During his 1997 inaugural address, Vermont's Governor offered the Vermont General Assembly a moment of telling candor: "As I've said before, money does buy access and we're kidding ourselves and Vermonters if we deny it. Let us do away with the current system." The General Assembly responded by promulgating Act 64, a comprehensive campaign finance reform package. The testimony and statements made during the General Assembly's debate demonstrated that Vermont lawmakers were concerned with more than just the quid pro quo corruption that pre-occupies much of campaign finance reform. Typically, this fear of corruption has involved the danger that politicians will sell their votes for campaign funds. The Vermont debate highlighted something else that public officials can, and apparently do, offer in exchange for funds: time and access. The General Assembly, together with the State's chief executive, concluded that Vermont needed limitations governing its campaigns for state office with respect to both expenditures and contributions.

This appeal arises from a consolidated suit which brings a First Amendment challenge to key sections of Act 64. The plaintiffs have argued that Vermont's reform violates the First Amendment guarantee of free speech and association in the political realm. At the conclusion of a bench trial, the District Court enjoined the enforcement of Act 64's limitations on expenditures, gifts by non-resident contributors, and contributions by political parties to candidates. The District Court upheld all of Act 64's other contribution limitations, including limits of between $200 and $400 on contributions to candidates by individuals and political action committees, limits of $2000 on contributions to political parties and political action committees, and regulations treating coordinated expenditures by third parties as contributions to a candidate.

All parties have appealed that decision. We are therefore asked to determine whether the First Amendment rights of free speech and political association forbid each of the challenged provisions, including (1) Vermont's campaign expenditure limitations; (2) the contribution limits applied to candidates; (3) the contribution limits applied to political parties and political associations; (4) the limit on contributions by non-residents; and (5) the regulation of coordinated expenditures by political parties.

After issuance of the original opinion in this case, see Landell v. Sorrell, Nos. 00-9159(L), 00-9180(CON), 00-9231(XAP), 00-9139(XAP), and 00-9240(XAP) (2d Cir. Aug. 7, 2002) (slip op.), in which we upheld in large part both Act 64's contribution limits and its expenditure limits, plaintiffs filed a petition for rehearing in banc. We withdrew our original opinion on October 3, 2002, pending further proceedings. Landell v. Sorrell, Nos. 00-9159(L), 00-9180(CON), 00-9231(XAP), 00-9139(XAP), and 00-9240(XAP), 2002 WL 31268493 (2d Cir. Oct. 3, 2002). Having reconsidered our holding and taking serious note of the views presented during the rehearing process, we now issue this amended opinion, modifying our holding only with regard to Act 64's expenditure limits. In both instances, our colleague, Judge Winter, has dissented.

As we did in our original opinion, we hold today that the Supreme Court, in Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976) (per curiam), did not rule campaign expenditure limits to be per se unconstitutional, but left the door ajar for narrowly tailored spending limits that secure clearly identified and appropriately documented compelling governmental interests.1 In applying the narrow tailoring test, we hold that the State has established that the challenged expenditure limits are supported by its compelling interests in safeguarding Vermont's democratic process from (1) the corruptive influence of excessive and unbridled fundraising2 and (2) the effect that perpetual fundraising has on the time of candidates and elected officials. The evidence considered by the District Court and the Vermont legislature demonstrates that, absent expenditure limitations, the fundraising practices in Vermont will continue to impair the accessibility to elected officials which is essential to any democratic political system. The race for campaign funds has compelled public officials to give preferred access to contributors, essentially requiring candidates to sell their time in order to raise campaign funds. In addition, we affirm the District Court's finding that effective campaigns can be run under Act 64's limits.

Nevertheless, although we reaffirm these aspects of our original holding, we now conclude that a remand is necessary for further fact-finding on an aspect of the narrow tailoring inquiry that was not fully considered by the District Court: the crucial question of whether Act 64's expenditure limits provision was the "least restrictive means" of furthering the State's compelling anti-corruption and time-protection interests — or whether there are other less restrictive mechanisms available that might be as effective in satisfying the compelling interests established by Vermont. On remand, the District Court should also consider another question that it did not reach in its original examination of this case — whether treating related expenditures as candidate expenditures is constitutional. We therefore leave in place the District Court's injunction, while remanding for further proceedings.

As for the remaining issues regarding Act 64's contribution limitations, our decision remains the same in all material respects. We hold that all of Vermont's provisions limiting the size of contributions survive scrutiny, including the treatment of a third party's related expenditures as contributions and the application of contribution limitations to political party donations to candidates. We thus affirm the District Court's rulings on contribution limits in part, but vacate and remand for further proceedings insofar as the District Court's injunction prohibits enforcement of the political party limit. We also vacate the judgment and remand for further proceedings on (1) whether the provisions of Act 64 regulate wholly independent expenditures by political action committees ("PACs") and, if so, whether those provisions are constitutional; and (2) the constitutionality of the law's regulation of funds transferred from national political parties to state and local party entities.

Finally, we affirm the District Court's holding that the First Amendment forbids Vermont's attempt to limit campaign contributions by non-residents to no more than 25 percent of the total contributions received. Vermont has asserted no governmental interest sufficient to justify such a rule.

Due to the number of issues involved in this case, we set out the following table of contents:

BACKGROUND .......................................................................  99
             A. Act 64 ...................................................................  99
             B. Procedural History ....................................................... 102
             C. The District Court's Decision ............................................ 103
        DISCUSSION ....................................................................... 105
        
          I. Act 64's Expenditure Limitations ............................................ 106
             A. The Rule of Buckley ...................................................... 106
             B. The Requisite Level of Scrutiny .......................................... 110
             C. Compelling Interests ..................................................... 114
                1. Anti-Corruption ....................................................... 115
                2. Time Protection ....................................................... 119
                3. Conclusion: Two Compelling Interests .................................. 124
             D. Narrow Tailoring ......................................................... 125
                1. Are Vermont's time-protection and anti-corruption interests
                     advanced by campaign spending caps? ................................. 126
                2. Do spending limits at these levels allow for "effective advocacy"? .... 128
                3. Are mandatory expenditure limits the least restrictive means of
                    advancing the State's interests? ..................................... 131
                   a. Type of Regulation ................................................. 132
                   b. Basis for Spending Cap Limits ...................................... 133
             E. Conclusion: Remand for Further Findings .................................. 135
        
         II. Act 64's Contribution Limitations ........................................... 137
             A. Limitations on Contributions by Individuals to Candidates ................ 137
             B. Limitations on Contributions to and by PACs and Political Parties ........ 139
             C. The Related Expenditure Provision is Constitutional as to
                  Contributions .......................................................... 145
             D. The 25 Percent Limit on Out-of-State Donations is Unconstitutional ....... 146

CONCLUSION ....................................................................... 148

In 1997, Vermont passed a comprehensive campaign reform act known as Act 64. 1997 Vermont Campaign Finance Reform Act, codified at Vt. Stat. Ann. tit. 17, §§ 2801-2883 ("Act 64" or "the Act"). As enacted, Act 64 is a comprehensive campaign finance reform package, regulating contributions, expenditures, and disclosures related to candidates for state office in Vermont and political organizations that participate in Vermont elections. Section 2805a limits the expenditures that a candidate for office may make during a two-year election cycle. Candidates for state-wide office are restricted to varying amounts depending on the position sought, with a candidate for governor limited to $300,000, for lieutenant governor to $100,000, and other statewide offices to $45,000. See id. § 2805a(a)(1)-(3). Candidates for governor and lieutenant governor also have the option of receiving public financing for their campaigns, provided they receive a certain number and amount of "qualifying contributions." See id. §§ 2851-2856. Candidates for state senator and county office are limited to $4000 in expenditures, with state senators permitted an additional $2500 per seat in multi-seat districts. See id. § 2805a(a)(4). Candidates for state representative in single-member districts can spend no more than $2000, and those in two-member districts no more than $3000. See id. § 2805a(a)(5). Incumbent candidates may spend only 85 percent of the permitted amounts, except for incumbents of the General Assembly who may spend 90 percent. See id. § 2805a(c).

The Act also limits the size of contributions which candidates, political committees, and political parties may receive from a single source during a two-year election cycle. Candidates for state representative or local office may accept no more than $200 from a single source, political party, or political action committee. See id. § 2805(a). Slightly higher limits apply to candidates for state senate or county office ($300) and to candidates for statewide office ($400). See id. Political action committees and political parties may accept no contribution greater than $2000. See id. For the purpose of all of these contribution limits, a political party's state, county, and local branches (and national and regional affiliates of the party) count as a single unit. See id. § 2801(5).

