The Sweep of the Electoral Power

36 Const. Comment. 1

Nicholas O. Stephanopoulos*

Congress is on the cusp of transforming American elections. The House recently passed a bill that would thwart voter suppression, end gerrymandering, and curb the undue influence of the rich. Something like this bill could soon become law. In this Article, I provide a multilayered foundation for such sweeping electoral legislation. From a theoretical perspective, first, I argue that Congress poses less of a threat to democratic values than do the states or the courts. It’s more difficult for a self-interested faction to seize control of federal lawmaking than to capture a state government or a judicial body. Second, surveying the history of congressional electoral regulation, I contend that it’s remarkably benign. Most federal interventions have advanced democratic values—in marked contrast to many of the states’ and the courts’ efforts.

Third, I show that current law grants Congress the expansive electoral authority that, normatively, it ought to possess. In particular, the Elections Clause, the Guarantee Clause, and the Fourteenth Amendment’s Enforcement Clause combine to empower Congress over most electoral levels and topics. And fourth, returning to the House’s recently passed bill, I maintain that its most controversial elements are constitutional under the applicable doctrine. In fact, Congress could venture considerably further than, to date, it has tried to go. Together, these points should hearten legislators when they next turn to the project of electoral reform. Not only is aggressive federal action permissible in the American political system—it may be the only way to save it.


Although it came and went without much fanfare, March 8, 2019 was a momentous day in the history of American election law. On that day, the House of Representatives passed the For the People Act, the most sweeping electoral reform bill ever to win the support of a majority of a chamber of Congress.[1] Among (many) other things, the Act would have mandated automatic voter registration,[2] ended felon disenfranchisement,[3] required states to redraw district lines using independent commissions,[4] established a system of public financing for congressional candidates,[5] and restructured the agency responsible for regulating money in politics.[6] This set of proposals was more ambitious than Congress’s most recent election laws—the National Voter Registration Act of 1993[7] and the Help America Vote Act of 2002[8]—which were limited to relatively minor aspects of registering to vote and casting ballots. The For the People Act was also more far-reaching than the post-Watergate reforms of campaign finance,[9] which didn’t offer public funds to congressional (as opposed to presidential) candidates. The For the People Act even swept further than the Voting Rights Act of 1965[10] and its Reconstruction antecedents a century earlier.[11] Those monumental laws targeted only one evil (racial discrimination in voting) primarily in only one part of the country (the deep South).

The lack of hubbub over the For the People Act—despite its extraordinary content—had a simple explanation. It was just a bill. It wasn’t an enacted law. After being passed by the House, it wasn’t even debated, let alone ratified, by the Senate. Senate Majority Leader Mitch McConnell declared, “I get to decide what we vote on,” and the For the People Act wasn’t on his agenda.[12] McConnell’s opposition to the bill had a simple legal explanation, too.[13] He thought it exceeded Congress’s constitutional authority. In remarks on the Senate floor, he stated, “the Constitution clearly gives State legislatures primary responsibility” for running elections.[14] However, “[d]ecision after decision that our Constitution properly leaves to the States just melts away in this proposal.”[15] So it amounts to a “Federal takeover of elections across the nation” that “upsets th[e] constitutional balance.”[16]

A chorus of conservative commentators echoed McConnell’s view that Congress lacked the power to enact the For the People Act. Former Federal Election Commission member Hans von Spakovsky told a House committee that the bill unconstitutionally “interferes with the ability of states to determine the qualifications of their voters, to secure the integrity of the election process, and to determine the districts and boundary lines of their representatives.”[17] Cato Institute scholar Ilya Shapiro labeled the bill “an unconstitutional abomination” because it “undermines basic principles of federalism.”[18] For the same reason, former Department of Justice official J. Christian Adams testified that the bill was “grotesquely offensive to the Constitution that vests power in the state legislatures to determine the manner of choosing Representatives.”[19] Summing up this position, the National Review, the house organ of contemporary conservatism, editorialized that the bill was “a frontal assault on the Constitution”—“the most comprehensively unconstitutional bill in modern American history.”[20]

To say the least, then, the battle has been joined over Congress’s authority to pass sweeping electoral reforms like those in the For the People Act. This question of congressional clout is my subject in this Article. Surprisingly, scholars haven’t previously examined the full scope of Congress’s power to regulate elections, under the whole array of relevant constitutional provisions.[21] These provisions really are an array; they number more than ten, in total, fixing the metes and bounds of Congress’s electoral authority in more detail than any other congressional power. Yet the existing literature has mostly neglected this unusual profusion of constitutional text, focusing instead on Congress’s ability to legislate under particular clauses or on particular topics.[22] This work is helpful, but it fails to convey the totality of the authority that Congress should and does possess over elections under the entire Constitution. The work tends to shine a narrow beam when what’s needed is a floodlight.

You may have noticed my use of the normative: the electoral power that Congress should enjoy. Before addressing the influence that Congress does wield, under current doctrine, I think it’s important to explain why, in a vast nation like ours characterized by checks and balances, the separation of powers, and federalism, we should prefer congressional electoral regulation to the alternatives: elections regulated by the states or by the courts. The reason is essentially Madisonian. It’s very difficult for a faction, in the Framers’ sense of a group that pursues its own interest at the expense of the public interest,[23] to seize control of the federal government. To do so, the group’s members would have to win election after election in districts, states, and the country as a whole; and they would have to remain allied despite their cultural, political, and geographic differences. It’s therefore unlikely that Congress would enact factional electoral legislation, suppressing or diluting votes to benefit a certain group. It’s more plausible that congressional electoral supervision, when it occurs at all, would arise under more benign conditions: either bipartisan consensus that action should be taken or supermajority control by a single party committed to democratic progress.

In contrast, factional electoral legislation is easier to imagine at the state level. States lack the scale and diversity of the country in its entirety. States’ legislative chambers are also both structured on the same basis—equal district population—and usually don’t allow minorities to block measures from advancing. These features make states more susceptible to capture by a single group that can then manipulate the electoral process to consolidate its hold on power. Factional control of the courts is more readily achievable, too. It takes just five appointments, in particular, to gain a majority on the Supreme Court. These appointments can’t be stopped by the House, the congressional chamber that more accurately reflects the will of the electorate. And when a likeminded President and Senate manage to make these appointments, the Court’s consequent decisions are final under all but the most exceptional circumstances. The checks and balances that restrain the other branches of the federal government, that is, mostly don’t apply to the Court.

It may seem odd to subject the courts to Madisonian factional analysis. The courts, after all, are the heroes of Carolene Products: the impartial arbiters tasked with intervening when self-interested politicians “restrict[] those political processes which can ordinarily be expected to bring about repeal of undesirable legislation.”[24] But one of this Article’s goals is to resist this way of thinking. Yes, the courts can step in when elected officials threaten democratic principles. (For example, one Court, the Warren Court, did so regularly.[25]) But the courts don’t necessarily heed Carolene’s admonition. (In fact, most Courts, including the current Court, have ignored it.[26]) And the courts shouldn’t even be expected to do a better job than Congress defending and improving American democracy. Because Congress is less vulnerable than the Court to factional takeover, it’s less likely to “restrict[] those political processes,”[27] including elections, and more apt to enhance them. So Congress is better suited to playing the role that Carolene once envisioned for the Court.

Madisonian analysis is fine as far as it goes, but more than two hundred years have now passed since the Framing. Does this history substantiate the claim that congressional electoral regulation is generally more constructive than the activities of the states and of the courts? I think it does. The states have been the sources of almost all of America’s democratically subversive policies. In earlier periods these included property and gender requirements for voting, poll taxes, literacy tests, and flagrant malapportionment. Today the states continue to require photo IDs to cast ballots, to deny and dilute minority votes, and to ruthlessly gerrymander their district maps. With the shining exception of the Warren Court, similarly, the Supreme Court has often eroded democratic values in its election law decisions. During Reconstruction, the Court famously neutered the landmark statutes that sought to incorporate African Americans into the political community.[28] In our own time, as I have discussed elsewhere, the Roberts Court has never ruled in favor of a plaintiff alleging a franchise burden, dismantled the nation’s campaign finance and voting rights laws, and announced its impotence to fight gerrymandering.[29]

By comparison, I named most of Congress’s major electoral interventions in the Article’s opening paragraph.[30] I think reasonable observers would agree that these policies serve important democratic ends: the political participation of racial minorities in the case of the laws of the First and Second Reconstructions, the prevention of corruption in the case of the post-Watergate reforms, and better election administration in the case of Congress’s most recent efforts. Even a critic would have to concede that there isn’t a democratic atrocity to be found in the roster of congressional action. There’s no attempt to disenfranchise or to disempower. Nor is there any ratification of these practices, averring their compliance with the Constitution.

To be sure (and to return to Madison[31]), members of Congress are no more angelic than state politicians and judges. So my contention isn’t that Congress’s electoral interventions have been driven exclusively by disinterested considerations. In fact, I want to acknowledge that most of these statutes have had significant partisan bases. By enfranchising African Americans, Reconstruction Republicans hoped to preserve their party’s influence in the South; by curbing money in politics, post-Watergate Democrats thought their party’s less amply funded candidates would benefit; and so on. My argument, then, is that on the occasions when Congress has regulated elections, it has typically done so for both democratic and partisan reasons. Not being angels, members of Congress have rarely sacrificed their self-interest for the wellbeing of American democracy. But nor have they commonly compromised American democracy for their own advantage.

Accordingly, theory suggests, and history confirms, that congressional electoral regulation is generally preferable to the activities of the states and of the courts. So it would be sensible for the law to endow Congress, the most trustworthy institution in this context, with sweeping authority over elections. Fortunately, that’s exactly what current doctrine does. I’ll just mention three of the relevant constitutional provisions here—the Elections Clause, the Guarantee Clause, and the Fourteenth Amendment’s Enforcement Clause—because they establish the basic architecture of Congress’s electoral power. The Elections Clause of Article I, first, enables Congress to “make or alter . . . Regulations” of “[t]he Times, Places and Manner of [congressional] Elections.”[32] It thus gives Congress close to full control of congressional elections: the “‘authority to provide a complete code’” for these races,[33] to quote one of the most unexpected passages ever penned by Justice Scalia.

Second, if the Elections Clause empowers Congress directly with respect to congressional elections but only incidentally with respect to state elections, the Guarantee Clause of Article IV does just the opposite. Through that provision, Congress “guarantee[s] to every State . . . a Republican Form of Government.”[34] How states run their own elections plainly bears on whether they respect the principle of popular sovereignty that forms the core of republicanism.[35] In contrast, how states administer another sovereign’s elections—those of the federal government—is less closely tied to the essence of republican rule. This distinction isn’t critical, though, because the courts have held for almost two centuries that Guarantee Clause issues are nonjusticiable.[36] Consequently, when Congress legislates under the Guarantee Clause, not only is its power substantively broad, it’s judicially unreviewable.

And third, the Fourteenth Amendment’s Enforcement Clause undergirds the interlocking plates of the Elections Clause and the Guarantee Clause by authorizing Congress to prevent and remedy Fourteenth Amendment violations at any level, state or federal.[37] The list of electoral practices that can violate the Fourteenth Amendment is long; in fact, it comprises most of the field of election law. The provision can be offended by burdens on voting, restrictions on political parties, racially discriminatory measures, malapportionment, partisan gerrymandering, and geographically variable voter treatment.[38] All these potential breaches are thus fair game for congressional regulation, even under the current Court’s dim view of the Enforcement Clause.[39] There only has to be “congruence and proportionality” between Congress’s policies and the many Fourteenth Amendment harms that can arise in the electoral domain.[40]

While incomplete, this tour of the legal landscape suffices to show that the For the People Act’s detractors are wrong. Even the bill’s most controversial elements lie within Congress’s electoral authority, and Congress could actually reach considerably further, if it were so inclined. Consider the bill’s enfranchisement of ex-felons in federal elections,[41] and assume it extended to state elections, too. Under current doctrine,[42] the Elections Clause enables Congress to regulate essentially all aspects of congressional elections, including being able to vote after completing a prison sentence. Under the Guarantee Clause, likewise, Congress could reasonably conclude that the disenfranchisement of ex-felons in state elections is inconsistent with republican government, and in any event, its judgment would be nonjusticiable. And to the (significant[43]) extent that states stripped ex-felons of the franchise for racially discriminatory reasons, the Fourteenth (and Fifteenth) Amendments would reinforce the other pillars of congressional electoral power, with respect to state and federal elections alike.

Or take the For the People Act’s mandate that congressional districts be designed by independent commissions,[44] and say this requirement also applied to state legislative districts. How congressional districts are crafted involves both the “Places” and the “Manner” of congressional elections, and so is in the heartland of Congress’s Elections Clause authority.[45] Similarly, how state legislative districts are drawn implicates the norm of popular sovereignty, meaning that Congress could plausibly (and unreviewably) decide that the people are not sovereign when politicians are free to gerrymander. And while the current Court recently reversed decades of precedent in holding that partisan gerrymandering is nonjusticiable,[46] it still conceded that the practice, when extreme enough, transgresses the Fourteenth Amendment.[47] Congress would thus prevent and remedy Fourteenth Amendment violations by insisting on the use of commissions for congressional and state legislative districts.

An important caveat is in order here. Both in these examples and throughout the Article, I only address Congress’s authority to enact electoral regulations. I don’t comment on the distinct constitutional doctrines that these laws might violate. Put another way, the Constitution’s power-conferring provisions are my present focus while its power-limiting elements are beyond this project’s scope.[48] These power-limiting elements include the First Amendment, which has been construed to bar many campaign finance restrictions[49]; the equal protection principle of colorblindness, which threatens race-conscious remedies[50]; and the federalism values of anti-commandeering[51] and equal sovereignty,[52] which sometimes prevent Congress from ordering the states or treating them unequally. Thanks to these and other doctrines, electoral regulations that Congress has the authority to enact aren’t necessarily constitutional. So my thesis here is just that Congress’s electoral power should be, and is, vast. I don’t maintain that every exercise of this power is valid.

The Article’s structure mirrors the discussion to this point. I begin with constitutional theory in Part I: why, in a regime organized like ours, we should expect congressional electoral regulations to be less democratically corrosive than the activities of the states and of the courts. Next, in Part II, I survey the history of Congress’s electoral interventions, showing that they have typically been driven by a mix of lofty democratic and less high-minded partisan motives. In Part III, I then turn to current law, explaining that it endows Congress with enormous authority over elections, at both the state and federal levels, under a range of constitutional provisions. Lastly, in Part IV, I consider a number of electoral reforms, some of them included in the For the People Act, and contend that Congress’s sweeping electoral power encompasses them all.

  1. Constitutional Theory

I’m not an originalist; a clause’s meaning at the time it was written is probative, in my view, but not dispositive. Nor am I an admirer of certain aspects of the Constitution; just think of the unrepresentative Senate, the Rube Goldberg Electoral College, the premodern omission of positive rights, and so on. But I do find persuasive one of Madison’s most crucial arguments about one of the Constitution’s most crucial institutions. Congress, Madison asserted, would rarely pass bad legislation—laws abridging people’s rights—for two reasons. One was the huge scale and diversity of the country from which Congress was drawn, which would make it difficult for a single nefarious faction to command a congressional majority. The other was the series of hurdles that would have to be overcome for bad bills to become bad statutes: approval by two differently constituted legislative chambers as well as the consent of the President.

In this Part, I push the familiar Madisonian argument in two new directions. First, I apply it to the electoral context, adding to the list of bad legislation that Congress should seldom pass laws that undermine democratic values. Second, I make the argument comparative rather than absolute: Congress should be less likely than other actors, like the states and the courts, to imperil American democracy. I further respond to the objection that the constitutional commitment to federalism requires the limitation of Congress’s electoral authority. Not so, I contend, if federalism matters because of its purported benefits and not just for its own sake. These benefits aren’t diminished, and may even be amplified, by an expansive congressional electoral power.

One more note before proceeding: When I speak here of “American democracy” and “democratic values,” I’m largely agnostic as to their content. I have expressed my view elsewhere that the alignment of the government’s outputs with the people’s preferences should be the guiding star of election law.[53] But for present purposes, I’m willing to group alignment with several other democratic principles: political participation, civic deliberation, governmental responsiveness, minority representation, the prevention of corruption, the efficient administration of elections, and the like. My claim is that Congress is less apt to threaten this full set of tenets than are the states and the courts.

  1. Congress and the States

In The Federalist No. 10, Madison ingeniously linked the scope of a representative government to the quality of its output.[54] The government of a large, heterogenous nation would “take in a great[] variety of parties and interests.”[55] These many groups would continuously form and break alliances, negotiate and renegotiate with one another. In this unending, multipronged struggle, it would be “less probable that a majority of the whole [would] have a common motive to invade the rights of other citizens.”[56] Even if an invidious purpose did animate a majority, “it [would] be more difficult for all who feel it to discover their own strength, and to act in unison.”[57] “Extend the sphere,” then, and you reduce the risk of governmental “oppression.”[58]

This logic plainly applies to Congress: the representative assembly of the vast continental democracy that is the United States.[59] Does it also apply to the electoral domain? I think so. Laws that disenfranchise certain people or dilute their electoral influence are prime examples of governmental “oppression” that “invade[s] the rights of other citizens.”[60] It’s plausible, too, that a faction would want to pass these laws (because they would augment its political power), but that it would be unable to do so. The malignant group might not be sufficiently large to comprise a congressional majority. Or even if big enough, its members might be too culturally and ideologically diverse, drawn as they would have to be from the whole country, to legislative effectively. In either case, the democratically offensive bills would fail to become statutes, stymied by Madison’s extended sphere.

While this rationale for trusting Congress not to enact abusive policies would reach any extended republic, Madison’s other reason was specific to the U.S. Constitution. The Constitution separated powers not just among the branches of the federal government, he pointed out in The Federalist No. 51, but also within the legislative branch. It “divide[d] the legislature into different branches” and “render[ed] them, by different modes of election and different principles of action, as little connected with each other” as possible.[61] A faction intent on oppressive action would thus have to capture majorities of both the House, with its members allocated based on states’ populations, and the Senate, with its members representing all states equally. The Constitution further checked and balanced congressional authority, Madison continued, by authorizing the President to veto bills. This “negative on the legislature” was another “natural defense” against a faction with tyrannical aims.[62] To implement its agenda, the group would have to control not just both congressional chambers but the presidency as well.[63]

Again, it’s obvious that this point holds in the electoral context. A faction hoping to deny or dilute certain people’s votes would have to dominate not just the House, constituted one way, but also the Senate, organized another way, and the President, too, selected along still other lines. In most cases, the group would be unable to achieve this much power in this many institutions. So democratic values would usually survive intact thanks to the Constitution’s separation of powers and checks and balances.[64]

A logical question here is whether Madison’s arguments are still valid. Is it still hard for a faction to win enough influence to pass democratically objectionable legislation? In one respect, it’s harder. There are now more veto points in Congress than there were at the Framing due to committees and the Senate filibuster. In both congressional chambers, bills must generally begin in particular committees and win their approval before advancing to chamber-wide votes.[65] So a faction would need majority support not just in both full chambers but also in their relevant committees. Similarly, almost all electoral bills can be filibustered in the Senate, in which case their passage requires sixty votes rather than fifty.[66] So a faction would have to command a Senate supermajority to successfully subvert American democracy.[67]

However, another modern development—the rise of polarized parties—may cut the other way, at least in some circumstances. Madison was a proto-pluralist who thought that many shifting groups would perpetually jostle for advantage in Congress.[68] In recent times, though, Congress has looked nothing like this. To the contrary, it has been comprised of two partisan blocs that are increasingly monolithic internally and divergent ideologically from each other.[69] In this polarized environment, it could be easier for a faction with undemocratic aspirations to enact its preferred policies. The group would no longer have to compete with myriad other interests in the separate arenas of the House, the Senate, and the presidency. Instead, it would only have to convince one of the major parties of its views, and then wait for this party to gain unified control of the federal government. With unified control would come the opportunity for action, notwithstanding Madison’s extended sphere and the Constitution’s separation of powers and checks and balances.[70]

But the faction’s wait for unified control might be a long one. Over the last half-century, divided government has been almost three times as common in Washington as unified government.[71] The moments when a single party has earned unified control and a Senate supermajority have been even more fleeting.[72] Moreover, even in this era of polarized parties, they’re not perfectly internally homogeneous. So it might not suffice for a group with undemocratic goals to dominate a party with supermajority control of the federal government. A few dissenting voices within the party could still sink the group’s legislative ambitions. Lastly, polarization isn’t an exclusively federal phenomenon. It has also transformed state governments in recent years—in fact, to a greater degree than Washington in many cases.[73] So even if polarization has made it easier for Congress to erode democratic values than in the past, it hasn’t necessarily made Congress more likely to do so than the states. It’s to this issue of relative institutional threat that I now turn.

Madison certainly saw the states as greater dangers to people’s rights than the newly created federal government. In the spring before the Philadelphia convention, he wrote an entire pamphlet on the “vices” of the states’ “political system[s].”[74] Chief among these was the “injustice of the laws of the States,” which manifested itself in “base and selfish measures,” “unjust violations of the rights and interests of the minority,” and “notorious factions and oppressions.”[75] So “alarming” was this “injustice” that it “question[ed] the fundamental principle of republican Government,” namely, that it is “the safest Guardian[] both of public Good and private rights.”[76] In more modern times, Justice Scalia has channeled Madison in another surprising passage. “Dispassionate objectivity,” he wrote in a 1989 concurrence, is “substantially less likely to exist at the state or local level,” where “discrimination against any group finds a more ready expression.”[77] A “heightened danger of oppression from political factions” thus looms “in small, rather than large, political units.”[78]

The reasons for Madison’s and Justice Scalia’s distrust of the states are straightforward. They’re simply the opposite of the reasons for expecting Congress generally not to imperil American democracy. First, even the largest of the states are still smaller and less diverse than the nation as a whole. So state politics must feature fewer, bigger factions than federal politics; and it must be less difficult for a group keen on vote denial or dilution to seize control of the government. As Madison put it in The Federalist No. 10, “the same advantage . . . in controlling the effects of faction” that is “enjoyed by a large over a small republic” is also “enjoyed by the Union over the States composing it.”[79]

Second, not only do the states lack the country’s extended sphere, they don’t fully share its separation of powers and checks and balances either. Since the reapportionment revolution of the 1960s, state senates have been constituted on exactly the same basis—one person, one vote—as state houses.[80] A party that earns a majority in one state legislative chamber can thus usually anticipate winning a majority in the other chamber, too. At present, notably, control of just one state legislature is divided.[81] Additionally, the gubernatorial veto is often less potent than its presidential counterpart. In more than a dozen states, governors’ objections can be overridden by less than the two-thirds vote that’s necessary in Congress.[82] In a few states, governors can’t veto certain critical electoral bills—new district plans—at all.[83] And no state currently recognizes an equivalent of the Senate filibuster: an almost universally applicable supermajority vote threshold. Some states do require supermajorities for tax increases.[84] But these criteria never extend to any and all legislation. In Madison’s words, this time in The Federalist No. 51, even if the federal Constitution doesn’t “perfectly” “give to each department [the] power of self-defense” against legislative overreach, the states’ constitutions are “infinitely less able to bear such a test.”[85]

To be clear, my claim here is a negative one: that Congress is less likely than the states to undermine democratic values, not that it’s more apt than them to improve American democracy.[86] Unfortunately, the same structural attributes that thwart most anti-democratic legislation also prevent the adoption of many pro-democratic laws. Madison’s extended sphere means that proponents of desirable reforms must bargain, frequently unsuccessfully, with many other groups. Likewise, the Constitution’s separation of powers and checks and balances ensure that these advocates often lack enough influence, across enough institutions, to pass their beneficial policies. To illustrate, consider the For the People Act, which, among other things, would have facilitated voting, curbed candidates’ dependence on private financing, and ended the scourge of gerrymandering.[87] For the first time ever, the bill’s backers won majority support in the House: the agreement of many different interests (albeit all within the Democratic Party) that the bill was worthwhile. But this remarkable achievement still left the bill nowhere near enactment. There weren’t sixty senators (enough to break a filibuster) willing to vote for the bill; in fact, there weren’t even fifty senators ready to debate it. The President also adamantly opposed the bill. The For the People Act thus died a quiet and predictable death: another victim, like so much anti- and pro-democratic legislation, of our Madisonian system.[88]

By the same token, the states can be fonts of good ideas in addition to bad ones. Their more limited spheres make it easier for groups with pro-democratic agendas to acquire enough clout to implement their programs. So do the states’ less separated powers and less stringent checks and balances. It shouldn’t be a shock, then, that several elements of the For the People Act—the same bill that Congress couldn’t pass—have been adopted by numerous states. Nineteen states provide for automatic voter registration[89]; twenty states automatically restore ex-felons’ voting rights upon their release from prison[90]; fourteen states entrust redistricting to independent commissions[91]; five states offer public financing to legislative candidates[92]; and so on. These policies couldn’t run the fearsome gauntlet that’s required for congressional authorization. But they were able to clear the lower hurdles for enactment in quite a few states.[93]

Even with these caveats, my argument remains fundamentally Madisonian, linking design features of the federal government and of the states with their respective odds of jeopardizing democratic values. But my position can also be usefully reframed through the lens of Carolene Products and political process theory. Recall that Carolene condemned regulations that “restrict[] those political processes” (like elections) “which can ordinarily be expected to bring about repeal of undesirable legislation.”[94] John Hart Ely, the father of process theory, similarly bemoaned situations where “the ins are choking off the channels of political change” (again like elections) “to ensure that they will stay in and the outs will stay out.”[95] In these terms, my view is that Congress is less likely to “restrict[] those political processes” and to “chok[e] off the channels of political change,” and that the states are more apt to do so. I also concede that Congress should be expected to enhance those political processes and channels of political change—to make elections freer and fairer—less regularly than the states.

