fallback definition, see BCRA § 201(a); FECA § 304(f)(3)(A)(i); 2 U.S.C. § 434(f)(3)(A)(i), however, the definition of candidate advocacy in Section 301(20)(A)(iii) is not unconstitutionally vague because: (1) potential party speakers can simply avoid regulation by not identifying a candidate; (2) the formulation Congress chose to use -- even for those not expert in the subtleties of campaign advocacy -- is anchored in every day words that have to be linked to a specific candidate's election, or re-election, to a particular office; (3) to paraphrase defendant's counsel during oral argument, as long as the communication is not neutral -- as to the candidate's election or defeat -- it is covered by the definition, see Intervenors Opp'n Br. at 66; and (4) the opportunity to seek advisory opinions to clarify any ambiguity"mitigates whatever chill may be induced by the statute and argues against constitutional adjudication on a barren record," see Martin Tractor Co. v. FEC, 627 F.2d 375, 384-85 (D.C. Cir. 1980); see also United States Civil Serv. Comm'n v. Nat'l Ass'n of Letter Carriers, 413 U.S. 548, 580 (1973).*fn89 Indeed, Section 301(20)(A)(iii), as crafted,"give[s] the person of ordinary intelligence a reasonable opportunity to know what is prohibited,""provide[s] explicit standards for those who apply them," and where First Amendment rights are implicated, does not induce"citizens to'steer far wider of the unlawful zone'... than if the boundaries of the forbidden areas were clearly marked." Grayned v. City of Rockford, 408 U.S. 104, 108-09 (1972) (quoting Baggett v. Bullitt, 377 U.S. 360, 372 (1964)).

Even if the above factors were not enough to save Section 201 from a vagueness attack, those factors, combined with the identity of the speaker being restricted and the fact that it is a restriction on a contribution (as opposed to an expenditure), would still rescue Section 301(20)(A)(iii). Because political parties are sophisticated participants in the election arena, they are much more likely to recognize the line between candidate advocacy and genuine issue advocacy. See Colorado II, 533 U.S. at 453 ("[T]he party marshals [the power to speak] with greater sophistication than individuals generally could.").*fn90 Political parties, more than others, are also more likely to assume the risk that they have crossed the line.*fn91 In short, the argument that Section 301(20)(A)(iii) is unconstitutionally vague rings hollow as it applies to parties who are experts in the business of discerning what combination of words and images help, or harm, candidates.*fn92

Vagueness concerns are also less pronounced with regard to contributions than expenditures. When restrictions on expenditures are at issue, one must be concerned that if the potential speaker cannot discriminate between issue advocacy and candidate advocacy, then the speaker will shy away from speaking at all. The restrictions at issue here, however, only require that Section 301(20)(A)(iii) communications be funded with federal money, so if the political parties are afraid that a communication may promote or attack a candidate, they can simply resort to federal funds. In any event, if despite all of the mitigations, some slight potential for vagueness remains,"uncertainty at the periphery" does not render a provision unconstitutionally vague. See FEC v. National Right to Work Comm., 459 U.S. 197, 211 (1982).

In sum, Section 301(20)(A)(iii), in its application to Sections 323(a) and 323(b), is neither unconstitutionally vague, nor insufficiently tailored to serve the compelling government interest of combating the appearance of corruption. By seeking only to restrict donations that can give rise to actual or apparent corruption, it sweeps neither too far nor too near. To the extent that"issue ads" are about issues in fact, as well as in name, political parties are free to raise as much soft money as they like for such advertisements. In contrast, soft money donations to political parties used for communications that support or oppose a clearly identified candidate for federal office -- that is, candidate advocacy -- directly affect a federal election and give rise to an appearance of corruption that Congress has a substantial interest in combating. Such candidate advocacy must be funded with federal money.

A final note on Section 301(20)(A)(iii). Requiring parties to fund Section 301(20)(A)(iii) communications with federal money not only serves the government interest of preventing actual and apparent corruption, but also prevents circumvention of campaign laws and principles. See NCPAC, 470 U.S. at 500 (noting the Court's"deference to a congressional determination of the need for a prophylactic rule").*fn93 Under Section 441b of FECA, corporations and unions are prohibited from using general treasury funds for independent expenditures expressly advocating the election or defeat of a particular federal candidate. 2 U.S.C. § 441b; see also MCFL, 479 U.S. at 241-64. If the backup definition of"electioneering communications" under Section 203 of BCRA is upheld, BCRA § 203(a); FECA §§ 316(a), (b)(2); 2 U.S.C. §§ 441b(a), (b)(2), corporations and unions would be additionally prohibited from using their general treasuries to fund candidate advocacy pieces which promote, oppose, attack, or support federal candidates. BCRA § 201(a); FECA § 304(f)(3)(A)(ii); 2 U.S.C. § 434(f)(3)(A)(ii). Most of the largest nonfederal money contributions to parties are by corporations and unions. See supra note 77. Thus, if national and state parties were allowed to spend unlimited amounts of nonfederal money on candidate advocacy, corporations and unions could largely circumvent the new BCRA Section 203 by funneling unlimited funds through political parties to pay for advertisements that they are now prohibited from broadcasting under Section 441b of FECA. To allow such a circumvention in that manner would be ludicrous. Finally, to permit national (or state or local) parties to convert nonfederal donations, whether from the general treasuries of corporations and unions, or from wealthy individuals and MCFL groups, to the federal purpose of directly influencing a federal election through candidate advocacy of the type defined in Section 301(20)(A)(iii) is, in and of itself, a direct circumvention of a fundamental premise of the FECA regulatory scheme relating to donations, blessed by the Supreme Court in Buckley and its progeny: only federal money may be used to directly influence a federal election. As long as the Buckley line of cases remains the law, it would make a mockery of those Supreme Court cases and the current regulatory scheme to allow political parties to use soft money to engage in what Justice Kennedy has aptly described as"covert advocacy." Shrink Missouri, 528 U.S. at 406, 407 (Kennedy, J., dissenting).

C. New FECA Section 323(d): Nonfederal Fund Restrictions on Tax-Exempt Organizations

Section 323(d) prohibits national, state, and local parties from donating either soft or hard money to, or soliciting for, either: (1) a Section 527 organization; or (2) a Section 501(c) organization if that organization makes expenditures in connection with a federal election, including expenditures for"federal election activity" as defined in Section 301(20)(A). BCRA § 101(a); FECA § 323(d); 2 U.S.C. § 441i(d). Judge Henderson strikes down the entire section as unconstitutional. I concur in her judgment, but for different reasons.

First, I would note that the defendants argue that the donation and solicitation restrictions should be reviewed under closely-drawn scrutiny. See Intervenors Opp'n Br. at 21-23. While I agree with the defendants that Buckley's closely-drawn scrutiny should apply to donation limitations to organizations as well as candidates, it is less clear which scrutiny should apply to the solicitation restriction. See J. Henderson Op. at Part IV.D.4. However, if the restriction on solicitation cannot even withstand closely-drawn scrutiny, there is no need to choose between the standards of review. Cf. Blount v. S.E.C., 61 F.3d 938, 942-43 (D.C. Cir. 1995).

As to the statute itself, Section 323(d) prevents national parties from donating any funds to certain Section 501(c) organizations. This complete ban on donations infringes the various parties' abilities to effectuate their members voices, and because it is not even closely drawn to serve a sufficient government interest, it cannot withstand scrutiny. Section 323(d) is not closely drawn as to Section 501(c) organizations because it prohibits solicitation for and donations to those organizations merely because they have made, in effect, expenditures for federal purposes in the past, and regardless of whether those donations will be used again for that very purpose. By not specifying the purpose for which the money will be put, Congress, in effect, is prohibiting solicitation for and donations to these Section 501(c) organizations that might in turn be used for nonfederal or mixed purposes. Congress, of course, can only do this if it could show that a sufficient government interest was being served by doing so. It has not. As discussed at length earlier, the only restrictions on uses of nonfederal funds that can be constitutionally regulated are uses that directly affect a federal election. See supra Parts I.A.2 & I.B.2. Any other use of the donation is too tangential to give rise to the risk of corruption, or appearance of corruption, that is necessary to warrant this congressional infringement on First Amendment rights. To say the least, the defendants have not produced sufficient evidence to demonstrate that an appearance of corruption, let alone corruption itself, arises when organizations of this type use donations from national, state, and local parties for nonfederal or mixed purposes.