The Act further imposes limits on the source of such contributions. Although candidates, political parties, and political action committees may accept contributions from out-of-state residents and political organizations, the sum of such amounts may not exceed 25 percent of the total contributions received. See id. § 2805(c).

Finally, the Act treats coordinated expenditures by third parties as both contributions to a candidate (subject to the applicable contribution limits) and expenditures by the candidate (counted against the candidate's permissible budget). See id. §§ 2809(a)-(b). The Act creates a rebuttable presumption that expenditures made by political parties or political action committees that recruit or endorse candidates are related expenditures if they primarily benefit six or fewer candidates. See id. § 2809(d).

The Vermont General Assembly promulgated Act 64 after extensive legislative consideration. Numerous committees considered the Act, holding over 65 hearings with more than 145 witnesses testifying. Moreover, Act 64 was the latest installment of Vermont's century-long effort to safeguard the accessibility and accountability of its elected officials.3

The General Assembly closely investigated the history of campaign financing for state races by examining campaign finance summaries for various Senate, House, and statewide races during the period 1978-1996, and reports of spending and contribution patterns in Vermont races. Members of the General Assembly analyzed the current status of Vermont's campaign finance law, including the disintegration of Vermont's voluntary expenditure limits. They also spoke with a range of experienced candidates and experts who provided testimony and data regarding the cost of campaigning, including the cost of travel, staff, materials, mailings, phone calls, and television and radio advertisements. Some of these witnesses described the widespread use of manipulative contribution devices, such as "bundling," which enable special interests to direct large quantities of money by way of individual contributions to particular candidates. Polls demonstrated that citizens held deep reservations and suspicions about the influence of money on the political system, particularly the influence of large contributions. Some witnesses provided testimony detailing the role that big donors have played in advocating or blocking particular pieces of legislation in Vermont.

The record considered by the General Assembly demonstrated how the Vermont system of unbridled expenditures has created a situation where public officials are functionally compelled to sell privileged access through the fundraising system. The Vermont legislature explained that this results in a number of related phenomena, including (1) candidates being forced to spend too much time fundraising; (2) fundraising requiring candidates to give preferred access to contributors over non-contributors; and (3) the system of increasing expenditures hindering the robust debate of issues, candidate interaction with the electorate, and public involvement and confidence in the electoral process.

The evidence adduced in those hearings also demonstrated broad and powerful support among the Vermont electorate for fundamental reform to the State's campaign financing scheme. These legislative hearings culminated in passage of the Act by an overwhelming majority and with strong bipartisan support.

Based on these hearings, reports and data, the General Assembly set forth specific findings which, in its view, indicated the need for comprehensive reform that includes contribution and expenditure limitations in Vermont electoral campaigns.

(1) Election campaigns for statewide and state legislative offices are becoming too expensive. As a result many Vermonters are financially unable to seek election to public office and candidates for statewide offices are spending inordinate amounts of time raising campaign funds.

(2) Some candidates and elected officials, particularly when time is limited, respond and give access to contributors who make large contributions in preference to those who make small or no contributions.

(3) In the context of Vermont, contributions larger than the amounts specified in this act are considered by the legislature, candidates and elected officials to be large contributions.

(4) Robust debate of issues, candidate interaction with the electorate, and public involvement and confidence in the electoral process have decreased as campaign expenditures have increased.

(5) Increasing campaign expenditures require candidates to seek and rely on a smaller number of larger contributors, often outside the state, rather than a large number of small contributors.

(6) In the context of Vermont, contributions scaled in proportion to the size of the electoral district of the office and up to the amounts specified in this act adequately allow contributors to express their opinions, level of support and their affiliations.

(7) In the context of Vermont, candidates can raise sufficient monies to fund effective campaigns from contributions no larger than the amounts specified in this act.

(8) Limiting large contributions, particularly from out-of-state political committees or corporations, and limiting campaign expenditures will encourage direct and small group contact between candidates and the electorate and will encourage the personal involvement of a large number of citizens in campaigns, both of which are crucial to public confidence and the robust debate of issues.

(9) Large contributions and large expenditures by persons or committees, other than the candidate and particularly from out-of-state political committees or corporations, reduce public confidence in the electoral process and increase the appearance that candidates and elected officials will not act in the best interests of Vermont citizens.

(10) Citizen interest, participation and confidence in the electoral process is lessened by excessively long and expensive campaigns.

(11) Public financing of campaigns, conditioned on an appropriate number of qualifying contributions, will increase citizen participation and will limit the time spent soliciting contributions, and will reduce the need of elected officials to respond to, and provide access to, contributors. As a result candidates will be freed to devote more time and energy to debate of the issues and elected officials will be able to spend more time responding to constituents and to performing their official duties.

(12) Public financing of campaigns, coupled with generally applicable contribution and expenditure limitations, will level the financial playing field among candidates and provide resources to independent candidates, both of which will increase the debate of issues and ideas.

(13) In Vermont, campaign expenditures by persons who are not candidates have been increasing and public confidence is eroded when substantial amounts of soft money are expended, particularly during the final days of a campaign.

(14) Identification of persons who publish political advertisements assists in enforcement of the contribution and expenditure limitations established by this act.

(15) Because it is essential for all candidates to have their names and positions on issues known to the electorate and because incumbents have a substantial advantage in these areas, public grants and campaign expenditures must be reduced for incumbents.

1997 Vt. Laws P.A. 64 (H. 28). On June 26, 1997, Vermont's Governor signed Act 64 into law.

The current suit was consolidated from three separate civil actions. On May 18, 1999, Marcella Landell, Donald R. Brunelle, and the Vermont Right to Life Committee, Inc., sued Vermont's Attorney General, Secretary of State and fourteen state's attorneys ("Vermont"). On August 13, 1999, Neil Randall, George Kuusela, Steve Howard, Jeffrey A. Nelson, John Patch, and the Vermont Libertarian Party also brought suit, as did the Vermont Republican State Committee on February 15, 2000. The remaining defendants, including the Vermont Public Interest Research Group, the League of Women Voters of Vermont, and numerous members of Vermont's General Assembly (collectively "Defendant-Intervenors"), successfully intervened in the consolidated action.4

Plaintiffs argued that the challenged provisions unconstitutionally infringe their First Amendment rights to free speech and political association.5 The District Court held a ten-day bench trial between May 8, 2000 and June 2, 2000. An array of former and current public office holders, private citizens, and electoral experts testified about Vermont's interest in campaign finance legislation, the history of elections and campaign finance reform in Vermont, the cost of campaigning in Vermont, and the likely effect of Act 64's challenged provisions on Vermont races, candidates and political actors. As we discuss in more detail below, the ten-day bench trial resulted in the District Court's upholding most of the challenged provisions, but striking down Act 64's expenditure limitations, its limitations on contributions by parties to candidates, and its restriction on contributions from out-of-state sources. Vermont and the other defendant-appellants timely appeal from the District Court's order holding those portions of Act 64 unconstitutional. Vermont is joined by amici, the Brennan Center for Justice at New York University School of Law and the States of Colorado, Connecticut, Maryland, New York, and Oklahoma. The plaintiffs have cross-appealed, contending that the District Court should have also enjoined the enforcement of the other disputed provisions of the Act.

After receiving post-trial submissions, the District Court issued an opinion containing its findings of fact and conclusions of law. See Landell v. Sorrell, 118 F.Supp.2d 459 (D.Vt.2000). First, the District Court held that the plaintiffs have standing to challenge the subject provisions of the Act. Id. at 475. As to the merits, although the District Court found that Vermont had generally demonstrated several compelling justifications for Act 64's comprehensive reform of the campaign finance system, the court concluded that some of Act 64's provisions violated the First Amendment. With the exception of the expenditure limitations, the District Court applied the standard of review of "exacting scrutiny," inquiring whether the provision is narrowly tailored to serve a sufficiently important governmental interest. With regard to the expenditure limits, the District Court interpreted Buckley v. Valeo, 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976) (per curiam), as forbidding such limitations per se and held that any contrary decision would violate the doctrine of stare decisis. 118 F.Supp.2d at 483. The District Court rejected the expenditure limitations despite its findings that Vermont had established several compelling interests in their favor, namely: (1) freeing office holders from the requirements of excessive fundraising so that they can perform their duties; (2) preserving faith in democracy; (3) protecting access to the political arena for those unable to access large sums of money; and (4) diminishing the importance of repetitive 30-second commercials. Id. at 482-83. Despite holding that the expenditure limitations are illegal under Buckley, the District Court did find that the expenditure limits would permit effective campaigning. Id. at 471-72.

