2 - Democracy

Summary

This is the first of two chapters that relate the myriad ways in which state government has impeded true democracy in the United States. In this chapter, the focus is on those counter-majoritarian distortions that are hardwired into the Constitution. These structural barriers to majority rule include equal representation of small states and large states in the US Senate; the Electoral College; the constitutional provision under which a majority of state delegations to the US House of Representatives choose the president when there is no Electoral College majority; and the processes for appointing federal judges and amending the Constitution.

In recent years, the literature has called particular, and much-needed, attention to the cracks in the democratic foundations of the United States.¹ Some of those cracks are structural, the product of a constitutional design that elevates states over the people they are supposed to serve. At the heart of all these structural sources of counter-majoritarianism is the central constitutional theory of dual (federal and state) sovereignty. On that subject, most of the writings have singled out equal Senate representation of the states, the Electoral College, or the constitutional amendment process.² Several authors have more comprehensively covered two or more of those constitutional blind spots.³

Other writers have focused on the purposeful counter-majoritarian actions of state legislators, other state officials, and their enablers in both the public and the private sectors. There are separate massive literatures, for example, on gerrymandering, on specific voter suppression strategies, and on other state-regulated election practices.

Still other writers have maintained that, to the contrary, state government affirmatively advances democratic norms. Those arguments will be deferred to Chapter 5, which examines more generally the contention that the benefits of state government outweigh the many democracy-related and other harms described in this and the next two chapters.

Three preliminary comments are necessary: a definition, a point of emphasis, and an acknowledgment.

I’ll start by defining my terms. What, exactly, does it take for a country to qualify as a democracy? Importantly too, what isn’t required?

Although there is no universally accepted definition of democracy, I go here with the most commonly understood – and in my judgment the most helpful – definition: A democracy is a system that embodies popular sovereignty (rule by the people), majority rule (more on that in a moment), and political equality.⁴

The main difficulty in applying that definition of democracy is that the various elements can be either hard to enforce or even in conflict. What if a constitution, like that of the United States, systematically gives some voters drastically different per capita representation than others, in both the legislative and the executive branches of the national government? There might well be popular sovereignty, but no political equality and, from time to time, a loss of majority rule as well. What if the majority of the people’s chosen representatives in a state legislature enact an election process that makes it disproportionately harder for members of a racial or partisan minority to vote? In that scenario, the appearance of majority rule comes at the expense of actual political equality. As one eminent scholar put it, “Two visions of political malfunction – one stressing fear of the many and the other stressing fear of the few – coexist in our traditional views of government.”⁵

To insist on perfection would rule out calling any country a democracy. Some leeway is needed. But how much and what kind?

Whether democracy requires that judges be elected by the people, as they are under most US state constitutions, will be taken up in Chapter 6. Important as the issue is in debates over the meaning of democracy, I do not see the difference between the various methods of selecting judges as a factor favoring either the retention or the abolition of state government. Any of those selection methods would be possible in either a federation or a unitary republic. So I will put off that debate for now. Rather, the focus in this and Chapter 3 will be on the officially political branches – the applicable constitutional parameters and the practices of state officials that deviate from the principle of majority rule.

My view, as noted in the Introduction, is that departures from the strict requirement of majority rule are consistent with democratic theory only when they are necessary to protect one of the other two elements of democracy itself – popular sovereignty or political equality – or are otherwise critical to the protection of any other rights that are fundamental in a democracy or to the basic institutions of government.⁶ As will be seen, the various constitutional departures from pure majoritarian rule that are discussed in this chapter were not designed for the benign purposes of safeguarding popular sovereignty, political equality, fundamental rights, or essential government institutions. They were simply unavoidable concessions to state sovereignty. Similarly, abusive state actions in the contexts of districting maps, voter qualifications, and election processes are not aimed at, and are often deliberately intended to subvert, political equality.

The world’s democracies come in all shapes and sizes. Among the modern variations, the broadest distinction is probably between what Robert Dixon and others⁷ have described as majoritarian democracy and consensus democracy. In their starkest forms, majoritarian democracies allow majorities to govern without regard to minority representation. Consensus democracies also allow majorities to govern, but in ways that simultaneously assure meaningful representation of significant racial, political, or other minorities.