The benefit of this reframing is that it invokes the language of Carolene and of process theory without referring to the courts. The courts are the traditional protagonists of this school of thought: the institutions that are supposed to step in to fix malfunctions in American democracy.[96] But, this perspective reminds us, the courts’ efforts are generally unnecessary at the federal level. Congress is quite unlikely to endanger democratic values and so to create a need for judicial intervention. Further, this account continues, the courts aren’t the only bodies that can correct democratic malfunctions at the state level. Congress can do so, too, via legislation instead of litigation. True, congressional action may not be optimally frequent or potent thanks to the country’s extended sphere and the Constitution’s separation of powers and checks and balances. But when congressional action does occur, it’s every bit as good as judicial involvement.[97]

  1. Congress and the Court

Actually, congressional action is often better than judicial involvement for the same Madisonian reason that it’s usually preferable to state action. Congress is less vulnerable than the Supreme Court to factional takeover. So Congress is less likely than the Court to threaten democratic values (in the relatively rare circumstances when Congress even acts).[98] I explained above why it’s so hard for a faction with undemocratic aims to gain control of Congress,[99] and I discuss below why it’s easier for such a group to become ascendant in the Court.[100] But first, it’s worth considering whether this mode of analysis is even applicable to the judiciary. The Framers, for their part, didn’t seem to think so. Madison failed to mention the Court in his famous disquisition on “[a]mbition [being] made to counteract ambition” in The Federalist No. 51.[101] Likewise, Hamilton labeled the judiciary the “least dangerous” branch in The Federalist No. 78, adding that it would serve as “an essential safeguard against the effects of occasional ill humors in the society.”[102]

But this credulous view of the Court isn’t tenable. The Court, no less than the House, the Senate, and the presidency, is a governmental institution. Like all such bodies, the Court is staffed not by angels but by people. These people necessarily have all kinds of professional, ideological, and partisan objectives—some good, some bad, some pro-democratic, some anti-democratic—which they don’t abandon as soon as they don black robes. Now, it could be that particular characteristics of the judiciary render it relatively (or even wholly) impervious to the “ill humors” of faction. But it could also be that certain qualities make the Court more susceptible to these forces. The point is that the Court has to be carefully examined in the same light as other federal and state institutions. It can’t just be exempted from scrutiny—treated as the lone neutral referee on the playing field where all other actors’ ambitions are endlessly in contest.[103]

Undertaking this analysis, then, the most striking aspect of the Court’s structure is its small size. Under current law, the Court is made up of just nine Justices, of whom just five, a bare majority, can issue rulings backed by the full judicial power of the United States. With its membership this limited, the Court can’t reap the anti-factional benefits of Madison’s extended sphere. “[A] great[] variety of parties and interests” can’t assemble and reassemble into one configuration after another in a collective body with just nine participants.[104] It’s also straightforward for “a majority of the whole” to emerge with “a common motive.”[105] In fact, such a majority does emerge in most cases. And when it does, it’s not “difficult for all those who feel it to discover their own strength, and to act in unison.”[106] All the Justices in the majority have to do to recognize their numerical clout is look around their conference room.

Compounding the problem of the Court’s small size is its method of selection. For a faction to enact democratically corrosive laws, it needs the support of half of the House, three-fifths of the Senate, as well as the President. In contrast, appointments to the Court require just the President’s nomination and the Senate’s confirmation by a simple majority. Approval by the House, the larger and more representative congressional chamber, is unnecessary, as is the assembly of a big enough coalition to overcome the Senate’s filibuster.[107] Moreover, Congress’s composition changes every two years, and the President’s identity every four or eight. But tenures on the Court are indefinite, with contemporary Justices serving for a median of nearly three decades.[108] So when a group secures control of a Court appointment, its influence lasts further into the future than if it had merely dominated the elected branches.

The polarization that has reshaped the rest of American government has fostered factionalism in the Court, too.[109] In earlier periods, the legal elite from which Justices were drawn was more ideologically homogeneous, meaning that fewer lawyers had both impeccable credentials and non-mainstream views on most legal issues.[110] A faction controlling a Court appointment could thus easily misfire, choosing a Justice who wouldn’t further the group’s fringe agenda. Today, however, there are distinct conservative and liberal legal elites, with separate institutional affiliations, social circles, and stances on the law.[111] Consequently, a controlling faction’s risk of error is substantially lower. Odds are, a newly appointed Justice will behave the way the group expects—including casting democratically troublesome votes if that’s part of the group’s program.[112]

Lastly, there are fewer checks and balances on the Court than on the other branches of the federal government. No Justice has ever been removed from office via impeachment.[113] The Court’s constitutional judgments are reversible only through the onerous process of constitutional amendment: an impossibility in most circumstances. Even the Court’s statutory rulings can be overturned only through legislation—that is, if a majority of the House, a supermajority of the Senate, and the President stand united against the Court. This degree of agreement is rarely attainable under modern polarized conditions.[114] As a result, when the Court decides cases anti-democratically, there’s little that any other actor can do. Effective resistance through constitutional mechanisms is all but off the table. Softer sanctions like criticisms by prominent elected officials are also unlikely to have much impact in today’s divided political environment. Such censure is typically offset by praise from the other side of the ideological spectrum.[115]

To be sure, the same features that expose the Court to factional takeover can, on occasion, incline it to act in more benign ways.[116] In particular, when a likeminded President and Senate manage to appoint five Justices whom they hope will advance democratic values, this goal may well be realized, especially with polarization making Justices’ records more foreseeable. This pro-democratic majority may then intercede in the political process with few constraints on its freedom to maneuver. In my view, this scenario (except for the polarization) accurately describes the performance of the Warren Court (but no other Court in American history).[117] For a brief, glittering period, the Court did staunchly protect the right to vote,[118] fight racial discrimination in voting,[119] and end the systemic distortion of malapportionment.[120]

I also don’t want to suggest that law and politics are one and the same. I agree that most Justices, in most cases, give at least some weight to nonpartisan and nonideological considerations. But I think it’s equally plain that partisanship and ideology play a role, too, and that they sometimes cause the Court to make anti-democratic decisions. My claim, then, isn’t that the Court usually or always abridges democratic values. Instead, it’s that the Court does so more often than Congress because of its greater vulnerability to factional domination. This higher likelihood of anti-democratic action is why I argue that the Court should be relegated from its starring role under Carolene Products and political process theory. Congress is less apt to threaten the democratic tenets prized by this school of thought. As I elaborate below, Congress has also launched more pro-democratic interventions than the Court.[121] So Congress should be the theory’s lead and not its understudy.

  1. Federalism Values

Based on the discussion so far, you might be ready to grant my premises but not my conclusion. Maybe Congress is less likely to imperil American democracy than the states or the courts. But why does it necessarily follow that Congress should have expansive authority over elections? After all, one of our Constitution’s core structural principles is federalism, the idea that the states, too, and not just the federal government, are sovereigns worthy of dignity and respect.[122] Wouldn’t a sweeping congressional electoral power conflict with our commitment to state sovereignty?

Not if we care about federalism because of the benefits it’s said to bring. Obviously, if our commitment to federalism were categorical, then broad congressional authority over elections would be objectionable. Such power would impinge on the states’ autonomy to structure their electoral processes as they see fit. But most observers—including the Supreme Court—defend federalism on instrumental rather than absolute grounds, on the basis of the positive consequences that supposedly flow from it. To take the best-known example, in the 1991 case of Gregory v. Ashcroft, the Court stated that federalism (1) “assures a decentralized government that [is] more sensitive to the diverse needs of a heterogeneous society,” (2) “increases opportunity for citizen involvement in democratic processes,” (3) “allows for more innovation and experimentation in government,” (4) “makes government more responsive by putting the states in competition for a mobile citizenry,” and (5) provides “a check on abuses of government power.”[123] These alleged advantages, I contend here, are perfectly consistent with a congressional electoral power that’s extensive but not exclusive. In fact, endowing Congress with this kind of authority better serves the ends of federalism than denying Congress this clout.[124]

Start with greater sensitivity to the diverse needs of a heterogeneous society.[125] Also assume that Congress promotes democratic values through legislation like the For the People Act or the landmark statutes I describe in the next Part.[126] Congressional action of this type improves states’ electoral processes. More people are able to cast ballots, vulnerable groups’ voices are less diluted, elections run more smoothly, and so on. Better-performing electoral processes, in turn, yield more sensitivity to a diverse society’s varied needs. These needs are more likely to be heard by elected officials when elections are freer and fairer and more people can participate politically. Elected officials are more apt to respond to these needs, too, when the government is structured to reflect the people’s views more accurately. Responsive politicians may boost their reelection odds while unresponsive ones risk lowering theirs.

In contrast, consider the democratically offensive policies that, on a Madisonian account, states are more prone to enacting than Congress.[127] When states deny or dilute the franchise, they reduce the government’s sensitivity to society’s diverse needs. People who can’t vote can’t expect their elected officials to listen attentively to their concerns. Similarly, groups whose influence is impaired through electoral mechanisms—at-large elections, cleverly drawn districts, and the like—receive worse representation than other segments of the electorate. In these ways, certain state actions negate the federalism-based rationale for allowing states to act in the first place. This rationale is enhancing the government’s responsiveness, which instead is undermined by anti-democratic state laws.

Of course, other state actions can heighten governmental sensitivity to varied societal needs. Just as pro-democratic congressional legislation can make elected officials more responsive to their constituents’ interests, so can state policies that further democratic values.[128] This is why I referred above to a congressional electoral power that’s extensive but not exclusive. Congress may need broad authority to regulate elections, especially when states attenuate the link between society’s needs and the government’s outputs. But Congress doesn’t have to be the only player on this stage. In fact, Congress shouldn’t occupy the field because, if it did, it would frustrate state efforts that are just as conducive to greater responsiveness as Congress’s own exertions.

These same points apply squarely to Gregory’s second federalism value: increasing opportunities for citizens to be involved in democratic processes.[129] When Congress makes it easier to vote by banning racial discrimination in voting,[130] ending literacy tests,[131] enabling people to register when getting their driver’s licenses,[132] and requiring higher-quality voting machines[133] (to cite some past congressional measures), Congress necessarily facilitates electoral involvement. People are better able to participate in democratic processes thanks to Congress’s interventions. Conversely, when states restrict the franchise, they push in exactly the opposite direction. They make it harder for (some of) their citizens to engage in self-governance. But when states expand and expedite voting, they produce the same participational benefits as congressional action. The possibility of obtaining these benefits through state action, too, is why it shouldn’t be precluded in favor of complete federal control.

Turning to Gregory’s third objective, more innovation and experimentation in government,[134] it has to be conceded that this goal is compromised by congressional electoral regulation. In particular, when Congress steps in to advance democratic values, it prevents the states from creatively thinking of ways to subvert them. A congressional ban on racial discrimination in voting, for instance, stops the states from pioneering ever more devious tactics to deny or dilute minority votes.[135] However, pro-democratic congressional action doesn’t impede the states from innovating further in the same vein.[136] The states remain free to adopt more robust safeguards for minority voters, to induce greater participation by all citizens in novel ways, to guard against racial and partisan vote dilution through policies with no federal analogues, and so on. Pro-democratic congressional action is thus asymmetric in its implications for subsequent state experimentation. It bars the states from racing to the anti-democratic bottom while permitting them to vie for the pro-democratic top.

Notably, Madison embraced this sort of asymmetry in The Federalist No. 43. “In a confederacy founded on republican principles, and composed of republican members,” he wrote, “the superintending government ought clearly to possess authority to defend the system against aristocratic or monarchial innovations.”[137] These “innovations” both clash with “republican principles”—what I call democratic values—and endanger the cohesion of the nation. “Governments of dissimilar principles and forms have been found less adapted to a federal coalition . . . than those of a kindred nature.”[138] But as long as the states refrain from unrepublican experiments, Madison continued, they can regulate their elections however they wish. “The only restriction imposed on them is, that they shall not exchange republican for antirepublican Constitutions.”[139]

This response also holds for Gregory’s fourth aim, augmenting governmental responsiveness by putting the states in competition for mobile citizens.[140] Again, pro-democratic congressional regulation of elections does deprive the states of one lever they might want to use to attract (certain) people: the denial or dilution of (other) people’s votes. But again, the loss of this lever is normatively unproblematic, and the states can still compete to draw mobile people by implementing pro-democratic measures. Moreover, election law is an unlikely driver of people’s residential choices. People may plausibly move to or from states because of their tax levels, their welfare benefits, the strength of their economies, and the like. But it’s hard to believe many people vote with their feet based on states’ periods of early voting,[141] their use (or not) of independent redistricting commissions,[142] or their provision (or not) of public campaign financing,[143] to name some areas where state approaches currently diverge. Gregory’s fourth aim thus seems largely inapplicable to the electoral context.

As for Gregory’s fifth and final end, checking abuses of governmental power,[144] it’s the fundamental reason to prefer congressional to state electoral regulation. To reiterate, Congress is less apt than the states to abridge the right to vote because of its more extended sphere and the Constitution’s more rigidly separated powers and stricter checks and balances.[145] So in the electoral domain, abuses of governmental authority are more likely to come at the hands of the states than at Congress’s. Further, if congressional control over elections is extensive but not exclusive, then a “balance of power” exists “between the States and the Federal Government.”[146] This “balance,” the Gregory Court argued, “reduce[s] the risk of tyranny and abuse from either front.”[147] “In the tension between federal and state power lies the promise of liberty.”[148]

  1. Congressional History

The Madisonian theory underpinning the Constitution, then, generates a hypothesis about the relative threat posed to democratic values by different institutions: Congress should be less of a menace to these values than the states or the courts. Or putting this proposition in the terms of Carolene Products and political process theory, Congress should be less inclined than the states or the courts to “restrict[] those political processes which can ordinarily be expected to bring about repeal of undesirable legislation.”[149] What’s more, federalism theory presents no objection to the implication of the Madisonian hypothesis: that Congress should have sweeping authority over elections. The tenets of federalism theory actually support expansive congressional power in this area.

But is the Madisonian hypothesis correct? Now that more than two centuries have gone by since the Framing, has Congress proven to be less hazardous to democratic principles than the states and the courts? I think history confirms that Congress has indeed been the least dangerous branch.[150] In fact, I would push the point further: Not only has Congress imperiled American democracy less often than its institutional competitors, but most congressional electoral regulations have affirmatively promoted democratic values. This performance wasn’t anticipated by Madison, who merely hoped to restrain the “impetuous vortex” of the “legislative department.”[151] But in a fortuitous development, Congress has compiled a generally pro-democratic record in its forays into the electoral arena.

I can’t possibly cover all these forays here. To do justice to the full history of congressional electoral activity would take several volumes. So instead, I focus in this Part on what I consider to be the five most important moments when Congress has regulated the electoral process: (1) the Apportionment Act of 1842, which introduced the requirement of single-member districts for Congress; (2) the cluster of Reconstruction statutes that sought to guarantee the franchise for African Americans; (3) the Voting Rights Act of 1965, which returned to the unfinished business of incorporating blacks into the political community; (4) the 1974 amendments to the Federal Election Campaign Act, the most far-reaching federal restrictions of money in politics; and (5) the National Voter Registration Act of Act of 1993 and the Help America Vote Act of 2002, which involved Congress in the nuts and bolts of election administration for the first time.[152] My thesis is the same for all these laws: that their enactment served both abstract democratic and practical political ends. In none of these pivotal moments, that is, did Congress sacrifice the wellbeing of American democracy for the sake of factional advantage.

I also don’t discuss in any detail the histories of the states’ and the courts’ electoral actions.[153] Suffice it to say that these histories are, at best, mixed. While there have been pro-democratic highlights, there have been many instances, too, when democratic values were discounted or cast aside entirely. Take the states. In the nation’s early years, they frequently barred blacks, women, and men without sufficient property from voting.[154] Over the century-long period between the first and second Reconstructions, southern states inhibited African Americans’ political participation through grandfather clauses, poll taxes, literacy tests, and outright violence.[155] To this day, states across the country restrict voting through photo ID requirements, limits on voter registration, and purges of voter rolls.[156] States also continue to dilute the votes of minority and political groups through the practices of racial and partisan gerrymandering.[157]

The story is similar for the Supreme Court, which has regularly failed to step in when democratic principles were under siege and stymied other actors’ efforts to vindicate these principles. In the nineteenth century, the Court held that challenges to unrepublican state policies are nonjusticiable[158] and struck down several statutory provisions, passed by Congress during Reconstruction, that tried to stop racial discrimination in voting.[159] Before the Warren Court, the Justices acquiesced in southern states’ violent refusals to enfranchise blacks[160] and allowed rampant malapportionment to make a mockery of representation.[161] And as I have explained elsewhere, our own Roberts Court has never come across a franchise burden it thought was excessive,[162] torn gaping holes in the nation’s campaign finance[163] and voting rights[164] laws, and recused itself from the fight against gerrymandering.[165] These are the state and judicial benchmarks, then, to which congressional electoral regulation should be compared. Not perfection, but rather ambiguous records replete with anti-democratic actions.

  1. The 1842 Apportionment Act

Starting with the Apportionment Act of 1842,[166] it’s the most obscure of the laws I address, but it’s still a milestone in the history of Congress’s regulation of elections.[167] Prior to the Act, Congress imposed no restrictions on the construction of congressional districts. States were thus free not even to use geographically demarcated districts, and instead to elect their House members through winner-take-all statewide elections. Under this approach, known as the “general ticket,” each voter in a state cast ballots for as many candidates as the state had House seats, and that many candidates—the ones receiving the most votes—were elected. The Act banned the general ticket in congressional elections and required the states to elect their House members through geographically bounded single-member districts. Each state’s House members “shall be elected by districts composed of contiguous territory equal in number to the number of Representatives to which said State is entitled, no one district electing more than one Representative.”[168]

The Apportionment Act furthered (and was advocated precisely because it furthered) the democratic value of a legislature that better reflects the will of the electorate. Under the general ticket, a partisan plurality could generally sweep every House seat in a state.[169] This windfall would lead to massive overrepresentation for the victorious plurality and extreme underrepresentation—no seats at all—for every other group. In contrast, single-member districts don’t guarantee proportional representation, and are actually compatible with quite disproportionate outcomes.[170] But they do enable statewide minorities that are concentrated in particular regions to win seats in these areas. By the same token, single-member districts usually prevent mere pluralities from claiming all of a state’s seats. Single-member districts are thus a waypoint between winner-take-all elections and full proportionality—and a clear improvement over the former from the perspective of accurate representation.

This was just the argument made by the Apportionment Act’s proponents. “The general ticket system destroy[s] the principles of democratic government,” declared Representative Thomas Arnold.[171] Under the general ticket, “the voice of the people [is] completely stifled,” agreed Representative Garret Davis.[172] On the other hand, according to Representative George Summers, congressional delegations “elected by the people, allotted into districts, come here representing the opinions of the constituency which sends them.”[173] So they satisfy “the essential feature of representative democracy,” which is that “the representatives shall reflect the will and know the wants of [their] constituents.”[174] “[A]nother great benefit which would arise from the district system,” in the words of Senator William Huntington, “would be, to give a voice to the minority, that otherwise would be effectually silenced under the general ticket system.”[175] Minorities “have rights too, which ought to be protected,” and which “cannot and will not be so [protected], except the election be by districts.”[176]

But recall that my claim here is that most congressional electoral regulations have simultaneously advanced democratic and partisan aims. In addition to enhancing legislative representation, the Apportionment Act’s backers, almost all of them Whigs, hoped to boost their own electoral prospects. Many Whigs had been outraged by a stratagem recently executed in Alabama. In 1840, the state’s Democratic legislature had switched from single-member districts to the general ticket, thereby eliminating the Whigs’ representation when they trailed the Democrats in the election held that year.[177] Examining the electoral landscape that awaited them in 1842 and beyond, many Whigs also noticed that general ticket states with Democratic majorities would be gaining seats in the next round of reapportionment, while Whig-majority general ticket states would be losing representation. By prohibiting the general ticket, then, the Whigs saw an opportunity to staunch their losses in the Democratic states and even to gain some seats in pro-Whig areas.[178] Unsurprisingly, since the Whigs’ partisan motivation was known to their opponents, the Act passed on almost a pure party line. Nearly every Whig voted for the bill and nearly every Democrat opposed it.[179]

Do the Whigs’ dual objectives tarnish the view of the Apportionment Act as a democratic step forward? I don’t think so. While the Act may have electorally benefited the Whigs,[180] it still eradicated a highly objectionable practice, the general ticket, from congressional elections. This democratic boon was real even if it came with a partisan fillip for the Whigs. Members of Congress, moreover, are political animals. So it’s unrealistic, in most circumstances, to expect them to incur partisan costs to ameliorate American democracy. When members of Congress find and pass laws that serve both democratic and partisan ends, they’re doing all that can reasonably be asked of them. They’re also doing more—per my brief survey of the states’[181] and the courts’[182] mixed records—than their institutional competitors have often done.

