The Donor Class

The Hundred-Year Consensus

In his annual message to Congress on December 5, 1905, Theodore Roosevelt called for the prohibition of corporate contributions to federal political campaigns.¹ His call was personally motivated. Roosevelt’s own 1904 reelection campaign had accepted, by the conservative reckoning of the time, between approximately two and four hundred thousand dollars in corporate contributions, including significant sums from Standard Oil, E. H. Harriman, and the J. P. Morgan banking interests.² Joseph Pulitzer, whose New York World had been investigating Roosevelt’s campaign finances, had begun publishing the disclosures in the months before the message; the issue was politically active and personally exposed. Roosevelt, who had spent his first term constructing the trust-busting reputation that would become his historical legacy, recognized the structural problem the disclosures revealed and offered Congress a structural response. “All contributions by corporations to any political committee or for any political purpose,” he proposed, “should be forbidden by law.”

Senator Benjamin Tillman of South Carolina, an avowed white supremacist whose racial politics have been correctly remembered as among the worst of the late Reconstruction-era Senate, was the bill’s lead sponsor.³ His motivations for sponsorship combined a Southern Democratic populist hostility to corporate power with a recognition that the Republican Party of his day was the principal beneficiary of corporate political contributions. The bill’s coalition was therefore unlikely on its face — a Republican president had called for it, a Southern Democrat had championed it, and the Republican-controlled Senate of the Sixty-Ninth Congress would pass it. The Tillman Act, signed into law on January 26, 1907, prohibited “any corporation whatever” from making “money contribution in connection with any election to any political office.”⁴ It was the first federal campaign finance statute, and it established a principle that would define American campaign finance regulation for the next sixty-nine years: corporate political spending was understood as structurally distinct from individual political speech, and the federal government could legitimately constrain it.

The principle was extended through a sequence of statutes that, taken together, produced a coherent regulatory regime by the early 1970s. The Federal Corrupt Practices Act of 1910, amended in 1911 and again in 1925, added disclosure requirements and per-candidate spending caps.⁵ The Hatch Act of 1939, amended substantially in 1940, capped individual contributions to a single federal candidate at five thousand dollars per year and capped total annual political committee expenditures at three million dollars; both caps were straightforwardly evaded through the proliferation of multiple committees, but the principle of contribution limits and committee spending caps was now established law.⁶ The Smith-Connally Act of 1943, a wartime measure, extended the corporate contribution ban to labor unions for the duration of World War II; the Taft-Hartley Act of 1947 made the union ban permanent and extended it from contributions to independent expenditures.⁷ The Federal Election Campaign Act of 1971 replaced the toothless 1925 statute with a modern regulatory framework: detailed disclosure requirements, the formal recognition of political action committees, limits on candidate spending on broadcast advertising, and the Revenue Act of the same year created the dollar-checkoff funding the Presidential Election Campaign Fund.⁸ The 1974 amendments to FECA, passed in the immediate post-Watergate period, completed the structure: the Federal Election Commission, contribution limits of one thousand dollars per individual to a candidate per election and five thousand dollars per political action committee, expenditure limits on candidates and on independent expenditures, an aggregate annual contribution cap of twenty-five thousand dollars, and a system of public matching funds for presidential primaries paired with full public financing for the general election.

Across the sixty-nine years between Tillman and the 1974 amendments, the regulatory framework had been built and rebuilt by both major parties under roughly half a dozen presidencies of both partisan affiliations. The framework was neither liberal nor conservative; it was a structural recognition that elections were a category of activity in which the unlimited participation of concentrated wealth produced a recognizable failure mode that representative government could not survive. The framework’s specific provisions were repeatedly contested at the margins; the principle that the framework was legitimate, that Congress could regulate the financing of federal elections in the public interest, was not. Through 1975, no Supreme Court decision had questioned that principle. The constitutional foundation of campaign finance regulation was, by every available indication, settled.

It was settled in 1975. It was unsettled in 1976. The unsettling has continued, with brief partial pauses, for the forty-nine years since.

The Doctrine Constructed

Buckley v. Valeo, decided on January 30, 1976, is the foundational decision of contemporary American campaign finance jurisprudence and the doctrinal source of every subsequent decision in the line.⁹ The case was a constitutional challenge, brought by a coalition of plaintiffs led by Senator James Buckley of New York and Senator Eugene McCarthy of Minnesota, to the 1974 amendments to FECA. The challenge was sweeping: it argued that nearly every operative provision of the 1974 amendments — contribution limits, expenditure limits, disclosure requirements, the appointment mechanism for the FEC, the public financing system — violated the First Amendment.

The Supreme Court issued a per curiam opinion, with no single named author, addressing each of the contested provisions individually. The decision did not strike down FECA wholesale; it produced a complex pattern of upheld and stricken provisions that reflected the Court’s effort to draw a constitutional distinction between two categories of political money. Contribution limits, the Court held, were constitutionally permissible because contributions involved “speech by proxy” — the contributor was supporting another person’s speech, not engaging in speech directly — and could be regulated to prevent quid pro quo corruption or its appearance.¹⁰ Expenditure limits, by contrast, were impermissible because expenditures themselves constituted core political speech, and any restriction on the volume or duration of that speech violated the First Amendment.

The doctrine the Court used to draw this distinction was the proposition that “money is speech.” The proposition was not, in Buckley, an explicit holding so much as an inference from a more careful argument: that the modern political speech of any practical scale necessarily required the expenditure of money on the means of its dissemination — broadcast time, newspaper advertising, mailings, staff salaries, transportation — and that restrictions on the expenditure of money for those purposes were therefore restrictions on the speech itself. Restrictions on contributions, by contrast, were restrictions only on the contributor’s expression of support for someone else’s speech, not on the contributor’s own speech, and were therefore permissible at lower scrutiny. The contribution-versus-expenditure distinction would become the doctrinal axis on which all subsequent campaign finance jurisprudence turned.