In most discussions of either democratic theory or electoral reform, that distinction is essential. But for present purposes, a different point is more crucial: Whichever of these broad political philosophies one favors, neither of them contemplates what we see all too often in the United States – a systematic bias in favor of governance by a partisan minority.

Nor is majority rule the only democratic casualty of the constitutional flaws and state actions that are the subject of this book. Political equality has also fallen prey. Indeed, the two fault lines are causally related, for it is political inequality that has bred minority rule. Only by awarding some citizens more say in the electoral process than others has it been possible to achieve minority rule on so regular a basis.

From this discussion, I draw two conclusions: First, these sorts of deviations from both majority rule and political equality are both serious and inconsistent with any credible definition of democracy. Second, most of those deviations – again, not all – are ones that would not occur but for either the constitutionally assigned powers of states or the ways in which state legislatures or individual state officials have exercised those powers.

Next, a point of emphasis: As laid out in the Introduction, I am proposing here a true unitary republic that retains ample space for decentralized decision-making – not the wholesale transfer of all, or even most, state power to an ever-expanding national government. But instead of decentralization occurring through a combination of state action and whatever powers each state chooses to delegate to its local political subdivisions, decentralized power would be exercised by the local political subdivisions directly. They could act on their own or in partnership with either the national government or other local governments. Either way, they would no longer need the permission of their state governments, because the latter would no longer exist. Their only legal constraints would be the US Constitution and other sources of national law.

In this respect, my prescription differs from that of the several other writers who have similarly embraced expansion of local government powers. Their models all contemplate the retention of state government.⁸ In an especially thoughtful article, Heather Gerken takes the expansion of local government power a step further. She argues that even those scholars who have encouraged greater devolutions of power from state governments to local governments have been remiss to stop with cities and towns. She advocates “federalism all the way down.”⁹ By this, Gerken means to include all the governmental “special purpose institutions” that, like cities and towns, lack sovereignty – for example, “juries, zoning commissions, local school boards, locally elected prosecutors’ offices, state administrative agencies, and the like.”¹⁰ She makes a strong case. But if, as I propose, state government were abolished and the bulk of its current functions transferred to local governments, I would leave it to those local governments themselves to decide how best to utilize these special purpose institutions.

The final preliminary note is an acknowledgment. State government is not all that ails our democracy. As the Introduction concedes,¹¹ there are multiple other causes: single-member districts for the US House; the Senate’s manipulation of the judicial appointment process; the weakening of traditional separation of powers constraints by modern political parties; racial, social, and economic inequalities that skew the political process; the roles of money and lobbying in exacerbating those inequalities; unprincipled elected officers; and, today, the threats of mob violence by the losers of elections.

Fine. Not all the threats to our democracy are the fault of the states. But most of them are. As the following discussion will show, the overwhelming majority of the problems that imperil our democracy can be traced, in most cases directly and in some cases indirectly, to either the states’ constitutionally hardwired powers (this chapter) or the behaviors of state legislatures or state executive branch officials (Chapter 3).

So let’s start with those counter-majoritarian attributes of states that are baked into the Constitution. This chapter identifies five of them. Unlike the examples considered in Chapter 3 (behaviors of state legislatures), these are examples to which one might respond as follows:

Yes, these constitutional features are counter-majoritarian, but you don’t have to abolish states in order to purge the counter-majoritarianism. That’s overkill. If you want to fix the problem of small and large states getting the same number of Senators, for example, you could keep states and just apportion US Senate districts by population, as we do with House districts.¹² Sure, maybe that solution would introduce other sources of counter-majoritarian outcomes, like urban versus rural residential patterns and single-member legislative districts,¹³ as is the case with House elections. But even without states that would be equally true, because the senators would still be elected from subdivisions of some kind.¹⁴

Similarly, one might observe, you don’t have to get rid of states just to replace the Electoral College with a national popular vote or to reduce the dominant role of states in the counter-majoritarian constitutional amendment process. And so on.