Section 323(d) is also constitutionally problematic as a result of its treatment of Section 527 organizations. Section 527 organizations, according to the applicable IRS definition, exist, in part, to influence the selection, nomination, election, or appointment of individuals to state and local public offices, as well as to various political organizations. 26 U.S.C. §§ 527(e)(1) and (2). Accordingly, since Section 527 organizations could expend the solicited or donated funds for one of these nonfederal or mixed purposes,*fn94 the blanket prohibition in Section 323(d) on national, state, and local parties from assisting them in this manner is similarly not closely drawn. Perhaps if Congress had structured Section 323(d) like Section 323(b) and prohibited these organizations from using nonfederal funds -- which they had received from national, state, and local parties -- to directly affect federal elections, it might have been able to demonstrate that the restriction was closely drawn to serve the sufficient government interest required under any standard of review to justify these infringements. But it did not. Therefore, for all of the above reasons, I conclude that Section 323(d) is unconstitutionally overbroad.

D. New FECA Section 323(e): Nonfederal Fund Restrictions on Federal Candidates

Judge Henderson and Judge Kollar-Kotelly uphold Section 323(e) in its entirety. While I concur to the extent that this section prohibits federal officeholders and candidates from receiving, directing, transferring, or spending any nonfederal funds in connection with any election, including the kind of election activity defined in Section 301(20)(A),*fn95 I dissent with regard to the prohibition on a federal candidate, or officeholder, from soliciting funds for the benefit of his national party, especially since Congress can and has legally prohibited the parties (national, state, and local) from using nonfederal funds to directly influence federal elections. Accordingly, I am writing separately to dissent in part from, and concur in part in, their opinions.

With regard to the First Amendment rights of federal candidates and officeholders to solicit soft money funds for the use of their parties, I would dissent principally on the basis that the defendants have failed to demonstrate that soliciting nonfederal funds to be used by parties for purposes that, at most, indirectly affect federal elections is regulable by Congress. For example, donations used for such activities as party building, newsletters, genuine issue advocacy, and generic voter mobilization so indirectly affect federal elections, if at all, that they do not give rise to the minimally necessary appearance of corruption to warrant congressional intervention.

Indeed, soliciting donations to be used for the national party on such mixed (and/or nonfederal) purposes is the kind of conduct by officeholders which the public not only would expect them to participate in, but which is fundamental to the successful operation of the major national parties. To the extent that helping their parties in this way provides officeholders with some added status among other party officials, in my judgment, is too attenuated a benefit to give rise to the appearance of corruption necessary to warrant congressional intervention. See supra Part I.A.3. Accordingly, I concur in part in, and dissent in part from, the holding of my colleagues.

E. New FECA Section 323(f): Nonfederal Fund Restrictions on State Candidates

Section 323(f) prohibits state candidates from spending funds for a communication of the type discussed in Section 301(20)(A)(iii) unless the funds are subject to the limitation, prohibitions, and reporting requirements of FECA (i.e., hard money). Restricting the use of soft money by state candidates for communications that directly affect federal elections is, of course, consistent with my preceding discussion of the constitutionality of Section 301(20)(A)(iii). By placing this restriction on state candidates Congress is simply guarding against similar conversions of soft money donations to fund communications that are designed to accomplish the federal purpose of directly influencing a federal election. Since this section is closely drawn to uses of soft money by state candidates exclusively for that purpose, I similarly uphold its constitutionality.

II. Title II: Noncandidate Campaign Expenditures

Sections 201, 204, 213

I join with Judge Henderson, but for different reasons, in holding that the primary definition of electioneering communications set forth in Section 201 is unconstitutional. I do not, however, join in her holding that the backup definition also provided in Section 201 is unconstitutional.

To the contrary, I uphold the backup definition's constitutionality, as does Judge Kollar-Kotelly who joins in my opinion as a necessary alternative to Judge Henderson's and my finding the primary definition unconstitutional. The following are my reasons for rejecting the primary definition and upholding the backup definition.

A. Section 201: The Primary Definition

Title II of BCRA is Congress's attempt to close certain loopholes in FECA and to regulate certain conduct arising from the use, both directly and indirectly, of corporate and union treasury funds to finance electioneering communications that directly affect federal elections, even though masquerading otherwise as"pure issue advocacy." Because, in part, Judge Henderson believes that Buckley and its progeny set forth a bright line test requiring the presence of certain"magic words" advocating the election or defeat of a federal candidate in order for this advocacy to be regulable by Congress, she strikes down both the primary and backup definitions.*fn96 Judge Kollar-Kotelly and I disagree with this reasoning, and I join in the portion of her opinion analyzing why the Supreme Court never intended the so-called express advocacy test to be a constitutional rule of law limiting the power of Congress to regulate expenditures for certain uncoordinated advocacy that directly affects federal elections, notwithstanding the absence of these words.*fn97

While Judge Kollar-Kotelly and I agree regarding Congress's power to regulate the source of the funds used for these communications and the disclosures that donors must file, I do not agree that the primary definition, as crafted by Congress, is sufficiently tailored to regulate the electioneering advocacy it seeks to cover. To the contrary, the primary definition, which regulates communications referring to clearly identified federal candidates based upon when and where they are broadcast, rather than their effect on federal elections, sweeps so broadly that it captures too much First Amendment protected speech that Congress, in the absence of a demonstrated compelling government interest, has no power to regulate. What type of speech is that?

The plaintiffs have clearly demonstrated that corporations, interest groups, labor unions and other entities air genuine issue advertisements in the periods immediately preceding general and primary elections, the sole purpose of which is educating the viewers about an upcoming vote on pending legislation, and encouraging them to inform their elected representative to vote for or against the bill (i.e., legislation-centered advertisements). Edward Monroe, Director of Political Affairs for Associated Builders and Contractors ("ABC"), explained that"serious legislative initiatives or regulatory proposals often are considered near the time of elections." Monroe Decl. ¶¶ 18-19. Laura Murphy, legislative director for the ACLU, seconds this:"[The] 60 days before a general election and 30 days before a primary... are often periods of intense legislative activity. During election years, the candidates stake out positions on virtually all of the controversial issues of the day. Much of the debate occurs against the backdrop of pending legislative action or executive branch initiatives." Murphy Decl. ¶ 12. See also Mann Cross Exam. at 176 (explaining that a flurry of legislative activity occurs near the end of a congressional session, therefore, often within the 60 day period preceding a general election); Finding 359.*fn98

Notwithstanding defendants' expert testimony to the contrary,*fn99 interest groups believe, and plaintiff's expert agrees,*fn100 that the periods immediately preceding elections are the most effective times to run issue advertisements discussing pending legislation because the public's interest in policy is at its peak.*fn101 Edward Monroe of ABC explains that"it is... clear that members of the public are generally more receptive to and engaged in considering government policy ideas and issues as elections near. If that is the time when people will listen, that is the time to speak. And once an election occurs, there seems to be a period of fatigue during which political matters are of less interest, making issue ads then less effective." Monroe Decl. ¶¶ 18-19; see also Finding 359. Likewise, Paul Huard, Senior Vice President for Finance and Administration for the National Association of Manufacturers ("NAM"), finds that"Americans tend to have greater interest in political matters as an election approaches. At the same time, elected officials are most attuned to the views of their constituents in the pre-election period. Thus, for many purposes, the pre-election season is a critical time for issue ads." Huard Decl. ¶ 10; see also Finding 359.

The mere fact that these issue advertisements mention the name of a candidate (i.e., the elected representative in whose district the advertisement ran) does not necessarily indicate, let alone prove, that the advertisement is designed for electioneering purposes. To the contrary, the testimony of various plaintiffs' witnesses indicates that, in their experience, there are many reasons why it is helpful, if not necessary, to mention a candidate's name in these advertisements in order to focus the public's attention on a particular pending piece of legislation. For example, Paul Huard of NAM states"[t]here are many reasons that an issue ad may need to refer to the name of an elected official or candidate. Many bills are identified with particular sponsors and may be known by the sponsors' names. Also, both incumbents and candidates may be prominent people whose support or opposition to a bill or policy may have important persuasive effect.... Also, if an issue ad is used to explain why a legislative position of a particular Member of Congress is good for his or her district or state, the member generally must be mentioned. The same is true if the purpose of the ad may be to induce viewers to contact the Member and communicate a policy position." Huard Decl. ¶ 12 (emphasis added); see also Finding 293.