The most consequential single sentence of the Buckley opinion was its rejection of one of the principal arguments the government had advanced in defense of expenditure limits. The government had argued that limiting expenditures was necessary to “equalize the relative ability of all citizens to affect the outcome of elections” — the structural-equality rationale that had been part of the political consensus underlying the FECA amendments. The Court rejected this argument categorically: “the concept that government may restrict the speech of some elements of our society in order to enhance the relative voice of others is wholly foreign to the First Amendment.”¹¹ The sentence has been quoted in nearly every subsequent campaign finance decision and is the doctrinal ground from which the line of cases extending through Citizens United and McCutcheon was eventually constructed. It established, as a matter of constitutional law, that the structural-equality rationale that had animated American campaign finance regulation since Tillman was not available as a justification for expenditure restrictions.

The Buckley decision struck down FECA’s expenditure limits on candidates, on candidate self-financing from personal funds, and on independent expenditures by individuals and groups. It upheld the contribution limits, the disclosure requirements, and the public financing system. It also struck down the FECA’s mechanism for appointing FEC commissioners on separation-of-powers grounds, requiring that mechanism to be restructured. The pattern of upheld and stricken provisions produced an immediate, structural asymmetry in the post-Buckley campaign finance regime: contributions were capped, but expenditures were not. Anyone who could spend their own money directly on political advocacy could spend without limit. Anyone whose preferred mode of political participation involved supporting a candidate through a contribution was constrained.

The asymmetry favored, by construction, two specific kinds of political actors. The first was the wealthy candidate who could self-finance — Buckley’s own coalition included candidates whose opposition to the FECA amendments was partly a defense of their right to spend personal wealth on their own campaigns without limit. The second was the wealthy non-candidate who preferred to spend independently rather than through a candidate’s campaign — the implicit founder of what would, three and a half decades later, become the super PAC.

In 1976, the second category was not yet active at scale. The legal infrastructure that would make independent expenditures by non-candidate actors operationally easy did not yet exist; the political culture had not yet developed the practice; the donor class that would eventually populate the category was still adjusting to the post-Watergate disclosure regime. The asymmetry the Buckley decision created sat largely dormant in the political economy of the late 1970s. It would be activated, in stages, over the next three decades, through a sequence of subsequent decisions that extended Buckley’s reasoning to new categories of actors and new categories of political activity. By the time the activation was complete, the regulatory framework that Tillman, Roosevelt, and the post-Watergate Congress had constructed would have been substantially dismantled, and the structural-equality rationale that had motivated the framework would have been formally read out of American constitutional law.

The Doctrine Extended

Two years after Buckley, in First National Bank of Boston v. Bellotti, the Court took the first step toward extending its First Amendment analysis from individual political speech to corporate political speech.¹² Bellotti concerned a Massachusetts statute that prohibited corporations from spending general treasury funds on referenda unrelated to the corporation’s business. The state’s argument was structural: the statute was designed to prevent the use of corporate wealth, accumulated through state-conferred privileges of incorporation, to dominate ballot-measure campaigns in which the corporation had no business interest. The Court, in a 5-4 majority opinion by Justice Lewis Powell, struck down the statute. First Amendment protection, the Court held, did not depend on the identity of the speaker; corporate speech on referenda was protected to the same extent as individual speech.

Bellotti was a narrow holding that left open the broader question of corporate political activity in candidate elections. For the next twelve years, the Court worked through that question incrementally. In FEC v. Massachusetts Citizens for Life (1986), the Court created an exception for ideological nonprofit corporations that took no business income and existed solely for political advocacy; such corporations could not, the Court held, be barred from independent expenditures.¹³ The decision was the first crack in the Tillman-derived framework’s prohibition on corporate independent political spending. In Austin v. Michigan Chamber of Commerce (1990), the Court considered a Michigan statute prohibiting business corporations from spending general treasury funds on candidate elections.¹⁴ The Court, in a 6-3 majority opinion by Justice Thurgood Marshall, upheld the Michigan statute. The state interest the Court recognized was novel and structural: “the corrosive and distorting effects of immense aggregations of wealth that are accumulated with the help of the corporate form and that have little or no correlation to the public’s support for the corporation’s political ideas.” Austin established, in 1990, that the structural-equality rationale Buckley had rejected for individual expenditures could be re-introduced under a different doctrinal framing — anti-distortion rather than anti-domination — for corporate expenditures specifically. The decision preserved the Tillman framework, in a constitutionally narrowed form, for the corporate political spending that had been Tillman’s principal target.

In McConnell v. FEC (2003), the Court considered the Bipartisan Campaign Reform Act of 2002, the McCain-Feingold legislation that had banned soft-money contributions to political parties and restricted “electioneering communications” — broadcast advertisements that named federal candidates within sixty days of an election — funded from corporate or union treasuries.¹⁵ The Court largely upheld BCRA, including the soft-money ban and the electioneering-communication restrictions, in a joint majority opinion by Justices John Paul Stevens and Sandra Day O’Connor. The vote was 5-4. McConnell extended the Austin framework into the electioneering-communication context and, in retrospect, marked the high-water mark of post-Buckley campaign finance regulation. The framework that had been built and partly preserved across the previous century survived through 2003 in a form that was constitutionally constrained at the margins but operationally intact at the core.

The retrenchment that began in 2010 was sudden and decisive. Citizens United v. FEC, decided on January 21, 2010, was a 5-4 decision authored by Justice Anthony Kennedy.¹⁶ The case concerned a documentary, Hillary: The Movie, produced by the conservative nonprofit Citizens United, which the FEC had ruled was an electioneering communication subject to BCRA’s funding restrictions. The Court could have decided the case narrowly, on the question of whether the specific film was the kind of communication BCRA was meant to cover, or whether Citizens United could be classified as an MCFL-eligible ideological nonprofit. The Court chose, instead, to expand the case at reargument and decide the broader constitutional question: whether the government could prohibit corporate or union independent expenditures in candidate elections. The majority answered no. Austin was overruled. The relevant portion of McConnell was overruled. The structural-equality and anti-distortion rationales that had underwritten the corporate-spending restrictions for over a century were declared inadequate to support those restrictions under the First Amendment.