  1. Reconstruction

Moving ahead by a generation, in the aftermath of the Civil War, Congress launched its most sustained intervention into the electoral realm. In a series of landmark statutes—three in 1867–68 reconstructing the former Confederate states, and five more in 1870–71 enforcing the newly ratified Fifteenth Amendment—Congress strived to rid elections of racial discrimination in voting as well as fraud. These laws’ provisions are startling in their scope, and so worth summarizing at some length. In the Reconstruction Acts, Congress required the former Confederate states to hold elections for delegates to constitutional conventions without regard to citizens’ “race, color, or previous condition [of servitude].”[183] This same nondiscrimination criterion applied to all elections (federal, state, and local) regulated by the states’ new constitutions[184] as well as to the elections to ratify the constitutions.[185] Congress further instructed military commanders to register eligible voters,[186] to apportion convention delegates using the one-person, one-vote rule,[187] to announce the results of the delegate elections,[188] and to appoint three-member boards to assist them in these tasks.[189] These boards were empowered to administer the voter registration process, to supervise the delegate elections, and to tabulate the votes and determine the winners.[190]

In the Enforcement Acts, Congress reached beyond the former Confederate states to target racial discrimination in voting and election fraud nationwide.[191] These laws mandated that all citizens, regardless of their race, have “the same and equal opportunity” to satisfy conditions for voting.[192] The laws stipulated that any citizen who offered to satisfy a voting requirement, but was denied this chance by an election official, would be deemed in compliance with it.[193] The laws banned private parties from using “force, bribery, threats, [or] intimidation” to interfere with qualifying to vote or voting.[194] The laws forbade several species of fraud, including voting “in the name of any other person,” voting “more than once at the same election,” and inducing election officials to violate their duties.[195] The laws converted election officials’ breaches of state statutory obligations into federal crimes.[196] And the laws authorized federal judges to appoint supervisors, aided by federal marshals, for congressional elections.[197] These supervisors monitored voter registration, inspected the voter rolls, observed the voting process, challenged unqualified voters, ensured the security of ballot boxes, and examined and counted ballots.[198]

The democratic values promoted by the Reconstruction Acts and the Enforcement Acts are self-evident. These statutes’ overarching goal was to end racial discrimination in voting—to safeguard the franchise for newly liberated African Americans, to welcome them as full members of the political community, and to redeem the promise of the Fifteenth Amendment. As the Supreme Court put it in an 1884 case (during a period when it was no friend of Reconstruction legislation[199]), Congress sought “to provide against these evils,” “common in one quarter of the country,” of “lawless violence,” “violence and outrage,” and “brute force” against blacks trying to vote.[200] Or as Senator Carl Schurz stated when Congress debated the first Enforcement Act, Congress wanted to place blacks’ voting rights “under the shield of national protection.”[201] That way, “the whole people, and not [] a part of them only,” would have “the right and the means to cooperate in the management of their common affairs.”[202]

The Enforcement Acts had the additional democratic aim of preventing election fraud. The Court recognized this purpose, too, in its 1884 decision, noting that Congress hoped to stop elections from being “poisoned by corruption” and “at the mercy of . . . unprincipled corruptionists.”[203] So did members of Congress like Senator Adam Sherman, who called election fraud a problem “of greater magnitude even than denial of right to vote to colored people.”[204] Sherman elaborated that fraud was “a national evil so great, so dangerous, and so alarming in character” that Congress had no choice but to act.[205]

Again, though, my claim isn’t that the Reconstruction Acts and the Enforcement Acts were enacted for entirely benevolent reasons. Rather, the radical Republicans who pushed through these laws—over the furious opposition of President Johnson[206] and what remained of the Democratic Party—pursued partisan advantage as well. In the South, the Republicans’ calculus was obvious. Most of their supporters were freed slaves, so if the Republican Party was to have a future in this part of the country, African Americans had to be able to vote freely and fairly.[207] In the North, rampant fraud by Democratic machines in urban centers was a major threat to Republican electoral fortunes.[208] Curbing ballot box tampering, noncitizen voting, and the like thus bolstered Republicans’ political position.[209]

And again, I don’t think the democratic appeal of the Reconstruction Acts and the Enforcement Acts is undermined by the laws’ concurrent partisan motives. Yes, the radical Republicans “linked principle and expedience together when they [legislated] issues of black suffrage and federal enforcement,” as historian Xi Wang has written.[210] But principle remains principle even when it’s tied to expedience. Democratic values don’t lose their force just because they’re intertwined with (and advanced through) self-interested politics.

  1. The Voting Rights Act

Progressing to modern times, Congress’s more recent electoral regulations are more familiar and so can be sketched in broader strokes. The most renowned of these regulations is undoubtedly the Voting Rights Act of 1965 (VRA), the so-called “crown jewel” of the Civil Rights Era.[211] Congress packed the VRA with one measure after another facilitating voting by African Americans. These provisions included the deployment of federal examiners to register eligible voters[212] and federal observers to monitor elections,[213] the suspension of literacy and moral character tests in the deep South,[214] the congressional judgment that poll taxes are unconstitutional,[215] and the criminalization of threatening or intimidating people for voting.[216] Congress also instituted a preclearance regime for certain deep southern jurisdictions, requiring them to prove to federal authorities that changes to their election rules weren’t racially discriminatory before they could implement these policies.[217]

The VRA shared its democratic aspiration with its Reconstruction antecedents a century earlier: protecting the franchise for all racial minorities (but especially blacks) throughout the country (but especially in the deep South). The impetus for the VRA’s enactment was the appalling violence in Selma, Alabama, where protesters marching for African American voting rights were savagely beaten by state troopers on Bloody Sunday. The VRA was Congress’s effort to show that the protesters’ sacrifices hadn’t been in vain.[218] President Johnson also articulated the VRA’s high-minded objective in a famous address to Congress. “Every American citizen must have an equal right to vote,” he told the chamber.[219] Consistent with this tenet, his administration’s bill would “eliminate illegal barriers to the right to vote” and so “[a]llow men and women to register and vote whatever the color of their skin.”[220]

Interestingly, the VRA didn’t have a clear partisan purpose; in fact, it was expected to cost the Democrats, who then enjoyed unified control of the federal government, support in the South. In the roll call votes in the House and Senate, southern Democrats comprised the bulk of the opposition to the bill.[221] In contrast, more than eighty percent of House Republicans, and more than ninety percent of Republican senators, voted in favor.[222] President Johnson is also rumored to have said to an aide, around the time of the VRA’s passage, “we just delivered the South to the Republican Party for a long time to come.”[223] His reasoning was that conservative southern whites couldn’t tolerate being part of the same party as newly enfranchised (and politically empowered) African Americans. So they would migrate to the Republican Party, making it viable below the Mason-Dixon line for the first time since Reconstruction and eroding, if not eliminating, what had long been the Democrats’ Solid South.[224]

President Johnson’s prophecy came true almost immediately at the presidential level. In the 1968 election—just three years after the VRA’s enactment—the Democratic ticket lost every southern state except Texas.[225] Republican presidential candidates have continued to dominate in the South ever since.[226] At lower levels of government, though, the transition from Democratic to Republican control took much longer. Not until 2002, for example, did Republicans win a majority of southern state house seats.[227] But while President Johnson’s prediction wasn’t instantly validated in all elections, it was accurate. The VRA was thus that rarest of congressional electoral interventions: a law that pursued a great democratic goal despite the likelihood of significant harm for the party that passed it. If Congress’s Reconstruction legislation fused principle and expedience,[228] the VRA severed them, and embraced the former over the latter.

  1. The Federal Election Campaign Act

Jumping forward another decade, while the Watergate scandal consumed American politics, Congress enacted its most aggressive ever regulations of electoral funding. The Federal Election Campaign Act Amendments of 1974 (FECA) limited campaign contributions to federal candidates by private persons and political committees.[229] FECA also capped campaign expenditures by federal candidates, political parties, and all other groups and individuals.[230] FECA further mandated the disclosure of contributions and expenditures exceeding certain thresholds.[231] FECA offered voluntary public financing to presidential (but not congressional) candidates as well.[232] And FECA created an independent agency, the Federal Election Commission, to administer and enforce the new rules.[233]

Because FECA’s passage was followed by litigation over almost all its provisions, the law’s democratic aims were recounted in unusual detail in judicial opinions. The “primary interest” served by the statute, according to the Supreme Court, was “the prevention of corruption and the appearance of corruption spawned by the real or imagined coercive influence of large financial contributions on candidates’ positions and on their actions if elected to office.”[234] An “ancillary interest underlying” FECA was “to mute the voices of affluent persons and groups in the election process and thereby to equalize the relative ability of all citizens to affect the outcome of elections.”[235] Another secondary objective was to “act as a brake on the skyrocketing cost of political campaigns and thereby . . . to open the political system more widely to candidates without access to sources of large amounts of money.”[236] Lastly, Congress sought to “provide[] the electorate with information . . . in order to aid the voters in evaluating those who seek federal office.”[237]

These democratic ends—stopping corruption, promoting equality, increasing access, and educating the electorate—are consistent with the history of FECA’s enactment. As historian Julian Zelizer has observed, by the late 1960s and early 1970s, a loose coalition had emerged, “composed of legislators, experts, philanthropists, foundations, and public interest groups,” who “believed that representative government could be improved.”[238] Prior to Watergate, this coalition had enough clout to put sweeping campaign finance reform on the congressional agenda, but not actually to pass it. With the eruption of a massive scandal that involved (among other things) illegal donations to the 1972 Nixon campaign, though, the coalition’s moment arrived. Public opinion focused like never before on governmental corruption.[239] Politicians previously averse to change found themselves unable to oppose (at least publicly) restrictions on electoral funding.[240] Watergate thus triggered a popular “groundswell,” in the words of political scientist Robert Mutch, which in turn brought “partisanship and principle together, merge[d] the principled arguments made in floor debate with the concerns about short-term partisan advantage expressed in cloakroom conversations.”[241]

But note Mutch’s references to “partisanship” and “short-term partisan advantage.” While the Democrats who controlled Congress in 1974 had genuine democratic rationales for enacting FECA, they had political justifications, too. One of these was highlighting an issue, campaign finance reform, that reminded the public of Watergate and divided the Republican Party, which had mostly resisted curbs on electoral funding in the past.[242] More importantly, Democrats had been massively outspent in recent elections, especially 1972. FECA promised to move them closer to parity by capping both the contributions that could be made to campaigns and the amounts that campaigns themselves could spend.[243] On the Republican side, the new and unelected president, Gerald Ford, had a practical motive of his own for signing FECA into law. Although he had “opposed the legislation,” “[a] veto was now politically dangerous and Congress might override it.”[244]

The story of FECA’s passage, then, is the usual one of Congress regulating elections: a confluence of democratic and partisan factors, all pushing in the same direction. This isn’t a wholly altruistic account, but it still rebuts the common criticism that when Congress legislates about money in politics, it does so for anti-democratic reasons—to entrench incumbents in office against the will of the electorate. Justice Scalia expressed this view best, in a 2003 opinion more in line with his antigovernment ideology. “The first instinct of power is the retention of power,” he wrote.[245] Campaign finance restrictions supposedly serve this end by “prohibit[ing] the criticism of Members of Congress” and thus providing “incumbents [with] an enormous advantage.”[246] Supposedly. In fact, most members of Congress had little interest in limiting electoral funding until the reform coalition’s lobbying and the shock of Watergate forced their hand.[247] Empirical evidence also shows that most campaign finance regulations benefit challengers, not incumbents.[248] So the legislators who enacted FECA weren’t the schemers imagined by Justice Scalia. Instead, they were regular politicians operating within the Madisonian system, yoking together democratic and partisan goals, not pitting them against each other.

  1. Modern Election Administration

Turning finally to a pair of recent congressional interventions, the National Voter Registration Act of 1993 (NVRA)[249] and the Help America Vote Act of 2002 (HAVA)[250] involved Congress in the details (some would say the minutiae) of election administration. The NVRA required states to allow voters to register for federal elections (1) when they apply for driver’s licenses; (2) when they interact with state welfare agencies; and (3) by mail rather than in person.[251] The statute also provided rules for the upkeep of state voter rolls.[252] HAVA discouraged the use of obsolete punch card and lever voting machines and subsidized their replacement by newer technologies.[253] It also authorized voters whose names are missing from the voter rolls to cast provisional ballots, whose validity is determined after the election.[254] HAVA further directed each state to create a computerized statewide voter registration list.[255] And the law stipulated that first-time registrants by mail must provide identification when they go to the polls.[256]

The NVRA had a straightforward democratic objective: boosting voter turnout. The statute aimed to “increase registration of eligible citizens in elections for Federal office,” stated its Senate report, and thus to “enhance[] the participation of eligible citizens as voters in [these] elections.”[257] Consequently, “[t]he declining numbers of voters who participate in Federal elections” might be reversed.[258] The primary purpose of HAVA, in turn, was making elections more efficient and less prone to errors, especially in the tabulation of votes and the registration of voters. Both of these had emerged as problems in Florida’s disputed 2000 election; as the law’s House report elaborated, “[t]he circumstances surrounding [that] election . . . brought an increased focus on the process of election administration, and highlighted the need for improvements.”[259] Secondarily, HAVA sought to prevent voter fraud through its identification requirement.[260]

It should come as no surprise, by now, that partisanship also played a role in the passage of the NVRA and HAVA. A bill similar to the NVRA had been approved by both houses of Congress in 1992, only to be vetoed by President Bush.[261] After the Democrats took unified control of the federal government in the 1992 election, they quickly enacted an analogous statute on a mostly party line basis.[262] The Democrats’ (potentially flawed[263]) calculation was that higher turnout would help them electorally because it would mean that more of their lower-propensity voters (young, poor, and minority citizens) had gone to the polls.[264] Most Republicans opposed the bill for the same (maybe inaccurate) reason, believing that their candidates did better when fewer voters turned out.[265]

HAVA, in contrast, wasn’t pushed through by a single party. It was passed, rather, by a divided Congress, with “[c]hanges thought to be beneficial to one party . . . balanced by provisions backed by, and assumed to aid, the other.”[266] In particular, the Democrats favored the aspects of HAVA that would modernize voting machines and enable voters to cast provisional ballots. The Democrats thought that outdated technologies and voter registration errors had cost them Florida’s 2000 election (and with it the presidency), and were eager to avoid these issues going forward.[267] The Republicans, on the other hand, insisted on HAVA’s identification requirement for first-time registrants by mail. This provision was intended to deter voter fraud, which the Republicans (wrongly) maintained was a widespread phenomenon that usually advantaged Democratic candidates.[268]

Notably, a strict identification requirement could reduce voter turnout and so undermine the democratic value of political participation. But HAVA’s ID requirement wasn’t very onerous at all. It applied to only the small population of people who hadn’t previously voted in a jurisdiction and registered to vote by mail.[269] It could also be satisfied by a photo or nonphoto ID, including “a current utility bill, bank statement, government check, paycheck, or other government document that shows the name and address of the voter.”[270] Accordingly, HAVA’s ID requirement is consistent with my claim here that most congressional electoral regulations simultaneously advance democratic and partisan ends. The provision tried to stop fraud (a democratic goal) and to aid Republicans (a partisan one), and it did so without curtailing participation (another element of a vibrant democracy).

* * *

Lest this discussion seem overly pollyannish, I want to concede that Congress has, on a few occasions, compromised democratic values for the sake of partisan gain. In the 1892 elections, the Democrats won full control of the federal government for the first time since before the Civil War.[271] They promptly repealed every statutory provision enacted by Congress during Reconstruction that related directly to elections.[272] The Democrats thus opened the door to the disenfranchisement of African Americans in the South and to widespread fraud in the North. And they did so for the nakedly partisan reason that excluding southern blacks and corrupting northern elections would lift their electoral prospects.[273] In 1929, likewise, members of both parties came together to eliminate almost every requirement that had previously applied to the design of congressional districts: contiguity, compactness, and equal population.[274] Had these criteria been followed in the next round of redistricting, in the face of rapid demographic shifts, many incumbents’ districts would have changed in unfavorable ways. But by discarding the criteria—and the value of fair representation they promoted—the law’s backers hoped to maintain their hold on office.[275]

One could try to excuse these statutes, perhaps, on the ground that they merely repealed earlier congressional acts. Congress didn’t, say, affirmatively bar African Americans from voting or mandate that congressional districts be unequally populated. But this argument is weak tea. Compared to the law on the books, these measures plainly subverted democratic values for partisan purposes. Incidents like these are why I have been careful to present a non-maximalist thesis in this Part. My position hasn’t been that Congress always furthers democratic ends when it regulates elections. Instead, it has been that Congress usually does so, and that Congress threatens democratic values less often than the states and the courts. This more circumspect claim, I believe, would hold even after scrutinizing these actors’ electoral records at greater length than is possible here.

III. Current Doctrine

The history of congressional electoral regulation, then, confirms what the theory of congressional electoral regulation suggests: that Congress is less likely to imperil American democracy than the states or the courts. Given this history and theory, it would be prudent for the law to endow Congress with sweeping authority over elections. Congress would be less apt to abuse this power than its institutional competitors, and if past performance is any guide, most of its efforts would serve democratic ends. In fact, that’s exactly what current constitutional doctrine does. As I explain in this Part, the law grants Congress expansive authority over most aspects of elections—the franchise itself, party primaries, election administration, redistricting, campaign finance, and so on—at both the federal and state levels.

This doctrinal discussion is complicated by the sheer number of constitutional provisions that relate to Congress’s electoral power: almost a dozen, in total. Because of this unusual efflorescence of constitutional text, parsing any single clause in isolation is unproductive. Each clause must always be understood as a piece of a larger jigsaw puzzle, a thread in a broader tapestry. It’s the assembled puzzle, the woven tapestry, that defines the full scope of congressional electoral authority.[276] The volume of relevant provisions also makes redundancy and reinforcement the touchstones of the legal analysis in this area. Very often, when Congress legislates about elections, its power to do so stems from multiple clauses simultaneously. Together, these clauses form a sturdier basis for congressional intervention than any provision standing alone. They fill in one another’s gaps, offer distinct but overlapping rationales for intervention, and make clear that congressional action is a feature (not a breach) of the constitutional design.[277]

The sheer number of electoral clauses, further, means that my commentary on each provision is necessarily abbreviated. I capture the state of current doctrine on each clause. But I don’t trace the evolution of this doctrine. Nor do I address its consistency with the original meaning at the time of each provision’s enactment.[278] And nor do I speculate how the law might continue to change under the supervision of the Roberts Court, no friend of sweeping congressional electoral authority.[279] To avoid a dry recitation of precedents, I also focus less on what the cases say and more on how they fit together. I argue that they create a sort of lattice—extending horizontally to give Congress power over most electoral matters, and reaching vertically to provide it with multiple bases for most actions it takes.

One last preliminary point: As I noted earlier,[280] I’m only interested here in the Constitution’s power-conferring clauses. Its power-limiting elements, which include the First Amendment, the Equal Protection Clause,[281] and various nontextual federalism principles, are outside the lines of this particular project. Accordingly, it could well be that a law Congress has the authority to pass is nevertheless unconstitutional. A campaign finance regulation might impermissibly restrict electoral spending, which the current Court thinks is core political speech.[282] A voting rights statute might be deemed overly race-conscious and so in conflict with the Constitution’s supposed commitment to colorblindness.[283] Any electoral law might offend the states’ ineffable dignity by commandeering them[284] or differentiating among them in violation of their equal sovereignty.[285] My claim, then, is just that Congress is constitutionally empowered to enact all these (and many more) measures. They may be unlawful because they infringe some other constitutional provision. But they aren’t ultra vires.[286]

  1. The Elections Clause

The logical place to begin this survey is the Elections Clause of Article I: the Constitution’s clearest articulation of (a part of) Congress’s electoral authority. The Clause authorizes Congress to “make or alter . . . Regulations” of “[t]he Times, Places and Manner of . . . Elections for Senators and Representatives.”[287] The provision thus applies to congressional elections but not to presidential, state, or local races. The provision also doesn’t define its key terms—the times, places, and manner of elections—leaving it to the courts to supply their content. Stepping into the breach, the courts have construed this language exceptionally broadly.[288] Under the Clause, “Congress has plenary and paramount jurisdiction over the whole subject” of congressional elections, declared the Supreme Court in an 1879 case.[289] This power “may be exercised as and when Congress sees fit,” and “so far as it extends and conflicts with the regulations of the State, necessarily supersedes them.”[290] The Court added in a 1932 case that the Clause’s “comprehensive words embrace authority to provide a complete code for congressional elections”—“to enact the numerous requirements . . . which experience shows are necessary.”[291] Even the Roberts Court agreed, in a 2013 case, that “[t]he Clause’s substantive scope is broad.”[292] Moreover, “the federalism concerns” that arise elsewhere “are somewhat weaker here,” since whenever Congress regulates congressional races, “it necessarily displaces some element of a pre-existing legal regime erected by the States.”[293]

Consistent with these pronouncements, the Court has held that Congress can legislate about almost every conceivable aspect of congressional elections. Congress can make it a crime for private, nongovernmental actors to interfere with voting in congressional races.[294] Congress can regulate congressional general as well as primary elections.[295] Congress can make rules for elections in which congressional and non-congressional candidates are on the ballot—and that apply to both types of races.[296] Congress can order states to administer congressional elections in certain ways, notwithstanding the usual ban on commandeering.[297] As the Court put it in 1932, in its best-known summary of the Election Clause’s scope, Congress can reach “not only [the] times and places” of congressional races, but also their “notices, registration, supervision of voting, protection of voters, prevention of fraud and corrupt practices, counting of votes, duties of inspectors and canvassers, and making and publication of election returns.”[298]

Maybe most controversially,[299] Congress can even regulate who votes (not just how voting is conducted) in congressional elections. In 1970, Congress passed a law that, among other things, lowered the minimum voting age in congressional elections from twenty-one to eighteen.[300] In the Court’s subsequent decision in Oregon v. Mitchell,[301] four Justices would have struck down this provision and another four would have upheld it on a broader equal protection theory.[302] But in his controlling opinion for the Court, Justice Black sustained it squarely on Elections Clause grounds. “[T]he powers of Congress to regulate congressional elections[] includ[e] the age and other qualifications of the voters,” he wrote.[303] “Congress has ultimate supervisory power over congressional elections.”[304] Since Mitchell, Congress has both required states to allow former residents of theirs who have moved abroad to vote in congressional elections[305] and made American citizenship a prerequisite for voting for Congress.[306] Thanks to Mitchell, neither of these congressionally imposed voting qualifications has ever been challenged.[307]

Stepping back from these cases, we can start to see the structure of Congress’s electoral authority emerging. Imagine a horizontal plane encompassing every possible electoral topic and level. The Elections Clause doesn’t fill this plane. It doesn’t give Congress plenary power over every electoral issue. But it does demarcate a portion of the plane—that involving congressional elections—in which Congress enjoys essentially complete control. The Clause also provides Congress with some electoral power outside this portion. As noted above, Congress can regulate non-congressional elections, too, to the extent they’re held concurrently with (and can’t be disentangled from) congressional elections.[308]

Think of a vertical dimension as well, registering different legal rationales for congressional electoral activity. The Elections Clause doesn’t occupy all of this dimension either. In fact, it supplies just a single basis for Congress to regulate elections: that under a particular constitutional provision, Congress explicitly has the authority to set or change the times, places, and manner of congressional elections.[309] As we’ll see below, this justification for congressional intervention is usually undergirded by several more, each corresponding to a different legal theory. The sources of congressional power thus often stack up, forming a pillar stronger than the sum of its parts.