The District Court upheld the provisions imposing limitations on amounts that individuals may contribute to political campaigns, Vt. Stat. Ann. tit. 17, §§ 2805(a)(b). The District Court found that the Vermont provision, like the statutory provision upheld in Buckley, served the governmental interest in preventing actual and perceived corruption in the political system. Id. at 476-79. As evidence of the existence of such an interest, the District Court relied on citizen polls, comments by public officials, and media accounts of citizen concern with the state of the political system, as well as direct testimony from citizens regarding their views of the political system. Id. at 465-70. The evidence indicated that the current financing scheme eroded public confidence in the democratic system and contributed to a waning public interest in elections. Id. Finally, the evidence supported the public's perception that large contributions won actual influence over the legislative process. Again, the District Court relied not only on trial testimony, but also on studies showing how the pressure to raise money made legislative initiatives less likely to succeed if contrary to the wishes of well-organized interest groups who frequently contribute to candidates. Id.

The District Court further analyzed the amounts of the contribution limitations, and held that they were narrowly tailored to serve this anti-corruption purpose. In support of the narrow-tailoring conclusion, the court relied upon the cost of previous elections in Vermont, the size of Vermont electoral districts and the corresponding cost-per-voter, the effect of the limitations on the Burlington mayoral election held after the passage of Act 64, the widely-held public view that donations in excess of the Act's limitations were suspicious, and the fact that the limitation did not inhibit "effective campaigning." Id. at 470-72, 476-80.

The District Court rejected the contention that PACs merit special treatment; it thus upheld the restrictions on contributions by and to PACs pursuant to Act 64. See Vt. Stat. Ann. tit. 17, §§ 2805(a)-(b). If contributions by individuals may be restricted, the court reasoned, then so too may gifts by individuals to associations that in turn give funds to candidates. The District Court reasoned that Vermont has the same anti-corruption interest in limiting PAC contributions as those by individuals. The contribution limit closes a loophole which individuals could exploit to evade individual contribution limitations. Id. at 488-89.

The District Court held, however, that political parties deserve greater freedom in their ability to make contributions to political candidates. Although the District Court upheld the $2000 limitation on contributions to political parties pursuant to Vt. Stat. Ann. tit. 17, § 2805(a), it struck down the provision limiting contributions to candidates insofar as it applies to those candidate's own political parties pursuant to Vt. Stat. Ann. tit. 17, § 2805(b). Id. at 486-87. Regarding contributions to political parties, the court relied on Vermont's anti-corruption interest, noting that unrestrained contributions to parties provided a loophole to individuals wishing to evade restrictions on direct contributions. Quoting Nixon v. Shrink Mo. Gov't PAC, 528 U.S. 377, 397, 120 S.Ct. 897, 145 L.Ed.2d 886 (2000)(Shrink), the District Court found that the limit imposed by the statute is not "so radical in effect as to render political association ineffective, drive the sound of [a political party's] voice below the level of notice, and render contributions pointless." 118 F.Supp.2d at 485. Moreover, the District Court found that, given Vermont's electoral situation, the $2000 limit did not inhibit the strength of political parties. The court relied on the evidence specifically concerning Vermont campaigns and politics, a comparison of limits on contributions to candidates in other jurisdictions, and the ability of the Republican Party to raise substantial sums while subject to Act 64's limitations. Id. at 484-86. The District Court, however, did not address the constitutionality of transfers of money to state and local parties from the national affiliated party, which are apparently subject to the $2000 limitation.

The District Court held that Vermont's limits on how much a political party could give to its own candidates for various state offices ($400, $300, and $200, respectively) were unconstitutionally low. Id. at 487. The court recognized that the anti-corruption interest may justify some limitations, given that corruption may "filter[] through the party machine." Id. at 486. But according to the District Court, those limitations must be balanced against the special role political parties play in the American electoral system. Without much factual discussion, the court concluded that the limits would reduce the party's voice to a whisper — since political parties speak through their candidates and the restrictions were too stringent even for the small scale of Vermont's electoral races. Id. at 487.

The District Court also upheld the treatment of state and local parties as a single entity for the purpose of calculating the contribution limitations pursuant to Vt. Stat. Ann. tit. 17, §§ 2801(5) & 2301-20. The court relied on a number of factors, including the fact that notwithstanding its adamant assertions, the defendant Vermont Republican State Committee had never acted as a loose confederation of entities in the conduct of the litigation. Id. at 487-88.

The District Court upheld the provision of Act 64 that treats third party expenditures "intentionally facilitated by, solicited by or approved by the candidate or the candidate's political committee" as contributions to the candidate pursuant to the Act. See Vt. Stat. Ann. tit. 17, § 2809(a) & (c). The purpose of the provision is to close a loophole which would otherwise permit evasion of the legitimate contribution limitations by engaging in coordinated expenditures. The District Court further upheld the provision establishing a rebuttable presumption that any third party expenditure benefiting six or fewer candidates is a related expenditure. See id. § 2809(d). The court explained that the presumption is a guideline to assist in compliance, and that since Vermont's Secretary of State has determined that the presumption is rebuttable, it does not unduly chill otherwise protected speech activity. Id. at 492. Although the District Court upheld the provision treating related expenditures as contributions to candidates, it struck down the provision treating related expenditures as expenditures by candidates. Vt. Stat. Ann. tit. 17 § 2809(a) & (b).

The District Court found unconstitutional the provision that caps out-of-state funds at 25 percent of total contributions received by a candidate, political party, or PAC pursuant to Vt. Stat. Ann. tit. 17, § 2805(c). The court found that the factual record did not establish any legitimate governmental interest in limiting such contributions. Id. at 483-84. Instead, the record only supported an inference that such contributions raise the risk of corruption when they are large — a problem solved by the contribution limits. The fact that a donor is a resident of another state is not an important factor in either increasing the risk of corruption or the public's perception of corruption. Moreover, the mechanics of the ban indicated a lack of proper tailoring because it acts as a complete bar to contributions by some would-be contributors.

Finally, the court held that, under Vermont law, the unconstitutional provisions may be severed from the rest of Act 64. Id. at 492-93.

As a threshold matter, the defendants have challenged the plaintiffs' standing to assert this facial challenge to Act 64's expenditure and contribution limitations. In order to present a "case or controversy" within the meaning of Article III of the Constitution, the plaintiffs seeking relief must have a sufficient "personal stake in the outcome of the controversy." Buckley v. Valeo, 424 U.S. 1, 11, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976) (internal quotation marks omitted). The District Court provided careful analysis demonstrating that each of the challenged provisions arguably affects the First Amendment rights of one or more of the plaintiffs. See Landell v. Sorrell, 118 F.Supp.2d 459, 474-76 (D.Vt. 2000). For the reasons set forth by the District Court, we uphold its determination that the plaintiffs have standing to assert their challenge to Act 64's expenditure and contribution limits.

Although we review the District Court's factual findings for clear error pursuant to Rule 52(a) of the Federal Rules of Civil Procedure, see Bose Corp. v. Consumers Union of U.S., Inc., 466 U.S. 485, 498, 104 S.Ct. 1949, 80 L.Ed.2d 502 (1984), the District Court's legal conclusions regarding the campaign finance reform legislation are subject to de novo review. Indeed, the breadth of review of factual issues is greater in cases raising First Amendment issues: "an appellate court has an obligation to `make an independent examination of the whole record' in order to make sure that `the judgment does not constitute a forbidden intrusion on the field of free expression.'" Id. at 499, 104 S.Ct. 1949 (quoting New York Times Co. v. Sullivan, 376 U.S. 254, 284-286, 84 S.Ct. 710, 11 L.Ed.2d 686 (1964)). The appellate court must also be vigilant for errors of law that "may infect a so-called mixed finding of law and fact, or a finding of fact that is predicated on a misunderstanding of the governing rule of law." Bose Corp., 466 U.S. at 501, 466 U.S. 485.

In reviewing campaign finance regulations, "the level of scrutiny is based on the importance of the political activity at issue to effective speech or political association." Federal Election Comm'n v. Beaumont, 539 U.S. 146, 161, 123 S.Ct. 2200, 156 L.Ed.2d 179 (2003) (internal quotation marks omitted). Campaign contributions advance political association by allowing one to affiliate with a political candidate, and "enabl[ing] like-minded persons to pool their resources in furtherance of common political goals." Buckley, 424 U.S. at 22, 96 S.Ct. 612. However, restrictions on contributions have been treated as merely "marginal" speech restrictions because contributions "lie closer to the edges than to the core of political expression." Beaumont, 539 U.S. at 161, 123 S.Ct. 2200. As a result, contribution limits pass muster if they are "closely drawn to match a sufficiently important interest." Id. at 162, 123 S.Ct. 2200 (citations and internal quotation marks omitted). And, as the Supreme Court recently observed, its cases "have made clear that the prevention of corruption or its appearance constitutes a sufficiently important interest to justify political contribution limits." McConnell, 540 U.S. at ___, 124 S.Ct. at 660; see also id. at ___, 124 S.Ct. at 657 n. 40 (explaining that since Buckley, the Court has "consistently applied less rigorous scrutiny to contribution restrictions aimed at the prevention of corruption and the appearance of corruption") (collecting cases).