Justice John Paul Stevens, writing the principal dissent (joined by Justices Ginsburg, Breyer, and Sotomayor), produced one of the longest and most cited dissents in modern Court history.¹⁷ His central argument was that the majority had misread its own precedent, ignored the empirical record of corporate political activity, and misunderstood the nature of corporate personhood under the First Amendment. “Corporations have no consciences, no beliefs, no feelings, no thoughts, no desires,” he wrote. “Their ‘personhood’ often serves as a useful legal fiction. But they are not themselves members of ‘We the People’ by whom and for whom our Constitution was established.” The dissent’s specific empirical claim — that the majority’s decision would “unleash” a flood of corporate political spending that would distort American democracy — was, at the time, contested by the majority and by the decision’s defenders, who argued that the existing political marketplace already accommodated corporate expression through PACs and that the practical effects of overruling Austin would be modest.

Two months after Citizens United, the United States Court of Appeals for the District of Columbia Circuit decided SpeechNow.org v. FEC.¹⁸ The decision, en banc and unanimous, applied Citizens United’s reasoning to political committees that engaged only in independent expenditures, rather than in direct contributions to candidates. If, the SpeechNow court reasoned, contributions to independent-expenditure-making organizations could not corrupt candidates (because the organizations did not contribute to candidates), then the government had no anti-corruption interest in limiting contributions to such organizations. The committees that emerged from SpeechNow — independent-expenditure-only committees, capable of accepting unlimited contributions from any source and making unlimited expenditures in support of or in opposition to federal candidates — became known, after a series of FEC advisory opinions formally recognized them in mid-2010, as super PACs. Citizens United alone had not created the super PAC. Citizens United plus SpeechNow had. The Supreme Court declined to review the SpeechNow decision; the super PAC’s constitutional foundation was thus established without a Supreme Court ruling specifically endorsing it.

In McCutcheon v. FEC (2014), the Court completed the line.¹⁹ The case concerned the FECA-derived aggregate cap on what a single donor could contribute across all federal candidates and committees in a two-year cycle, then approximately one hundred twenty-three thousand dollars. The Court, in a 5-4 plurality opinion by Chief Justice John Roberts, struck down the aggregate cap. Per-candidate base contribution limits — the original Buckley-upheld provision — were left intact; but the structural restraint that had prevented a single wealthy donor from giving the maximum permissible contribution to every federal candidate of his preferred party was eliminated. Justice Clarence Thomas concurred separately, indicating he would have gone further and overruled Buckley’s contribution-limit analysis as well. The four-justice dissent, by Justice Stephen Breyer, argued that the plurality had narrowed the recognized governmental interest in preventing corruption to a definition so cramped that “even Buckley would not survive its application.”

Between Buckley in 1976 and McCutcheon in 2014, the structural-equality rationale that had underwritten American campaign finance regulation for the seventy years from Tillman to the 1974 amendments was systematically read out of American constitutional law. The Tillman Act remains formally in force; the prohibition on direct corporate contributions to federal candidates remains. The framework that Tillman, Roosevelt, the Hatch Act Congress, the Taft-Hartley Congress, and the post-Watergate Congress had built around that core prohibition has been substantially dismantled. The independent-expenditure regime that emerged from Buckley, and the corporate-and-union extension of that regime that emerged from Citizens United and SpeechNow, has produced a campaign finance system whose empirical structure the next section describes.

The Empirical Result

The campaign finance system that emerged from the post-Buckley line of decisions has now operated at full extension for nearly fifteen years, and the empirical record of its consequences is no longer ambiguous. The four most consequential consequences are documented in the political science and election finance literatures, and are summarized in this section.

The first consequence is the absolute scale of money flowing through federal elections. The Federal Election Commission and the standard aggregator of campaign finance data, OpenSecrets, report that total federal election spending grew from approximately three point one billion dollars in the 2000 cycle to approximately fourteen point four billion in 2020 and to approximately fifteen to sixteen billion in the 2024 cycle.²⁰ The growth is not principally inflationary; over the same period, the consumer price index roughly doubled, while federal election spending grew by a factor of approximately five. The growth is concentrated in the post-Citizens-United period: outside spending, the principal post-2010 growth category, grew from approximately three hundred thirty-eight million dollars in the 2008 cycle (the last pre-Citizens-United presidential) to approximately one billion forty million in 2012 (the first post-Citizens-United presidential), to approximately three point three billion in 2020 and approximately four point five billion in 2024.²¹ Outside spending, which had been a marginal category through 2008, became a primary structural feature of federal election finance after 2010.

The second consequence is the concentration of the donor base. Research by Adam Bonica of Stanford University, drawn from the Database on Ideology, Money in Politics, and Elections (DIME) — the most comprehensive academic dataset on American campaign finance — has documented that a vanishingly small fraction of the American population provides a disproportionate share of campaign contributions.²² The specific share varies across cycles and methodological choices, but the pattern is consistent: the top one-hundredth of one percent of the adult population provides a substantial share of all individual contributions to federal candidates and committees, with the share rising over time. Bonica’s broader research, conducted with Nolan McCarty, Keith Poole, and Howard Rosenthal in a 2013 Journal of Economic Perspectives article, established that the donor class is not merely small but ideologically distinct from the median voter; donors are systematically more conservative on economic policy than non-donor citizens of the same income level, and they exert pressure on candidates of both parties to align with donor preferences.²³

The third consequence is the empirical demonstration that single donors can sustain candidacies that traditional fundraising would not. The 2012 Republican presidential primary produced the canonical case. Sheldon and Miriam Adelson, casino magnates whose wealth had been concentrated in the Las Vegas Sands Corporation, provided approximately twenty million dollars to a super PAC called Winning Our Future, which was organized to support the candidacy of former Speaker of the House Newt Gingrich.²⁴ Gingrich, whose campaign had been financially marginal through Iowa and New Hampshire, won the South Carolina primary on January 21, 2012 — the second anniversary, to the day, of the Citizens United decision — in substantial part on the strength of advertising paid for by the Adelson super PAC. He went on to win the Georgia primary in March before withdrawing in May. A separate super PAC funded by Wyoming investor Foster Friess, the Red, White and Blue Fund, sustained the candidacy of former Senator Rick Santorum through eleven primary victories. Pre-Citizens-United campaign finance law would have made both candidacies untenable by mid-February 2012; post-Citizens-United law made them sustainable through the spring on the strength of single-donor patronage. The pattern has scaled. In the 2024 cycle, Elon Musk’s America PAC reportedly spent in excess of two hundred million dollars in support of Donald Trump’s general-election campaign, and Timothy Mellon, an heir to the Mellon banking and oil fortune, gave fifty million dollars in a single contribution to MAGA Inc. in May 2024 alone.²⁵ The 2024 cycle had the highest concentration of top-donor giving in modern American history, by Brennan Center analysis.²⁶