  1. The Electors Clause

You may have noticed that, in my discussion of the Elections Clause, I carefully referred to congressional rather than federal elections. I did so because presidential elections are federal elections, too, but are outside the heartland of Congress’s Elections Clause authority.[310] Congress’s power to regulate presidential elections derives, instead, from the Electors Clause of Article II.[311] This provision states that “Congress may determine the Time of ch[oo]sing the Electors” (presidential electors, of course, being the group that officially selects the President under the Constitution’s complicated Electoral College system).[312]

As a textual matter, the Electors Clause is plainly narrower than the Elections Clause. It only authorizes Congress to set the time of presidential elections. It’s silent about the places, and more importantly the manner, of these races. Nevertheless, the courts have construed the Electors Clause coextensively with the Elections Clause, holding that the former endows Congress with the same authority over presidential elections that the latter grants it over congressional races.[313] In a 1934 case, for instance, the Supreme Court ruled that Congress has the “power to pass appropriate legislation to safeguard [a presidential] election . . . . from impairment or destruction.”[314] “[T]he choice of means to that end presents a question primarily addressed to the judgment of Congress.”[315] In Mitchell, similarly, Justice Black’s controlling opinion announced that “it is the prerogative of Congress to oversee the conduct of presidential and vice-presidential elections.”[316] This prerogative is identical in the presidential and congressional contexts: “It cannot be seriously contended that Congress has less power over the conduct of presidential elections than it has over congressional elections.”[317]

Unsurprisingly, given these views, the Court has held that Congress can mandate the disclosure of campaign contributions and expenditures,[318] restrict campaign donations, and institute a system of public financing for presidential elections.[319] As at the congressional level, the Court has also ruled that Congress can change the qualifications for voting in presidential elections. The 1970 law that lowered the congressional voting age to eighteen did the same for the presidential voting age.[320] The statute further abrogated state residency requirements for voting for President, directed states to allow voters to register for presidential elections until thirty days before election day, and established that voters can vote absentee for President as long as they request their ballots at least seven days before the election.[321] All these measures were upheld in Mitchell—the ones unrelated to age by an eight-to-one vote.[322] As Justice Black explained, “[e]ssential to the survival and to the growth of our national government is its power to fill its elective offices and to insure that the officials who fill those offices are as responsive as possible to the will of the people whom they represent.”[323]

Returning to the dimensions of congressional electoral authority, the Electors Clause delimits an additional region in the horizontal plane: that involving presidential elections. Within this region, Congress possesses the same power as in the zone for congressional elections, which is to say almost complete control. On the vertical axis of rationales for congressional electoral activity, the Clause also provides another justification: that a specific constitutional provision authorizes Congress to fix certain aspects of presidential elections.[324] But this basis is best understood as paralleling, not supplementing, the reason for congressional intervention supplied by the Elections Clause. In other words, a given electoral regulation can be grounded in either the Elections Clause (if it relates to congressional elections) or the Electors Clause (if it addresses presidential elections) but not both provisions.[325]

  1. The Guarantee Clause

What about a regulation of state or local elections? Congress can’t impose one under the Elections Clause or the Electors Clause (except to the extent the rule is necessarily intertwined with regulations of federal elections). Rather, Congress’s broadest source of power to legislate about nonfederal elections is the Republican Guarantee Clause of Article IV. “The United States shall guarantee to every State in this Union a Republican Form of Government,” the Clause reads.[326] States’ elections (indeed, their entire systems of government) must thus be republican. If and when they become unrepublican, the federal government is authorized to step in and restore democracy at the state or local level.

Why just at the state or local (and not at any) electoral level? One could argue, perhaps, that if a state poorly administers its federal elections, then the state’s form of government isn’t republican. But that seems implausible. A state’s federal elections determine its representatives (congressional members and presidential electors) in the government of a different sovereign: the United States. These elections don’t select the officials who actually govern the state itself. No matter how deficient a state’s federal elections, then, it’s hard to say the state has fallen below the threshold of republicanism if its nonfederal elections still top this bar. As long as these state and local elections are suitably republican, the people of the state continue to be able to govern themselves democratically. They may not be able to elect their federal representatives democratically, but their own form of government remains intact.[327]

Republicanism, of course, is a hotly contested concept.[328] It could mean quite little: just the absence of authoritarian rule.[329] Or it could mean rather a lot: free and fair elections, various civil, social, and political rights for the people, maybe even minimum levels of health, education, or welfare.[330] The Supreme Court has never defined a republican form of government, so none of these options commands judicial support. Scholars, however, have arrived at a rough consensus that, at least as an originalist matter, the essence of republicanism is popular sovereignty. “The central pillar of Republican Government . . . is popular sovereignty,” Akhil Amar concluded in one well-known study.[331] “[T]he people’s right to alter or abolish [laws], and popular majority rule in making and changing [laws]”—these “were bedrock principles in the Founding, Antebellum, and Civil War eras.”[332] Echoing Amar, Fred Smith reviewed Framing-era materials and summarized that “[t]he ‘republican principle’ is the cardinal and indispensable axiom that the ultimate sovereignty . . . rests in the hands of the governed, not persons who happen to govern.”[333]

Under this conception of republicanism, Congress can regulate nonfederal elections to prevent and remedy infringements of popular sovereignty. States can violate this principle in many ways: by restricting the franchise and so stopping certain people from having a say in their own governance, by diluting the influence of targeted racial or partisan groups, by failing to curb the corruption of the electoral process, and so on. All these practices (or absences of practices) can impede a popular majority from electing its preferred candidates and ultimately from enacting its preferred laws. All these policy choices, that is, can undermine the value of popular sovereignty that’s at the heart of the Guarantee Clause. And when this tenet is threatened, it’s clear from the Clause’s text that Congress is empowered to act. Congress is the legislative arm of the government of the United States. Under the Clause, it’s the United States that’s the guarantor of republicanism in every state.[334]

Crucially, any congressional judgment that intervention is necessary to safeguard popular sovereignty can’t be second-guessed by the courts. The Supreme Court hasn’t said much about the Guarantee Clause. But the one holding it has reiterated for almost two centuries is that issues arising under the Clause are nonjusticiable, incapable of judicial resolution. This is the case when a plaintiff invokes the Clause to argue that some state law is unrepublican and thus unconstitutional.[335] More relevant here, it’s also the case when Congress legislates pursuant to the Clause: No litigant can allege that the statute exceeds Congress’s authority to guarantee a republican form of government. As the Court put it in a foundational 1849 decision, “Congress must necessarily decide . . . whether [a state government] is republican or not.”[336] “And its decision is binding, on every other department of the government, and could not be questioned in a judicial tribunal.”[337] Or as the Court affirmed in an 1871 case, “[t]he action of Congress upon the subject cannot be inquired into.”[338] “[T]he judicial is bound to follow the action of the political department of the government, and is concluded by it.”[339]

This 1871 case involved the Reconstruction Acts, which, as described earlier, barred the former Confederate states from disenfranchising citizens on the basis of race and prescribed detailed rules for these states’ elections for their constitutional conventions.[340] Notably, the Reconstruction Acts weren’t passed under the Fourteenth or Fifteenth Amendments (which hadn’t yet been ratified). Instead, Congress enacted these statutes pursuant to its Guarantee Clause power. The first of the laws explained that “peace and good order” had to be congressionally enforced until “republican State government can be legally established.”[341] The Court also recognized that Congress’s “authority was derived from the obligation of the United States to guarantee to every State in the Union a republican form of government.”[342] Consequently, the Guarantee Clause isn’t some obsolete constitutional oddity, a “sleeping giant” (in Senator Charles Sumner’s words[343]) that has never awoken. To the contrary, in one of the most fateful periods in American history, Congress based several of its most controversial actions squarely on the Clause. At least during Reconstruction, as Sumner observed, the Clause “c[a]m[e] forward with a giant’s power.”[344]

Revisiting the dimensions of congressional electoral authority, the Guarantee Clause occupies the rest of the horizontal plane: the area corresponding to state and local elections. However, the boundaries between the Guarantee Clause’s territory (nonfederal elections) and the zones of the Elections Clause (congressional elections) and the Electors Clause (presidential elections) aren’t entirely fixed. While the best reading of the Guarantee Clause may be that it empowers Congress only with respect to nonfederal elections, if Congress nevertheless sought to regulate federal elections on this ground, its decision to do so would be judicially unreviewable. On the vertical axis of rationales for congressional electoral activity, the Guarantee Clause supplies one more parallel justification for legislation: the need to prevent or remedy states’ lapses into unrepublicanism. Again, Congress can generally cite this reason instead of, but not in addition to, the Elections Clause or the Electors Clause when it tackles electoral matters.

  1. The Fourteenth Amendment

I turn next to a series of constitutional amendments (starting with the Fourteenth) that bear some resemblance to the Guarantee Clause. Like the Clause, these amendments authorize Congress to avoid or cure certain evils. These evils, though, are distinct from the unrepublican rule that’s the Clause’s concern. Whether congressional action is consistent with the amendments is also a justiciable issue. And the amendments aren’t limited to certain types of elections but rather enable Congress to reach federal, state, and local races alike.

The provisions of the Fourteenth Amendment that are relevant here are the Equal Protection Clause of Section 1 and the Enforcement Clause of Section 5. The former prohibits each state from “deny[ing] to any person within its jurisdiction the equal protection of the laws,”[345] while the latter states that “Congress shall have power to enforce, by appropriate legislation, the provisions of this article.”[346] Historically, the courts were highly deferential toward Congress’s exercises of its Fourteenth Amendment enforcement authority. As long as the courts could “perceive a basis upon which the Congress might resolve the conflict as it did,” they would uphold the challenged legislation.[347] In a 1997 case, however, the Supreme Court changed the operative standard to require “a congruence and proportionality between the injury to be prevented or remedied and the means adopted to that end.”[348] So there must now be a closer fit between the policy chosen by Congress and the underlying constitutional harm.

It’s well-understood that “congruence and proportionality” is a more demanding test—one more likely to result in the invalidation of a congressional statute—than rational basis review.[349] But scholars haven’t fully grasped just how many “injur[ies] to be prevented or remedied” there are in the electoral context.[350] In fact, much of election law revolves around causes of action that stem from different kinds of violations of the Equal Protection Clause. All these transgressions are evils that Congress can try to avoid or cure through its Fourteenth Amendment enforcement power.

Consider burdens on the right to vote ranging from mere inconvenience to outright disenfranchisement. It’s under the Equal Protection Clause that the courts have established a system of sliding-scale scrutiny that adjusts the rigor of judicial review based on the severity of the burden.[351] Or take the principle, announced in Bush v. Gore, that ballots can’t be treated differently in one part of a state than in another. This is also an equal protection doctrine.[352] The Equal Protection Clause is the source, too, of the theories of racial vote dilution[353] (intentionally reducing the electoral influence of minority voters) and racial gerrymandering[354] (drawing districts with a predominant racial motive). While it has now been deemed nonjusticiable,[355] extreme partisan gerrymandering is unconstitutional, as well, because it offends the equal protection norm of treating each party’s voters symmetrically.[356]

That the congruence-and-proportionality standard is relatively more restrictive, then, matters less in the electoral domain than in most other areas. There are simply so many potential violations of the Equal Protection Clause—spanning franchise burdens, unequal ballot treatment, racial vote dilution, racial gerrymandering, partisan gerrymandering, and more—that Congress can seek to prevent or remedy. Put another way, the sphere of arguably unconstitutional conduct is unusually large in the electoral realm. So even if the congruence-and-proportionality standard doesn’t allow Congress to venture far beyond this sphere, what Congress can lawfully regulate is still quite a lot.[357]

The saga of the literacy test nicely illustrates the point. The Supreme Court first held that requirements that voters be literate don’t necessarily contravene the Equal Protection Clause since they sometimes have legitimate, non-invidious purposes.[358] Congress then passed a law banning any citizen who has completed the sixth grade from being denied the franchise due to an inability to read or write.[359] The Court subsequently sustained this provision, though it diverged from the Court’s earlier decision, because there was a sufficient link between Congress’s chosen policy and an underlying constitutional problem. Congress knew that “prejudice played a prominent role in the enactment of [literacy] requirement[s].”[360] The tests also weren’t very effective in “furthering the goal of an intelligent exercise of the franchise.”[361] So Congress could validly conclude that the tests “constituted an invidious discrimination in violation of the Equal Protection Clause.”[362]

Updating the matrix of congressional electoral authority, the Fourteenth Amendment doesn’t fill any more space in the horizontal plane. That is, it doesn’t empower Congress with respect to any particular category of electoral topics or levels. But on the vertical axis of rationales for congressional electoral activity, the Fourteenth Amendment does provide the first justification that can supplement, not just parallel, the reasons I have covered so far. When Congress legislates to avoid or cure breaches of the Equal Protection Clause, it can do so while also invoking its authority under the Elections Clause, the Electors Clause, or the Guarantee Clause. The same statute can be grounded in both Congress’s power over a certain class of elections and its ability to enforce the Fourteenth Amendment.

  1. The Voting Amendments

The next group of relevant amendments (the Fifteenth, Nineteenth, Twenty-Fourth, and Twenty-Sixth) can be considered as a group thanks to their shared structure. They all stipulate, using nearly identical language, that the right of citizens of the United States to vote shall not be denied or abridged on a certain basis. This forbidden basis is “race, color, or previous condition of servitude” in the Fifteenth Amendment,[363] “sex” in the Nineteenth Amendment,[364] “failure to pay any poll tax or other tax” in the Twenty-Fourth Amendment (for federal elections only),[365] and “age” in the Twenty-Sixth Amendment (for citizens eighteen or older).[366] All four amendments also describe Congress’s enforcement authority in almost the same terms (which themselves are almost the same as the Fourteenth Amendment’s Enforcement Clause). Under each provision, “Congress shall have power to enforce this article by appropriate legislation.”[367]

The Supreme Court has only commented on Congress’s enforcement authority under the Fifteenth Amendment. According to the Court, for a measure to be lawful, Congress must reasonably think the step will serve the goal of stopping racial discrimination in voting. “Congress may use any rational means to effectuate the constitutional prohibition of racial discrimination in voting,” the Court held in an important 1966 case.[368] Notably, the Roberts Court implicitly affirmed this view in the 2013 case of Shelby County v. Holder,[369] which otherwise marked the nadir of the Court’s understanding of Congress’s Fifteenth Amendment enforcement power. Over and over, the Court called it “irrational” for Congress to subject certain states to a preclearance requirement before they could change their election laws, based on a formula that relied on data from the 1960s and 1970s.[370] Not once in Shelby County did the Court hint that the formula’s flaw was its lack of congruence and proportionality with the underlying evil of racial discrimination in voting.[371]

Presumably, highly deferential rational basis review would also apply to exercises of Congress’s enforcement authority under the Nineteenth, Twenty-Fourth, and Twenty-Sixth Amendments. It would certainly be odd if Congress had less power under these strikingly similar provisions than under the Fifteenth Amendment.[372] This is speculation, though, since the Court has never addressed Congress’s ability to fight voting discrimination on the basis of sex, wealth, or age. Another note of caution is that the Court has sometimes suggested (albeit without ever holding) that only intentional racial discrimination in voting violates the Fifteenth Amendment.[373] If this dismissal of disparate impact liability became law, it, too, would presumably extend to the Nineteenth, Twenty-Fourth, and Twenty-Sixth Amendments. In that case, the standard of review might remain highly deferential, but the sphere of arguably unconstitutional conduct that Congress could target would contract. Congress would then need a rational basis for thinking that a given policy would reduce purposeful voting discrimination on account of race, sex, wealth, or age.

But this rational basis might be available surprisingly often. In one of the cases where the Court intimated that the Fifteenth Amendment only bars intentional racial discrimination in voting, the Court evaluated a provision that prevented states from revising their election laws in ways that diminished the influence of minority voters.[374] Although this was a disparate impact provision, the Court upheld it, reasoning that it was “an appropriate method of promoting the purposes of the Fifteenth Amendment.”[375] “Congress could rationally have concluded that, because electoral changes by jurisdictions with a demonstrable history of intentional racial discrimination in voting create the risk of purposeful discrimination, it was proper to prohibit changes that have a discriminatory impact.”[376] Under this logic, Congress could also attack state electoral practices that differentially burden women compared to men, the poor compared to the rich, and the young compared to the old. Again, Congress could reasonably believe that these practices’ disparate impacts indicate the presence of at least some deliberate voting discrimination, which Congress is unquestionably authorized to thwart.[377]

Like the Fourteenth Amendment, the Fifteenth, Nineteenth, Twenty-Fourth, and Twenty-Sixth Amendments don’t claim any more ground in the horizontal plane of congressional electoral power. They don’t empower Congress to legislate about specific electoral topics or levels. But also like the Fourteenth Amendment, the Fifteenth, Nineteenth, Twenty-Fourth, and Twenty-Sixth Amendments add to the vertical axis of rationales for congressional electoral activity. They supply another justification (actually, four more justifications) for Congress to regulate elections: that Congress is trying to stop voting discrimination on the basis of race, sex, wealth, or age. These reasons can be cited alongside all the ones discussed already. So Congress can root its electoral interventions not just in the Elections Clause, the Electors Clause, or the Guarantee Clause, and not just in its effort to prevent or remedy equal protection violations, but also in its resolve to end discrimination in voting.


  1. Other Provisions

This leaves a number of provisions—the Commerce Clause, the Spending Clause, and the Necessary and Proper Clause—that aren’t primarily about elections but still have significant applications in the electoral context. Beginning with the Commerce Clause, it has been construed to authorize Congress to regulate activities that have “substantial effects” on interstate commerce.[378] In an earlier time, this formulation could perhaps have endowed Congress with essentially plenary power over elections. After all, elections, well, elect candidates to office, who then implement policies about taxation, health care, immigration, and a host of other issues. These policies plainly have substantial effects on the national economy.[379] But this line of argument is now foreclosed. In more recent cases, the Supreme Court has clarified that Congress can only address commercial activities under the Commerce Clause.[380] Many aspects of elections, however, are quintessentially noncommercial. In particular, casting a ballot isn’t just a noneconomic act on its face; it’s forbidden from being made an economic transaction by the federal ban on vote buying.[381]

Other parts of the electoral process, though, are commercial in nature and so regulable under the Commerce Clause. For example, jurisdictions purchase voting machines that voters then use to cast their ballots. These machines are conventional items of commerce that cross state lines and comprise a national market. So Congress can pass laws about them, say, requiring them to be accessible to disabled people or secure against hacking.[382] Similarly, campaign finance is, as its name confirms, financial. When donors make contributions to candidates, or people or organizations spend independently to express their own electoral views, they necessarily engage in economic transactions with substantial interstate effects. Congress can thus legislate about all money in politics, at all electoral levels, pursuant to its Commerce Clause authority.[383]

Speaking of money, the Spending Clause enables Congress to offer funds to the states with strings attached: conditions the states must satisfy to receive the federal dollars. These conditions can be any measures that aren’t themselves unconstitutional—even measures that Congress wouldn’t have the right to impose directly.[384] But the conditions must be related to the purpose of the funds; Congress can’t require the states to do X to receive money earmarked for Y.[385] The overall terms of Congress’s proposals to the states also can’t be unduly coercive. The states, that is, must be genuinely free to turn down the federal dollars, along with the obligations that come with them.[386]

Under its Spending Clause power, then, Congress could affix all kinds of electoral conditions to funds offered to the states to administer their elections. These funds would have the aim of making elections freer and fairer—a better instrument for accurately registering the will of the people. So any requirement that also served this goal would be related to the funds’ purpose. Moreover, it’s unlikely that electoral legislation could run afoul of the prohibition on overly coercive federal programs. At present, the states manage to run their elections without any consistent federal subsidization. So if Congress did suddenly make federal money available, it would hardly be implausible for the states to say no, if they didn’t like the attached strings. By rejecting the congressional proposal, the states would simply be opting to maintain the status quo.[387]

Lastly, the Necessary and Proper Clause allows Congress to adopt policies that aren’t explicitly authorized by its enumerated powers but that advance these powers’ ends. Congress’s chosen measures must merely be “convenient” or “useful” or “conducive” to accomplishing goals recognized by other constitutional clauses.[388] “[T]he statute [need only] constitute[] a means that is rationally related to the implementation of a constitutionally enumerated power.”[389] However, the steps taken by Congress can’t “undermine the structure of government established by the Constitution” or amount to a new “great substantive and independent power.”[390] If they do, no matter how necessary the congressional policies may be, they aren’t proper ways of achieving constitutional objectives.

In the electoral domain, the Necessary and Proper Clause has often been invoked to justify congressional regulations of presidential elections, including the Enforcement Acts,[391] the VRA,[392] and campaign finance restrictions.[393] Recall that the only constitutional provision that mentions presidential elections, the Electors Clause, is textually quite meager.[394] The Necessary and Proper Clause has compensated for this sparse language in three different ways. Most obviously, congressional regulations of presidential elections are “convenient” or “useful” or “conducive” to the exercise of Congress’s Electors Clause authority to set the time for choosing electors.[395] Second, congressional legislation about presidential elections is rationally related to Congress’s expansive power over congressional elections. In particular, Congress’s efforts to safeguard the integrity of congressional elections would be frustrated if fraudsters could claim they were only trying to subvert concurrently held presidential elections.[396] And third, the Supreme Court has held that Congress implicitly has the authority to “preserve the departments and institutions of the general government from impairment or destruction, whether threatened by force or by corruption.”[397] Congressional rules for presidential elections further this unenumerated end, rendering them “a question primarily addressed to the judgment of Congress.”[398]

In the matrix of congressional electoral power, the Commerce Clause and the Spending Clause function just like the Fourteenth, Fifteenth, Nineteenth, Twenty-Fourth, and Twenty-Sixth Amendments. They don’t occupy any more of the horizontal plane since they don’t empower Congress to regulate any particular electoral topics or levels. But they do add notches to the vertical axis since they provide further rationales for congressional electoral activity: namely, that Congress is legislating about economic aspects of elections that have substantial effects on interstate commerce, or that Congress is linking electoral conditions to electoral funds that it’s offering to the states.

In contrast, the Necessary and Proper Clause doesn’t have a distinct place of its own in either the horizontal or the vertical dimension. Instead, it operates as a sort of supplement for all the other provisions, expanding their coverage to include not just explicitly authorized means but also all other measures that relate to the provisions’ purposes. This enhancement applies in the horizontal plane to the Elections Clause, the Electors Clause, and the Guarantee Clause, allowing each provision to spread beyond its apparent boundaries. The boost applies to the clauses that comprise the vertical axis, too, widening and deepening each justification for congressional electoral regulation.

* * *

I realize that this discussion of planes, axes, and other spatial concepts may be confusing. For visually oriented readers, I therefore present the below figure, which shows graphically how all the pieces of Congress’s electoral authority fit together. The horizontal plane on top represents the universe of American elections and indicates which provision enables Congress to regulate which category of elections. Again, this universe is divided among the Elections Clause for congressional elections, the Electors Clause for presidential elections, and the Guarantee Clause for state and local elections. All the levels below that top plane then correspond to complementary legal reasons—capable of being asserted concurrently—for congressional electoral activity. Once more, these are preventing or remedying equal protection violations under the Fourteenth Amendment; fighting voting discrimination due to race, sex, wealth, and age under the Fifteenth, Nineteenth, Twenty-Fourth, and Twenty-Sixth Amendments; addressing economic aspects of elections with substantial effects on interstate commerce under the Commerce Clause; and tying electoral conditions to electoral funds under the Spending Clause. Finally, all the borders in the diagram are somewhat indistinct because of the Necessary and Proper Clause. As just explained, it extends the reach of each enumerated power to additional means related to the power’s function. Like cement, the Clause thus fills the spaces between the other provisions, bonding them into an even firmer foundation for congressional action.