However, "limits on political expenditures deserve closer scrutiny than restrictions on political contributions." Federal Election Comm'n v. Colorado Republican Federal Campaign Comm., 533 U.S. 431, 440, 121 S.Ct. 2351, 150 L.Ed.2d 461 (2001) (Colorado Republican II); see also McConnell, 540 U.S. at ___, 124 S.Ct. at 655. The Supreme Court has treated limits on campaign spending as a direct restraint on speech, and thus, expenditure limits must be narrowly tailored to serve a compelling state interest. See Austin v. Michigan Chamber of Commerce, 494 U.S. 652, 657, 110 S.Ct. 1391, 108 L.Ed.2d 652 (1990) (addressing corporate expenditures).

With these standards in mind, we review each of the challenged provisions in turn.

Buckley v. Valeo remains the seminal case governing the constitutional review of campaign finance reform efforts, including expenditure limitations. 424 U.S. 1, 96 S.Ct. 612, 46 L.Ed.2d 659 (1976). The Buckley Court considered and rejected a variety of expenditure limitations, including a ceiling on independent, campaign-related expenditures, a ceiling on a candidate's use of personal or family resources, and a ceiling on a candidate's campaign expenditures. Like the federal statute reviewed in Buckley, Act 64 limits the total amount of campaign funds that a candidate may spend.

Although the clear language of Buckley requires that courts should review expenditure limits with exacting scrutiny, the District Court in this case (and it is by no means alone) apparently felt that Buckley categorically prohibits expenditure limitations. See, e.g., Homans v. City of Albuquerque, 366 F.3d 900, 914-21 (10th Cir.2004); Kruse v. City of Cincinnati, 142 F.3d 907, 918-19 (6th Cir.), cert. denied 525 U.S. 1001, 119 S.Ct. 511, 142 L.Ed.2d 424 (1998); see also post at 151, 152, 159, 172, 185 (Winter, dissenting). We disagree. The Buckley Court's rejection of particular federal campaign expenditure limitations was rooted in Congress' purported reasons for such legislation and the failures of those interests to demonstrate any need for expenditure limits. 424 U.S. at 55-58, 96 S.Ct. 612. Ultimately, the Court concluded that the federal government had failed to assert any sufficiently important interest that its expenditure limitations served. See id. at 55, 96 S.Ct. 612.

Examining the federal government's interest in eliminating corruption from federal elections, the Buckley Court concluded that the government's asserted rationale only applied to large contributions — that is, eliminating large contributions fully satisfied the government's anti-corruption interest. See id. at 56-57, 96 S.Ct. 612. The federal government claimed that expenditure limitations were necessary to make contribution limitations easier to enforce, arguing that when candidates cannot spend large quantities of money, they have a weaker incentive to accept illegally large contributions. But the Court concluded that the contribution limitations promised to be sufficiently effective on their own. See id. Based on the Court's review of the record, "[t]here [was] no indication that the substantial criminal penalties" attached to violations of contribution limits, as well as the "political repercussion of such violations," would not suffice to realize this anti-corruption interest. Id.

Nor was the Court persuaded that the federal government had a sufficient interest in utilizing expenditure limitations to equalize the financial resources of candidates competing for office. See id. at 56-57, 96 S.Ct. 612. The contribution limits would assure that any difference in resources "var[ies] with the size and intensity of the candidate's support." Id. at 56, 96 S.Ct. 612. Finally, the Court addressed the argument that expenditure limitations served the federal government's interest "in reducing the allegedly skyrocketing costs of political campaigns." Id. at 57, 96 S.Ct. 612. The Court rejected the idea that the state had a sufficient interest in setting the appropriate scope of the "quantity and range of debate on public issues in a political campaign." Id. In other words, Buckley held that large campaign expenditures, in and of themselves, are not inherently suspect.

We conclude, then, that Vermont cannot sustain Act 64 by asserting a need to control excessive campaign spending per se. But critically, the Buckley Court did not conclude that the Constitution would always prohibit expenditure limits, regardless of the reasons asserted and the record supporting the limitations. It simply held that based on the record before it, "[n]o governmental interest that has been suggested is sufficient to justify" the federal expenditure limits. Id. at 55, 96 S.Ct. 612. Accordingly, after Buckley, there remains the possibility that a legislature could identify a sufficiently strong interest, and develop a supporting record, such that some expenditure limits could survive constitutional review.

We are not alone in concluding that Buckley does not permanently foreclose any consideration of campaign expenditure limitations. In Shrink, Justices Breyer, Ginsburg and Stevens all recognized that our post-Buckley experiences with campaign finance have demonstrated that we need a flexible approach to the constitutional review of campaign finance rules. Justice Breyer, who was joined by Justice Ginsburg, concluded that courts must resist a static reading of Buckley's mandate, which may require reinterpretation in light of subsequent experience, including a legislature's "political judgment that unlimited spending threatens the integrity of the electoral process." 528 U.S. at 403-04, 120 S.Ct. 897 (Breyer, J., concurring). Legislatures may protect the electoral process not only from quid pro quo corruption, but also from the threat that campaign funding may pose to the "integrity of the electoral process." Id. at 401, 120 S.Ct. 897. Justice Stevens also articulated the need for "a fresh reexamination" of Buckley, and concluded that "Money is property; it is not speech." Id. at 398, 120 S.Ct. 897 (Stevens, J., concurring). And although Justice Kennedy argued from a different perspective that the post-Buckley experience requires a wholesale abandonment of the approach adopted in Buckley, he too left open the possibility that "Congress, or a state legislature, might devise a system in which there are some limits on both expenditures and contributions thus permitting officeholders to concentrate their time and effort on official duties rather than on fundraising." Shrink, 528 U.S. at 409, 120 S.Ct. 897 (Kennedy, J., dissenting); see also McConnell, 540 U.S. at ___, 124 S.Ct. at 745 (Kennedy, J., concurring in part and dissenting in part) (indicating by implication that Buckley did not make expenditure limits per se invalid).6

Indeed, some judges have noted that reconsideration might be required were a court faced with compelling evidence that unlimited expenditures posed great dangers to the very political process that Buckley sought to safeguard. Justices Stevens and Ginsburg have supported the constitutionality of spending limits on political parties for, among other reasons, the likelihood that such limits would improve, rather than inhibit, a flourishing political system:

It is quite wrong to assume that the net effect of limits on contributions and expenditures — which tend to protect equal access to the political arena, to free candidates and their staffs from the interminable burden of fundraising, and to diminish the importance of repetitive 30-second commercials — will be adverse to the interest in informed debate protected by the First Amendment.

Colorado Republican Federal Campaign Comm. v. Federal Election Comm'n, 518 U.S. 604, 649-50, 116 S.Ct. 2309, 135 L.Ed.2d 795 (1996) (Colorado Republican I) (Stevens, J., dissenting). In part, they reached this conclusion because of the comparative competency of the different branches of government: "Congress surely has both wisdom and experience in these matters that is far superior to ours." Id. at 650, 116 S.Ct. 2309. Moreover, one judge sitting on the Sixth Circuit has pointed out that Buckley was "decided on a slender factual record" and that a fuller record might satisfy the constitutional requirement that expenditure limits be narrowly tailored to a compelling interest. Kruse, 142 F.3d at 919 (Cohn, J., concurring); cf. LAURENCE H. TRIBE, AMERICAN CONSTITUTIONAL LAW § 13-27, at 1133 n.1 (2d ed. 1988) ("One consequence of th[e] expedited review [in Buckley] was that the Supreme Court, working in a factual vacuum, was forced to indulge in more than a little empirical speculation about such issues as the circumvention of expenditure limits and the impact of those limits on campaign speech."); Burt Neuborne, One Dollar-One Vote: A Preface to Debating Campaign Finance Reform, 37 WASHBURN L.J. 1, 30 (1997) ("Since the Buckley Court's judgment was made without the benefit of a factual record, critics have argued that it is time for a factually based study of the potential for corruption inherent in large, independent expenditures."); David R. Lagasse, Note, Undue Influence: Corporate Political Speech, Power and the Initiative Process, 61 BROOK. L. REV. 1347, 1357 (1995) ("The Supreme Court granted certiorari in Buckley v. Valeo without either party to the action having the opportunity to develop a strong factual record on which the Court could base its ultimate decision. Thus, the Court faced the issue of Congress's power to regulate campaign expenditures purely on theoretical grounds, without the benefit of developing an adequate factual record.") (citing BOB WOODWARD & SCOTT ARMSTRONG, THE BRETHREN 469-70 (1979)).