The fourth consequence is structural rather than transactional. Heather Gerken, the dean of Yale Law School, has argued in a sequence of articles that the post-2010 super PAC regime has produced what she calls “shadow parties” — political organizations that perform the coordination, message-discipline, and resource-allocation functions historically performed by formal political parties, but with substantially less institutional accountability.²⁷ Shadow parties accept unlimited contributions from individual donors; formal parties cannot. Shadow parties are not bound by the internal coalition disciplines that historically caused party leadership to moderate factional extremes; their leadership answers to the donors who fund them. Shadow parties operate alongside the formal parties, in nominal independence from candidates but in practical close coordination, and have absorbed an increasing share of the strategic functions that formal parties once monopolized. The result is that the formal Democratic and Republican Parties of 2025 are weaker, relative to the political environment they operate in, than the formal parties of 2005 were; and the donor-funded shadow parties that have absorbed their displaced functions are accountable to a substantially narrower constituency than the formal parties they have partially replaced.

The four consequences taken together describe a campaign finance system in which a small population of very wealthy donors and a corresponding population of corporate and union actors have come to occupy a position of structural prominence that would have been recognizable to the Tillman-era Congress as the precise problem the Tillman Act was passed to prevent. The framework that prevented the problem from 1907 to 1976, partially through 2010, has been dismantled. The problem the framework was constructed to prevent is now operating, at scale, across federal elections and the broader political economy that federal elections shape. The Buckley-era constitutional doctrine that the structural-equality rationale was “wholly foreign to the First Amendment” has proven, over half a century, to be a durable constraint on legislative response. The doctrine has not been revisited.

The Lobbying Layer

The campaign finance system described above is one half of the structure through which concentrated wealth has acquired its contemporary position in American government. The other half is the federal lobbying complex, which operates on a parallel financial scale and through a complementary mechanism.

Total federal lobbying spending, as reported under the Lobbying Disclosure Act of 1995, has grown from approximately one and a half billion dollars in 2000 to approximately four point three billion in 2023.²⁸ The figure understates the true scale of influence spending substantially: the LDA’s reporting requirements cover only persons who spend twenty percent or more of their professional time on direct lobbying contacts with covered officials, exempting strategic communications, grassroots organizing, think-tank funding, and a category of “shadow lobbying” that political scientist Tim LaPira and Herschel Thomas have argued may be roughly equal in magnitude to the reported activity.²⁹ The reported total of registered federal lobbyists has hovered between approximately eleven and thirteen thousand in recent years; the unreported total may be twice that. The figure of approximately twenty-two registered lobbyists per voting member of Congress, frequently cited in the press, is a useful shorthand. The actual ratio, including shadow lobbying, is substantially higher.

The structural argument about the contemporary lobbying complex was made by Lee Drutman in his 2015 book The Business of America Is Lobbying.³⁰ Drutman’s central empirical claim, drawn from comprehensive analysis of LDA filings, is that business interests outspend diffuse and public-interest groups on federal lobbying by a ratio of approximately thirty-four to one. The asymmetry is not a recent development; it has been the structural condition of American federal lobbying since at least the early 1990s. What has changed in the post-2000 period is the absolute scale of the activity and the institutional sophistication with which corporate political activity is now conducted. The contemporary corporate lobbying operation, in Drutman’s account, is not a defensive posture against regulatory imposition but a proactive policy-shaping enterprise. It does not principally aim to stop legislation it opposes; it aims to shape the legislative agenda from the beginning, defining which problems get raised as problems, which solutions get drafted, which compromises survive markup. The thirty-four-to-one resource asymmetry is what makes the agenda-shaping effective. On any given legislative question, the constituency of organized business interests has more lawyers, more lobbyists, more former officials, more capital, and substantially longer attention span than the constituency of public-interest opposition. The legislative outcome reflects the asymmetry.

The mechanism by which the asymmetry is sustained is the revolving door between federal service and private-sector lobbying. The Honest Leadership and Open Government Act of 2007, passed in the wake of the Jack Abramoff scandal, established a two-year cooling-off period for departing senators and senior executive-branch officials and a one-year cooling-off period for departing House members, during which the former official was prohibited from “lobbying contacts” with the relevant chamber or department.³¹ The statute has been substantially circumvented through the proliferation of titled positions — “strategic adviser,” “senior counselor,” “policy adviser,” “vice chairman” — that involve substantial influence activity but do not technically meet the LDA’s twenty-percent threshold for registration as lobbyists. Public Citizen’s longitudinal analyses of post-congressional employment have found that approximately forty percent of members of Congress who leave office and remain in the workforce subsequently work in lobbying or government-relations capacities, with the figure rising to over fifty percent for senators specifically.³² The pattern is industry-specific. Defense contractors employ thousands of former Department of Defense officials, including senior military officers and Pentagon political appointees; the Project On Government Oversight’s Brass Parachutes report identified six hundred forty-five instances of senior government and military personnel moving to the top twenty defense contractors in a single year.³³ Pharmaceutical and medical-device companies employ former Food and Drug Administration officials, including a substantial share of the agency’s senior medical reviewers. Wall Street firms employ former Treasury, SEC, and CFTC officials. The pattern is general across the industries that lobby federal regulators.