Figure 1: The Dimensions of Congressional Electoral Authority

  1. Policy Examples

At this point, having explored the theory, the history, and the doctrine of congressional electoral regulation, we can revisit the For the People Act: the far-reaching electoral reform bill passed in 2019 by the House of Representatives.[399] The Act included so many provisions that a blow-by-blow account of each element’s constitutionality would be impractical here.[400] In my view, the Act also didn’t push the limits of Congress’s electoral authority, making it less interesting whether each of its components was actually lawful. So instead, in this Part, I consider a series of proposals that are partly derived from the Act but that exceed its ambition: (1) enfranchising ex-felons in federal and state elections; (2) requiring independent redistricting commissions for congressional and state legislative districts; (3) regulating money in politics at the state level; and (4) requiring the states to use the same proportional system to allocate their presidential electors.[401] I argue that, under the law as it currently stands, Congress has the power to enact all these policies. In fact, consistent with the previous section’s emphasis on the redundancy of Congress’s electoral tools,[402] Congress could assert multiple, mutually reinforcing justifications for each regulation.

  1. Ex-Felon Enfranchisement

The For the People Act stated that a citizen’s right to vote in federal elections “shall not be denied or abridged because that individual has been convicted of a criminal offense unless such individual is serving a felony sentence in a correctional institution or facility at the time of the election.”[403] The Act would thus have barred the states from disenfranchising ex-felons—individuals who have completed their prison sentences—in federal elections. Suppose the Act extended this prohibition to state elections, too. Would Congress be authorized to adopt these measures? Or to broaden the question somewhat, state exclusions of ex-felons from the electorate are classic voting qualifications. So would Congress have the power to revise the qualifications that certain states have chosen, for voting in both federal and state elections?

It should be clear from the earlier discussion that the answer is yes. At the federal level, Mitchell involved a pair of congressionally mandated changes to certain states’ voting qualifications: setting the minimum voting age at eighteen for congressional and presidential elections, and eliminating residency requirements for voting for President.[404] Again, both these changes were upheld by the Supreme Court. According to Justice Black’s controlling opinion, Congress has the authority under the Elections Clause to amend the states’ voting qualifications for congressional elections.[405] And pursuant to the Electors Clause (supplemented by the Necessary and Proper Clause), Congress has the same power with respect to voting in presidential elections.[406] These rulings logically extend to the enfranchisement of ex-felons in federal elections. This expansion of the electorate would be materially identical to the enfranchisement of individuals aged eighteen to twenty-one, or of people who moved to a state less than a month before an election. If Congress could enable young adults and recent movers to vote, it could also secure the federal franchise for ex-felons.

At the state level, likewise, Congress could reasonably conclude that the disenfranchisement of ex-felons is an unrepublican practice that requires congressional reversal. Congress made exactly this kind of determination in the Reconstruction Acts when it used its Guarantee Clause authority to stop the ex-Confederate states from disenfranchising their African American citizens in state elections.[407] Because of the black citizens’ exclusion, the people—all the people—were not truly sovereign in the ex-Confederate states, meaning that the core value of the Guarantee Clause was undermined. So, too, with the rejection of ex-felons from the electorate: Popular sovereignty can’t be realized when a substantial slice of the citizenry is unable to participate in the political process. Or at least Congress could rationally make that judgment, which would then (like the Reconstruction Acts’ enfranchisement of black citizens) be judicially unreviewable.[408]

With respect to any voting qualification other than the disenfranchisement of ex-felons, Congress’s power to annul the restriction would be bolstered by the Fourteenth Amendment. As noted above, one of the many equal protection doctrines pertaining to election law creates a system of sliding-scale scrutiny for burdens on voting.[409] State voting qualifications often levy substantial burdens on the franchise without advancing legitimate state interests. So Congress could invoke its Fourteenth Amendment enforcement authority to attack these qualifications, arguing that its intervention is congruent and proportional to the restrictions’ violations of the Equal Protection Clause. However, this approach is inapplicable to the disenfranchisement of ex-felons. In a 1974 case, the Supreme Court held that, because Section 2 of the Fourteenth Amendment bars any state’s congressional representation from being reduced on the ground that the state has disenfranchised citizens “for participation in rebellion, or other crime,” such disenfranchisement can’t contravene the Equal Protection Clause of Section 1.[410] If the exclusion of ex-felons can’t breach the Equal Protection Clause, then Congress also can’t try to prevent or remedy this exclusion through its enforcement power. There simply isn’t an underlying equal protection problem for Congress to avoid or cure.

But this unique textual limitation has no bearing on Congress’s enforcement power under the Fifteenth Amendment. To exercise that power, Congress merely needs a rational basis to think that its chosen policy will fight racial discrimination in voting.[411] It would certainly be reasonable for Congress to believe that the enfranchisement of ex-felons would make American elections less racially discriminatory. In numerous states, ex-felons were disenfranchised in the late nineteenth and early twentieth centuries as elements of larger efforts to intentionally prevent African Americans from voting.[412] Even in states without Jim Crow histories of voter suppression, the disenfranchisement of ex-felons disproportionately excludes minority members from the electorate.[413] So the rational basis for the congressional enfranchisement of ex-felons is either obvious (that these individuals were barred from voting for invidious reasons) or the same as in prior cases (that the disparate impact of ex-felons’ disenfranchisement suggests a racially discriminatory motive).[414] Either way, the predicate for the use of Congress’s Fifteenth Amendment enforcement authority is satisfied.

Lastly, to the extent there are holes in these arguments under other constitutional provisions, the Necessary and Proper Clause helps to fill them. For instance, say that voting qualifications aren’t part of the “Manner” of congressional elections that Congress is entitled to regulate under the Elections Clause (a view probably held by the current Court[415]). The enfranchisement of ex-felons could still be rationally related to the Elections Clause’s purpose: uniformly administering congressional elections in the ways specified by Congress. Or assume (also plausibly for the current Court[416]) that the disparate impact of ex-felons’ exclusion isn’t highly probative of racially discriminatory intent. Again, including these individuals in the electorate might still be “convenient” or “useful” or “conducive” to achieving the Fifteenth Amendment’s aspiration of elections untainted by racial prejudice.[417]

  1. Redistricting Commissions

Turning from the franchise itself to the districts in which people vote, the For the People Act would have required the states to redraw their congressional districts using independent commissions. These commissions would have been comprised of fifteen members (five Democrats, five Republicans, and five independents) selected through a complex process.[418] The commissions would then have designed districts based on the criteria of equal population, compliance with the VRA, further protections for minority voting rights, respect for communities of interest, and partisan fairness.[419] Imagine that these provisions applied to state legislative districts, too. Would Congress have the power to enact them?

This is an easy call at the federal level. Congressional redistricting is in the heartland of Congress’s Elections Clause authority—a subject about which Congress has legislated many times before. Just as Congress has previously required congressional districts to elect a single member and to be contiguous, equipopulous, and compact,[420] it could specify additional criteria for these districts and insist they be crafted by commissions instead of politicians. The only twist is that, as drafted, the For the People Act would have compelled the states to create commissions themselves (rather than having the federal government establish them for unwilling states).[421] This could be seen as commandeering the states, which is generally a constitutional problem—but which isn’t proscribed when Congress acts pursuant to the Elections Clause. As observed earlier, Congress can dictate “the manner in which a State is to fulfill its pre-existing constitutional obligations” to administer federal elections.[422]

At the state level, the Guarantee Clause is the natural basis for congressionally mandated redistricting commissions. When politicians are free to carve their own districts, using the criteria of their choice, unrepublican abuses often ensue. Politicians insulate themselves from competition, gerrymander against the opposing party, dilute the votes of racial minorities, and so on. Independent commissions following fixed redistricting criteria help to prevent these evils from arising. They ensure that redistricting is impartial instead of self-interested, that its parameters are transparent rather than opaque—that the sovereignty of the people isn’t subverted by cleverly drawn districts.[423] Nor would it be unprecedented for Congress to regulate nonfederal redistricting under the Guarantee Clause. As pointed out above, in the Reconstruction Acts, Congress apportioned delegates to the ex-Confederate states’ constitutional conventions based on the one-person, one-vote rule.[424] Congress thus avoided the representational distortion caused by malapportionment: a skew very much like that caused by gerrymandering.

On the topic of gerrymandering, its deterrence is another justification for congressionally mandated redistricting commissions, at both the federal and state levels, under the Fourteenth Amendment. In a recent case, the Supreme Court held that partisan gerrymandering is nonjusticiable.[425] But even if the practice can’t be policed by the federal courts, the Court confirmed that “unconstitutional partisan gerrymandering” does exist—that district maps can be so tilted against a targeted party as to infringe the Equal Protection Clause.[426] This constitutional offense, like all equal protection violations, is one that Congress can try to prevent or remedy through its Fourteenth Amendment enforcement power. And independent commissions following fixed redistricting criteria indeed make gerrymandering less likely. They replace self-interested politicians with mapmakers more apt to pursue the public good. They also exclude partisan advantage as an aim, or even, like the For the People Act, affirmatively require that district plans not “unduly favor or disfavor any political party.”[427] Congressionally mandated redistricting commissions are thus a congruent and proportional response to an activity, gerrymandering, that can be unconstitutional even if it can’t be judicially curbed.

Analogous reasoning applies to Congress’s enforcement power under the Fifteenth Amendment. Left to their own devices, politicians commonly dilute the influence of racial groups by cracking or packing their members. But commissions like those prescribed by the For the People Act transfer redistricting authority to neutral mapmakers less inclined to engage in racial vote dilution. These commissions are also directed “not [to] dilute or diminish [minority members’] ability to elect candidates of choice.”[428] The commissions are therefore a rational way for Congress to combat racial discrimination in voting. And again, this (and each other) argument for the commissions’ validity is reinforced by the Necessary and Proper Clause. At the very least, the commissions are conducive to the achievement of goals recognized by other constitutional provisions: uniformly administered federal elections (the Elections Clause), republican state elections (the Guarantee Clause), elections without partisan gerrymandering (the Fourteenth Amendment), and elections without racial vote dilution (the Fifteenth Amendment).[429]

  1. State Campaign Finance Regulation

Next, the For the People Act included several new regulations of federal campaign finance. It would have strengthened disclosure requirements,[430] instituted a system of public financing for congressional elections,[431] and restructured the Federal Election Commission, the agency responsible for enforcing these rules.[432] No one doubts Congress’s power to legislate about federal campaign finance (even though this legislation’s compliance with the First Amendment is often controversial). But what if Congress sought to regulate money in politics at the state level? That is, what if Congress replaced the hodgepodge of existing state rules with a single national regime of disclosure requirements, contribution limits, and public financing?[433] Would this be a constitutionally authorized intervention?

Since it would be an intervention aimed at state elections, the Guarantee Clause is the logical place to start. Congress could reasonably think that, left unchecked, money in state politics threatens republican rule. It breeds corruption, it distorts election outcomes, and it induces officeholders to do the bidding of their funders (not their voters). Congress could also rationally believe that its policies would advance the Guarantee Clause’s fundamental value of popular sovereignty. When donations to candidates are disclosed and restricted, and when candidates don’t even need to solicit private contributions thanks to the availability of public financing, candidates have less incentive to prioritize their funders’ interests over those of their constituents.[434] In any event, any congressional determination to this effect would be judicially unreviewable. No court could conclude, contrary to Congress, that unregulated campaign finance doesn’t imperil republicanism or that limits on electoral funds don’t vindicate this principle.

Even given its constriction over the last generation, the Commerce Clause would further empower Congress to legislate about state campaign finance. As explained earlier, contributions to candidates and independent expenditures about elections are indisputably commercial transactions.[435] They inevitably involve the transfer of funds from one party to another. When aggregated, money in politics also plainly has substantial effects on interstate commerce. Billions of dollars are raised and spent in every American election.[436] These sums often flow into a given race from out of state, and then stream back out into the accounts of consultants, ad makers, and other nonresident campaign professionals.[437] State campaign finance is therefore at the center of Congress’s Commerce Clause authority. It’s an inherently economic activity that typically occurs at a national level.

Depending on how the legislation was structured, the Spending Clause could provide another basis for Congress to regulate state campaign finance. In particular, suppose that Congress offered the money for public financing not directly to candidates but instead to the states, which could then distribute the funds in a number of ways. In that case, Congress could attach conditions to the money like the imposition of disclosure requirements and contribution limits. These conditions would be lawful since they would serve the same purpose as the funds themselves, namely, a less corrupt government that better reflects the will of the people. And once more, the Necessary and Proper Clause would augment both this and each other rationale for the congressional regulation of state campaign finance. Congress’s new rules would only need a rational link to the other constitutional provisions’ objectives of republican rule, national control of the national economy, and governmental outlays that mirror congressional priorities.

  1. Proportional Allocation of Electors

Finally, the For the People Act contained a grab bag of provisions about presidential elections. The Act would have strengthened disclosure requirements for presidential inaugural committees,[438] revised the existing system of public financing for presidential elections,[439] and instituted various ethics reforms for the President and other executive branch officials.[440] However, the Act didn’t address the actual mechanisms by which the states allocate their presidential electors: winner-take-all elections everywhere except Maine and Nebraska.[441] But imagine that Congress legislated about this topic, too. Specifically, say that Congress required the states to allocate their presidential electors proportionately based on the share of the statewide vote received by each candidate. Would Congress have the power to take this step?

The argument that it would begins with the Electors Clause (fortified by the Necessary and Proper Clause). Recall that these provisions have been construed to endow Congress with the same broad authority over presidential elections that it enjoys over congressional elections under the Elections Clause.[442] Under the Elections Clause, Congress can compel the states to elect their House members in certain ways (like through single-member districts) but not in others (like using the general ticket).[443] So if Congress’s powers over presidential and congressional elections are in fact the same, then Congress can also dictate how the states allocate their presidential electors. The general ticket, in particular, is simply a congressional analogue of the winner-take-all system that almost all states have adopted for assigning their electors. If Congress can forbid the general ticket, then it can also ban winner-take-all presidential elections.

This claim, though, runs into a textual retort. Before it authorizes Congress to fix the time of choosing electors, Article II, Section 1 stipulates that each state may appoint its electors “in such Manner as the Legislature thereof may direct.”[444] The discretion conferred to the states by this provision would allegedly be infringed by congressional legislation telling the states how to allocate their electors.[445] But this rejoinder has no more force than the view that Congress can’t regulate the qualifications for voting in congressional elections because Article I, Section 2 empowers the states to set these qualifications in the first instance.[446] As discussed above, Justice Black’s controlling opinion in Mitchell rejected this position, holding that while the states can adopt congressional voting qualifications, Congress can supersede these criteria if it wishes.[447] If Congress’s presidential and congressional electoral powers are indeed identical, then the same result should follow here. The states should be able to select their preferred mechanisms for assigning their electors—but Congress should be free to displace these choices as it sees fit. Any other outcome would disrupt the parallelism between Congress’s presidential and congressional electoral powers.

Both other bases for Congress to prohibit winner-take-all presidential elections involve their potential constitutional violations. First, several ongoing suits assert that these elections contravene the Equal Protection Clause by flouting the one-person, one-vote rule and diluting the influence of political minorities.[448] According to these suits, the ballots of voters who support candidates failing to earn a plurality of the vote end up carrying no weight, and groups of voters backing these less popular candidates are entirely unrepresented among states’ electors.[449] These arguments may or may not ultimately succeed,[450] but they’re at least colorable. Under its Fourteenth Amendment enforcement authority, Congress could therefore try to prevent and remedy these potential equal protection problems. Replacing winner-take-all presidential elections with the proportional allocation of electors would do just that. If electors were assigned proportionately, then each citizen’s vote for President would count no more and no less than any other citizen’s vote. The sway of each group of voters would also rise and fall in tandem with the group’s size—not toggle between complete control and none whatsoever based on whether the group constituted a statewide plurality or not.

Second, winner-take-all presidential elections dilute the votes of racial minorities in certain states. Virtually everywhere in America, most minority members vote for Democratic presidential candidates.[451] But if these individuals happen to live in states where Republican candidates generally win a plurality of the vote, then their ballots earn them no representation among the states’ electors. Instead, all these electors espouse candidates opposed by the minority community.[452] Under its Fifteenth Amendment enforcement power, then, Congress could rationally target this disparate racial impact, which could be indicative of intentional racial discrimination. Congress could also reasonably conclude that the proportional allocation of electors would eliminate the racial vote dilution caused by winner-take-all presidential elections. If electors were assigned proportionately, then some electors—not none—would share the preferences of most minority voters, even in Republican-dominated states. Moreover, this fraction of politically aligned electors wouldn’t be arbitrary; it would be precisely the share of the vote received by the minority community’s favored candidate.

* * *

I want to emphasize that all the analysis in this Part is based on current law—constitutional doctrine as it presently stands. Of course, current law is just that: current. It could change in the future. And it likely will change since the Roberts Court has a dimmer view of Congress’s electoral authority than that manifested in the existing case law. My claim, then, isn’t that this Court would necessarily uphold Congress’s enfranchisement of ex-felons, institution of redistricting commissions, regulation of state campaign finance, or proscription of winner-take-all presidential elections. In fact, this Court would likely amend the applicable doctrine to cast doubt on the validity of these measures. My argument, rather, is that Congress has the power to enact these policies under the best reading of current law; and equally importantly that Congress should have this power because it’s a more trustworthy electoral regulator than the states or the courts. Put differently, it’s a happy coincidence that current law grants Congress the expansive authority over elections that, in my view, it ought to possess as a normative matter. This fortuity might fade in the future, especially as the Roberts Court decides more election law cases, but for now it still holds.


American democracy is at the brink of a profound—and profoundly positive—transformation. For the first time, a chamber of Congress has passed a bill that would address the most glaring flaws of the American electoral process: voter suppression, racial discrimination, gerrymandering, the undue influence of the wealthy, and so on. Depending on political developments, this bill could become law in the near future. In this Article, I have provided a multilayered foundation for such sweeping congressional legislation. Specifically, I have advanced four propositions: (1) that, from a theoretical perspective, Congress is less apt to threaten democratic values than are the states or the courts; (2) that, historically, most congressional electoral regulation has actually promoted democratic values; (3) that current law endows Congress with broad power over most electoral levels and topics; and (4) that, under this doctrine, Congress could venture even further than the For the People Act tries to go. Together, these points should hearten legislators when they next turn to the project of electoral reform. Not only is aggressive federal action permissible in the American political system—it may be the only way to save it.




         *    Professor of Law, Harvard Law School. I am grateful to Rick Hasen, Michael Morley, and Larry Solum for their helpful comments. My thanks also to the workshop participants at the University of Pittsburgh, where I presented an earlier version of this Article.

        [1].    See For the People Act of 2019, H.R. 1, 116th Cong. (2019). For other observers noting the bill’s significance, see Franita Tolson, The Elections Clause and the Underenforcement of Federal Law, 129 Yale L.J. F. 171, 174 (2019) (“If enacted, H.R. 1 would be the most expansive exercise of federal power over elections since the VRA and the most aggressive assertion of federal authority under the Elections Clause since Reconstruction.”); For the People: Our American Democracy: Hearing Before the H. Comm. on House Administration, 116th Cong. 14 (2019) (statement of Chiraag Bains) (“H.R. 1 [is] the boldest and most comprehensive proposal to strengthen our democracy since the aftermath of Watergate.”).

As this Article goes to press (May 2021), Congress has again begun to consider omnibus electoral reform. See For the People Act of 2021, H.R. 1, 117th Cong. (2021). This Article doesn’t explicitly address this more recent debate, whose twists and turns remain in the future.

        [2].    See H.R. 1 §§ 1011–21. All these reforms were for federal elections only.

        [3].    See id. §§ 1401–08.

        [4].    See id. §§ 2400–15.

        [5].    See id. §§ 5101–16.

        [6].    See id. §§ 6001–10.

        [7].    52 U.S.C. §§ 20501–11 (2018).

        [8].    52 U.S.C. §§ 20901–21145 (2018).

        [9].    See Federal Election Campaign Act Amendments of 1974, Pub. L. No. 93-443, 88 Stat. 1263 (1974).

      [10].    Pub. L. No. 89-110, 79 Stat. 437 (1965).

      [11].    See, e.g., An Act to Provide for the More Efficient Government of the Rebel States, 14 Stat. 428 (1867).

      [12].    Kate Ackley, HR 1 Debate Gets Under Way as GOP Sharpens Attacks, Roll Call (Mar. 6, 2019, 5:50 PM),

      [13].    Equally importantly, McConnell opposed the bill for the political reason that it would supposedly “rewrite the rules to favor Democrats and their friends.” 165 Cong. Rec. S730 (daily ed. Jan. 29, 2019).

      [14].    165 Cong. Rec. S776 (daily ed. Jan. 31, 2019) (statement of Sen. McConnell).

      [15].    Id.

      [16].    Id.

      [17].    For the People Act of 2019: Hearing on H.R. 1 Before the H. Comm. on the Judiciary, 116th Cong. 2 (2019).

      [18].    Ilya Shapiro & Nathan Harvey, What Left-Wing Populism Looks Like, Nat’l Rev. (Mar. 7, 2019, 6:30 AM),


      [19].    For the People Act of 2019: Hearing on H.R. 1 Before the H. Comm. on the Judiciary, 116th Cong. 6 (2019).

      [20].    The Democrats’ Election-Reform Bill Is an Unconstitutional, Authoritarian Power Grab, Nat’l Rev. (Mar. 10, 2019, 9:32 PM),

      [21].    To be sure, some scholars have analyzed multiple electoral provisions in tandem—though not all the clauses that shape Congress’s authority over elections. See, e.g., Pamela S. Karlan & Daniel R. Ortiz, Congressional Authority to Regulate Elections, in The Federal Regulation of Elections: Background Report of the Task Force on Legal and Constitutional Issues 14, 14 (2001) (considering “[f]our major constitutional provisions [that] give Congress power to regulate how states and localities conduct elections”); Michael T. Morley, Prophylactic Redistricting – Congress’s Section 5 Power and the New Equal Protection Right to Vote, 59 Wm. & Mary L. Rev. 2053, 2061 (2018) (considering the Elections Clause and the Enforcement Clauses of the Fourteenth and Fifteenth Amendments); Franita Tolson, The Spectrum of Congressional Authority Over Elections, 99 B.U. L. Rev. 317, 323 (2019) (also considering a “combination of the Elections Clause and the Fourteenth and Fifteenth Amendments”).

      [22].    But see supra note 21 and accompanying text (noting some exceptions).

      [23].    See The Federalist No. 10 (James Madison).

      [24].    United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4 (1938).

      [25].    See generally John Hart Ely, Democracy and Distrust: A Theory of Judicial Review (1980).

      [26].    See generally Nicholas O. Stephanopoulos, The Anti-Carolene Court, 2019 Sup. Ct. Rev. 111.

      [27].    Carolene Prods., 304 U.S. at 152 n.4.