The academic literature also contains persuasive analyses that our post-Buckley understanding of campaign finance requires a careful evaluation of the evidence in support of expenditure limits. See, e.g., Richard Briffault, Nixon v. Shrink Missouri Government PAC: The Beginning of the End of the Buckley Era?, 85 MINN. L. REV. 1729, 1765-69 (2001) (arguing that fair and competitive elections may require some form of expenditure limitations); Vincent Blasi, Free Speech and the Widening Gyre of Fund-Raising: Why Campaign Spending Limits May Not Violate the First Amendment After All, 94 COLUM. L. REV. 1281, 1288-89 (1994) (noting that changed circumstances and never-before considered governmental interests, including the protection of candidates' time, might be sufficiently compelling to support expenditure limits).

Although we recognize that there is considerable dissatisfaction with Buckley's approach, we still premise our conclusions on the assumption that Buckley continues to govern the constitutional review of campaign finance laws. However, we do not accept an unyielding interpretation of Buckley that expenditure limits are per se unconstitutional, because such a static approach to Buckley's import would require us to ignore not only Buckley's own language, but also over three decades of experience as to how the campaign funds race has affected public confidence and representative democracy.7 In sum, like the federal expenditure limitations considered in Buckley, Act 64's expenditure limitations rise or fall on whether they have been narrowly tailored to a compelling governmental interest. It is to that question that we now turn.

As a regulation of the amount that a candidate can spend on speech made "for the purpose of influencing an election," Vermont's expenditure limits are a content-based restriction on speech. See Burson v. Freeman, 504 U.S. 191, 197, 112 S.Ct. 1846, 119 L.Ed.2d 5 (1992) (treating election provision as content-based because "whether individuals may exercise their free speech rights ... depends entirely on whether their speech is related to a political campaign").8 "Content-based regulations are presumptively invalid," R.A.V. v. St. Paul, 505 U.S. 377, 382, 112 S.Ct. 2538, 120 L.Ed.2d 305 (1992), and the government bears the burden of rebutting that presumption. United States v. Playboy Entertainment Group, Inc., 529 U.S. 803, 817, 120 S.Ct. 1878, 146 L.Ed.2d 865 (2000). To uphold a content-based restriction on speech, the government must prove the existence of a compelling state interest to support the restriction, and that the restriction is narrowly tailored to advance that interest.

In the context of expenditure limits, then, the level of scrutiny applied is akin to the "strict scrutiny" standard frequently employed in the equal protection context, in terms of the required degree of "fit" between means and ends. Cf. Guido Calabresi, Antidiscrimination and Constitutional Accountability (What the Bork-Brennan Debate Ignores), 105 HARV. L. Rev. 80, 112-13 n.94 (citing cases) (noting that, traditionally, judicial review has been at its strongest in protecting against infringement on First Amendment rights). Our application of this standard is informed both by the particular First Amendment right implicated by the challenged restrictions, as well as by the degree of deference owed to the supporting legislative findings.

Turning to the first of these factors, there is no doubt that "[p]olitical speech is the primary object of First Amendment protection." Shrink, 528 U.S. at 410-11, 120 S.Ct. 897 (Thomas, J., dissenting). Moreover, in our representative democracy, the free exchange of political information "should receive the most protection when it matters the most — during campaigns for elective office." Id. at 411, 120 S.Ct. 897. However, the precise object of First Amendment protection in this case, for most plaintiffs, is the ability to spend money on political speech — not the speech itself. See Shrink, 528 U.S. at 400, 120 S.Ct. 897 (Breyer, J., concurring) (money is not speech; it "enables speech"). To be sure, the Supreme Court has consistently held that the expenditure of money is so critical in enabling political speech in today's mass society, that it should receive the same First Amendment protection as the speech itself. We do not question this proposition — and indeed apply it in this case — but, particularly in light of at least one Supreme Court Justice's willingness to rethink the money equals speech equation (J. Stevens concurring in Shrink, 528 U.S. at 398, 120 S.Ct. 897), think it important to define the protected interest as precisely as possible.

Although most of the plaintiffs are persons or organizations that want to spend money on speech, plaintiff Marcella Landell is a voter who wants to receive political speech. Her First Amendment right to receive such speech is the equivalent of the right of the speakers. See Virginia State Board of Pharmacy v. Virginia Citizens Consumer Council, Inc., 425 U.S. 748, 756, 96 S.Ct. 1817, 48 L.Ed.2d 346 (1976) (the First Amendment protection afforded is to "the communication, to its source and to its recipients both"). As Landell describes her interest in her brief, she "does not wish her ability to cast a wise and informed vote to be restricted by the State of Vermont imposing a direct barrier on the amount of candidate speech she may receive." Restrictions of political speech that "hamstring[ ] voters seeking to inform themselves about the candidates and the campaign issues" are unconstitutional. Eu v. San Francisco County Democratic Central Comm., 489 U.S. 214, 223, 109 S.Ct. 1013, 103 L.Ed.2d 271 (1989). Plaintiffs argue that this high level of protection, as applied in Buckley, dictates that the expenditure limit provision must automatically be struck down.

On the other hand, Vermont appears to argue that deference to the legislature — on whether the interests asserted in favor of expenditure limits are compelling, and whether expenditure limits are necessary to achieve these goals — is warranted. Vermont cites several Supreme Court cases in support of its view of legislative deference, including Federal Election Comm'n v. National Right to Work Comm., 459 U.S. 197, 210, 103 S.Ct. 552, 74 L.Ed.2d 364 (1982) (it is improper to "second-guess a legislative determination as to the need for prophylactic measures where corruption is the evil feared"); Turner Broadcasting System, Inc. v. FCC, 512 U.S. 622, 665, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994) (Turner I) ("courts must accord substantial deference to the predictive judgments" of the legislature); Turner Broadcasting System, Inc. v. FCC, 520 U.S. 180, 196, 117 S.Ct. 1174, 137 L.Ed.2d 369 (1997) (Turner II) ("We owe Congress' findings an additional measure of deference out of respect for its authority to exercise the legislative power."); and Walters v. National Association of Radiation Survivors, 473 U.S. 305, 330-31 n. 12, 105 S.Ct. 3180, 87 L.Ed.2d 220 (1985) (Congress' factual findings are entitled to "a great deal of deference, inasmuch as Congress is an institution better equipped to amass and evaluate the vast amounts of data bearing on" an issue). Indeed, the District Court concluded that "[a]lthough legislative findings are not entirely isolated from review," it was "required to exercise considerable deference to such findings." 118 F.Supp.2d at 476 (citing Turner II). Accordingly, the court adopted the fifteen official findings excerpted supra and in the District Court opinion, but made clear that it was also considering the other evidence presented at trial. Id. at 468-74.

As to plaintiffs' position, we disagree that the high level of protection accorded political speech or the money enabling it dictates that the provision must automatically be struck down. Cf. McConnell, 540 U.S. at ___, 124 S.Ct. at 706 ("Many years ago we observed that `[t]o say that Congress is without power to pass appropriate legislation to safeguard ... an election from the improper use of money to influence the result is to deny to the nation in a vital particular the power of self protection.'") (quoting Burroughs v. United States, 290 U.S. 534, 545, 54 S.Ct. 287, 78 L.Ed. 484 (1934)); Storer v. Brown, 415 U.S. 724, 729-30, 94 S.Ct. 1274, 39 L.Ed.2d 714 (1974) (compelling interest in the integrity and stability of the election process means that "every substantial restriction on the right to vote or to associate" should not automatically be invalidated). In our view, this level of protection is the starting point, not the endpoint, for scrutiny of Vermont's expenditure limits.