The structural claim about the revolving door is not principally a claim about corruption in the conventional sense. It is a claim about the incentive structure facing currently-serving federal officials. A senior congressional staffer who anticipates a post-government career in the lobbying complex has powerful career incentives to avoid antagonizing the corporate interests with whom she will eventually negotiate her own employment. A House member who anticipates a multimillion-dollar lobbying career upon retirement has powerful career incentives to avoid voting in ways that would foreclose that career. The incentives operate prospectively, in the absence of any explicit transactional conduct, and are not constrained by the cooling-off period or by any other component of the existing legal regime. The constituency the official is most attentive to is, in important part, the constituency the official expects to work for in the future.

Lawrence Lessig, whose 2011 book Republic, Lost developed the most influential contemporary framework for understanding this dynamic, calls the resulting condition “dependence corruption.”³⁴ The framework distinguishes the contemporary American political economy from the corruption traditionally policed by federal anti-bribery and anti-gratuity statutes. Quid pro quo bribery is rare, illegal, and prosecuted; dependence corruption is structural, legal, and pervasive. Members of Congress and their senior staff develop a working dependency on the small population of donors and lobbyists who fund their campaigns and shape their post-government career options. The dependency does not bribe specific votes; it shapes which questions get asked, which proposals get drafted, which compromises survive markup, and which issues never reach the agenda at all. The resulting structure is, in Lessig’s language, a republic that has been “lost” — not in the sense that the formal institutions of representative government have collapsed, but in the sense that the substantive representation those institutions provide has been substantially captured by a constituency much smaller than the citizenry the institutions are nominally accountable to.

The two halves — the campaign finance system that produces the donor class, and the lobbying complex that operationalizes the donor class’s influence in the period between elections — are mutually reinforcing. The campaign finance system supplies the donors; the lobbying complex supplies the agenda; the revolving door supplies the personnel; the post-government career market supplies the incentive. Each component supports the others. The structure as a whole is what the Buckley-McCutcheon line of decisions has, in effect, made constitutionally irreducible. Statutory reform of any one component is constrained by the constitutional treatment of the others. A statute that closed the revolving door without addressing campaign finance would address only a minor component of the structure; a statute that addressed campaign finance at the contribution level without addressing the independent-expenditure and corporate-spending regime would address only a minor component of the structure. The structure has been, by judicial decision, moved largely beyond the reach of ordinary legislative response.

The Policy That Won’t Come

The pattern previous essays in this series have identified — that the structurally indicated policy response to the diagnosed problem cannot be enacted within the current configuration of American politics — operates in this case with particular force, because the configuration that prevents enactment is itself the diagnosed problem. The campaign finance system that has produced the donor class is the system that funds the political coalitions that would have to vote to constrain the donor class. The structural feedback loop is closed, and there is, by the explicit terms of the post-Buckley constitutional jurisprudence, no immediate route to opening it.

The legislative record of attempted reform is extensive and consistent in its failures. The DISCLOSE Act, introduced in 2010 in direct response to Citizens United, would have required corporations, unions, and 501(c)(4) organizations engaging in political spending to disclose the donors funding their political activity above a modest threshold; the act would have closed the dark-money loophole through which corporate political spending now occurs without public knowledge of the underlying funders. The act passed the House of Representatives in 2010 by a vote of two hundred nineteen to two hundred six. The Senate cloture vote on the same legislation failed fifty-seven to forty-one — three votes short of the sixty required to overcome the Senate filibuster.³⁵ The act has been reintroduced in essentially every subsequent Congress. The Senate has never passed it. The most recent vote, in September 2022, failed forty-nine to forty-nine on a party-line basis.

The For the People Act, designated H.R. 1 in both the 116th and 117th Congresses, was introduced by Representative John Sarbanes of Maryland as a comprehensive democracy-reform bill encompassing campaign finance, voting rights, and ethics.³⁶ Its campaign finance components included a six-to-one federal small-donor matching program for House candidates who refused PAC money and capped individual contributions, the DISCLOSE Act provisions on dark-money disclosure, the Honest Ads Act on online political advertising disclosure, and a restructuring of the Federal Election Commission to break the partisan deadlock that has rendered the agency largely inactive since the early 2010s. The act passed the House in March 2019 by a vote of two hundred thirty-four to one hundred ninety-three. The Senate Majority Leader, Mitch McConnell, declined to bring it to the Senate floor. The act was reintroduced in the 117th Congress as both H.R. 1 and S. 1 in January 2021, passed the House in March 2021 by a vote of two hundred twenty to two hundred ten, and reached the Senate floor in June 2021. The motion to proceed failed fifty to fifty on a party-line vote, with no Republican senator voting in support and Democratic senators Joe Manchin and Kyrsten Sinema declining to support the necessary filibuster carve-out. The Freedom to Vote Act, a slimmed-down successor negotiated by Senator Manchin to address his procedural objections, retained the small-donor matching program and the DISCLOSE provisions; it failed cloture in October 2021 and again in January 2022.³⁷ The campaign-finance reform legislation that has been the principal contemporary vehicle for restoring the pre-Buckley regulatory framework has, in three Congresses across two presidential administrations, not advanced beyond a Senate procedural vote.

The constitutional amendment route has been pursued in parallel and has failed in the same pattern. Senator Tom Udall of New Mexico introduced S.J. Res. 19 in the 113th Congress in 2014, an amendment that would have authorized Congress and the states to regulate campaign spending. The amendment was approved by the Senate Judiciary Committee in July 2014 by a vote of ten to eight. The Senate floor cloture vote on September 11, 2014 failed fifty-four to forty-two — six votes short of the sixty required to proceed and twenty-five votes short of the sixty-seven required to actually pass a constitutional amendment.³⁸ Udall reintroduced the amendment in the 115th and 116th Congresses; neither version received a floor vote. Roughly twenty-two states have, through their legislatures or by ballot initiative, passed resolutions calling on Congress to propose a constitutional amendment to overturn Citizens United.³⁹ Article V of the Constitution requires two-thirds of the state legislatures to call a constitutional convention in order to bypass the congressional proposal route; that threshold has not been approached. The amendment route, like the statutory reform route, has not produced an outcome.