      [28].    See, e.g., United States v. Reese, 92 U.S. 214 (1875).

      [29].    See generally Stephanopoulos, supra note 26.

      [30].    See supra notes 1–11 and accompanying text.

      [31].    See The Federalist No. 10 (James Madison).

      [32].    U.S. Const. art. I, § 4.

      [33].    Arizona v. Inter Tribal Council of Ariz., Inc., 570 U.S. 1, 8 (2013) (quoting Smiley v. Holm, 285 U.S. 355, 366 (1932)).

      [34].    U.S. Const. art. IV, § 4.

      [35].    See generally Akhil Reed Amar, The Central Meaning of the Republican Government: Popular Sovereignty, Majority Rule, and the Denominator Problem, 65 U. Colo. L. Rev. 749 (1994).

      [36].    See, e.g., Luther v. Borden, 48 U.S. (7 How.) 1 (1849).

      [37].    See U.S. Const. amend. XIV, § 5.

      [38].    For cites to representative cases, see infra Part III.D.

      [39].    See, e.g., City of Boerne v. Flores, 521 U.S. 507 (1997).

      [40].    Id. at 520.

      [41].    See For the People Act of 2019, H.R. 1, 116th Cong. §§ 1401–08 (2019).

      [42].    See, e.g., Oregon v. Mitchell, 400 U.S. 112, 120–31 (1970) (opinion of Black, J.).

      [43].    See, e.g., Hunter v. Underwood, 471 U.S. 222, 229 (1985) (striking down an Alabama provision that “was enacted with the intent of disenfranchising blacks”).

      [44].    See For the People Act of 2019, H.R. 1, 116th Cong. §§ 2400–15 (2019).

      [45].    U.S. Const. art. I, § 4.

      [46].    See Rucho v. Common Cause, 139 S. Ct. 2484 (2019).

      [47].    See, e.g., id. at 2506 (acknowledging that “gerrymandering is incompatible with democratic principles” (internal quotation marks omitted)).

      [48].    See Richard Primus, The Limits of Enumeration, 124 Yale L.J. 576, 578 (2014) (also distinguishing between “internal limits” and “external limits” on congressional powers).

      [49].    See, e.g., Citizens United v. FEC, 558 U.S. 310 (2010).

      [50].    See, e.g., Shaw v. Reno, 509 U.S. 630 (1993).

      [51].    See, e.g., Printz v. United States, 521 U.S. 898 (1997).

      [52].    See, e.g., Shelby Cnty. v. Holder, 570 U.S. 529 (2013).

      [53].    See Nicholas O. Stephanopoulos, Elections and Alignment, 114 Colum. L. Rev. 283 (2014).

      [54].    I’m not the only one who finds this argument ingenious. See, e.g., Frank H. Easterbrook, The State of Madison’s Vision of the State: A Public Choice Perspective, 107 Harv. L. Rev. 1328, 1333 (1994) (“Recognizing the simultaneous terror, inevitability, and desirability of faction, and proposing conquest by division (the strategy of faction itself), is genius.”).

      [55].    Madison, supra note 23.

      [56].    Id.

      [57].    Id.

      [58].    Id. Madison returned to this theme in The Federalist No. 51. “[B]y comprehending in the society so many separate descriptions of citizens as will render an unjust combination of a majority of the whole very improbable,” the likelihood of “oppression” or “injustice” is lowered. Madison, supra note 31.

      [59].    See, e.g., Alan Gibson, Madison’s “Great Desideratum”: Impartial Administration and the Extended Republic, 1 Am. Pol. Thought 181, 201 (2012) (making the easy extension of The Federalist No. 10 to Congress).

      [60].    Madison, supra note 23.

      [61].    Madison, supra note 31.

      [62].    Id.

      [63].    For an example of a scholar echoing Madison, see Neal Riemer, James Madison’s Theory of the Self-Destructive Features of Republican Government, 65 Ethics 34, 38 (1954), noting that the “separation of powers . . . would provide another safeguard against factional abuse of power.”

      [64].    Hamilton made this exact argument in The Federalist No. 60. The federal government would not abuse its “power over the elections,” he maintained, because of “the dissimilar modes of constituting the several component parts of the government.” The Federalist No. 60 (Alexander Hamilton).These dissimilar modes meant “there would be little probability of a common interest to cement these different branches in a predilection for any particular class of electors.” Id.

      [65].    For a classic study, see William L. Morrow, Congressional Committees (1969). However, “[b]oth chambers have procedures for calling up measures that have not been reported by a committee,” so the committee veto point can be circumvented in some circumstances. Judy Schneider, Cong. Rsch. Serv., House and Senate Rules of Procedure: A Comparison 4 (2008),

      [66].    The only exception would be electoral bills that primarily involve revenue increases or decreases, which could be passed through the budget reconciliation process. See generally Bill Heniff Jr., Cong. Rsch. Serv., The Budget Reconciliation Process: The Senate’s “Byrd Rule” (2021),

      [67].    Of course, congressional committees and the Senate filibuster aren’t prescribed by the Constitution. But their lack of constitutional grounding is irrelevant to my argument that it’s very difficult for an undemocratic faction to seize control of Congress. The result is the same whether the faction is stymied by constitutional or non-constitutional structures.

      [68].    See generally John G. Gunnell, The Genealogy of American Pluralism: From Madison to Behavioralism, 17 Int’l. Pol. Sci. Rev. 253 (1996).

      [69].    See, e.g., Nicholas O. Stephanopoulos, The Dance of Partisanship and Districting, 13 Harv. L. & Pol’y Rev. 507, 521–22 (2019) (surveying the political science literature on polarization).

      [70].    See, e.g., Daryl J. Levinson & Richard H. Pildes, Separation of Parties, Not Powers, 119 Harv. L. Rev. 2312, 2338 (2006) (“Under unified governments, smaller partisan majorities will be able to effect major policy change without the full range of checks and balances that are supposed to divide and diffuse power in the Madisonian system.”). In contemporary American politics, it’s plainly the Republican Party that’s more vulnerable to takeover by an antidemocratic faction. Indeed, given the high level of Republican support for measures like photo ID requirements for voting, limits on voter registration, and purges of voter rolls, the Party is arguably already committed to policies that impede democratic participation.

      [71].    See Divided Government in the United States, Wikipedia, Divided_government_in_the_United_States (last visited Aug. 1, 2020).

      [72].    The only such periods over the last half-century were 1977–79 and 2009–11. See Party Division, U.S. Senate, (last visited Aug. 1, 2020).

      [73].    See, e.g., Boris Shor & Nolan McCarty, The Ideological Mapping of American Legislatures, 105 Am. Pol. Sci. Rev. 530, 546 (2011) (showing the levels of legislative polarization in Congress and all fifty states).

      [74].    Vices of the Political System of the United States, in Selected Writings of James Madison 35, 35 (Ralph Ketcham ed., 2006).

      [75].    Id. at 39–41.

      [76].    Id. at 39. For examples of scholars noting Madison’s distrust of the states, see Larry D. Kramer, Madison’s Audience, 112 Harv. L. Rev. 611, 634 (1999); James S. Liebman & Brandon L. Garrett, Madisonian Equal Protection, 104 Colum. L. Rev. 837, 843 (2004).

      [77].    City of Richmond v. J.A. Croson Co., 488 U.S. 469, 522–23 (1989) (Scalia, J., concurring in the judgment) (internal quotation marks omitted).

      [78].    Id. at 523.

      [79].    Madison, supra note 23; see, e.g., Adam Winkler, Free Speech Federalism, 108 Mich. L. Rev. 153, 161 (2009) (arguing that, because “the range of represented interests tends to be smaller and the constituencies more homogenous” in the states, “oppressive legislation is easier to achieve”).

      [80].    See Reynolds v. Sims, 377 U.S. 533, 568 (1964) (“[T]he seats in both houses of a bicameral state legislature must be apportioned on a population basis.”). The irony isn’t lost on me that Congress may be less likely to enact undemocratic policies in part because of an undemocratic institution (the malapportioned Senate).

      [81].    This state legislature is Minnesota’s. See State Partisan Composition, Nat’l Conf. State Legislatures (May 13, 2021),

      [82].    See Veto Overrides in State Legislatures, Ballotpedia, state_legislatures (last visited Aug. 1, 2020).

      [83].    See Who Draws the Lines?, All About Redistricting, (last visited Aug. 1, 2020).

      [84].    See Letter from James R. White, Assoc. Dir., U.S. Gen. Acct. Off., to Sen. Larry E. Craig (June 2, 1998), assets/90/87978.pdf.

      [85].    Madison, supra note 31.

      [86].    My claim also isn’t absolute. As I discuss briefly in the next Part, it’s certainly possible for Congress to undermine democratic values. It plainly did so, for example, when it repealed the Reconstruction-era voting statutes in 1894 and when it eliminated most criteria for congressional redistricting in 1929. See infra notes 271–275 and accompanying text.

      [87].    See supra notes 1–6 and accompanying text.

      [88].    For other scholars noticing this aspect of our constitutional architecture, see Levinson & Pildes, supra note 70, at 2325–26 (describing Woodrow Wilson’s view that “Madisonian government was dramatically ineffective and vulnerable to paralysis and stalemate”); and Riemer, supra note 63, at 38 (“Madison’s theory is essentially negative: that it may block legislation in the public interest as well as legislation opposed to the public interest.”).

      [89].    See Automatic Voter Registration, Nat’l Conf. State Legislatures
(Feb. 8, 2021),

      [90].    See Felon Voting Rights, Nat’l Conf. State Legislatures (June 1, 2021), research/elections-and-campaigns/felon-voting-rights.aspx.

      [91].    See Redistricting Commissions: State Legislative Plans, Nat’l Conf. State Legislatures (Apr. 30, 2021), (reporting the figure for commissions with primary responsibility for state legislative plans).

      [92].    See Public Financing of Campaigns: Overview, Nat’l Conf. State Legislatures (Feb. 8, 2019),

      [93].    For other scholars noting the potential of states to pass pro-democratic policies, see generally Joshua A. Douglas, Vote for US: How To Take Back Our Elections and Change the Future of Voting (2019); Jessica Bulman-Pozen & Miriam Seifter, The Democracy Principle in State Constitutions, 119 Mich. L. Rev. 859 (2021).

      [94].    United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4 (1938); see also supra note 24 and accompanying text.

      [95].    Ely, supra note 25, at 103.

      [96].    I’m as guilty as anyone of this traditional way of thinking. See generally Stephanopoulos, supra note 26.

      [97].    For other scholars noting the capacity of Congress to regulate elections in pro-democratic ways, thereby avoiding any need for judicial intervention, see Pamela S. Karlan, Democracy and Disdain, 126 Harv. L. Rev. 1, 15 (2012) (“[C]ourts should exercise special restraint . . . when the political system itself is ‘clearing the channels of political change’ . . . .”); Ethan J. Leib, Redeeming the Welshed Guarantee: A Scheme for Achieving Justiciability, 24 Whittier L. Rev. 143, 147 (2002) (“[N]eo-republicans [should] shift their attention from the justification for judicial action, to the justification for Congress to act in service of republican ends.”).

      [98].    I only address the Supreme Court here, though a good deal of the analysis would also apply to the lower federal courts.

      [99].    See supra notes 54–73 and accompanying text.

    [100].    See infra notes 104–115 and accompanying text.

    [101].    Madison, supra note 31; see, e.g., David F. Epstein, The Political Theory of the Federalist 191 (2008) (also observing that the judiciary “did not appear in Federalist 51’s description of ambition counteracting ambition”).

    [102].    The Federalist No. 78 (Alexander Hamilton).

    [103].    For a similar perspective on the Court, see Eric A. Posner & Adrian Vermeule, Inside or Outside the System?, 80 U. Chi. L. Rev. 1743, 1753 (2013) (“[T]he judiciary is just another of the branches struggling to encroach upon the others or to aggrandize itself at the expense of the others; judges are just part of the invisible-hand system, not some sort of external regulator of the system.”).

    [104].    Madison, supra note 23.

    [105].    Id.

    [106].    Id.

    [107].    See Matt Flegenheimer, Senate Republicans Deploy ‘Nuclear Option’ to Clear Path for Gorsuch, N.Y. Times (Apr. 6, 2017), 04/06/us/politics/neil-gorsuch-supreme-court-senate.html.

    [108].    See Russell Wheeler, Should We Restructure the Supreme Court?, Brookings (Mar. 2, 2020),

    [109].    I don’t claim that the Court has been more affected by polarization than other governmental institutions.

    [110].    See, e.g., Neal Devins & Lawrence Baum, Split Definitive: How Party Polarization Turned the Supreme Court into a Partisan Court, 2016 Sup. Ct. Rev. 301, 346 (describing “a milieu of people and groups,” including “the American Bar Association, legal academics, Supreme Court reporters, and elite social circles in Washington, D.C.,” that “influenced the behavior of the Justices” in the past).

    [111].    See, e.g., id. at 304 (“Polarization is reflected in the social networks that are critical to grooming and identifying appointees to the federal courts . . . [And] also pervades the social networks of which the Justices themselves are a part . . . .”).

    [112].    At least, this is so if the faction corresponds to one or another side of the ideological spectrum, and if this side’s stances are anti-democratic. See Amelia Thomson-DeVeaux, Laura Bronner & Anna Wiederkehr, What the Supreme Court’s Unusually Big Jump to the Right Might Look Like, FiveThirtyEight (Sept. 22, 2020), (showing the impressive ideological consistency over time of recent Supreme Court Justices).

    [113].    One Justice, Samuel Chase, was impeached by the House but not convicted by the Senate. See generally William H. Rehnquist, Grand Inquests: The Historic Impeachments of Justice Samuel Chase and President Andrew Johnson (1992).

    [114].    See Richard L. Hasen, End of the Dialogue? Political Polarization, the Supreme Court, and Congress, 86 S. Cal. L. Rev. 205, 217–19 (2013) (documenting the decline in congressional overrides in recent years).

    [115].    See generally Richard H. Pildes, Is the Supreme Court a Majoritarian Institution, 2010 Sup. Ct. Rev. 103 (2010) (discussing the Court’s largely unfettered room to maneuver in most cases).

    [116].    This is the same point I made earlier about the states: that they’re more likely than Congress to act in anti- and pro-democratic ways. See supra notes 89–93 and accompanying text.

    [117].    And not just in my view. See generally Ely, supra note 25.

    [118].    See, e.g., Harper v. Va. State Bd. of Elections, 383 U.S. 663 (1966) (striking down the poll tax).

    [119].    See, e.g., South Carolina v. Katzenbach, 383 U.S. 301 (1966) (upholding the Voting Rights Act).

    [120].    See, e.g., Reynolds v. Sims, 377 U.S. 533 (1964) (striking down state legislative malapportionment).

    [121].    See infra Part II.

    [122].    See, e.g., Alden v. Maine, 527 U.S. 706, 709 (1999) (“Federalism requires that Congress accord States the respect and dignity due them as residuary sovereigns and joint participants in the Nation’s governance.”).

    [123].    501 U.S. 452, 458 (1991); see also FERC v. Mississippi, 456 U.S. 742, 788–90 (1982) (O’Connor, J., concurring in the judgment in part and dissenting in part) (endorsing federalism on similar grounds); Heather K. Gerken, Our Federalism(s), 53 Wm. & Mary L. Rev. 1549, 1552 (2012) (“Federalism theory has long exhibited a healthy pluralism with regard to the ends federalism promotes . . . . The Supreme Court reels off these arguments as easily as scholars do.”). These pluralistically inclined scholars include Richard H. Fallon, Jr., Barry Friedman, and Daniel Halberstam. See Richard H. Fallon, Jr., The Conservative Paths of the Rehnquist Court’s Federalism Decisions, 69 U. Chi. L. Rev. 429, 440–41 (2002) (listing several benefits of federalism); Barry Friedman, Valuing Federalism, 82 Minn. L. Rev. 317, 389, 394, 397, 402 (1997) (same); Daniel Halberstam, Federalism: A Critical Guide 14–15 (Mich. L. Sch. Pub. L. & Legal Theory Working Paper, Paper No. 251, 2011) (same).

    [124].    However, I don’t contend that Gregory’s federalism values support extensive congressional power in other, non-electoral areas. How these values might apply elsewhere is beyond this project’s scope.

    [125].    See Gregory, 501 U.S. at 458.

    [126].    See infra Part II.

    [127].    See supra Part I.A (discussing this account).

    [128].    Moreover, these pro-democratic state policies are enacted more frequently than analogous congressional laws. See supra notes 89–93 and accompanying text.

    [129].    See Gregory, 501 U.S. at 458.

    [130].    See, e.g., An Act to Enforce the Right of Citizens of the United States to Vote in the Several States of This Union, and for Other Purposes, 16 Stat. 140 (1870).

    [131].    See, e.g., 52 U.S.C. § 10303(e).

    [132].    See id. § 20504.

    [133].    See id. § 20902.

    [134].    See Gregory, 501 U.S. at 458.

    [135].    Though only if Congress means what it says; otherwise certain states wouldn’t have engaged in almost a hundred years of continued (and clever) racial discrimination in voting after Congress’s dramatic interventions during Reconstruction. See infra Part II.B; see, e.g., Winkler, supra note 79, at 183 (“[C]hoice and diversity are not necessarily values that should be encouraged when it comes to fundamental rights.”).

    [136].    At least, it doesn’t necessarily do so, as long as the pro-democratic congressional legislation is drafted in the right non-preclusive terms.

    [137].    The Federalist No. 43 (James Madison).

    [138].    Id.

    [139].    Id.; see, e.g., William M. Wiecek, The Guarantee Clause of the U.S. Constitution 26 (1972) (explaining how Montesquieu, who heavily influenced Madison, “also insisted that in a confederation all governments had to be republican because in a mixed confederacy a monarchy would swallow up its republican neighbors”).

    [140].    See Gregory v. Ashcroft, 501 U.S. 452, 458 (1991).

    [141].    See State Laws Governing Early Voting, Nat’l Conf. State Legislatures (Oct. 22, 2020),

    [142].    See Redistricting Commissions: State Legislative Plans, supra note 91.

    [143].    See Public Financing of Campaigns: Overview, supra note 92.

    [144].    See Gregory, 501 U.S. at 458.

    [145].    See supra Part I.A.

    [146].    Gregory, 501 U.S. at 458. Of course, this balance is one in which Congress ultimately has the upper hand, at least officially.

    [147].    Id.

    [148].    Id. at 459.

    [149].    United States v. Carolene Prods. Co., 304 U.S. 144, 152 n.4 (1938).

    [150].    Cf. Hamilton, supra note 102 (using this phrase (erroneously in my view) for the judiciary).

    [151].    The Federalist No. 48 (James Madison).

    [152].    While subjective, I doubt that this list of the most significant congressional interventions is too controversial. Perhaps other observers would add post-1842 laws that imposed additional requirements on congressional redistricting, see, e.g., An Act for the Apportionment of Representatives to Congress Among the Several States According to the Ninth Census, § 2, 17 Stat. 28, 28 (1872); the 1982 amendments that turned Section 2 of the Voting Rights Act into a disparate impact provision, see 52 U.S.C. § 10301 (2018); or the Bipartisan Campaign Reform Act’s prohibition of soft money, see 52 U.S.C. § 30125.

    [153].    The most important reason for this omission is space; it’s simply impossible, in a single Article, to summarize adequately congressional and state and judicial electoral regulation. The states and the courts are also much more numerous than the single institution of Congress, and so more difficult to describe accurately. And for present purposes, the precise details of state and judicial electoral regulation are insignificant. What matters is that a substantial part of this regulation has plainly been undemocratic, while very little of Congress’s has been.

    [154].    See, e.g., Alexander Keyssar, The Right to Vote: The Contested History of Democracy in the United States 3–76 (2000) (describing these abuses); Chilton Williamson, American Suffrage: From Property to Democracy, 1760–1860, at 92–300 (1960) (same).

    [155].    See generally Paul E. Herron, Framing the Solid South: The State Constitutional Conventions of Secession, Reconstruction, and Redemption, 1860–1902 (2017); J. Morgan Kousser, The Shaping of Southern Politics: Suffrage Restriction and the Establishment of the One-Party South, 1880–1910 (1974).

    [156].    See, e.g., New Voting Restrictions in America, Brennan Ctr. for Just.
(Nov. 19, 2019),

    [157].    For works of mine on these topics, see Nicholas O. Stephanopoulos, Race, Place, and Power, 68 Stan. L. Rev. 1323 (2016) (addressing racial vote dilution); Nicholas O. Stephanopoulos & Eric M. McGhee, Partisan Gerrymandering and the Efficiency Gap, 82 U. Chi. L. Rev. 831 (2015) (addressing partisan gerrymandering).

    [158].    See, e.g., Luther v. Borden, 48 U.S. (7 How.) 1 (1849).

    [159].    See, e.g., United States v. Cruikshank, 92 U.S. 542 (1875); United States v. Reese, 92 U.S. 214 (1875).

    [160].    See, e.g., Giles v. Harris, 189 U.S. 475 (1903).

    [161].    See, e.g., Colegrove v. Green, 328 U.S. 549 (1946).

    [162].    See Stephanopoulos, supra note 26, at 160.

    [163].    See, e.g., Citizens United v. FEC, 558 U.S. 310 (2010).

    [164].    See, e.g., Shelby Cnty. v. Holder, 570 U.S. 529 (2013).

    [165].    See, e.g., Rucho v. Common Cause, 139 S. Ct. 2484 (2019).

    [166].    An Act for the Apportionment of Representatives Among the Several States According to the Sixth Census, 5 Stat. 491 (1842).

    [167].    See, e.g., Jay  K. Dow, Electing The House: The Adoption and Performance of the US Single-Member District Electoral System 110 (2017) (“The Apportionment Act of 1842 is arguably one of the most important pieces of legislation in US history.”).

    [168].    § 2, 5 Stat. at 491.

    [169].    See Erik J. Engstrom, Partisan Gerrymandering and the Construction of American Democracy 25 (2013) (showing that the plurality party won every House seat in 95% of congressional general ticket elections between 1800 and 1840).

    [170].    The whole point of partisan gerrymandering, of course, is to achieve lopsided outcomes in favor of the line-drawing party through the manipulation of single-member districts’ boundaries.

    [171].    Michele Rosa-Clot, The Apportionment Act of 1842: ‘An Odious Use of Authority’, 31 Parliaments, Ests. & Representation 33, 46 (2011) (internal quotation marks omitted).

    [172].    Id. (internal quotation marks omitted).

    [173].    Dow, supra note 167, at 130 (internal quotation marks omitted).

    [174].    Id. (internal quotation marks omitted).

    [175].    Rosa-Clot, supra note 171, at 46 (internal quotation marks omitted).

    [176].    Dow, supra note 167, at 131 (internal quotation marks omitted); see, e.g., id. (“To the Whigs, the most important feature of single-member district elections was that this system protects political minorities.”); Engstrom, supra note 169, at 47 (“One explanation [for the Apportionment Act] is that it was a good government reform aimed at creating uniform electoral standards to protect the interests of minority constituencies.”).

    [177].    See, e.g., Rosa-Clot, supra note 171, at 45; Johanna Nicol Shields, Whigs Reform the “Bear Garden”: Representation and the Apportionment Act of 1842, 5 J. Early Republic 355, 362 (1985).

    [178].    For a detailed discussion of the Whigs’ partisan calculations, see Engstrom, supra note 169, at 43–51.