Those responses would be fair. You don’t have to abolish state government to neutralize their constitutionally enshrined counter-majoritarian effects. But it’s also true that state government is a but for cause of those effects; without it, none of those particular counter-majoritarian effects could occur. Moreover, while there exist less drastic alternative solutions to the constitutionally enshrined counter-majoritarian effects discussed in this chapter, abolition of state government is the only way to excise some of the counter-majoritarian state legislative behaviors taken up in Chapter 3. Abolition is also the only way to eliminate the fiscal waste discussed in Chapter 4 of this book. And as Chapter 5 illustrates, the countervailing affirmative benefits claimed for state government turn out to be minimal, if any. So, why not kill two birds with one stone?

Abraham Lincoln asked us to resolve “that government of the people, by the people, for the people, shall not perish from the earth.”¹⁵ From Lincoln’s simple statement of popular sovereignty, it is but a short step to two other guiding principles: Every citizen should have an equal voice in the selection of the country’s leaders. And, subject only to the exceptions described above, the majority should be able to get elected and then to govern.

Thus, the criticisms in this and Chapter 3 deserve to resonate with everyone who believes in representative democracy. In his book advocating abolition of the Electoral College, Jesse Wegman hit the nail on the head when he implored the reader to “approach [his] book not as a liberal or a conservative, not as a Republican or a Democrat, not as a Texan or a Californian or a Kansan or a New Yorker, but as an American.”¹⁶ I humbly offer the same plea here.

With that, it’s time to consider the major constitutional barriers to majority rule:

A Two Senators per State

Article I, Section 3, Clause 1 of the US Constitution reads: “The Senate of the United States shall be composed of two Senators from each State …”¹⁷ Article V, which lays out the procedure for amending the Constitution, contains an exception to that process: “[N]o State, without its Consent, shall be deprived of its equal Suffrage in the Senate.” As discussed in Section D, a literal reading of this exception would make it impossible ever to amend the equal Senate suffrage provision.

Neither the principle of equal state suffrage in the Senate nor the decision to preserve that principle for eternity was an accident. The Constitutional Convention was marked by a fierce battle between those who wanted every citizen’s vote to count equally in electing members of Congress and those who wanted the states to be counted equally. As noted in Chapter 1, the low-population states (along with the slave states, large and small) generally feared that, under a one-person-one-vote system, their influence would be overwhelmed by the larger delegations from the high-population states. Leaders of the high-population states believed that, in the democracy that they were proposing to create, each of their citizens should enjoy the same voting power and the same per capita representation as everyone else (as long as they were white male landowners).

In his landmark treatment of the equal representation principle, Robert Dixon distinguished two ways of viewing the right that is at stake. It can be seen as the individual right of each voter to have the same say as voters in other districts. Or (in the context of legislative elections), it can be viewed as the collective right of the people in a given district to equal representation in the legislature.¹⁸

With the impasse threatening to undo the Convention, Benjamin Franklin urged both sides to seek middle ground. A majority of the delegates ultimately voted for the “Great Compromise,” a proposal that had long been advanced by Roger Sherman of Connecticut. The compromise was a bicameral national legislature. Each state would have two senators, but in the House of Representatives the size of each state’s delegation would be proportional to its population.¹⁹

If they wanted the Constitution to become a reality, those who believed in the principle of one-person-one-vote had no choice but to capitulate. But they were not happy about it. In Federalist 37, Madison observed that the interests of large states and small states diverged. He lamented that, as a result, the delegates to the constitutional convention were “compelled to sacrifice theoretical propriety to the force of extraneous considerations.”

As Robert Dahl points out, several others among the most distinguished framers of the Constitution were also “bitterly opposed” to the counter-majoritarian concept of equal representation of states in the Senate. In Hamilton’s words,

As states are a collection of individual men, which ought we to respect most, the rights of the people composing them, or the artificial beings resulting from the composition? Nothing could be more preposterous or absurd than to sacrifice the former to the latter. It has been sd. that if the smaller States renounce equality, they renounce at the same their liberty. The truth is it is a contest for power, not for liberty [punctuation added].²⁰

James Wilson, speaking at the Constitutional Convention, put the point concisely: “Can we forget for whom we are forming a government? Is it for men, or for the imaginary beings called States?”²¹

I have previously written that “countries don’t immigrate; people do.”²² By way of analogy, and in keeping with the emphases of Hamilton and Wilson on states as “artificial” or “imaginary” creations, I would add here: “States don’t have interests; people do.” And from that, it seems a short step to “States shouldn’t vote; people should.”