Indeed, the Supreme Court has been clear in its rejection of the view that "strict scrutiny is `strict in theory, but fatal in fact.'" Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 237, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995) (quoting Fullilove v. Klutznick, 448 U.S. 448, 519, 100 S.Ct. 2758, 65 L.Ed.2d 902 (1980) (Marshall, J., concurring)) (explaining that "[w]hen race-based action is necessary to further a compelling interest, such action is within constitutional constraints if it satisfies the `narrow tailoring' test this Court has set out in previous cases"). This observation has proven true in the First Amendment context, as the Supreme Court has validated a number of electoral regulations against First Amendment challenge even while applying strict scrutiny. See, e.g., Burson v. Freeman, 504 U.S. 191, 112 S.Ct. 1846, 119 L.Ed.2d 5 (1992) (plurality opinion) (upholding state ban on electioneering activity near polling places); Austin v. Michigan Chamber of Commerce, 494 U.S. 652, 110 S.Ct. 1391, 108 L.Ed.2d 652 (1990) (upholding statute restricting independent expenditures by corporations on campaigns). Careful analysis is particularly important in applying strict scrutiny, where, as Justice Breyer has put it, "a law significantly implicates competing constitutionally protected interests in complex ways." Shrink, 528 U.S. at 402, 120 S.Ct. 897 (Breyer, J., concurring). As we will explain, this is a case where "constitutionally protected interests lie on both sides of the legal equation," preventing a simple equation of strict scrutiny with constitutional infirmity. Id. at 400, 120 S.Ct. 897; see also, e.g., Burson, 504 U.S. at 199, 112 S.Ct. 1846 (plurality opinion) (recognizing compelling interest in preserving integrity of electoral process); id. at 213, 112 S.Ct. 1846 (Kennedy, J., concurring) ("[T]here is a narrow area in which the First Amendment permits freedom of expression to yield to the extent necessary for the accommodation of another constitutional right."); Storer, 415 U.S. at 736, 94 S.Ct. 1274 (allowing some restrictions on ballot access in order to further the "State's interest in the stability of its political system").

Nor should we adopt total legislative deference as the appropriate level of scrutiny. Deference to legislative findings may well be warranted on certain issues relating to the constitutionality of election-related laws, such as the precise level of contribution limits, as in Buckley, 424 U.S. at 30, 96 S.Ct. 612, or whether 100 feet, as opposed to 50 or 75 feet, is an adequate radius surrounding a polling place to ban electioneering, as in Burson, 504 U.S. at 209-10, 112 S.Ct. 1846. Some degree of deference on the issue of whether there are state interests that justify legislative changes to the State's electoral system may also be appropriate. See, e.g., Federal Election Comm'n v. Beaumont, 539 U.S. 146, 155, 123 S.Ct. 2200, 156 L.Ed.2d 179 (2003) ("[D]eference to legislative choice is warranted particularly when Congress regulates campaign contributions, carrying as they do a plain threat to political integrity and a plain warrant to counter the appearance and reality of corruption and the misuse of corporate advantages."). But total deference is not warranted on the core questions of whether those interests are truly compelling enough, in a constitutional sense, to justify the expenditure limits, and whether this regulation places an undue burden on the First Amendment rights of those who bring this challenge. See, e.g., Metromedia, Inc. v. City of San Diego, 453 U.S. 490, 519, 101 S.Ct. 2882, 69 L.Ed.2d 800 (1981) (plurality opinion) ("[I]t has been this Court's consistent position that democracy stands on a stronger footing when courts protect First Amendment interests against legislative intrusion, rather than deferring to merely rational legislative judgment in this area."); Schneider v. State, 308 U.S. 147, 161, 60 S.Ct. 146, 84 L.Ed. 155 (1939) ("This court has characterized the freedom of speech and that of the press as fundamental personal rights and liberties.... [T]he delicate and difficult task falls upon the courts ... to appraise the substantiality of the reasons advanced in support of the regulation of the free enjoyment of the rights.").

We read the District Court opinion as consistent with this view. It gives "considerable deference" to the legislative findings on the need for the law only, but not to the legislature's assessment of whether its solution is narrowly tailored. Cf. Regents of University of California v. Bakke, 438 U.S. 265, 299, 98 S.Ct. 2733, 57 L.Ed.2d 750 (1978) (Powell, J.) ("Political judgments regarding the necessity for the particular classification may be weighed in the constitutional balance, but the standard of justification will remain constant"), quoted in Adarand Constructors, Inc. v. Pena, 515 U.S. 200, 224-25, 115 S.Ct. 2097, 132 L.Ed.2d 158 (1995). This approach is consistent with Justice Breyer's concurrence in Shrink, where he indicated that the Court should "defer to [the Missouri legislature's] political judgment that unlimited spending threatens the integrity of the electoral process," but not with respect to whether "its solution, by imposing too low a contribution limit, significantly increases the reputation-related or media-related advantages of incumbency and thereby insulates legislators from effective electoral challenge." 528 U.S. at 403-04, 120 S.Ct. 897.

Although we bear in mind Justice Breyer's observations in Shrink, we cannot adopt his conclusion, in light of the extensive Supreme Court precedent to the contrary, that the interests must be balanced here, and that there is therefore "no place" for a "strong presumption against constitutionality of the sort often thought to accompany the words `strict scrutiny.'" Id. at 400, 120 S.Ct. 897. Such a presumption is proper, at least until the Supreme Court tells us otherwise, and it means that the burden of persuasion at trial was on the State to defend Act 64 — i.e., to establish that there was a compelling state interest to support the expenditure limit provision and that the provision was narrowly tailored to advance that interest. See Burson, 504 U.S. at 226, 112 S.Ct. 1846 (Stevens, J., dissenting, joined by O'Connor and Souter) (noting that "a core premise of strict scrutiny" is that "the heavy burden of justification is on the State"). But this burden does not excuse the courts from actually applying the scrutiny that the First Amendment demands, and the State of Vermont deserves.

Therefore, although we do not question the validity of the factual findings developed by the legislature in support of Act 64,9 our system of judicial review provides plaintiffs the opportunity to present competing evidence, assigns to the District Court the responsibility for making findings of fact and conclusions of law after weighing the evidence, and leaves to the Court of Appeals the independent responsibility to assess the legal significance of these factual findings. This responsibility is particularly important here, where, as plaintiffs claim, complete deference to the legislature could "risk such constitutional evils as permitting incumbents to insulate themselves from effective electoral challenge." Shrink, 528 U.S. at 402, 120 S.Ct. 897 (Breyer, J., concurring).

Put differently, this level of scrutiny is a serious barrier for expenditure limits, but it is not impenetrable. Rather, an independent court must be convinced that the legislature was serving the people's interest and not its own. See, e.g., Burson, 504 U.S. at 213, 112 S.Ct. 1846 (Kennedy, J., concurring) (discussing the use of the compelling-interest test as "one analytical device to detect, in an objective way, whether the asserted justification is in fact an accurate description of the purpose and effect of the law"), quoted in R.A.V. v. City of St. Paul, 505 U.S. 377, 395, 112 S.Ct. 2538, 120 L.Ed.2d 305 (1992); Cf. Croson, 488 U.S. at 493, 109 S.Ct. 706 (noting in the equal protection context that "the purpose of strict scrutiny is to `smoke out' illegitimate uses of race by assuring that the legislative body is pursuing a goal important enough to warrant use of a highly suspect tool," with the narrow tailoring analysis helping to ensure that there is "little or no possibility that the motive for the classification was illegitimate").

In Shrink, the Court indicated that "[t]he quantum of empirical evidence needed to satisfy heightened judicial scrutiny of legislative judgments will vary up or down with the novelty and plausibility of the justification raised." 528 U.S. at 391, 120 S.Ct. 897. For example, the Court in Shrink accepted a relatively minimal evidentiary showing of Missouri's interest in preventing corruption or the appearance thereof, because in its view, there was "little reason to doubt that sometimes large contributions will work actual corruption of our political system, and no reason to question the existence of a corresponding suspicion among voters." Id. at 395, 120 S.Ct. 897; see also McConnell, 540 U.S. at ___, 124 S.Ct. at 661 ("The idea that large contributions to a national party can corrupt or, at the very least, create the appearance of corruption of federal candidates and officeholders is neither novel nor implausible."). Similarly, the Shrink Court relied in large part on Buckley's conclusion on the fit between contribution limits and the anti-corruption interest, to decide that the contribution limits in Missouri were sufficiently tailored and not "so different in kind as to raise essentially a new issue about the adequacy of the Missouri statute's tailoring to serve its purposes." 528 U.S. at 395, 120 S.Ct. 897.

With Shrink's guidance on "the quantum of empirical evidence needed" in mind, we turn then to the interests asserted by Vermont in support of Act 64's expenditure limits to assess whether any of the interests might be sufficiently compelling to support this regulation of political speech. Vermont offers five interests that it argues are sufficiently compelling to support the spending limits on candidates: (1) "avoiding the reality and appearance of corruption in elective politics and government"; (2) "assur[ing] that candidates and officeholders will spend less time fund-raising and more time interacting with voters and performing official duties"; (3) promoting "electoral competition and in protecting equal access to political participation"; (4) "bolster[ing] voter interest and engagement in elective politics"; and (5) "enhanc[ing] the quality of political debate and voters' understanding of the issues."