    [179].    See Dow, supra note 167, at 133.

    [180].    Any benefit the Whigs received was probably quite small, given their crushing defeat in the next congressional election. See 1842 and 1843 United States House of Representatives Elections, Wikipedia, _United_States_House_of_Representatives_elections (last visited Aug. 1, 2020) (showing that the Whigs lost sixty-nine seats, as well as their House majority, in the 1842 election).

    [181].    See supra notes 154–157 and accompanying text.

    [182].    See supra notes 158–165 and accompanying text.

    [183].    An Act to Provide for the More Efficient Government of the Rebel States, § 5, 14 Stat. 428, 429 (1867).

    [184].    See id. (“[T]he elective franchise shall be enjoyed by all such persons as have the qualifications herein stated for electors of delegates . . . .”).

    [185].    See id. (“[P]ersons voting on the question of ratification [shall be] qualified as electors for delegates . . . .”).

    [186].    See An Act Supplementary to an Act Entitled “An Act to Provide for the More Efficient Government of the Rebel States,” § 1, 15 Stat. 2, 2 (1867).

    [187].    See § 2, 15 Stat. at 3 (“[G]iving to each [district] representation in the ratio of voters registered . . . as nearly as may be”).

    [188].    See § 3, 15 Stat. at 3.

    [189].    See § 4, 15 Stat. at 3.

    [190].    See id.

    [191].    For thorough descriptions of these laws, see Richard M. Valelly, The Two Reconstructions: The Struggle for Black Enfranchisement 106–09 (2004); Xi Wang, The Trial of Democracy: Black Suffrage and Northern Republicans, 1860–1910, at 57–92 (1997).

    [192].    An Act to Enforce the Right of Citizens of the United States to Vote in the Several States of This Union, and for Other Purposes, § 2, 16 Stat. 140, 140 (1870).

    [193].    See § 3, 16 Stat. at 140.

    [194].    § 4, 16 Stat. at 141; see also § 5, 16 Stat. at 141 (creating similar criminal offenses with respect to African Americans “to whom the right of suffrage is secured . . . by the fifteenth amendment”).

    [195].    § 19, 16 Stat. at 144–45; see also § 20, 16 Stat. at 145 (forbidding fraud in the voter registration process).

    [196].    See § 22, 16 Stat. at 145–46.

    [197].    See An Act to Amend an Act Approved May Thirty-One, Eighteen Hundred and Seventy, Entitled “An Act to Enforce the Rights of Citizens of the United States to Vote in the Several States of This Union, and for Other Purposes,” § 2, 16 Stat. 433, 433–34 (1871) (authorizing the appointment of supervisors for municipalities with more than twenty thousand residents).

    [198].    See §§ 4–5, 16 Stat. at 434–45 (describing the supervisors’ duties).

    [199].    See supra note 159 and accompanying text.

    [200].      Ex parte Yarbrough (The Ku-Klux Cases), 110 U.S. 651, 667 (1884); see also Oregon v. Mitchell, 400 U.S. 112, 254 (1970) (Brennan, J., dissenting in part and concurring in part) (describing one motive for these laws as “an idealistic determination that the gains of the Civil War not be surrendered”).

    [201].    Wang, supra note 191, at 63 (internal quotation marks omitted).

    [202].    Id. (internal quotation marks omitted); see also id. at 47 (“The idea of equality as pronounced by the Declaration of Independence was the main ideological source for Reconstruction politics . . . .”).

    [203].    The Ku-Klux Cases, 110 U.S. at 667; see also Ex parte Siebold, 100 U.S. 371, 382 (1879) (observing that “fraud, corruption, and irregularity [] have frequently prevailed [in recent] elections”).

    [204].    Wang, supra note 191, at 64 (internal quotation marks omitted).

    [205].    Id. at 65 (internal quotation marks omitted).

    [206].    The Reconstruction Acts were passed over President Johnson’s vetoes, but he was out of office by the time the Enforcement Acts were enacted.

    [207].    See, e.g., Oregon v. Mitchell, 400 U.S. 112, 254 (1970) (Brennan, J., dissenting in part and concurring in part) (noting that Congress’s Reconstruction legislation could be seen as “a grasp for partisan political power”); Vikram David Amar & Alan Brownstein, The Hybrid Nature of Political Rights, 50 Stan. L. Rev. 915, 943 (1998) (observing that “extending the franchise promised to benefit the Republican Party” since “[n]ew black voters were expected to vote Republican”).

    [208].    See, e.g., Wang, supra note 191, at 68 (“[T]he reality that northern fraud could hurt the party as much as southern black disfranchisement, and eventually hurt the whole system of congressional elections, alarmed Republicans.”).

    [209].    See, e.g., id. (“One serious fraudulent practice in northern cities . . . was that of recent immigrants casting votes by using false naturalization papers as their certificates of citizenship.”).

    [210].    Id. at xxiv.

    [211].    See, e.g., Heather Gerken, Goodbye to the Crown Jewel of the Civil Rights Movement, Slate (June 25, 2013, 3:50 PM),

    [212].    See Voting Rights Act of 1965, Pub. L. No. 89-110, §§ 3, 6, 7, 79 Stat. 437, 437, 439–41 (1965).

    [213].    See § 8, 79 Stat. at 441.

    [214].    See § 4, 79 Stat. at 438–39.

    [215].    See § 10, 79 Stat. at 442–43.

    [216].    See § 11, 79 Stat. at 443.

    [217].    See § 5, 79 Stat. at 439.

    [218].    For an accessible history of the VRA’s enactment, see Ari Berman, Give Us the Ballot: The Modern Struggle for Voting Rights in America 13–38 (2015).

    [219].    President Lyndon B. Johnson, Special Message to the Congress: The American Promise (Mar. 15, 1965),

    [220].    Id.; see also South Carolina v. Katzenbach, 383 U.S. 301, 308 (1966) (“The Voting Rights Act was designed by Congress to banish the blight of racial discrimination in voting . . . .”). Note that racial vote denial was the VRA’s original target. Concerns about racial vote dilution through at-large elections, cleverly drawn districts, and the like didn’t emerge until the late 1960s. See Allen v. State Bd. of Elections, 393 U.S. 544 (1969) (holding that Section 5 of the VRA extended to racial vote dilution).

    [221].    See To Pass H.R. 6400, the 1965 Voting Rights Act, GovTrack, congress/votes/89-1965/h87 (last visited Aug. 1, 2020) (showing that, in the House, sixty-two Democrats voted against the VRA compared to twenty-three Republicans); To Pass S. 1564, the Voting Rights Act of 1965, GovTrack, (last visited Aug. 1, 2020) (showing that seventeen Democrats voted against the VRA in the Senate compared to two Republicans).

    [222].    See To Pass H.R. 6400, the 1965 Voting Rights Act, supra note 221 (showing that 112 out of 135 voting House Republicans supported the VRA); To Pass S. 1564, the Voting Rights Act of 1965, supra note 221 (thirty out of thirty-two Senate Republicans).

    [223].    Berman, supra note 218, at 38 (internal quotation marks omitted) (dating the quote a year earlier).

    [224].    See generally id. at 65–99 (describing the Republicans’ “southern strategy” in the late 1960s and 1970s). In the long run in a diversifying America, however, it’s possible that the VRA could end up benefiting Democrats. As minority voters increase in numbers and clout, it may be preferable for a party to command their support—even at the cost of conservative white backing.

    [225].    Five of these states were lost to the segregationist George Wallace, not to Richard Nixon. See 1968 United States Presidential Election, Wikipedia, election (last visited Aug. 1, 2020).

    [226].    See generally David Lublin, The Republican South: Democratization and Partisan Change (2007).

    [227].    See Stephanopoulos, supra note 157, at 1387.

    [228].    See Wang, supra note 191, at xxiv.

    [229].    See Federal Election Campaign Act Amendments of 1974, Pub. L. No. 93-443, § 101(b), 88 Stat. 1263, 1263–64 (1974).

    [230].    See § 101(c), (e), (f), 88 Stat. at 1264–66.

    [231].    See § 201, 88 Stat. at 1272–75.

    [232].    See § 408, 88 Stat. at 1297–1303.

    [233].    See §§ 310–14, 88 Stat. at 1280–85.

    [234].    Buckley v. Valeo, 424 U.S. 1, 25 (1976). In the decision below, the court of appeals had referred to this interest more generally as “preserving the integrity of the system of elections.” Buckley v. Valeo, 519 F.2d 821, 835 (D.C. Cir. 1975), rev’d, 424 U.S. 1 (1976).

    [235].    Buckley, 424 U.S. at 25–26. The Court famously deemed this interest “wholly foreign to the First Amendment.” Id. at 49. In contrast, the court below (more persuasively, in my view) thought that “equaliz[ing] the relative ability of all voters to affect electoral outcomes” “affirmatively enhances First Amendment values” by “broaden[ing] the choice of candidates and the opportunity to hear a variety of views.” Buckley, 519 F.2d at 841.

    [236].    Buckley, 424 U.S. at 26.

    [237].    Id. at 66–67.

    [238].    Julian E. Zelizer, Seeds of Cynicism: The Struggle over Campaign Finance, 1956–1974, 14 J. Pol’y Hist. 73, 74 (2002).

    [239].    See id. at 100 (“Public-opinion polls confirmed that Watergate had heightened public interest in taking action against campaign corruption.”); see also Robert E. Mutch, Buying the Vote: A History of Campaign Finance Reform 137 (2014) (“[M]ore than 25 percent of all mail in the post-Watergate period was on campaign finance—far more than on any other issue.”).

    [240].    See, e.g., Mutch, supra note 239, at 187 (noting that “[m]embers of Congress” were “attentive to public opinion after Watergate,” and so “knew that open opposition to reform would be a short-term political liability”).

    [241].    Id. at 138; see also, e.g., Zelizer, supra note 238, at 99 (“Watergate offered a ‘focusing event’ for the coalition to push their proposals that had incubated over many years.”).

    [242].    See, e.g., Zelizer, supra note 238, at 97 (“Watergate turned campaign finance into a political liability for Republicans as many Democrats were preparing to use it as a campaign issue.”).

    [243].    See, e.g., Richard L. Hasen, The Nine Lives of Buckley v. Valeo, in First Amendment Stories 345, 350 (Richard W. Garnett & Andrew Koppelman eds., 2010) (observing that “Democrats became more interested in campaign finance reform as large donor money shifted from Democrats to Republicans following Nixon’s 1968 election”); Zelizer, supra note 238, at 88–89 (noting that Democrats “faced a large deficit” relative to Republicans and “were very aware of how this legislation could change their electoral fortunes”).

    [244].    Id. at 103.

    [245].    McConnell v. FEC, 540 U.S. 93, 263 (2003) (Scalia, J., concurring in part and dissenting in part).

    [246].    Id. at 248–49.

    [247].    See, e.g., Mutch, supra note 239, at 138 (“‘[T]hese people who were quite recalcitrant initially, they got taken over by public events . . . .’” (quoting prominent activist Susan King)).

    [248].    See, e.g., Nicholas O. Stephanopoulos, Accountability Claims in Constitutional Law, 112 Nw. U. L. Rev. 989, 1047–52 (2018) (surveying this evidence).

    [249].    52 U.S.C. §§ 20501–11 (2018).

    [250].    Id. §§ 20901–21145.

    [251].    See id. §§ 20503–06.

    [252].    See id. § 20507.

    [253].    See id. §§ 20901–06.

    [254].    See id. § 21082.

    [255].    See id. § 21083(a).

    [256].    See id. § 21083(b).

    [257].    S. Rep. No. 103-6, at 1 (1993).

    [258].    Id. at 2; see also Cong. Rsch. Serv., The National Voter Registration Act of 1993: History, Implementation, and Effects 6 (2013), (“Many proponents of [the NVRA] argued that by making it easier to register eligible citizens, the law would not only increase the number of persons who were registered to vote but also would encourage more voter turnout.”). The NVRA also achieved its goal by modestly increasing voter registration and turnout. See, e.g., Benjamin Highton & Raymond E. Wolfinger, Estimating the Effects of the National Voter Registration Act of 1993, 20 Pol. Behav. 79 (1998).

    [259].    H.R. Rep. No. 107-329, at 31 (2001); see also Daniel P. Tokaji, Early Returns on Election Reform: Discretion, Disenfranchisement, and the Help America Vote Act, 73 Geo. Wash. L. Rev. 1206, 1207 (2005) (“The legislation was intended to improve not only the equipment used to cast and count votes, but also the way that registration lists are maintained and polling place operations conducted.”).

    [260].    See, e.g., Tokaji, supra note 259, at 1214 (“[M]uch of HAVA is designed to make it harder to cheat, by preventing fraud in the voting process.”).

    [261].    See Cong. Rsch. Serv., supra note 258, at 25.

    [262].    See id. at 25 (noting that “[t]he NVRA was not supported by most Republicans”).

    [263].    A long line of political science scholarship finds that higher turnout doesn’t consistently or necessarily benefit either party. See generally Raymond E. Wolfinger & Steven J. Rosenstone, Who Votes? (1980).

    [264].    See, e.g., Marshall Ganz, Motor Voter or Motivated Voter?, Am. Prospect (Dec. 19, 2001), (“Many conservatives feared—just as many liberals hoped—that Motor Voter would produce a Democratic bonanza at the polls.”).

    [265].    See, e.g., id.

    [266].    Martin J. Siegel, Congressional Power over Presidential Elections: The Constitutionality of the Help America Vote Act Under Article II, Section 1, 28 Vt. L. Rev. 373, 418 (2004).

    [267].    See, e.g.,, Election Reform: What’s Changed, What Hasn’t, and Why, 2000–2006, at 13 (2006), (noting the Democrats’ support for “significant financial investment by the federal government in state election administration and rules mandating provisional voting”).

    [268].    See, e.g., id. (observing that the Republicans “sought universal nationwide voter ID requirements” and settled for HAVA’s “limited voter ID requirement”).

    [269].    See 52 U.S.C. § 21083(b)(1).

    [270].    Id. § 21083(b)(2).

    [271].    See 1892 United States Elections, Wikipedia, wiki/1892_United_States_ elections (last visited Aug. 1, 2020).

    [272].    See An Act to Repeal All Statutes Relating to Supervisors of Elections and Special Deputy Marshals, and for Other Purposes, 28 Stat. 36 (1894).

    [273].    See, e.g., Wang, supra note 191, at 254 (describing the Democrats’ argument that the laws they repealed “were partisan by nature and used for partisan purposes”).

    [274].    See An Act to Provide for the Fifteenth and Subsequent Decennial Censuses and to Provide for Apportionment of Representatives in Congress, 46 Stat. 21 (1929); see also Wood v. Broom, 287 U.S. 1, 6 (1932) (noting that this law “did not carry forward those requirements as previous apportionment acts had done”).

    [275].    See, e.g., Dow, supra note 167, at 168 (“The removal of the compact, contiguous, and equal population requirements from the apportionment act furthered the interests of both urban and rural legislators.”).

    [276].    Cf. Heather K. Gerken, Slipping the Bonds of Federalism, 128 Harv. L. Rev. 85, 110 (2014) (criticizing the Supreme Court because “it addresses each source of federal power seriatim” and thus overlooks “the larger ends they serve together”).

    [277].    Justice Ginsburg made a similar argument in her Shelby County dissent. She discussed six constitutional provisions empowering Congress to regulate elections, concluding, “[t]he implication is unmistakable: Under our constitutional structure, Congress holds the lead rein in making the right to vote equally real for all U.S. citizens.” Shelby Cnty. v. Holder, 570 U.S. 529, 567 n.2 (2013) (Ginsburg, J., dissenting). In the academy, Franita Tolson has repeatedly argued that courts should be more deferential toward congressional electoral regulations because of the multiple sources of power from which Congress draws when it legislates about elections. See, e.g., Franita Tolson, Reinventing Sovereignty? Federalism as a Constraint on the Voting Rights Act, 65 Vand. L. Rev. 1195, 1207 (2012) (maintaining that the combination of the Elections Clause with the Fourteenth and Fifteenth Amendments “should inform the level of deference that the Court uses to analyze congressional acts”); Tolson, supra note 21, at 326–27 (“[T]he presence of multiple sources of authority justifies increased deference towards the legislative record.”).

    [278].    Hamilton, for example, didn’t think the original Constitution authorized Congress to regulate state elections. If it had, he wondered, “would any man have hesitated to condemn it, both as an unwarrantable transposition of power and as a premeditated engine for the destruction of the State governments?” The Federalist No. 59 (Alexander Hamilton).

    [279].    I’m aware of the irony of relying on decisions of the Court—an institution I argue is less trustworthy than Congress—to define Congress’s electoral power. Unfortunately, this is an unavoidable irony in a system, like ours, that gives the Court the final word about the Constitution’s meaning.

    [280].    See supra notes 48–52 and accompanying text.

    [281].    The Equal Protection Clause is also power-conferring to the extent that Congress prevents and remedies violations of the provision through its Fourteenth Amendment enforcement power. See infra Part III.D.

    [282].    See, e.g., Citizens United v. FEC, 558 U.S. 310 (2010).

    [283].    See, e.g., Shaw v. Reno, 509 U.S. 630 (1993).

    [284].    See, e.g., Printz v. United States, 521 U.S. 898 (1997).

    [285].    See, e.g., Shelby Cnty. v. Holder, 570 U.S. 529 (2013).

    [286].    Although I don’t defend the claim further, my sense is that many (probably most) electoral regulations that Congress might plausibly enact wouldn’t run afoul of any of the Constitution’s power-limiting elements (at least as those elements are currently construed).

    [287].    U.S. Const. art. I, § 4. Unless and until Congress acts, the times, places, and manner of congressional elections “shall be prescribed in each State by the Legislature thereof.” Id.

    [288].    For scholars agreeing with this characterization, see Samuel Issacharoff, Comment, Beyond the Discrimination Model on Voting, 127 Harv. L. Rev. 95, 111 (2013) (explaining how the Court has “put the Elections Clause on a higher rung of full federal power than even the Commerce Clause”); Pamela S. Karlan, Section 5 Squared: Congressional Power to Extend and Amend the Voting Rights Act, 44 Hous. L. Rev. 1, 16 (2007) (noting that the Court’s decisions “have confirmed the longstanding interpretation of the clause as a grant of essentially plenary authority”).

    [289].    Ex parte Siebold, 100 U.S. 371, 388 (1879).

    [290].    Id. at 384.

    [291].    Smiley v. Holm, 285 U.S. 355, 366 (1932).

    [292].    Arizona v. Inter Tribal Council of Ariz., Inc., 570 U.S. 1, 8 (2013).

    [293].    Id. at 14. However, the Elections Clause doesn’t authorize Congress (or the states) “to dictate electoral outcomes” or “to favor or disfavor a class of candidates.” U.S. Term Limits, Inc. v. Thornton, 514 U.S. 779, 833–34 (1995).

    [294].    See, e.g., Ex parte Yarbrough (The Ku-Klux Cases), 110 U.S. 651, 666 (1884) (holding that “acts [that] have no sanction in the statutes of a state, or which are not committed by any one exercising its authority” can be criminalized under the Elections Clause).

    [295].    See, e.g., United States v. Classic, 313 U.S. 299, 317 (1941) (holding that Congress’s Elections Clause power “includes the authority to regulate primary elections when . . . they are a step in the exercise by the people of their choice of representatives in Congress”).

    [296].    See, e.g., The Ku-Klux Cases, 110 U.S. at 662 (asking rhetorically, “are [Congress’s Elections Clause] powers annulled because an election for state officers is held at the same time and place?”).

    [297].    See, e.g., Branch v. Smith, 538 U.S. 254, 280 (2003) (plurality opinion) (holding that when Congress issued certain directives under the Elections Clause, it “was not placing a statutory obligation on the state legislatures as it was in [the anti-commandeering cases]; rather, it was regulating (as the Constitution specifically permits) the manner in which a State is to fulfill its pre-existing constitutional obligations”).

    [298].    Smiley v. Holm, 285 U.S. 355, 366 (1932).

    [299].    The controversy stems from Article I’s statement that voters in U.S. House elections “shall have the Qualifications requisite” for voters in state house elections. U.S. Const. art. I, § 2, cl. 1. Arguably, this clause’s more specific language means that the Elections Clause’s more general reference to the “manner” of elections can’t encompass the setting of qualifications for voting in congressional races. In dicta in Arizona v. Inter Tribal Council, a majority of the Court embraced this view, opining that “the Elections Clause empowers Congress to regulate how federal elections are held, but not who may vote in them.” 570 U.S. 1, 16 (2013). But this was merely dicta, and as Richard Hasen has noted, Mitchell “remains good law unless overruled by the Court.” Richard L. Hasen, Too Plain for Argument: The Uncertain Congressional Power to Require Parties to Choose Presidential Nominees Through Direct and Equal Primaries, 102 Nw. U. L. Rev. 2009, 2018 (2008).

    [300].    See Voting Rights Act Amendments of 1970, Pub. L. No. 91-285, §§ 301–05, 84 Stat. 314, 318–19 (1970).

    [301].    400 U.S. 112 (1970).

    [302].    See id. at 117–18 (opinion of Black, J.) (describing the Justices’ votes).

    [303].    Id. at 122.

    [304].    Id. at 124.

    [305].    See 52 U.S.C. §§ 20301–11 (2018).

    [306].    See 18 U.S.C. § 611 (2018).

    [307].    Scholars, though, have complained about them. See generally Brian C. Kalt, Unconstitutional but Entrenched: Putting UOCAVA and Voting Rights for Permanent Expatriates on a Sound Constitutional Footing, 81 Brook. L. Rev. 441 (2016); Stephen E. Mortellaro, The Unconstitutionality of the Federal Ban on Noncitizen Voting and Congressionally-Imposed Voter Qualifications, 63 Loy. L. Rev. 447 (2017).

    [308].    See supra note 296 and accompanying text; see also Karlan, supra note 288, at 17 (“[T]he Elections Clause has long been interpreted to give Congress power over so-called ‘mixed elections’—that is, to permit Congress to regulate all aspects of an election (or an electoral process) used even in part to select members of Congress.”).

    [309].    This is just a paraphrase of the Election Clause’s text. The deeper rationale for the provision’s allocation of electoral power to Congress is that some institution must be able to regulate congressional elections, and Congress is a more trustworthy grantee for this power than are the states. See, e.g., Hamilton, supra note 278 (“[A]n exclusive power of regulating elections for the national government, in the hands of the State legislatures, would leave the existence of the Union entirely at their mercy.”).

    [310].    Congress can still regulate presidential elections to some degree, under the Elections Clause, because they’re held concurrently with congressional elections. See supra notes 296, 308 and accompanying text.

    [311].    This power also derives from the Necessary and Proper Clause. See infra Part III.F.

    [312].    U.S. Const. art. II, § 1, cl. 4.

    [313].    For scholars echoing this view, see Dan T. Coenen & Edward J. Larson, Congressional Power over Presidential Elections: Lessons from the Past and Reforms for the Future, 43 Wm. & Mary L. Rev. 851, 891 (2002) (“[T]he [presidential electoral] power is fully coextensive with Congress’s sweeping authority to regulate in any way the ‘Manner’ of House and Senate elections.”).

    [314].    Burroughs v. United States, 290 U.S. 534, 545 (1934).

    [315].    Id. at 547.

    [316].    Oregon v. Mitchell, 400 U.S. 112, 124 (1970) (opinion of Black, J.).