Montesquieu would almost certainly have echoed similar sentiments. Though more frequently cited for his insights on separation of powers, he also had a great deal to say about what today would be called federalism. In particular, he would not have been a fan of the US Senate. In one essay, speaking about federations generally, he said this:

It is difficult for the united states to be all of equal power and extent. The Lycian republic²³ was an association of twenty-three towns; the large ones had three votes in the common council, the middling ones two, and the small towns one. The Dutch republic consists of seven provinces of different extent of territory, which have each one voice. [He then offers additional examples of power exercised proportionately in the Lycian system but not in the Netherlands.] Were I to give a model of an excellent confederate republic, I should pitch upon that of Lycia.²⁴

In Wesberry v. Sanders²⁵ and Reynolds v. Sims,²⁶ the Supreme Court proclaimed the one-person-one-vote principle for elections to the US House and for elections to both houses of state legislatures, respectively. Requiring roughly equal per capita representation in those legislative bodies, the majority opinion in Reynolds, an 8-1 decision, contains eloquent language from Chief Justice Earl Warren:

Legislatures represent people, not trees or acres. Legislators are elected by voters, not farms or cities or economic interests. … [I]f a State should provide that the votes of citizens in one part of the State should be given two times, or five times, or 10 times the weight of votes of citizens in another part of the State, it could hardly be contended that the right to vote of those residing in the disfavored areas had not been effectively diluted. It would appear extraordinary to suggest that a State could be constitutionally permitted to enact a law providing that certain of the State’s voters could vote two, five, or 10 times for their legislative representatives, while voters living elsewhere could vote only once. … Of course, the effect of state legislative districting schemes which give the same number of representatives to unequal numbers of constituents is identical. … Weighting the votes of citizens differently, by any method or means, merely because of where they happen to reside, hardly seems justifiable. … Since legislators are responsible for enacting laws by which all citizens are to be governed, they should be bodies which are collectively responsive to the popular will.²⁷

Precisely. And from the standpoint of democratic theory, every one of those normative statements applies with the same logical force to the United States Senate. That point was not lost on the defendant states. Their response to the Court, an attempt to justify unequal representation, boiled down to “Then how come the US Senate gets to do it?”

The Court gave them the only answer it could. Translated into lay English, it was “Because the Constitution says so.” And the only reason the Constitution says so is that, more than 200 years ago, the counter-majoritarian Senate was the price the large sovereign states had to pay the small sovereign states to get them to vote for ratification.²⁸

As a legal justification, the Court’s answer was bulletproof. The Constitution is clear. As an account of the Constitution’s political history, it was also dead on. But as to the normative question whether the US Senate, like all the other legislative chambers of a democracy, should be built on a foundation of one-person-one-vote, the Court’s answer amounted to nothing more than “Sorry, this is just the hand we’ve been dealt.” This book suggests that the same answer could be given to “Why do we have state government at all?”

The consequences of this unequal representation scheme have been dramatic from the start. In 1790, four states comprised a majority of the national population but held only eight of the Senate’s then twenty-six seats, that is, 31 percent of the Senate.²⁹ And “the most populous State (Virginia) had 12.6 times the population of the least populous (Delaware).”³⁰

The imbalance has only grown worse – in fact, much worse. In part that is because, in the years leading up to the Civil War, the original constitutional compromise was followed by a series of additional compromises. When it came to new admissions, one slave state would be paired with one free state – often with very different population sizes.³¹ Since then, states have continued to be admitted with only minimal regard for their sizes. As per the most recent decennial Census in 2020, the 576,851 residents of Wyoming receive the same Senate representation as the 39,538,223 residents of California. Each Wyoming resident thus enjoys more than sixty-nine times as much Senate representation as each California resident. The nine most populous states together comprise a majority of the US population but receive only 18% of the Senate’s seats; that is, a minority of the US population receives 82% of the Senate’s seats. The five least populated states collectively comprise only 1% of the country but receive 10% of the Senators – an overrepresentation of 10 to 1. And the ten least populated states account for only 2.83% of the national population, but their residents get 20% of the Senators.³²