Similarly, Denise Mitchell, Special Assistant for Public Affairs to the AFL-CIO, concurred, explaining that it is often necessary to refer to a federal candidate by name because"[t]he express or implied urging of viewers or listeners to contact the policymaker regarding [an] issue is... especially effective by showing them how they can personally impact the issue debate in question." Mitchell Decl. ¶ 11; see also Finding 293. A quick review of a classic example of a legislation-centered genuine issue advertisement demonstrates these points.

In 1998, the AFL-CIO aired an advertisement entitled"Barker." The advertisement referred to a federal candidate by name only in the call-to-action line at the end of the advertisement where it urged viewers to call their Member of Congress and express their position regarding the pending Fast Track trade legislation. The legislation was scheduled for a vote in the House of Representatives on September 25, 1998 -- within 60 days of the general election. The text of the advertisement is as follows:*fn102

Paid for by the Working Men and Women of the AFL-CIO. [Barker speaking]: Okay ladies and gents, step right up and see if you can follow the ball. Is it here? Is it there? Where could it be? [Voice over]: They're playing games again in Washington. Without discussion or debate, they're planning another vote on the controversial Fast Track law -- special powers to ram through trade deals like NAFTA. Fast Track failed last year because working families don't want more trade deals that put big corporations first; deals that ignore our concerns about lost jobs; environmental problems on our borders, and dangerous, imported foods. But Newt Gingrich and the sponsors of Fast Track hope they can sneak it by this fall, while public attention is focused on other issues. [Barker speaking]: Keep your eyes on the ball now... [Voice over]: Call Representative _____ at xxx-xxx-xxxx and tell him to vote no on Fast Track. Tell him we're still paying attention. And Fast Track is still a bad idea. Mitchell Decl. Exhibit 116, Exhibit 1 at 86.

Defendants' own expert, David Magleby, who stated in his report that he would "presume" an advertisement is electioneering merely because it mentions a federal candidate by name,*fn103 candidly admitted after reviewing "Barker" that such an advertisement is a"genuine" issue advertisement. He concluded as such -- even though it mentions a federal candidate's name -- because "[t]he body of the ad has no referent to [a candidate] whatsoever. The only referent to [the candidate] is the call line." Magleby Cross Exam. at 104.

[A] generic call your Congressman, call your Senator, when then linked to a legislation and call your Congressman or Senator about this legislation without a referent to their position on the issue, seems to me substantively different than when they are mentioned in view of what their position is on that issue. Q. When you say substantively different, are you referring to a difference with respect to whether the advertisement communicates an electioneering message? A. Yes. Magleby Cross Exam. at 106.

Because genuine issue advertisements, like"Barker," have been, and will need to be, aired during periods of legislative activity leading up to elections, plaintiffs contend the primary definition, if upheld, will capture them as"electioneering communications."*fn104 As such, they will be subjected to a host of limitations and regulations which, according to the plaintiffs, will limit and chill their ability to engage in First Amendment protected speech. I agree.

The crux of the problem with the primary definition is that, unlike the backup definition, it does not depend on the effect of the communication's message on a candidate's election. As such, many genuine issue ads, like"Barker," will be treated the same as the sham"issue" ads Congress supposedly was intending to regulate. It is the absence of a link between the advocacy of an issue and a candidate's fitness, or lack thereof, for election that renders congressional intervention with respect to genuine issue ads of this type unconstitutional. Notwithstanding the absence of this link, defendants contend, and Judge Kollar-Kotelly agrees, that the evidence sufficiently demonstrates that even though some genuine issue advertisements that run in the months leading up to an election will be swept in under this definition, it is too insufficient a number to render the primary definition constitutionally defective. I disagree.

The plaintiffs have met their burden in this facial challenge because they have shown that BCRA's primary definition"reaches a substantial amount of constitutionally protected conduct." City of Houston v. Hill, 482 U.S. 451, 458 (1987) (quoting Village of Hoffman Estates v. Flip-side, Hoffman Estates, Inc., 455 U.S. 489, 494 (1982)) (emphasis added). Plaintiffs, in short, have demonstrated from the statute's"text... and from actual fact that a substantial number of instances exist in which the [l]aw cannot be applied constitutionally." New York State Club Ass'n v. City of New York, 487 U.S. 1, 14 (1988) (emphasis added). These instances are not merely"'some' overbreadth," Ashcroft v. ACLU, 535 U.S. 564, 584 (2002) (quoting Reno v. ACLU, 521 U.S. 844, 896 (1997)), or a "single" or"bare possibility" of an impermissible application. Members of City Council of the City of Los Angeles v. Taxpayers for Vincent, 466 U.S. 789, 800 & n.19 (1984) (quoting Broadrick v. Oklahoma, 413 U.S. 601, 630 (1973) (Brennan, J., dissenting)). Instead, the plaintiffs have shown, as the Supreme Court aptly put it in Vincent, that BCRA presents"a realistic danger that the statue will significantly compromise recognized First Amendment protections of parties not before the Court." Id. at 800-801.

The evidentiary basis of both the plaintiffs' and defendants' arguments concerning the primary definition's overbreadth is the Buying Time 1998 and Buying Time 2000 studies. While the defendants correctly contend that plaintiffs carry the burden to show that BCRA is substantially overbroad,*fn105 the studies defendants rely upon to show that it is narrowly tailored*fn106 are, in essence, a highly controversial"survey" of the ads run in the months leading up to the 1998 and 2000 elections. Judge Henderson correctly notes in her opinion, see J. Henderson Op. Part IV.A, that the parties"quarrel at length" regarding the significance of the Buying Time studies and, in particular, its conclusions regarding the percentage of genuine issue advocacy that would be improperly regulated by Section 201 of BCRA. Unlike Judge Henderson, I believe that the Buying Time studies are entitled to some evidentiary weight.*fn107

However, I do not believe that the studies' statistical conclusions are the last word in this Court's analysis of whether or not the primary definition is overbroad. With the respect to the percentage of protected, political speech that is, or will be, regulated by BCRA, it is, of course, impossible to quantify the exact percentage with absolute certainty. See New York v. Ferber, 458 U.S. 747, 773 (1982) (explaining that"[h]ow often, if ever" the statute at issue in Ferber would regulate protected speech"cannot be known with certainty. Yet we seriously doubt, and it has not been suggested, that these arguably impermissible applications of the statute amount to more than a tiny fraction of the materials within the statute's reach."). Thus, it is important not to overstate the significance of the Buying Time studies when using them as the basis of a finding of unconstitutional overbreadth. After all, the studies did not analyze every advertisement that ran in every market during the 1998 and 2000 elections. Instead, they analyzed the top 75 media markets, encompassing eighty percent of the advertisements runs during those elections. See Findings 315-17. I do not state this fact to suggest that every ad had to be reviewed before the studies could be deemed credible, rather, that the percentages produced by the studies may, in fact, be an overstatement, or understatement, of the statute's overbreadth. See Findings 317-18. In addition, I would note that the Buying Time 2000 study did not analyze advertisements run in the 30 days preceding a primary or preference election, even though such ads aired during that period are entirely regulable by BCRA's primary definition. See Finding 316.