Defendants-intervenors appear to rely primarily on the first two of these interests to support the spending limits — describing those interests as (1) deterring corruption and the appearance of corruption; and (2) permitting candidates and officeholders to spend less time fund-raising and more time interacting with voters and performing duties. Defendants-intervenors also argue that the third interest asserted by the State — "protecting political equality" — should be recognized as an "additional basis" to support the spending limits on candidates.10

We now consider the interests asserted by the defendants.

The Supreme Court has recently clarified that the anti-corruption interest, in the campaign finance context, is "not confined to bribery of public officials, but extend[s] to the broader threat from politicians too compliant with the wishes of large contributors." Shrink, 528 U.S. at 389, 120 S.Ct. 897; see also McConnell, 540 U.S. at ___, 124 S.Ct. at 660. Moreover, the Shrink Court reiterated that, in addition to the actual influence of campaign contributions on politicians' behavior, the perception of corruption was an important part of this compelling state interest because it "could jeopardize the willingness of voters to take part in democratic governance." Shrink, 528 U.S. at 390, 120 S.Ct. 897 (citing United States v. Mississippi Valley Generating Co., 364 U.S. 520, 562, 81 S.Ct. 294, 5 L.Ed.2d 268 (1961) (democracy works "only if the people have faith in those who govern")).

In terms of "the quantum of empirical evidence needed," we note that although the interest in avoiding corruption and the appearance thereof is well-established as sufficiently important in the context of contribution limits, the rejection in Buckley of the anti-corruption interest as a constitutional justification for spending limits dictates the need for considerable evidence to demonstrate that unlimited spending is part of the corruption problem, and that spending limits are a necessary and plausible solution.

In this case, the District Court found that Vermont had proven that the reality and perception of corruption in its political system was a legitimate concern. Specifically, it found that "[e]vidence at trial overwhelmingly demonstrated that the Vermont public is suspicious about the effect of big-money influence over politics," and "it appears they have reason to feel that way," 118 F.Supp.2d at 468, concluding that "[t]he record suggested that large contributors often have an undue influence over the legislative agenda." Id. In light of Shrink, this "undue influence" over the legislative agenda is properly considered part of the anti-corruption interest. See 528 U.S. at 389, 120 S.Ct. 897.11 For the reasons that follow, our independent review of the evidence supports these findings by the District Court.

First, citizens in Vermont have consistently demonstrated a belief that the attention of their public representatives may be available for a price. As a result, public faith in the democratic system has declined. The General Assembly described the effects of a need to raise ever growing amounts of funds: "Robust debate of issues, candidate interaction with the electorate, and public involvement and confidence in the electoral process have decreased as campaign expenditures have increased." 1997 Vt. Laws P.A. 64 (H. 28) (finding No. 4). "Large contributions and large expenditures by persons or committees, other than the candidate and particularly from out-of-state political committees or corporations, reduce public confidence in the electoral process and increase the appearance that candidates and elected officials will not act in the best interests of Vermont citizens." Id. (finding No. 9). At trial, one expert witness, Celinda C. Lake, concluded that "[v]oters are extremely concerned about the influence of special interests in the political process." In fact, according to polling data, nearly 75 percent of Vermont voters say that ordinary voters do not have enough influence over Vermont politics and government, and more than two thirds believe that "large corporations and wealthy individuals have too much influence." (expert report of Celinda C. Lake).

Testimony by Vermont's elected officials revealed that this disenchantment and loss of public faith played a critical role in their belief that expenditure limits are necessary. One sponsor of Act 64, Representative Karen Kitzmiller, presented the Vermont House of Representatives with evidence showing that 94 percent of Vermonters believe that too much money is spent in politics, and 76 percent believe that ending private contributions would "reduce the power of special interest groups." Another state legislator, Gordon Bristol, testified at trial about his concern "about the regular guy on the street, and I think if they feel that candidates are spending a modest amount of money, that they are going to get candidates in there who are representing issues and not a special interest. . . ." According to another legislator, citizens have reported that they do not vote because "`[a]ll the big money controls everybody in Montpelier anyways.' . . . They think it's all wrapped up and that the special interests control it and, quite frankly, they aren't that wrong." (testimony of Elizabeth Ready). Another legislator said: "[I]t's the monied interests that control the process, and that cynicism ... it keeps people from participating, from engaging. . . ." (testimony of Donald Hooper).

Second, because of the limited number of campaign contributors and the constant concern of being outspent, candidates and elected officials are significantly influenced in deciding positions on issues by a belief that they are unable to oppose too many special interests, no matter how unpopular, because they will be cut off from funds. The General Assembly described the effects of a need to raise ever growing amounts of funds: "Increasing campaign expenditures require candidates to seek and rely on a smaller number of larger contributors, often outside the state, rather than a large number of small contributors." 1997 Vt. Laws P.A. 64 (H.28) (finding No. 5). If legislation alienates one major special interest group, officials are reluctant to alienate others because the number of entities and people making political contributions is finite and small. One state legislator admitted that, when considering a piece of legislation, "You have to initially consider it as whether or not you want to risk losing the financial support or, in the worst case, having that financial support go to a primary opponent or to a person who opposes you in a general election." (testimony of Peter Smith). One candidate recalled being told by another lawmaker: "We've already lost the drug money [because of the pharmacy bill], and I don't need to lose the food manufacture money too. So I'm not going to sign the bill." (testimony of Cheryl Rivers).

Third, and perhaps most perniciously, the demands of fundraising also affect the behavior of elected officials in the context of agenda-setting, since officials pay attention to which contributor "wants what to happen in terms of language of the bill, in terms of calendaring the bill, in terms of writing the rules." (testimony of Peter Smith). That same witness also noted that a crucial part of any deliberation on a bill involves speculation about the reaction of contributors because they control the money: politicians are forever asking "what's the industry position, what's the union position, what's — you know, and what they're talking about is where [is] the money behind the issue, what does the money want, where is the conflict between and among the power brokers." Senator Rivers testified that campaign contributors, by virtue of their role as contributors, can dominate the attention of party leadership or a committee chair, and thereby influence the legislature's agenda. In her words, "there is kind of an atmosphere that is created that there is [an] assumption that phone calls [of contributors] will get taken and [their] policy issues will be considered." Another senator, Elizabeth Ready, recognized that "there is an agenda out there that is pretty much set by folks that are not elected."12 Candidates, often with great reluctance, accept the bargain with contributors so that they do not lose large sources of potential fundraising for the "arms race" in which they feel compelled to participate.

The evidence at trial established that candidates for public office rely on special interests for financial support, produced directly or by way of "bundling" smaller contributions from a particular company or industry.13 The Buckley Court seemed to assume that many small contributions could not raise the specter of corruption. "If a senatorial candidate can raise $1 from each voter, what evil is exacerbated by allowing that candidate to use all that money for political communication?" 424 U.S. at 56 n. 64, 96 S.Ct. 612 (internal quotation marks omitted). But the reality of campaign financing in Vermont is a far cry from this idyllic vision of political fundraising, in large part because not every voter has the financial ability to participate by giving campaign contributions. "[T]he average Vermonter has been, to some degree, disenfranchised because the average Vermonter cannot afford the price of admission." Senate J. of the State of Vt., at 1338 (Biennial Session, 1997) (statement of William T. Doyle).

Vermont has a compelling interest in safeguarding its political process from such contributor dominance, because it corrupts the process for achieving accessibility and accountability of state officials and candidates. The evidence at trial demonstrated that money — and the special interests that wield it — has a great influence on candidate behavior in Vermont, at the expense of the electorate as a whole, since candidates depend on it in order to run for office. Where access and influence can be bought, citizens are less willing to believe that the political system represents the electorate, exacerbating cynicism and weakening the legitimacy of government power. See Jacobus v. Alaska, 338 F.3d 1095, 1113 (9th Cir.2003) (describing the phenomenon of "access-peddling" and explaining that it "creates a danger of corruption and the appearance of corruption"). The accessibility and accountability of public officials — and the public's faith that Vermont's government is accessible and accountable — are fundamental to any democratic system.

In our view, such influence of campaign contributors is pernicious because it is bought. Certain private citizens and organizations should not be given greater access to public office holders — and thus greater influence — on account of those citizens' ability and willingness to pay for candidates' campaigns. Even with contribution limits, the arms race mentality has made candidates beholden to financial constituencies that contribute to them, and candidates must give them special attention because the contributors will pay for their campaigns. Quid pro quo corruption is troubling not because certain citizens are victorious in the legislative process, but because they achieve the victory by paying public officials for it.