    [317].    Id.; see also U.S. Term Limits, Inc. v. Thornton, 514 U.S. 779, 894 (1995) (Thomas, J., dissenting) (“[T]he treatment of congressional elections in Article I parallels the treatment of Presidential elections in Article II.”).

    [318].    See, e.g., Burroughs, 290 U.S. at 548 (respecting Congress’s “conclusion that public disclosure of political contributions . . . would tend to prevent the corrupt use of money to affect elections”).

    [319].    See, e.g., Buckley v. Valeo, 424 U.S. 1, 90 (1976) (“[P]ublic financing of Presidential elections as a means to reform the electoral process was clearly a choice within the granted power.”).

    [320].    See Voting Rights Act Amendments of 1970, Pub. L. No. 91-285, §§ 301–05, 84 Stat. 314, 318–19 (1970).

    [321].    See § 202, 84 Stat. at 316–17.

    [322].    See Mitchell, 400 U.S. at 118–19 (opinion of Black, J.) (describing the Justices’ votes).

    [323].    Id. at 134.

    [324].    As with the Elections Clause, see supra note 309, the deeper nontextual rationale for authorizing Congress to legislate about presidential elections is that the federal government must ultimately be able to regulate the elections that constitute it. To quote Justice Black again, “under its broad authority to create and maintain a national government,” Congress must have “power under the Constitution to regulate federal elections.” Mitchell, 400 U.S. at 134 (opinion of Black, J.).

    [325].    With the caveat that a presidential electoral regulation could be based on both the Electors Clause and Congress’s Elections Clause power to reach mixed elections. See supra notes 296, 308, 310 and accompanying text.

    [326].    U.S. Const. art. IV, § 4.

    [327].    For a scholar agreeing with this view, see Gabriel J. Chin, Justifying a Revised Voting Rights Act: The Guarantee Clause and the Problem of Minority Rule, 94 B.U. L. Rev. 1551, 1586 (2014) (“The Guarantee Clause benefits the states, not the federal government, so some other rationale will have to be found to the extent that [congressional legislation] applies to purely federal elections.”). For a scholar disagreeing, see Erwin Chemerinsky, Cases Under the Guarantee Clause Should Be Justiciable, 65 U. Colo. L. Rev. 849, 876 (1994) (“[Unrepublican] [e]lection practices in the state might relate not only to state government offices, but to elections of representatives and senators as well.”).

    [328].    This is precisely why some scholars argue that suits alleging unrepublican practices should remain nonjusticiable. See, e.g., Richard L. Hasen, Leaving the Empty Vessel of ‘Republicanism’ Unfilled: An Argument for the Continued Non-Justiciability of Guarantee Clause Cases 10–19 (Loyola L. Sch. Pub. L. & Legal Theory Working Paper No. 2003-10, 2003).

    [329].    Another limited definition could be a government no less democratic than that of states when they joined the Union. See, e.g., Minor v. Happersett, 88 U.S. (21 Wall.) 162, 175 (1874) (“All the States had [republican] governments when the Constitution was adopted. In all the people participated to some extent . . . .”).

    [330].    For a relatively expansive conception of republicanism, see Arthur E. Bonfield, The Guarantee Clause of Article IV, Section 4: A Study in Constitutional Desuetude, 46 Minn. L. Rev. 513, 516 (1962) (arguing that it extends to “the protection and advancement of all citizens’ political and civil rights”).

    [331].    Amar, supra note 35, at 749.

    [332].    Id.

    [333].    Fred O. Smith, Jr., Awakening the People’s Giant: Sovereign Immunity and the Constitution’s Republican Commitment, 80 Fordham L. Rev. 1941, 1949 (2012); see also e.g., David S. Louk, Reconstructing the Congressional Guarantee of Republican Government, 73 Vand. L. Rev. 673, 736 (2020) (“A broad consensus seems to be that at minimum, the will of the majority must be reflected in the selection of legislators and government representatives.”). It’s ironic, of course, that republicanism was originally understood to mean popular sovereignty when so many of the people—women, slaves, the poor—weren’t enfranchised and so weren’t truly sovereign at the Framing.

    [334].    For a different view, arguing that Congress can act under the Guarantee Clause only when it has been asked to do so by a state, see generally Ryan C. Williams, The “Guarantee” Clause, 132 Harv. L. Rev. 602 (2018).

    [335].    See, e.g., Pac. States Tel. & Tel. Co. v. Oregon, 223 U.S. 118 (1912) (involving a Guarantee Clause challenge to Oregon’s use of direct democracy); Luther v. Borden, 48 U.S. (7 How.) 1 (1849) (involving a Guarantee Clause challenge to actions taken by an allegedly illegitimate Rhode Island government).

    [336].    Luther, 48 U.S. at 42.

    [337].    Id.

    [338].    White v. Hart, 80 U.S. (13 Wall.) 646, 649 (1871).

    [339].    Id.; see also Ohio ex rel. Davis v. Hildebrant, 241 U.S. 565, 570 (1916) (noting that it’s Congress “upon whom the Constitution has conferred the exclusive authority to uphold the guaranty of a republican form of government”). But see New York v. United States, 505 U.S. 144, 186 (1992) (hinting that the Guarantee Clause could “provide[] a basis upon which a State or its subdivisions may sue to enjoin the enforcement of a federal statute”).

    [340].    See supra notes 183–190 and accompanying text.

    [341].    An Act to Provide for the More Efficient Government of the Rebel States, § 1, 14 Stat. 428, 428 (1867).

    [342].    Texas v. White, 74 U.S. (7 Wall.) 700, 727 (1868).

    [343].    Wiecek, supra note 139, at 290.

    [344].    Id.; see also Louk, supra note 333, at 705–22 (describing the use of the Guarantee Clause during Reconstruction). Outside of Reconstruction, Congress also frequently invoked the Clause in the context of the admission of new states. See generally Charles O. Lerche, Jr., The Guarantee of a Republican Form of Government and the Admission of New States, 11 J. Pol. 578 (1949).

    [345].    U.S. Const. amend. XIV, § 1.

    [346].    Id. § 5.

    [347].    Katzenbach v. Morgan, 384 U.S. 641, 653 (1966); see also Ex parte Virginia, 100 U.S. 339, 345–46 (1879) (“Whatever legislation is appropriate . . . is brought within the domain of congressional power.”).

    [348].    City of Boerne v. Flores, 521 U.S. 507, 520 (1997).

    [349].    This is precisely why many scholars have criticized the congruence-and-proportionality standard. See generally Michael W. McConnell, Comment,  Institutions and Interpretation: A Critique of City of Boerne v. Flores, 111 Harv. L. Rev. 153 (1997); Robert C. Post & Reva B. Siegel, Legislative Constitutionalism and Section Five Power: Policentric Interpretation of the Family and Medical Leave Act, 112 Yale L.J. 1943 (2003).

    [350].    Cf. Karlan & Ortiz, supra note 21, at 19 (considering “how far may Congress legislate to ‘enforce’ the equal protection principle announced in Bush v. Gore,” but not addressing any other equal protection doctrines); Richard H. Pildes, The Future of Voting Rights Policy: From Anti-Discrimination to the Right to Vote, 49 How. L.J. 741, 759 (2006) (same).

    [351].    See, e.g., Anderson v. Celebrezze, 460 U.S. 780, 789 (1983) (citing “the rights protected by the First and Fourteenth Amendments that the plaintiff seeks to vindicate”). Sliding-scale scrutiny also applies to burdens on candidates’ access to the ballot, see, e.g., id., and on parties’ ability to manage their internal affairs, see, e.g., Wash. State Grange v. Wash. State Republican Party, 552 U.S. 442, 451–52 (2008).

    [352].    See Bush v. Gore, 531 U.S. 98, 104 (2000) (“The right to vote is protected in more than the initial allocation of the franchise. Equal protection applies as well to the manner of its exercise.”).

    [353].    See, e.g., Rogers v. Lodge, 458 U.S. 613, 617 (1982) (“[M]ultimember districts violate the Fourteenth Amendment . . . by [intentionally] minimizing, cancelling out or diluting the voting strength of racial elements in the voting population.” (internal quotation marks omitted)).

    [354].    See, e.g., Shaw v. Reno, 509 U.S. 630, 649 (1993) (“[A] plaintiff challenging a reapportionment statute under the Equal Protection Clause may state a claim by alleging that the legislation . . . rationally cannot be understood as anything other than an effort to separate voters into different districts on the basis of race . . . .”).

    [355].    See Rucho v. Common Cause, 139 S. Ct. 2484 (2019).

    [356].    See Davis v. Bandemer, 478 U.S. 109, 143 (1986) (plurality opinion) (“[W]e hold that political gerrymandering cases are properly justiciable under the Equal Protection Clause.”).

    [357].    Of course, Congress can’t simply point to potential constitutional violations to legislate under its Fourteenth Amendment enforcement authority. Rather, it must create a record of documented violations that it’s trying to prevent or remedy.

    [358].    See Lassiter v. Northampton Cnty. Bd. of Elections, 360 U.S. 45 (1959).

    [359].    See Voting Rights Act of 1965, Pub. L. No. 89-110, § 4(e), 79 Stat. 437, 439 (1965).

    [360].    Katzenbach v. Morgan, 384 U.S. 641, 654 (1966).

    [361].    Id.

    [362].    Id. at 656; see also Oregon v. Mitchell, 400 U.S. 112, 133 (1970) (opinion of Black, J.) (holding that, based on “evidence that literacy tests reduce voter participation in a discriminatory manner,” “Congress was supported by substantial evidence in concluding that a nationwide ban on literacy tests was appropriate to enforce the Civil War amendments”). To be sure, Morgan and Mitchell significantly predated the Court’s adoption of the congruence-and-proportionality standard. To date, that standard hasn’t been applied by the Court in any case involving a congressional electoral regulation.

    [363].    U.S. Const. amend. XV, § 1.

    [364].    Id. amend. XIX, § 1.

    [365].    Id. amend. XXIV, § 1.

    [366].    Id. amend. XXVI, § 1.

    [367].    Id. amend. XV, § 2; id. amend. XIX, § 2; id. amend. XXIV, § 2; id. amend. XXVI, § 2.

    [368].    South Carolina v. Katzenbach, 383 U.S. 301, 324 (1966).

    [369].    570 U.S. 529 (2013).

    [370].    See, e.g., id. at 554 (noting “the irrationality of continued reliance on the § 4 coverage formula”); id. at 556 (deeming it “irrational for Congress to distinguish between States in such a fundamental way based on 40-year-old data”).

    [371].    In dissent, Justice Ginsburg also observed that the majority “does not purport to alter settled precedent establishing that the dispositive question is whether Congress has employed ‘rational means.’” Id. at 569 (Ginsburg, J., dissenting).

    [372].    See, e.g., Akhil Reed Amar, The Lawfulness of Section 5—and Thus of Section 5, 126 Harv. L. Rev. F. 109, 119 (2013) (“[E]very one of these amendments has contained express language suggesting that Congress should have a considerable choice of means in discharging its powers . . . .”). But see Richard L. Hasen & Leah M. Litman, Thin and Thick Conceptions of the Nineteenth Amendment Right to Vote and Congress’s Power to Enforce It, 108 Geo. L.J. 27, 65–68 (2020) (taking more seriously the possibility that the congruence-and-proportionality standard might apply to laws enacted under the Nineteenth Amendment).

    [373].    A plurality would have held that “racially discriminatory motivation is a necessary ingredient of a Fifteenth Amendment violation” in City of Mobile v. Bolden, 446 U.S. 55, 62 (1980) (plurality opinion). Subsequently, in City of Rome v. United States, the Court “assumed” (but didn’t hold) that the Fifteenth Amendment “prohibits only intentional discrimination in voting.” 446 U.S. 156, 158 (1980).

    [374].    See id. at 172–73.

    [375].    Id. at 177.

    [376].    Id.

    [377].    See, e.g., Hasen & Litman, supra note 372, at 55–63 (discussing an array of measures that Congress could enact under the Nineteenth Amendment to reduce differential voting burdens on women, such as a ban on photo ID requirements for voting).

    [378].    See, e.g., Wickard v. Filburn, 317 U.S. 111, 125 (1942) (holding that activity can be “reached by Congress if it exerts a substantial economic effect on interstate commerce”).

    [379].    See, e.g., Coenen & Larson, supra note 313, at 879 (agreeing that, “[i]n an earlier era, proponents of [electoral] reform . . . would have raced to embrace the Commerce Clause”).

    [380].    See, e.g., United States v. Morrison, 529 U.S. 598, 611 (2000) (holding that, for “federal regulation of intrastate activity” to be valid, “the activity in question” must be “some sort of economic endeavor”).

    [381].    See 18 U.S.C. § 597 (2018) (criminalizing “mak[ing] an expenditure to any person, either to vote or withhold his vote, or to vote for or against any candidate”).

    [382].    See, e.g., Stephanie Philips, The Risks of Computerized Election Fraud: When Will Congress Rectify a 38-Year-Old Problem?, 57 Ala. L. Rev. 1123, 1159 (2006) (agreeing that, “even under the recent, restrictive interpretations of Congress’s Commerce Clause power . . . Congress does have the authority to regulate voting systems sold to states”).

    [383].    See, e.g., Marlene Arnold Nicholson, Campaign Financing and Equal Protection, 26 Stan. L. Rev. 815, 824 (1974) (“Should Congress decide to reform state campaign financing practice, the commerce clause is probably the safest grant of authority upon which to rely.”).

    [384].    See, e.g., South Dakota v. Dole, 483 U.S. 203, 207 (1987) (“[O]bjectives not thought to be within Article I’s enumerated legislative fields may nevertheless be attained through the use of the spending power . . . .” (internal citations and quotation marks omitted)).

    [385].    See, e.g., id. (“[C]onditions on federal grants might be illegitimate if they are unrelated to the federal interest in particular national projects or programs.” (internal quotation marks omitted)).

    [386].    See, e.g., Nat’l Fed. Indep. Bus. v. Sebelius, 567 U.S. 519, 581 (2012) (invalidating terms for Medicaid expansion that allegedly amounted to “a gun to the head” of the states).

    [387].    Cf. id. (explaining that, under Congress’s terms for Medicaid expansion, a declining state would “lose not merely a relatively small percentage of its existing Medicaid funding, but all of it” (internal quotation marks omitted)). Unsurprisingly, given how hard it is for electoral conditions attached to electoral funds to be unduly coercive, no litigant ever thought to challenge the money that HAVA made available to the states on this basis. Cf. Jocelyn Friedrichs Benson, Democracy and the Secretary: The Crucial Role of State Election Administrators in Promoting Accuracy and Access to Democracy, 27 St. Louis U. Pub. L. Rev. 343, 350 (2008) (discussing how HAVA was passed, in part, pursuant to Congress’s Spending Clause power).

    [388].    United States v. Comstock, 560 U.S. 126, 133–34 (2010) (internal quotation marks omitted).

    [389].    Id. at 134.

    [390].    Sebelius, 567 U.S. at 559 (internal quotation marks omitted).

    [391].    See, e.g., Richard Primus & Cameron O. Kistler, The Support-or-Advocacy Clauses 25 (Michigan L. Sch. Pub. L. & Legal Theory Working Paper, Paper No. 666, 2020).

    [392].    See, e.g., id. at 25 n.101.

    [393].    See, e.g., Burroughs v. United States, 290 U.S. 534, 545–48 (1934).

    [394].    See supra Part III.B.

    [395].    United States v. Comstock, 560 U.S. 126, 133–34 (2010) (internal quotation marks omitted).

    [396].    See, e.g., Primus & Kistler, supra note 391, at 25 n.101 (“[S]ome legislation regulating presidential elections can be warranted as necessary and proper for carrying into execution the powers that Congress has over congressional elections under the Elections Clause.”).

    [397].    Burroughs, 290 U.S. at 545.

    [398].    Id. at 547.

    [399].    See supra notes 1–6 and accompanying text.

    [400].    For a diligent effort to analyze the 2021 version of the Act in its entirety, see Annotated Guide to the For the People Act of 2021, Brennan Ctr. for Just. (Mar. 18, 2021),

    [401].    These proposals are meant to be controversial in their reach (especially to state elections) and representative of ideas across the whole field of election law.

    [402].    See supra Part III.

    [403].    For the People Act of 2019, H.R. 1, 116th Cong. § 1402 (2019).

    [404].    See Oregon v. Mitchell, 400 U.S. 112, 117 (1970) (opinion of Black, J.) (summarizing these and other provisions). As noted above, though, Mitchell is a controversial decision disfavored by the current Court. See supra note 299.

    [405].    See Mitchell, 400 U.S. at 119–24 (opinion of Black, J.).

    [406].    See id. at 134.

    [407].    See supra notes 183–185 and accompanying text.

    [408].    See supra notes 335–339 and accompanying text.

    [409].    See supra note 351 and accompanying text.

    [410].    U.S. Const. amend. XIV, § 2; see Richardson v. Ramirez, 418 U.S. 24, 54 (1974) (“[T]he exclusion of felons from the vote has an affirmative sanction in s. 2 of the Fourteenth Amendment,” which “distinguish[es] such laws from those other state limitations on the franchise which have been held invalid under the Equal Protection Clause by this Court.”).

    [411].    See supra notes 368–371 and accompanying text.

    [412].    See, e.g., George Brooks, Felon Disenfranchisement: Law, History, Policy, and Politics, 32 Fordham Urb. L.J. 851, 854–59 (2005).

    [413].    See, e.g., State-by-State Data, Sent’g Project, https://www.sentencing (last visited Aug. 1, 2020) (showing the black-white and Hispanic-white disparities in incarceration in each state).

    [414].    See supra notes 374–376 and accompanying text.

    [415].    See supra note 299 and accompanying text.

    [416].    See, e.g., Dep’t of Homeland Sec. v. Regents of Univ. of Cal., 140 S. Ct. 1891, 1916 (2020) (inferring no anti-Latino animus from the fact that Latinos “ma[d]e up an outsized share of recipients of [a] cross-cutting immigration relief program” rescinded by the Trump administration).

    [417].    United States v. Comstock, 560 U.S. 126, 133–34 (2010) (internal quotation marks omitted).

    [418].    See For the People Act of 2019, H.R. 1, 116th Cong. § 2411 (2019).

    [419].    See H.R. 1 § 2413.

    [420].    See supra Part II.A (discussing the Apportionment Act of 1842); see also An Act for the Apportionment of Representatives to Congress Among the Several States According to the Ninth Census, § 2, 17 Stat. 28, 28 (1872) (imposing an equal population requirement for congressional districts for the first time); An Act Making an Apportionment of Representatives in Congress Among the Several States Under the Twelfth Census, § 3, 31 Stat. 733, 734 (1901) (same for a compactness requirement).

    [421].    See H.R. 1 § 2411(a)(1) (stating that a particular state agency “shall establish an independent redistricting commission for the State” (emphasis added)).

    [422].    Branch v. Smith, 538 U.S. 254, 280 (2003) (plurality opinion); see also Justin Weinstein-Tull, Election Law Federalism, 114 Mich. L. Rev. 747, 782 (2016) (“Congress may enact election legislation that forces a state to take action it might not otherwise take, without violating the anticommandeering doctrine.”).

    [423].    For a longer version of this normative argument for commissions, see Nicholas O. Stephanopoulos, Arizona and Anti-Reform, 2015 U. Chi. Legal F. 477, 487–91.

    [424].    See supra note 187 and accompanying text.

    [425].    See Rucho v. Common Cause, 139 S. Ct. 2484 (2019).

    [426].    Id. at 2504; see also id. at 2512 (Kagan, J., dissenting) (noting that “[t]he majority disputes none” of the ways in which “gerrymanders undermine democracy” and the Constitution).

    [427].    For the People Act of 2019, H.R. 1, 116th Cong. § 2413(a)(2) (2019).

    [428].    H.R. 1 § 2413(a)(1)(C).

    [429].    Note that racial vote dilution is also unconstitutional under the Fourteenth Amendment, see, e.g., White v. Regester, 412 U.S. 755, 765–67 (1973), and that its illegality under the Fifteenth Amendment hasn’t been explicitly confirmed, see, e.g., Voinovich v. Quilter, 507 U.S. 146, 159 (1993).

    [430].    See H.R. 1 §§ 4001–4801.

    [431].    See H.R. 1 §§ 5101–5501.

    [432].    See H.R. 1 §§ 6001–6401.

    [433].    Note that these potential policies don’t include campaign expenditure limits, which so clearly violate the First Amendment (as currently construed) that it’s irrelevant that Congress may have the authority to enact them.

    [434].    In this vein, I have previously argued that unchecked money in politics can cause a misalignment between voters’ preferences and governmental outputs, and that campaign finance regulations can reduce this noncongruence. See generally Nicholas O. Stephanopoulos, Aligning Campaign Finance Law, 101 Va. L. Rev. 1425 (2015).

    [435].    See supra note 383 and accompanying text.

    [436].    See, e.g., Niv M. Sultan, Election 2016: Trump’s Free Media Helped Keep Cost Down, but Fewer Donors Provided More of the Cash, (Apr. 13, 2017, 1:10 PM), 2017/04/election-2016-trump-fewer-donors-provided-more-of-the-cash/ (reporting a total cost of $6.5 billion for the 2016 federal election).

    [437].    See generally David Fontana, The Geography of Campaign Finance Law, 90 S. Cal. L. Rev. 1247 (2017) (discussing the interstate flows that characterize modern campaign finance).

    [438].    See For the People Act of 2019, H.R. 1, 116th Cong. §§ 4701–02 (2019).

    [439].    See H.R. 1 §§ 5200–21.

    [440].    See H.R. 1 §§ 8001–81, 10001.

    [441].    See, e.g., Maine & Nebraska, FairVote, maine_nebraska (last visited Aug. 1, 2020) (positively describing the Maine and Nebraska systems of allocating electors by congressional district).

    [442].    See supra notes 313–323 and accompanying text.

    [443].    See supra Part II.A (discussing the 1842 Apportionment Act).

    [444].    U.S. Const. art. II, § 1, cl 2.

    [445].    Cf. McPherson v. Blacker, 146 U.S. 1, 25 (1892) (holding that “the [state] legislature possesses plenary authority to direct the manner of [presidential electors’] appointment,” but not considering any potentially conflicting congressional regulation).

    [446].    See U.S. Const. art. I, § 2, cl. 1 (“[T]he Electors in each State shall have the Qualifications requisite for Electors of the most numerous Branch of the State Legislature.”).

    [447].    See supra notes 299–307 and accompanying text.

    [448].    See, e.g., Lyman v. Baker, 954 F.3d 351, 357 (1st Cir. 2020) (discussing these and other claims).

    [449].    See, e.g., id.

    [450].    They have failed to date. Of course, if the courts were to definitively hold that winner-take-all presidential elections don’t violate the Equal Protection Clause under any circumstances, then there would be no underlying constitutional violations to which congressional action could be congruent and proportional.

    [451].    See, e.g., Jowei Chen & Nicholas O. Stephanopoulos, The Race-Blind Future of Voting Rights, 130 Yale L.J. 862, 898–91(2021) (calculating voters’ preferences by racial group in the 2012 presidential election).

    [452].    Cf. Matthew M. Hoffman, The Illegitimate President: Minority Vote Dilution and the Electoral College, 105 Yale L.J. 935, 962–1010 (1996) (making the related argument that winner-take-all presidential elections dilute the votes of racial minorities in violation of the VRA).