The reason both routes have failed is the same reason both routes have been pursued: the constituency that benefits from the current campaign finance regime is the constituency that funds the politicians who would have to vote to change it. The Republican Party has, since approximately the mid-2000s, treated the post-Buckley regime as constitutive of its legislative coalition; corporate political spending, dark-money 501(c)(4) operations, and the donor-class small-population funding base are now structural to Republican congressional politics. The Republican coalition’s principal donors and the Republican coalition’s principal policy positions — corporate tax reduction, regulatory rollback, opposition to organized labor — are tightly linked through a network of donor-funded organizations, lobbying firms, and post-government employment opportunities that the existing regime makes possible. Republican defense of Citizens United, which is now explicit in party platform documents and in the on-record statements of Republican leadership, is consistent with this structural position. The Republican coalition is not going to vote to dismantle the regime that funds it.

The Democratic coalition’s relationship to the regime is more conflicted and is, in its own way, more revealing. The Democratic Party has, in its public communications, opposed Citizens United since the day the decision was issued; the party platform calls for its overruling; Democratic-introduced legislation in every recent Congress has attempted to restore some component of the pre-Buckley framework. The party’s operational relationship to the regime, however, has been one of accommodation rather than opposition. Democratic candidates raise large sums from the same donor class that funds Republican candidates; Democratic-aligned super PACs (American Bridge, Senate Majority PAC, Priorities USA, House Majority PAC) operate at scale alongside their Republican counterparts; the ActBlue small-donor platform, which represents a genuine alternative model of campaign finance, has not been paired with a categorical Democratic refusal of large-donor and super-PAC funding. The Democratic record has been one of rhetorical opposition to a regime the party operates within. The structural commitment to dismantling the regime — the willingness to forgo the funding the regime makes available, in the name of restoring the framework that would prohibit it — has not, in any major Democratic campaign, been evident. The closest approximation came in Bernie Sanders’s 2016 and 2020 primary campaigns, both of which refused PAC and corporate funding and built fundraising operations on small-donor contributions through ActBlue. Both campaigns lost the nomination. The model has not become institutional in the broader Democratic operation.

The third structural obstacle is the Supreme Court itself. The post-Buckley constitutional doctrine has not merely failed to evolve in directions that would permit broader regulation; it has continued to evolve in directions that further constrain regulation. Americans for Prosperity Foundation v. Bonta (2021) struck down a California requirement that nonprofit organizations disclose their major donors to the state attorney general for tax-enforcement purposes, on the theory that compelled disclosure burdened First Amendment associational rights even where the disclosure was not public.⁴⁰ The decision substantially narrowed the constitutional space for state-level disclosure requirements that might have substituted for the federal disclosure requirements the DISCLOSE Act has failed to enact. The Roberts Court’s general direction on campaign finance has been to extend Citizens United’s reasoning rather than to constrain it. A statutory reform that produced a campaign finance regime substantially more restrictive than the post-McCutcheon status quo would face a likely constitutional challenge in a court that has shown no inclination to revisit its post-Buckley line. A constitutional amendment, which would moot the Court’s role, faces the procedural and political obstacles described above.

The off-the-shelf reform that comes closest to immediate adoption — public small-donor matching, modeled on New York City’s eight-to-one program and on Seattle’s democracy-vouchers experiment⁴¹ — has been demonstrated to work, has produced documented effects on donor diversity and candidate viability, and is currently operational at the municipal scale in several American jurisdictions. Its adoption at the federal scale would not require a constitutional amendment; it would not require overturning Citizens United; it would not require restricting any existing form of political spending. It would simply create a parallel public-financing channel through which candidates who chose to participate could be competitive without large-donor funding. The For the People Act’s small-donor matching provision is an instance of this approach. The provision has not passed the Senate, in three Congresses, despite never having been the subject of a major constitutional objection. The obstacle is not constitutional. It is the same obstacle that has prevented every other component of campaign finance reform: the constituency that benefits from the current regime is the constituency whose votes would be required to change it.

The Cascade That Closes

The previous two essays in this series each ended at the same wall: the absence of a political coalition capable of enacting the structurally indicated reforms.⁴² The AI Implosion concluded that the political coalition required to enact a fourth-settlement response to the AI labor cascade has not assembled in the United States in fifty years. The Quiet Foreclosure concluded that the coalition required to enact the four-component housing-policy reform does not currently exist. Each essay, considered alone, presented the absence of the coalition as a particular feature of the particular policy area it addressed. Considered together, the absences point at a common cause that the present essay has now described.

The mechanism is not subtle. The AI labor cascade’s principal beneficiaries are the owners of the firms that the AI transition is making more profitable per worker. The housing carrying-cost cascade’s principal beneficiaries are the institutional holders of single-family residential who acquire the assets the cascade shakes loose. These two beneficiary populations overlap substantially with each other and overlap nearly completely with the donor class that funds the political coalitions that would have to enact the reforms either essay identified. The political-economy circle, viewed at the scale of the underlying mechanism, closes. The same population of very wealthy individuals and corporate actors that benefits from the structural transfers diagnosed in the previous essays is the population whose campaign contributions, super PAC funding, dark-money operations, and lobbying expenditures shape the legislative response to those transfers. The legislative response that would constrain the transfers is, in the existing configuration, the legislative response that the donor class will not fund. The legislative response that the donor class will fund is the response that does not constrain the transfers. The system selects, in equilibrium, for the second.

The structural feedback this produces is the principal cause of the policy paralysis the previous essays diagnosed without naming. It is not principally that elected officials are unaware of the AI labor cascade or of the housing carrying-cost cascade. The relevant officials are well-briefed; the relevant academic literature has been read; the relevant policy proposals have been drafted and circulated. The reason the proposals do not advance is that they are opposed, with the full intensity that the contemporary lobbying complex can mobilize, by the constituency that funds the politicians whose votes would be required to advance them. The opposition is not bribery and is not corruption in the prosecutorial sense; it is the dependence corruption Lessig diagnosed, operating as designed.