Furthermore, this Court, in my judgment, cannot rely upon the results of the two, Buying Time studies without analyzing, to some extent, the parties' dispute regarding the formulas used to produce those results. The defendants favor the formula used in the 1998 study,*fn108 which compares the total number of genuine issue ads regulated by BCRA to the total number of genuine issue ads run in a calendar year. See Findings 335, 336, 337. The result is the percentage of all genuine issue advocacy that would have been regulated by BCRA in the course of that year. The plaintiffs, however, contend that the 1998 formula misstates BCRA's impact because it includes ads that were run outside the 60-day period preceding a general election, and therefore would not have been subject to regulation under the primary definition. See Findings 336, 338. According to the plaintiffs, the formula applied in the Buying Time 2000 study more accurately reflects BCRA's impact by focusing on the exact period of time regulated by BCRA: the 60 days preceding a general election.*fn109

Looking at ads aired only during that 60-day period, the 2000 formula compares the number of genuine issue ads to the total number of ads, thereby calculating the percentage of all ads that would have been regulated by BCRA that were genuine issue ads.*fn110

As the Supreme Court in Broadrick v. Oklahoma stated that a statute's overbreadth must be"judged in relation to the statute's plainly legitimate sweep," I find that the 2000 formula more accurately measures BCRA's impact. 413 U.S. 601, 615 (1973). The results produced by the 1998 formula do not assist this Court in comparing BCRA's overbreadth to its"plainly legitimate sweep," because it measures ads that never would have been regulated by BCRA.*fn111 While the 1998 formula shows BCRA's impact on all genuine issue advocacy over the course of a calendar year, that information is of limited value when BCRA's primary definition applies only in the 30 days preceding a primary election and the 60 days before a general election.*fn112

Applying the 2000 formula to the data collected during the 1998 and 2000 studies shows that the primary definition's overbreadth is neither speculative nor hypothetical, but real and substantial. In 1998, of the ads that met BCRA's primary definition and were aired in the 60 days preceding a general election, 14.7 percent were genuine issue advocacy. See Finding 339. As to 2000, however, the Buying Time 2000 study concluded that figure was only 2.33 percent. See Finding 357. That percentage, however, was later increased by Professor Goldstein, who compiled the data base that served as the foundation of the Buying Time studies. Professor Goldstein testified on cross examination that he had reevaluated the results of the study for the purposes of this litigation and concluded that, in fact, 17 percent of the ads that met BCRA's primary definition and were aired in the 60 days preceding the 2000 general election were genuine issue advocacy. See Finding 357; see also Goldstein Cross Exam. at 160, 169; McConnell Reply Br. at 36-37. Percentage discrepancies aside, I find that 14.7 percent and 17 percent of the ads run in the months leading up to the 1998 and 2000 elections, respectively, represents a"substantial amount" of protected speech and renders the primary definition defective as constitutionally overbroad.

But these statistics, alone, do not present the full picture of BCRA's impact on genuine issue advocacy in the 60 days preceding a general election. Indeed, determining whether BCRA's primary definition reaches a substantial amount of constitutionally protected issue advocacy is not simply a function of calculating the percentage of pure issue ads that would have been captured by that definition during the 60-day period preceding the 1998 and 2000 federal elections. Because the total amount of issue advocacy likely to be generated in any given election year is a function of both the quantity and nature of the issues Congress chooses to address in that pre-election period, those numbers should not be viewed in a legislative vacuum. Ideally, this court should additionally assess whether the legislative agendas in 1998 and 2000 were unusually active, controversial or both. Regrettably, however, the record does not lend itself to such an analysis. Obviously, the more active and/or controversial the legislative schedule, the greater (or lesser) the amount of issue advocacy one would expect it to have generated. Simply put, the amount of pure issue advocacy captured in a particularly contentious, or active, legislative period, is likely to be higher than that captured in a slow, or routine, legislative period. Furthermore, restricting 14.7 percent of genuine issue advocacy in 1998 would have restricted otherwise protected speech that would have been seen in 30 million American homes, a number that brings into sharp relief the effect BCRA will have on the amount of information available to voters. See Finding 335. Accordingly, for the reasons stated previously, there is reason to believe that the amount of issue advocacy likely to be generated in future election cycles will be at least as substantial as it was during those years.

Ever mindful that overbreadth is"strong medicine" to be used"sparingly and only as a last resort," Broadrick, 413 U.S. at 613, I do not lightly find that the primary definition of electioneering communications is substantially overbroad. However, the realistic danger that the primary definition of electioneering communications will significantly compromise genuine issue advocacy necessitates such a finding. In circumstances such as these,"the possible harm to society in permitting some unprotected speech to go unpunished is outweighed by the possibility that protected speech of others may be muted." Ashcroft v. Free Speech Coalition, 535 U.S. 234, 255 (2002) (quoting Broadrick, 413 U.S. at 612). I therefore find that the primary definition of electioneering communications is unconstitutionally overbroad.

I do not, however, join in Judge Henderson's conclusion that both the primary and backup definitions of electioneering communications are unconstitutional due to underinclusion. Plaintiffs argue, and Judge Henderson agrees, that Section 201 must fail because it regulates only broadcast and cable communications, but does not regulate other mediums of communication such as print, direct mail, and the internet.*fn113 I disagree and join Judge Kollar-Kotelly's analysis on that issue. See J. Kollar-Kotelly Op. at Part III.I.F. However, one point bears emphasizing: Congress is infinitely more familiar than this Court with the circumstances and practical ramifications surrounding federal elections and campaign finance laws. The record thoroughly, and convincingly, demonstrates that Congress's decision to regulate broadcast communications, rather than other forms of communication, was well justified. Congress, as defendants argue,"need not regulate the entire universe of activity intended to influence federal elections" in order to constitutionally regulate electioneering communications. Gov't Opp'n Br. at 96. As Congress, in its expertise, has made this judgment and the record"demonstrates that the recited harms [justifying a statute] are real," plaintiffs' argument that the statute is unconstitutional for underinclusion is unavailing. Turner I, 512 U.S. at 664.

B. Section 201: The Backup Definition

Unlike the primary definition, the backup requires as a link between the identified federal candidate and his election to that office, certain language the purpose of which is advocacy either for, or against, the candidate. For the same reasons that the absence of that link doomed the primary definition, it sustains the backup.

Congress concluded, and the record more than adequately demonstrates, that in the twenty-eight years since Buckley, corporations, unions, and interest groups, have increasingly affected federal elections by funding out of their general treasuries uncoordinated"issue ads" that either they, or a political party, ran in the months leading up to an election. See Findings 280-84, 296-301; see also Annenberg Study at 1, 4, 3, 7-8 (showing that the number of issue advertisements has increased, as well as the number of groups airing them, from 1996 to 1998, and that corporations, interest groups, and unions began in 1996 to actively use treasury funds to sponsor issue advertisements that"looked and sounded like campaign ads."). In order to avoid regulation as express advocacy, those so-called"issue advertisements" did not contain certain"magic words" designed to support or oppose a specific candidate's election or re-election.*fn114 The ads, however, were constructed in such a way that they simultaneously presented their sponsors' stand on an issue, identified a specific candidate's positions or track record thereon, and under the guise of admonishing the viewer to inform the candidate of his view, suggested that a candidate who takes (or has taken) the candidate's position should (or should not) be elected to that office. Ads such as these have been typically described as"sham issue advertisements,"*fn115 or candidate-centered advertisements, and the factual record unequivocally establishes that they have not only been crafted for the specific purpose of directly affecting federal elections, but have been very successful in doing just that. See Findings 273-74. Indeed, they have been so successful that it is widely believed in the industry that old fashioned express advocacy of the"magic words" type is far less effective in winning over a viewer's vote. See Findings 273-74, 279; see also Bailey Decl. ¶¶ 1-2 (stating that"it is rarely advisable to use such words as'vote for' or'vote against.'").

Defendants contend that because these sham issue ads, in essence, directly affect federal elections, candidates are"beholden" to those who fund them.*fn116 It is that indebtedness, according to the defendants, that if left unregulated, will undermine the public's confidence in the integrity and fairness of our electoral process in the same way that unregulated express advocacy otherwise would, and thereby give rise to corruption or the appearance of corruption. See Mellman and Wirthlin at 9-10.*fn117 For essentially the same reasons that I uphold Congress's definition of federal election activity in Section 301(20)(A)(iii), see supra Part I.B.2, I believe that large soft money expenditures to fund this type of covert advocacy, regardless of whether corporations, labor unions, and interest groups produce them themselves, give rise to a public perception that the candidate is being directly benefitted and will naturally reciprocate with either favored access, or increased legislative influence, or both. See id. Accordingly, the source of the funds used and the identity of the sponsors are both, in my judgment, regulable by Congress.