In short, we believe, based on the District Court's findings and our own independent review of the record, that Vermont has proven the strength of this interest, and its relationship to unlimited campaign spending. And we believe that the factual record developed by Vermont in support of the anti-corruption interest, through the legislative process and at trial, may be sufficient to distinguish Buckley. Cf. Colorado Republican I, 518 U.S. at 617-18, 116 S.Ct. 2309 (plurality opinion) (indicating that "the lack of coordination between the candidate and the source of the expenditure... prevents us from assuming, absent convincing evidence to the contrary, that a limitation on political parties' independent expenditures is necessary to combat a substantial danger of corruption of the electoral system.") (emphasis added); Planned Parenthood of Southeastern Pennsylvania v. Casey, 505 U.S. 833, 863-64, 112 S.Ct. 2791, 120 L.Ed.2d 674 (1992) (citing West Coast Hotel and Brown v. Board of Education as examples of "applications of constitutional principle to facts as they had not been seen by the Court before"). Nonetheless, given Buckley's holding rejecting the anti-corruption interest as inadequate to support the expenditure limits at issue in that case, we are reluctant to conclude that the same general interest, standing alone, is sufficiently compelling to support Act 64's expenditure limits. See Buckley, 424 U.S. at 46-48, 96 S.Ct. 612; see also McConnell, 540 U.S. at ___, 124 S.Ct. at 647. We turn then to the second interest asserted by defendants.

Vermont also submits that it has a compelling interest in "assur[ing] that candidates and officeholders will spend less time fundraising and more time interacting with voters and performing official duties." Indeed, the District Court found that "the need to solicit money from large donors at times turns legislators away from their official duties." 118 F.Supp.2d at 468. The District Court also indicated that the State proved that this concern exists, and that Vermont's expenditure limits addressed this interest, among others. Id. at 482-83.

Again, we are mindful of Shrink's guidance that "[t]he quantum of empirical evidence needed to satisfy heightened judicial scrutiny of legislative judgments will vary up or down with the novelty and plausibility of the justification raised."14 528 U.S. at 391, 120 S.Ct. 897. On this score, the "time protection" rationale has been recognized as compelling, although not in the context of candidate spending limits. Indeed, the Buckley Court considered this interest in assessing, and deemed it sufficiently important to support, the public financing scheme for Presidential election campaigns — a provision it upheld. See 424 U.S. at 96, 96 S.Ct. 612 ("Congress properly regarded public financing as an appropriate means of relieving major-party Presidential candidates from the rigors of soliciting private contributions") (citing Senate Rep. No. 93-689); 424 U.S. at 91, 96 S.Ct. 612 ("Congress was legislating for the `general welfare' . . . to free candidates from the rigors of fundraising."). See also Republican Nat'l Committee v. Federal Election Comm'n, 487 F.Supp. 280, 284-86 (S.D.N.Y.) (three-judge District Court) (upholding constitutionality of expenditure limits as condition of accepting presidential public financing in part on ground that it would "give candidates the opportunity to lessen the `great drain on (their) time and energies' required by fundraising `at the expense of providing competitive debate of the issues for the electorate'") (quoting Senate Rep. No. 93-689), aff'd mem., 445 U.S. 955, 100 S.Ct. 1639, 64 L.Ed.2d 231 (1980).

Moreover, other circuits have more recently recognized the compelling nature of the time-protection interest in similar contexts. See Rosenstiel v. Rodriguez, 101 F.3d 1544, 1553 (8th Cir.1996), cert denied, 520 U.S. 1229, 117 S.Ct. 1820, 137 L.Ed.2d 1028 (1997) (upholding Minnesota's voluntary public financing scheme because the government has a compelling interest in reducing "the time candidates spend raising campaign contributions, thereby increasing the time available for discussion of the issues and campaigning"); Vote Choice, Inc. v. DiStefano, 4 F.3d 26, 39 (1st Cir.1993) (holding that statute survives exacting scrutiny because Rhode Island has "a valid interest in having candidates accept public financing because such programs `facilitate communication by candidates with the electorate' [and] free candidates from the pressures of fundraising.") (quoting Buckley, 424 U.S. at 91, 96 S.Ct. 612). Indeed, the Rosenstiel court determined that it is "well settled" that this interest is compelling. 101 F.3d at 1553 (collecting cases).

The Buckley Court, in determining that the expenditure limits in that case were unconstitutional, alluded to this time-protection interest only in passing. 424 U.S. at 91, 96, 96 S.Ct. 612 (mentioning generally Congress' desire to relieve political candidates from the "rigors" of soliciting and fundraising); see also Blasi, supra, at 1285-86 & n. 15 ("[D]uring the public and legislative debates that led to the passage in 1974 of mandatory spending limits for congressional races, and during the Buckley litigation which resulted in the invalidation of those limits, candidate time protection was almost wholly ignored as a justification for campaign spending limits."). Only Justice White, concurring in part and dissenting in part, observed that imposing "expenditure ceilings" would "ease the candidate's understandable obsession with fundraising, and so free him and his staff to communicate in more places and ways unconnected with the fundraising function." Buckley, 424 U.S. at 264-65, 96 S.Ct. 612 ("There is nothing objectionable — indeed it seems to me to be a weighty interest in favor of the provision — in the attempt to insulate the political expression of federal candidates from the influence inevitably exerted by the endless job of raising increasingly large sums of money."). One commentator explains that "candidate time protection was not at the center of either the reform agenda or the constitutional analysis" because Buckley was decided "[b]efore the advent of pervasive war chests and candidate-PAC merchandizing bazaars."15 Blasi, supra, at 1287.

Plaintiffs argue that this interest is no different than the goal of reducing the "skyrocketing costs of political campaigns," rejected by Buckley as an insufficiently compelling interest to support expenditure limits. 424 U.S. at 57, 96 S.Ct. 612. The Sixth Circuit agreed in Kruse, reasoning that "[t]he need to spend a large amount of time fundraising is a direct outgrowth of high costs of campaigns. However, because the government cannot constitutionally limit the cost of campaigns, the need to spend time raising money, which admittedly detracts an officeholder from doing her job, cannot serve as a basis for limiting campaign spending." 142 F.3d at 916-17. We are unpersuaded by this reasoning.

Indeed, we think the language of Buckley, as well as an examination of the Buckley briefs, oral argument, and subsequent commentary from judges and scholars, precludes such an interpretation. In its discussion of the federal campaign expenditure ceilings at issue in Buckley, the Buckley Court explained that the limits "appear to be designed primarily to serve the governmental interests in reducing the allegedly skyrocketing costs of political campaigns," and cited the statistics put forward by appellees and appellants on how the percentage increase in campaign spending in recent years compared to the rise in the consumer price index, gross national product, and total expenditures for commercial advertising over the same time period. 424 U.S. at 57, 96 S.Ct. 612. The Court concluded that, regardless of the import of such statistics, "the mere growth in the cost of federal election campaigns in and of itself provides no basis for governmental restrictions on the quantity of campaign spending and the resulting limitation on the scope of federal campaigns." Id. (emphasis added). Particularly in light of the recent statements of three Justices indicating that this "time protection" rationale may be a compelling interest, see supra at 108 (quoting Shrink, 528 U.S. at 409, 120 S.Ct. 897 (Kennedy, J., dissenting)); and Colorado Republican I, 518 U.S. at 649-50, 116 S.Ct. 2309 (Stevens, J., joined by Ginsburg, J., dissenting), we see no reason to read Buckley more broadly than its language indicates.

At trial, Vermont presented powerful evidence concerning the time pressures which the prospect of unlimited expenditures places on candidates for office. In particular, there is strong evidence that unlimited expenditures have compelled candidates to engage in lengthy fundraising in order to preempt the possibility that their political opponents may develop substantially larger campaign war chests. The Vermont General Assembly found that such fundraising by candidates requires an "inordinate[] amount of time." 1997 Vt. Laws P.A. 64 (H. 28) (finding No. 1). The large, and growing, campaign war chests in Vermont have created strong pressures on elected officials to ensure that they can raise funds comparable to any opponent. One witness, former State Senator and Lieutenant Governor Peter Smith, described the "stampede or nuclear arms race mentality that we currently have, which is just keep building the bank because you never know what's going to happen." Under the current system, Vermont candidates feel like "you had two races you were running. The first was for the money. . . ." (testimony of Donald Hooper).

Although there may be no inherent problem with candidates competing to raise large quantities of funds, the evidence in Vermont is clear that the pressure to raise large sums of money grea