The argument has comparative force. Every other major democracy operates a campaign finance regime substantially more restrictive than the contemporary American regime.⁴³ The United Kingdom prohibits paid political advertising on television and radio entirely; it caps per-constituency candidate spending and per-party national spending in the low tens of millions of pounds for a general election. Canada prohibits corporate and union contributions to federal parties and candidates outright, caps individual contributions in the low thousands of Canadian dollars per year, and operates a partial public-financing regime. Germany operates a mixed public-private regime with strict transparency requirements and an absolute cap on state subsidies. France prohibits corporate contributions to candidates and parties and operates strict spending caps with partial public reimbursement. None of these countries has collapsed into authoritarianism. None of them is rated as anything other than fully democratic by Freedom House, the V-Dem Institute, or the Economist Intelligence Unit. The empirical premise of the post-Buckley constitutional doctrine — that strict campaign finance regulation chokes democratic discourse and represents an authoritarian tendency that the First Amendment must guard against — is not supported by the comparative record. Every other democracy has done what the American jurisprudence has held the First Amendment forbids, and they remain democracies. The American system is the outlier.

The American system is also, by extensive empirical evidence, the system in which policy outcomes track elite preferences in ways that other systems do not. Martin Gilens of Princeton (now UCLA) and Benjamin Page of Northwestern University, in a 2014 Perspectives on Politics article that has become the most-cited contemporary work on American policy responsiveness, analyzed approximately one thousand seven hundred federal policy issues between 1981 and 2002 for which national survey data measured public preferences disaggregated by income.⁴⁴ Their finding, in a sentence the authors took some care to state precisely: “Economic elites and organized groups representing business interests have substantial independent impacts on U.S. government policy, while average citizens and mass-based interest groups have little or no independent influence.” The specific quantitative result is that, when the policy preferences of high-income citizens diverge from the preferences of median-income citizens, federal policy outcomes track the preferences of high-income citizens with statistical robustness, while the apparent influence of median-income preferences on those outcomes is statistically indistinguishable from random. Subsequent critique has narrowed the substantive significance of the finding by pointing out that high-income and median-income preferences agree on most issues and that the divergent-preference subset is a small share of the policy universe.⁴⁵ The narrowing critique is correct on its own terms and changes the picture less than its proponents have argued. The divergent-preference subset is small but is not random; it concentrates on questions of taxation, financial regulation, social spending, and labor policy — the areas in which the AI labor cascade and the housing carrying-cost cascade are operating. On precisely the policy questions that this series of essays has concerned, the system has been demonstrated, by the most comprehensive available empirical work, to produce outcomes that track elite preferences and to be statistically indifferent to ordinary-citizen preferences. The previous essays’ diagnosis of the absence of an enacting coalition is not an artifact of insufficient public attention. It is the system functioning as it has been demonstrated to function.

The same comparative and empirical record establishes, finally, that the equilibrium is not stable in perpetuity. Public opinion on the campaign finance system, across two decades of consistent polling, shows large bipartisan majorities favoring substantial reform. Pew Research Center polling has found, across multiple waves between 2015 and 2018, that approximately seventy-five to eighty percent of Americans support imposing limits on the amount of money individuals and organizations may spend on political campaigns; the bipartisan composition of the supportive majority routinely includes seventy percent or more of Republican identifiers.⁴⁶ Bloomberg Politics polling in 2015 found seventy-eight percent supporting a constitutional amendment to overturn Citizens United, with majorities of both party affiliations. The campaign finance system that has produced the contemporary donor class is not democratically legitimate by any conventional measure of democratic legitimacy; it is sustained against the express preferences of substantial bipartisan majorities by the constitutional doctrine and the political-economy feedback loop this essay has described. The conditions under which the equilibrium might be broken are not, on the historical record, available within ordinary politics. They have, in past cycles of comparable structural mismatch — the Progressive Era, the New Deal — been produced by external pressure that exceeded the system’s capacity to absorb. The pressure that the AI labor cascade and the housing carrying-cost cascade are now beginning to produce, on the trajectory the previous essays described, will, in the next decade, exceed the contemporary system’s absorptive capacity. The question is not whether the equilibrium will be challenged. It is what the political response will look like when the challenge arrives.

What’s at Stake

The American political economy has confronted the structural relationship between concentrated wealth and democratic government three times before, and each time produced a settlement distinctive to its moment.

The Progressive Era settlement of the 1900s and 1910s, the period in which Theodore Roosevelt called for and Benjamin Tillman sponsored the act that opened this essay, established the principle that the federal government could legitimately regulate corporate political activity in the public interest. The settlement extended through the Federal Trade Commission, the Clayton Act, the Federal Reserve, the Sixteenth and Seventeenth Amendments, and the broader institutional architecture of the regulatory state. It was the first settlement in which the American government formally constrained the political activities of accumulated capital.

The post-Watergate settlement of the 1970s extended the Progressive Era principle through the modern statutory framework — FECA, the FEC, contribution limits, public financing of presidential elections, the Honest Politics Act of the late 1980s — and produced the regulatory regime that operated, in attenuating form, through the early 2000s. The settlement was not principally a Democratic project; the post-Watergate Congress was bipartisan in its participation in the framework, and the FECA amendments were signed by President Ford. It reflected a shared institutional understanding that the Watergate-era abuses had revealed structural vulnerabilities in the campaign finance system that required a structural response.

The Buckley-Citizens-United-McCutcheon dismantling of the post-Watergate settlement constitutes a third settlement of a kind, though it has not been described in those terms in conventional accounts. It is a settlement constructed primarily through the judicial branch rather than the legislative; it has produced a regulatory regime that the major political parties’ platforms, and the express opinion of the relevant legislative majorities, would not have produced independently. The regime is, in important respects, judicially imposed against the legislative will. The settlement is incomplete in the sense that no constitutional amendment has formalized it; it is durable in the sense that no statutory reform within the Court’s permissible reading of the First Amendment can substantially undo it.