Plaintiffs raise a number of arguments as to why the backup definition of electioneering communications is"impermissibly vague." McConnell Opening Br. at 70. First, they claim that"reasonable people can, and emphatically do, disagree about whether virtually any particular advertisement meets the criteria of BCRA's fallback definition." Id. As proof of the definition's vagueness, plaintiffs offer the deposition testimony of one of BCRA's sponsors and one of defendants' expert witnesses, in which they disagree about whether a particular ad is a genuine issue ad, or whether it promotes or supports a particular candidate.*fn118 Plaintiffs also offer testimony of two Members of Congress, in which one concludes that a particular ad supports a candidate, while the other concludes that the same ad attacks the candidate.*fn119 While BCRA's sponsors may disagree about the purpose and effect of an ad, that fact alone does not demonstrate unconstitutional vagueness. Perfect clarity, of course, is not required when a law regulates speech. As the Supreme Court said in Grayned,"we can never expect mathematical certainty from our language." 408 U.S. at 110 (1972). For this reason, the Supreme Court has held that a statute's vagueness exceeds constitutional bounds only when"its deterrent effect on legitimate expression is... both' real and substantial' and... the statute is [not] readily subject to a narrowing construction by state courts." Young v. American Mini Theatres, Inc., 427 U.S. 50, 60 (1976) (quoting Erznoznik v. City of Jacksonville, 422 U.S. 205, 216 (1975)) (emphasis added). Moreover, if a statute"gives the person of ordinary intelligence a reasonable opportunity to know what is prohibited," it is not void for vagueness. Grayned, 408 U.S. at 108.

Next, plaintiffs contend that the definition's use of the words"promote,""support,""attack," and"oppose" to define the sponsor's message causes it to be unconstitutionally vague.*fn120 I disagree. The backup definition's language, specifically those words, is not void for vagueness because a person of ordinary intelligence would understand what is prohibited. See Grayned, 408 U.S. at 108. Indeed, one need only conclude, in effect, that the ad is not neutral as to both candidates for it to have satisfied the backup definition, and thereby have satisfied the objective First Amendment standard that a reasonable person considering the context and nature of the expression at issue is able to evaluate the speech. Such objective tests have routinely been applied in the First Amendment context. See, e.g., County of Allegheny v. ACLU, 492 U.S. 573 (1989) (religious expression); Hess v. Indiana, 414 U.S. 105 (1973) (fighting words); Miller v. California, 413 U.S. 15 (1973) (obscenity); Brandenburg v. Ohio, 395 U.S. 444 (1969) (fighting words). For example, the following advertisement entitled"No Two Way" and sponsored by the AFL-CIO is not neutral as to a federal candidate, as it attacks his or her position on the federal budget. Mitchell Decl. ¶ 41. While former Congresswoman Andrea Seastrand is named, this advertisement also ran in thirty-five other congressional districts, naming federal candidates from those districts.

CAROLYN: My husband and I both work. And next year, we'll have two children in college. And it will be very hard to put them through, even with the two incomes.

[Announcer]: Working families are struggling. But Congresswoman Andrea Seastrand voted with Newt Gingrich to cut college loans, while giving tax breaks to the wealthy. She even voted to eliminate the Department of Education. Congress will vote again on the budget. Tell Seastrand, don't write off our children's future.

CAROLYN: Tell her, her priorities are all wrong. Mitchell Decl. Exhibit 114.

Congress is not, and should not be, constitutionally confined to "bright line" tests such as the primary definition or the"magic words" formulation in enacting campaign finance restrictions. A person of ordinary intelligence can be expected to understand this test, and know what is prohibited.

However, while I do not believe that the words used to define the message are vague, I do believe that the backup definition's final clause, which requires the message to be"suggestive of no plausible meaning other than an exhortation to vote," is unconstitutionally vague. In my judgment, it is extremely difficult, if not impossible, for a speaker to determine with any certainty prior to airing an ad that it meets that requirement. Whether an ad is suggestive of no plausible meaning other than an exhortation to vote depends on a number of variables such as the context of the campaign, the issues that are the centerpiece of the campaign, the timing of the ad, and the issues with which the candidates are identified.*fn121 The"uncertain meaning[]" of this phrase in the backup definition will, as the Supreme Court stated in Grayned,"inevitably lead citizens to steer far wider of the unlawful zone... than if the boundaries of the forbidden areas are clearly marked." 408 U.S. at 109. The chilling effect of this language does not doom the backup definition as unconstitutionally vague, however, because it is susceptible to a saving construction. See Edward J. Debartolo Corp. v. Gulf Coast Bldg. & Constr. Trades Council, 485 U.S. 568, 575 (1998) ("[E]very reasonable construction must be resorted to, in order to save a statute from unconstitutionality." (quoting Hooper v. California, 155 U.S. 648, 657 (1895)); see also Carlin Communications, Inc. v. FCC, 837 F.2d 546, 560-61 (2d Cir. 1988) (finding that as"the presumption is in favor of severability," the court"dropped" the"invalid part" from an FCC regulation, i.e., the words"or indecent," because"the remainder of the statute is fully operative."), cert. denied, 488 U.S. 984 (1988).

The backup definition's susceptibility to a saving construction is a function of how it is written. Because the offending phrase is simply appended to the end of the definition, it can be excised without rewriting the entire definition. See Commodity Futures Trading Comm'n, 478 U.S. at 841 ("Although this Court will often strain to construe legislation so as to save it against constitutional attack, it must not and will not carry this to the point of perverting the purpose of statute... or judicially rewriting it." (quoting Aptheker v. Secretary of State, 378 U.S. 500, 515 (1964) (internal quotations omitted)). By so construing the backup definition to avoid vagueness, the definition assures that there will be no real, let alone substantial, deterrent effect on political discourse unrelated to federal elections. Genuine issue advocacy thereby remains exempt from both the backup definition and its attendant disclosure requirements and source restrictions. Similarly, genuine issue advocacy, specifically of the legislation-centered type, that mentions a federal candidate's name in the context of urging viewers to inform their representatives or senators how to vote on an upcoming bill will not be regulated by the backup definition because it does not promote, support, attack, or oppose the election of that candidate. See Findings 368-73 (providing examples of legislation-centered advertisements that do not promote, support, attack, or oppose the election of a federal candidate).

Indeed, the backup definition of electioneering communications with its final clause severed is very similar to the type of"public communication" defined by BCRA's Section 301(20)(A)(iii) as"federal election activity." See BCRA § 301(20)(A)(iii). While the arguments against vagueness applicable to Section 301(20)(A)(iii) are generally applicable to the backup definition, one distinction is important to note: the sophistication in electioneering communications of the parties being regulated is not equally applicable to the backup definition. That said, however, I do not believe it is necessary to make a similar finding here in regard to the comparative sophistication of corporations, labor unions, interest groups, and other participants in political speech. Additionally, whatever chilling effect, if any, the definitions of"public communications" and"electioneering communications" may have are minimized in two, substantial ways: (1) corporations, labor unions, national banks, individuals, and other entities can avoid regulation simply by not mentioning a candidate for federal office in its ad, and (2) those groups may seek an advisory opinion from the FEC to determine whether a communication is regulated by BCRA.*fn122

Thus, the plaintiffs have failed to demonstrate that the vagueness of the backup definition as severed is real and substantial enough to deprive a person of ordinary intelligence a reasonable opportunity to know what is prohibited. As such, the backup definition of electioneering communications is constitutional.

C. Section 204: The Wellstone Amendment

While I join Judge Henderson in holding that Section 204 is unconstitutional as it applies to MCFL nonprofit corporations, I do not agree that it is unconstitutional as it applies to those nonprofit corporations that do not qualify for MCFL status. Therefore, I am writing separately to explain my reasons for joining with Judge Henderson in part, and with Judge Kollar-Kotelly in part who upholds the constitutionality of Section 204 for different reasons.

Defendants concede that Section 204 does not contain an exemption for MCFL organizations. However, defendants argue that an exemption nonetheless exists for those organizations because the Supreme Court in MCFL"carve[d] out an as-applied exemption to restrictions on corporate election activity." Gov't Opening Br. at 167 (emphasis in original). It is therefore inconsequential, according to the defendants, that BCRA does not codify the MCFL exemption. The defendants' reasoning misses the mark.