A fourth settlement is now required, and it is structurally different from any of the three that preceded it. The Progressive settlement created federal regulatory capacity where none had existed. The post-Watergate settlement built statutory infrastructure on top of constitutional foundations the courts had not yet challenged. The Buckley-line settlement dismantled, through judicial interpretation, the statutory infrastructure the post-Watergate Congress had built. The settlement now required is a re-establishment of the structural-equality framework that the post-1976 jurisprudence has read out of American constitutional law, on a basis that the judicial doctrine cannot dismantle. The mechanism for that re-establishment, in practice, is the constitutional amendment route the Udall amendment and its successors have pursued. The political coalition required to assemble the two-thirds congressional majorities and three-quarters state ratifications such an amendment requires has not, in the contemporary period, been within reach. The conditions under which it might come into reach are the conditions the previous essays in this series have described as the trajectory of the next decade.

The fourth settlement, when and if it assembles, will have to do four things simultaneously. It will have to enact a constitutional amendment establishing that the structural-equality rationale Buckley rejected is, in fact, available to support legislative regulation of campaign spending — that, in the language The Intelligent Party’s policy documents propose, money is not speech and corporations are not people for First Amendment purposes in the campaign finance context.⁴⁷ It will have to dismantle the super PAC and 501(c)(4) regimes that the Citizens United and SpeechNow line of decisions established, through statutory contribution limits on independent-expenditure-only committees and through the disclosure requirements the DISCLOSE Act has repeatedly attempted to enact. It will have to establish public financing of federal elections at a scale that allows candidates to be competitive without large-donor funding — the small-donor-matching mechanism the For the People Act repeatedly proposed, scaled to a six-to-one or higher ratio, paired with the democracy-voucher system the Seattle program demonstrated and the New York City program has refined over thirty years of operation. It will have to address the revolving-door mechanism through which the lobbying complex sustains itself, through cooling-off periods substantially longer than the current one- and two-year statutory periods and through restrictions on the post-government employment categories that have been used to circumvent the existing rules. None of these is, on its own, a solution. The four together are a re-establishment of the regulatory framework the Tillman Act through the FECA amendments had built, retrofitted for the corporate, technological, and judicial conditions of the twenty-first century rather than the corporate conditions of 1907 or the post-Watergate conditions of 1974.

The Vienna of the campaign finance world, the comparative model that demonstrates what the alternative looks like, is in this case not a single foreign country but the broader comparative record. Every other major democracy operates within some version of the framework the United States operated within from 1907 to 1976; none of them has experienced the political-economy concentration that the post-Buckley American system has produced; none of them is regarded as authoritarian for having maintained the regulation. The American departure from the comparative pattern is not a matter of principled democratic theory; it is the outcome of a particular constitutional doctrine, established by a particular Court, that the post-Watergate Congress did not anticipate and that subsequent Congresses have been unable to reverse.

The political coalition that would produce the fourth settlement does not currently exist. It has not existed, in assembled form, since the post-Watergate moment of the mid-1970s, when the conditions for its assembly were the public revulsion at the documented abuses of a particular administration. The conditions for a comparable assembly in the contemporary period are not principally a matter of public revulsion — public opinion has been steadily and overwhelmingly in favor of reform for two decades, without producing the assembly — but a matter of structural rupture in the political-economy equilibrium that the post-Buckley jurisprudence has stabilized. The previous essays in this series have described what such a rupture might look like: the AI labor cascade compressing the wage base on which the political-economy equilibrium depends; the housing carrying-cost cascade liquidating the household wealth base on which the political-economy equilibrium depends. The arithmetic argument the AI Implosion ended on — that a multi-trillion-dollar economy cannot operate without wage-earning consumers, and that the AI transition will not, on its current trajectory, produce wage-earning consumers in numbers sufficient to sustain the economy — applies with the same force to the political-economy equilibrium that the campaign finance system supports. The donor class can fund the political coalitions that prevent reform. It cannot fund the consumer demand that the underlying economy requires. The mechanical contradiction the AI Implosion identified at the level of the macro economy is the same mechanical contradiction that operates at the level of the political economy of campaign finance. Both contradictions resolve, on their current trajectories, in the same direction: a structural break that exceeds the absorptive capacity of the system in which the contradiction has been operating.

The realism that the previous two essays in this series called for — the construction of frameworks that have not been built, against political coalitions that have not yet assembled — applies with particular force in this case, because the framework that needs to be built is the framework whose construction is the precondition for the construction of the others. Without a fourth settlement on campaign finance, the fourth settlement on AI labor and the fourth settlement on housing will not assemble; the donor class will continue to fund the politicians who block them. With a fourth settlement on campaign finance, the political-economy circle that has prevented the other settlements opens; the legislative coalitions that the previous essays have described as currently impossible become structurally available. The campaign finance settlement is not one item in the policy stack the next decade requires. It is the enabling condition for the rest of the stack.

The essay about the campaign finance dismantling that someone writes in 2040 will describe either the construction of the fourth settlement sketched above or its refusal. The constituency that would benefit from the settlement is the constituency the campaign finance system has been demonstrated to systematically under-represent: the eighty percent of Americans whose policy preferences, on the questions where they diverge from the preferences of the donor class, are statistically indistinguishable from random in their effect on federal policy. That constituency is, in 2026, the great majority of the American population. Its political assembly has not yet occurred. The mechanical conditions for its assembly, on the trajectory the previous essays have described, will be in place within the next decade.

Theodore Roosevelt closed his 1905 message with a sentence that has been quoted, in various paraphrases, throughout the century of campaign finance regulation that his message helped initiate: “There is no enemy of free government more dangerous and none so insidious as the corruption of the electorate.”⁴⁸ The corruption Roosevelt was describing was the specific contemporary practice of corporate contributions to political campaigns, the practice the Tillman Act was passed to address. The contemporary form of the corruption is structurally larger, judicially protected, and constitutionally insulated from ordinary legislative response. The framework Roosevelt called for and Tillman enacted operated, in attenuating form, for sixty-nine years. The framework that has replaced it has now operated for nearly fifty years. The question the next decade will answer is whether the American political economy, on the trajectory the previous essays have described, will produce the conditions under which the original framework can be reconstructed at the scale the contemporary problem requires, or whether the alternative — what the AI Implosion called the third path, systemic collapse — will arrive first.

The framework will be rebuilt. The question is what will have to be true of the country first.

---

Notes