Under FECA, nonprofit corporations were required to pay for express advocacy expenditures from a separately segregated fund, rather than general treasury funds, unless that corporation met the three requirements set forth by the Supreme Court in MCFL.*fn123 Section 203 of BCRA expands FECA's separately segregated fund requirement to apply not only to corporate expenditures for express advocacy, but to expenditures by corporations, labor unions, and national banks for electioneering communications (i.e., nonexpress advocacy), as defined by Section 201. Likewise, Section 203 expands the exemption for nonprofit corporations from the separately segregated fund requirement of FECA. Whereas only those nonprofit corporations that met the three requirements set forth in MCFL were exempt from FECA's separately segregated fund requirement, Section 203 of BCRA ignores completely the MCFL criteria and instead provides that all nonprofit corporations organized under either Section 501(c)(4) or Section 527(e)(1) of the Internal Revenue Code may pay for electioneering communications using general treasury funds, simply by virtue of their incorporation under those sections. Doing so, in my judgment, was perfectly consistent with both MCFL and Congress's power under the Elections Clause, U.S. Const. art. I, § 4.

What Section 203 provides, however, Section 204 takes away. As defendants acknowledge, Section 204 completely cancels out the exemption for all nonprofit corporations provided by Section 203. See Gov't Opening Br. at 164, n.115. Plaintiffs contend, and Judge Henderson agrees, that Section 204 is unconstitutional because, in essence, it contravenes the Supreme Court's decision in MCFL.*fn124 I, too, agree with this conclusion, but I also believe that Section 204 is unconstitutional only in its application to those nonprofit organizations exempted under MCFL. Congress, in my judgment, has the authority to withdraw the exemption it provided in Section 203 to those nonprofit corporations not exempted by MCFL. Thus, the fact that Section 204 applies to nonprofit corporations does not, in and of itself, impair its constitutionality; the critical factor is whether the nonprofit corporation that is being regulated is one protected from such regulation by the MCFL holding.

Congress's authority to withdraw the expansion of the exemption rests on the same basis as its authority to require non- MCFL, nonprofit corporations to use only separately segregated funds to purchase electioneering communications: the demonstrated appearance of corruption that arises if for-profit corporations and unions are able to funnel their general treasury funds through nonprofit corporations in order to purchase electioneering communications that they cannot otherwise purchase directly.*fn125 Judge Kollar-Kotelly sets forth this reasoning in the portions of her opinion upholding Section 203's prohibition on the use of corporate and union treasury funds, and Section 201's disclosure requirements, for corporations, unions, and nonprofit corporations. See J. Kollar-Kotelly Op. at Part III.I.E.1.a. I concur in her reasoning and it needs no further repetition here.

Accordingly, for the reasons stated above and the relevant portions of Judge Kollar-Kotelly's opinion, I hold that Section 204 is unconstitutional only in its application to MCFL, nonprofit corporations.

D. Section 213: The Party Choice Provision

Under FECA, as interpreted by the Supreme Court in Colorado I, and Colorado II, national party committees could make unlimited independent expenditures on behalf of a federal candidate, as well as coordinated expenditures up to the amount provided for by the Party Expenditure Provision, 2 U.S.C. § 441a(d).*fn126 BCRA, however, changes this by forcing national parties, in effect, to choose between making coordinated expenditures and unlimited independent expenditures on behalf of their federal candidates. The defendants contend that such a choice is necessary, because once a national party coordinates with a candidate, no expenditure thereafter is truly independent.*fn127 While I agree that Congress can constitutionally place restrictions on the ability of a national party to make coordinated expenditures, Section 213's restriction on national party independent expenditures flies in the face of the Supreme Court's holding in Colorado I, and is therefore unconstitutional.*fn128

The provisions of Section 213 are, as plaintiffs acknowledge,"relatively straightforward." McConnell Opening Br. at 85. After the date on which a party nominates its candidate, if it first makes an independent expenditure on behalf of that candidate, any subsequent coordinated expenditure will be subject to the same $5,000 limitation that exists for multicandidate committee expenditures, and not under the higher limitation set forth in formula form in the Party Expenditure Provision of Section 441a(d) of FECA. Conversely, if the party chooses to first make a coordinated expenditure on behalf of a candidate, thereby taking advantage of the higher, formula-based dollar limitation under the Party Expenditure Provision, it cannot thereafter make any independent expenditures on behalf of that candidate. Additionally, Section 213 treats"all political committees established and maintained by a national political party (including all congressional campaign committees) and all political committees established and maintained by a State political party (including any subordinate committee of a State committee)" as a"single political committee." Finally, Section 213 forbids transfers, assignments, and receipt of funds by a committee of a political party that has made a coordinated expenditure for a federal candidate to a political party committee"that has made or intends to make an independent expenditure with respect to the candidate." BCRA § 213; FECA § 315(d); 2 U.S.C. § 441a(d).

Plaintiffs argue that the practical effect of Section 213 is to"ban political parties from making independent expenditures at all (if they choose to make coordinated expenditures first), and to ban them from making coordinated expenditures at all (if they choose to make independent expenditures first)." McConnell Opening Br. at 86. As to the latter claim, plaintiffs undoubtedly exaggerate Section 213's reach. While it is clear that it does prevent national party committees that have made an independent expenditure on behalf of a federal candidate from making coordinated expenditures up to the increased amount as calculated under the formula in 2 U.S.C. § 441a(d), they are still able nonetheless to make coordinated expenditures (or contributions) up to the $5,000 limitation provided in 2 U.S.C. § 441a(a)(2), just like any other multicandidate committee. Thus, since the national political parties may still make coordinated expenditures after making an independent expenditure first, the real issue before this Court is the constitutionality of Congress's new provision prohibiting national political party committees from making independent expenditures if it has first made a coordinated expenditure on behalf of a federal candidate. Applying a strict scrutiny review to this prohibition,*fn129 I find that Section 213's restriction on independent expenditures by national political parties is unconstitutional.

The Supreme Court made clear in Colorado I that"[t]he independent expression of a political party's views is'core' First Amendment activity no less than is the independent expression of individuals, candidates or other political committees," 518 U.S. at 616, and it cannot therefore be restricted unless the expenditures pose"special dangers of corruption."*fn130 Id. Despite the FEC's contention in Colorado I that"all party expenditures should be treated as if they had been coordinated as a matter of law" because"'party officials will as a matter of course consult with the party's candidates before funding communications intended to influence the outcome of a federal election,'" the Supreme Court found that no such risk of corruption existed. 518 U.S. at 619, 620 (citing the brief of the FEC)."[T]he constitutionally significant fact" upon which the Supreme Court relied to determine that national political parties could, like any other political speaker, exercise their First Amendment right to make independent expenditures without posing a risk of corruption,"is the lack of coordination between the candidate and the source of the expenditure." Id. at 617 (emphasis added). Absent such evidence, the Supreme Court refused to accept"as a factual matter" the FEC's contention that all expenditures by a party were ipso facto coordinated.*fn131 Id. at 621.

Section 213 conditions a party's exercise of its First Amendment right to make independent expenditures upon its having refrained from engaging in coordinated expenditures up to the limit of the Party Expenditures Provision. Defendants' argument that"[p]roviding... an option, which a party committee is free to accept or decline, does not constitute a restriction of First Amendment rights" is, at best, fanciful. Gov't Opening Br. at 180.*fn132 By mandating that a party that has made a coordinated expenditure cannot thereafter make independent expenditures, BCRA leaves national political parties with the familiar predicament of only being able to make expenditures up to the Party Expenditure Provision limits set forth in 2 U.S.C. § 441a(d). Indeed, this is the very situation the Supreme Court found unconstitutional in Colorado I.

Defendants have produced no evidence of"special dangers of corruption" posed by national parties making an independent expenditure after having made one or more coordinated ones. At most this Court is presented with defendants' own speculation and suspicion that all expenditures made after a coordinated expenditure are necessarily coordinated between candidate and party. Of course, if they had such evidence of corruption or potential corruption, then restrictions on the independent expenditures of national parties might possibly be warranted. However, the nature of the defendants' concerns about coordination between parties and candidates*fn133 are such that they can file a complaint with the FEC challenging whether a party's expenditure on behalf of a candidate is, in fact, independent. A complete ban on all party independent expenditures after a coordinated expenditure has occurred is just not sufficiently tailored to deal with that type of problem. Moreover, Section 213's ban on independent expenditures treats national political parties differently than other political actors even though, as mentioned previously, a party's independent expenditures are no less protected as First Amendment activity than that of others. See Colorado I, 518 U.S. at 616.

Having rejected Section 213's ban on independent expenditures for parties that have first made coordinated expenditures, I find it unnecessary to reach the constitutionality of the two remaining provisions of Section 213: the provision treating all committees of a party as a single committee, and the provision forbidding transfers and receipt of funds between committees of a political party. Even if I found those provisions to be constitutional, I do not believe -- after finding the ban on independent expenditures unconstitutional -- that a saving construction can be applied to Section 213 so that it is consistent with Congress's clear intent to provide a choice to parties between unlimited independent expenditures and coordinated expenditures up to the Party Expenditure Provision limit.*fn134 Applying a saving construction to the statute that would allow only one option, but not the other, would flout that intent. Therefore, for the reasons set forth above, Section 213 of BCRA is, in our judgment, unconstitutional.

III. Section 318: Prohibition of Donations by Minors

I concur in the judgment of Judge Henderson and Judge Kollar-Kotelly that Section 318, which prohibits children from making contributions to candidates or donations to political parties,*fn135 is unconstitutional. I write, however, only to explain why I believe this Court should evaluate this provision under the strict-scrutiny standard of review.

If Section 318 were a complete ban on adults, instead of on minors, Supreme Court precedent would require us to review such a provision under the strictest of scrutiny. Indeed, one of the primary reasons that the Supreme Court reviewed contribution limitations in Buckley under the less-demanding closely-drawn standard of review was because contribution limitations permit some expression of support. 424 U.S. at 20-21. A complete ban on donations, on the other hand, prevents even a symbolic expression of support for a candidate or a party's agenda. Id. Similarly, Section 318 merits heightened scrutiny regardless of the age of those it silences. Simply stated, the standard of review applied by this Court should be a function of the constitutional rights being infringed upon, rather than the age of those whose rights are being infringed.

Children, like adults, are protected by the Constitution.*fn136 The Supreme Court, however, has made clear that minors in certain circumstances should not be treated in the same way as adults, repeatedly upholding laws that treat them differently.*fn137 Because the treatment of minors by the Supreme Court has varied across a wide range of circumstances, lower courts have had to wrestle with some uncertainty in assessing when the state's power can trump the rights of children, but not those of adults. Indeed, the Supreme Court itself acknowledged that uncertainty:"The question of the extent of state power to regulate conduct of minors not constitutionally regulable when committed by adults is a vexing one, perhaps not susceptible of precise answer." Carey v. Population Services International, 431 U.S. 678, 692 (1977). Two years later the Supreme Court shed greater light on the matter in Bellotti v. Baird, when it identified certain reasons that justify the disparate treatment:

We have recognized three reasons justifying the conclusion that the constitutional rights of children cannot equated with those of adults: the peculiar vulnerability of children; their inability to make critical decision in an informed, mature manner; and the importance of the parental role in child rearing. 443 U.S. at 634.

While these reasons provide clearer guidance on when children can be treated differently from adults, they do not, however, explain whether a court, having found one of those reasons present, should employ a lesser standard of review in evaluating children's constitutional rights or merely rely upon one of the reasons as a factor in recognizing a government interest justifying the regulation. In this jurisdiction, the D.C. Circuit seemingly answered that question for certain circumstances.

In Hutchins v. D.C., where the D.C. Circuit upheld the District of Columbia's imposition of a curfew on minors, 188 F.3d 531, 541 (1999) (en banc), the Circuit Court, in applying the Baird reasons, concluded that children may be more"vulnerable to harm during curfew hours than adults,""are less able to make mature decisions in the face of peer pressure, and"are more in need of parental supervision during curfew hours." Id. For those reasons, it decided to review the curfew statute under intermediate scrutiny and found the curfew was substantially related to the achievement of important government interests. Id. Here, however, I do not believe that we should lower the level of scrutiny. In Hutchins, and other cases in which courts applied a lesser standard of review, the courts identified paternalistic state interests in protecting children from harm, or acting in loco parentis. The D.C. Circuit, in Hutchins, was particularly concerned that children are"vulnerable to harm during curfew hours." Id. And in cases concerning decisions by children that affect procreation -- cases in which the Supreme Court has applied a lesser standard of review -- the Court has been equally concerned that children may make decisions that harm themselves.*fn138

Section 318 of BCRA was not drafted out of a concern to safeguard children against harming themselves in the absence of their parents (i.e., the state acting in loco parentis ), but to prevent parents from using their children to circumvent existing donation limitations to political parties and politicians. Because it involves political expression, this case is not unlike Tinker v. Des Moines Independent Community School District, in which the Supreme Court rejected a school district's attempt to discipline students wearing black armbands to express their opposition to the Vietnam War. 393 U.S. 503 (1969). Even in a case where the students were in the custody of the state, the Supreme Court recognized that the speech caused no harm to, nor interrupted the learning of, the state's charges. Id. at 508-09. It emphasized that"[s]tudents in school as well as out of school are'persons' under our Constitution. They are possessed of fundamental rights which the State must respect." Id. at 511. Such cases involving government interests, unrelated to the protection of children, do not warrant review under any scrutiny less than strict. The political expression of children, especially if they are old enough to work and pay taxes, surely deserves the same level of constitutional scrutiny as a young adult old enough to attend college and vote.

This does not mean, of course, that the government need entirely ignore the fact that some children do not possess"that full capacity for individual choice which is the presupposition of First Amendment guarantees," Ginsberg, 390 U.S. at 649-50, and that they sometimes cannot make"critical decisions in an informed, mature manner," Baird, 443 U.S. at 634. It simply means that the government, and the courts, can take into account these concerns when evaluating whether there is in fact a compelling government interest. Here, for example, the government could claim that children under thirteen who generally have no means independent of their parents and rarely make financial decisions independent of their parents, are vulnerable to being represented by their parents as having made donations which, in effect, circumvent contribution limits. If there were in fact more substantial evidence of such circumvention, see J. Kollar-Kotelly Op. at Part III.III, and the provision were more narrowly tailored to serve the circumvention concern, see J. Henderson Op. at Part IV.E, then restrictions, which are somewhat more onerous on children than on adults given a child's lack of independence, may be justifiable based on a compelling government interest. That is not the case here. For these reasons and for those set forth, in part, by Judge Kollar-Kotelly and Judge Henderson, I agree that the prohibition on donations by children is unconstitutional.

IV. Section 504: Public Access to Broadcasting Records

Section 504 requires broadcast licensees to collect and disclose records of any"request to purchase broadcast time" that"is made by or on behalf of a legally qualified candidate for public office" or that relates"to any political matter of national importance," including communications relating to"a legally qualified candidate,""any election to Federal office," and"a national legislative issue of public importance." BCRA § 504; FECA § 315(e)(1); 47 U.S.C. § 315(e)(1).*fn139 Since"compelled disclosure has the potential for substantially infringing the exercise of First Amendment rights," Buckley, 424 U.S. at 66, the Supreme Court insisted that any law compelling disclosure of campaign information must be reviewed under"exacting scrutiny," id. at 64, such that the government interest served is"substantial," id. at 80, and"sufficiently important to outweigh the possibility of infringement," id. at 66. For the following reasons, I find that the record does not establish the existence of a substantial governmental interest necessary to warrant the disclosure requirements set forth in Section 504.

Section 504 is different from the other disclosure provisions in BCRA in two important ways: (1) broadcast licensees, not the purchasers, are required to make the disclosures; and (2) the disclosures are required for broadcasts on"political matters of national importance" as well as on federal candidates. In relation to disclosure of communications by, and relating to, federal candidates, the government has provided no evidence that Section 504 serves any of the government interests specified in Buckley. *fn140

Indeed, Section 201, which this Court finds constitutional in part, already requires purchasers of"electioneering communications" to disclose a wide array of information, including the amount of each disbursement and the elections to which the electioneering communications pertain. BCRA § 201(a); FECA § 304(f)(2); 2 U.S.C. § 434(f)(2). These disclosure requirements for purchasers in Section 201 mirror many of the provisions for broadcast licensees in Section 504, and the defendants have provided no evidence that such a belt-and-suspender approach is necessary. Moreover, they have provided no evidence suggesting, let alone proving, that purchasers have evaded, or will evade, disclosure requirements. And though the defendants made an offhanded reference to the importance of ensuring that broadcast requests are processed in an"even-handed fashion," Gov't Opp'n Br. at 134-35, there is nothing in the record which demonstrates that broadcast licensees have treated purchasers unfairly.

As to noncandidate-focused communications which supposedly relate to"political matters of national importance" (e.g.,"a national legislative issue of public importance"),*fn141 disclosure requests are even more difficult to justify. First, such disclosure requirements fail to serve any of the three sufficient government interests set forth in Buckley because those interests are specific to evaluating, and preventing corruption of, federal candidates. See supra note 135. Second, the defendants' contention that there is a substantial government interest in helping the public identify the sponsors of political broadcasts and evaluate the credibility of the political message, see Gov't Reply Br. at 89-90, does not quite square with the holding in Buckley v. American Constitutional Law Foundation, Inc., 525 U.S. 182 (1999). In American Constitutional Law Foundation, the Supreme Court did recognize a substantial government interest in disclosing the names and addresses of ballot-initiative proponents who pay circulators to collect signatures, explaining that disclosure was"a control or check on domination of the initiative process by affluent special interest groups." Id. at 202. Here, however, the defendants have provided no evidence that"special interest groups" have, as yet, dominated or co-opted broadcast communications relating to political issues of national importance. See id. at 203 (refusing to require disclosure of petition circulators' names and amounts paid because the"lower courts fairly determined from the record as a whole" that the benefit of such information had not been demonstrated).

Absent such evidence, the government lacks a constitutionally acceptable justification to enact a disclosure provision that imposes an onerous collection and disclosure system on broadcast licensees; infringes the associational rights of groups and their members who engage in broadcasting; and potentially curtails political speech invaluable to an informed electorate. In Buckley v. Valeo, the D.C. Circuit rejected FECA Section 437a, a disclosure requirement that also encompassed"completely nonpartisan public discussion of issues of public importance." See Buckley v. Valeo, 519 F.2d 821, 870, 869-78 (D.C. Cir. 1975).*fn142 In so doing, the court explained:

[I]ssue discussions unwedded to the cause of a particular candidate hardly threaten the purity of elections. Moreover, and very importantly, such discussions are vital and indispensable to a free society and an informed electorate. Thus the interest of a group engaging in nonpartisan discussion ascends to a high plane, while the governmental interest in disclosure correspondingly diminishes. Id. at 873.

I could not agree more. In the absence of evidence that disclosure of nonelection-related broadcasts serves a substantial government interest, Section 504 cannot withstand scrutiny.

V. Findings of Fact

Despite our best efforts to produce a complete set of Findings of Fact, in which two or more members of this Court concur, we were unable to do so.

Because of the critical importance of the facts to the holdings in the case, and the sheer enormity of the record developed by the parties, I believe it necessary to set out, for the most part,*fn143 those facts which in my judgment are sufficiently relevant and probative to rely upon in reaching my conclusions. Accordingly, while there may be other relevant and probative facts in the record, I do not accord them sufficient weight to warrant either relying on them in my judgments, or including them in my Findings.

Moreover, in some instances where evidence was submitted by one side or another (or both) on a particular point, but in my judgment that evidence did not meet this standard, I have either commented specifically to that effect, or simply stated that there was"no probative evidence in the record" as to that point. In other instances, where I accorded great weight to the relevant and probative evidence submitted on a particular point, I have specifically acknowledged that that evidence was"substantial" or"overwhelming" as to that point. In those limited circumstances where the evidence submitted by one side was not controverted by the other, I have acknowledged the"uncontroverted" nature of the evidence on that point.

Finally, I would be remiss to not acknowledge the herculean effort expended by the parties in assembling this"elephantine"*fn144 record over a seven-month time period. It is a testament, indeed, to their sense of professionalism and duty on a matter of utmost importance to our system of government. Moreover, the job of reviewing and evaluating this record would have been substantially more difficult, and less reliable, in my judgment, if they had not assembled these factual materials with such extraordinary care.

That said, upon reviewing the record as a whole, I make the following Findings of Fact:

Table of Contents

I. Title I: Restrictions on Nonfederal Funds

A. National Parties and Their Congressional Campaign Committees

B. National Party Nonfederal Fundraising and Spending........... 120

C. Senator McConnell

D. Thomas McInerney

E. The Role of Political Parties in Democracies

F. Republican National Committee

G. The RNC's Federal Election Activities

H. National Party Use of Nonfederal Money to Benefit Federal Candidates

I. National Party Use of Nonfederal Money for Federal-Candidate Advocacy

J. The RNC's Involvement in Nonfederal Activities

K. The RNC's Involvement in"Mixed" Activities

L. The RNC's Assistance to State and Local Parties for Nonfederal and Mixed Activities

M. RNC Fundraising

N. The California Democratic Party ("CDP") and the California Republican Party ("CRP")

O. Proposition 34

P. The CDP's and CRP's Focus on State and Local Activities

Q. CDP & CRP Voter Registration Activities

R. CDP and CRP Direct Mail Activities

S. CDP and CRP Get-Out-the-Vote Activities

T. State Party"Issue Ads" that Directly Affect Federal Elections

U. CDP and CRP Cooperation with the National Parties

V. Reduction in CDP and CRP Fundraising and Voter Mobilization Efforts

W. Federal Officeholder Solicitation of Nonfederal Money

X. Political Parties and Nonprofit Groups

Y. Interest Groups Compared to Political Parties

Z. The Record Contains No Evidence of Quid Pro Quo Corruption

AA. While Some Federal Candidates Are Aware of the Identities of Party Contributors, Others Are Not

BB. There is No Probative Evidence That National Parties Use Nonfederal Donations to Induce Federal Officeholders to Support or Oppose Legislation

CC. The Relationship Between Access and Donations

1. The RNC

2. Statistical Evidence Regarding Donations and Access

3. Federal Officeholders Are Not More Likely to Meet With Nonfederal Donors Than Federal Donors

4. Lobbyists, Former Members of Congress, Business Leaders, and Donors Believe that Soft Money Donations to Parties Increase Access to Officeholders

5. Party Donation Programs Show That Increased Access Corresponds With Larger Donations

DD. Donors Make Donations to National Parties With the Expectation of Building Relationships and Receiving Access

1. Lobbyists Believe That Donations W ill Enable Donors to Establish Relationships With Officeholders

2. Current and Former Federal Officeholders Acknowledge That Donors Expect to Establish Relationships With and Obtain Access to Federal Officeholders

3. Donors Hope to Establish Relationships or Receive Access With Their Donations

EE. Effectiveness of Giving Nonfederal Donations as Opposed to Federal Donations

FF. Donors Often Times Give to Both National Parties in Order to Receive Access to Members of Both Parties

GG. Lobbying and Donations of Nonfederal Money

HH. Public Perception of Corruption

II. Title II: NonCandidate Campaign Expenditures

A. Issue Advocacy in Modern Campaigns

B. The Distinction Between Candidate-Centered Advocacy and Pure Issue Advocacy Is Not a Function of the Presence or Absence of the Buckley"Magic Words"

C. The Rise of"Issue Advocacy" Campaigns Funded by Corporate and Labor Union General Treasuries

D. Explaining the Shift Toward"Issue Advocacy"

E. Candidate-Centered"Issue Advertisements" That Do Not Contain Words of Express Advocacy Are Distinguishable ...

Buy This Entire Record For $7.95

Download the entire decision to receive the complete text, official citation,
docket number, dissents and concurrences, and footnotes for this case.

Learn more about what you receive with purchase of this case.