The inequities are likely to persist. As Pamela Karlan notes, “[B]y 2040, 70 percent of Americans will live in the fifteen largest states. So 70 percent of the population will elect only thirty senators, leaving less than a third of the population to control the selection of nearly three-quarters of the Senate.”³³

The Australian Senate follows the US model. It consists of twelve members from each state, regardless of population.³⁴ Rodney Hall, in his book advocating the abolition of the Australian states, pointed out that under that system “a Tasmanian vote [at the time was] ten times as valuable as a Victorian vote.”³⁵ Hall’s book was written in 1998. As in the United States, that imbalance has only grown worse. As of June 2022, Australia’s most populous state, New South Wales (which is home to Sydney), had 8,153,600 residents; Tasmania had 571,500.³⁶ This gives Tasmanians more than 14 times as much per capita Senate representation as residents of New South Wales. For perspective, even that extreme disparity pales in comparison to the 69-1 advantage given to Wyoming residents over Californians.

As a result, it is not unusual for the party that controls the US Senate to represent significantly fewer Americans than the Senate’s minority party. In the three Congresses that convened during the period 2015–21, the Republicans controlled the Senate despite representing states that collectively contained only 46.9%, 44.8%, and 46.8% of the national population, respectively.³⁷

Today, these patterns systematically benefit one political party – Republicans. Four of the five least populated states (Wyoming, Alaska, North Dakota, and South Dakota) are reliably red. Among the five, only Vermont is reliably blue.

It was not always this way. As Pamela Karlan observes, “For most of American history, while large-population and small-population states might have had distinctive interests, their differences did not map onto a partisan divide.” During the twentieth century, small states were diverse in “their interests and political alignments. … Overall, sparsely populated, low-population states now tilt decisively toward the Republican Party [emphasis in original].”³⁸

If the correlation between state populations and Senate representation seems out of whack, actual voting in senatorial elections reflects similar inequities. In 2016, Democratic Senate candidates nationwide received 53% of the total Senate ballots cast, to Republicans’ 42%. Yet, of the Senate seats that were filled in that election, Republican candidates came away with twenty-two, Democrats with only twelve.³⁹

Those 2016 Senate elections illustrate how extreme the counter-majoritarian outcomes can be. Perhaps the most probative comparison of the disconnect between the national popular vote and the resulting composition of the Senate, however, would focus on the three most recent (at this writing) Senate election cycles – 2018, 2020, and 2022. Three consecutive cycles cover all 100 Senate seats. And this particular sequence of election cycles has the additional advantage of rough partisan symmetry, since it includes a midterm with a Republican President, a Presidential election year, and a midterm with a Democratic President.

As Table 2.1 demonstrates, in each of those three election years the party for whom the greater number of Americans voted in senatorial elections did indeed win a greater number of those races – unlike in 2016. But there the symmetry ends. For the three cycles combined, people voted for Democratic Senate candidates over Republicans by a sizeable margin – 53.2%–46.8%. Yet the number of Senate seats the parties ended up with was virtually identical.⁴⁰

Table 2.1 National popular vote versus Senate outcomes⁴¹

	Votes		Senate seats won
Year	Democrats	Republicans	Democrats	Republicans
2018	52,224,867 58.2%	34,687,875 38.7%	22	11
2020	38,011,916 47.0%	39,834,647 49.3%	15	20
2022	39,802,675 49.0%	39,876,285 49.1%	15	20
Total	130,039,458 53.2%	114,398,807 46.8%	52	51

Sources: Wikipedia, 2018 United States Senate Elections, https://en.wikipedia.org/wiki/2018_United_States_Senate_elections; Wikipedia, 2020 United States Senate Elections, https://en.wikipedia.org/wiki/2020_United_States_Senate_elections; Wikipedia, 2022 United States Senate Elections, https://en.wikipedia.org/wiki/2022_United_States_Senate_elections.

In the modern era, that cycle is not an outlier. Steven Levitsky and Daniel Ziblatt demonstrate that “the Democrats have won an overall popular majority for the Senate in every six-year cycle since 1996–2002. And yet the Republicans controlled the Senate for most of this period” [emphasis in original].⁴²

As if the composition of the Senate didn’t provide a large enough unfair advantage to the minority over the majority, the Senate’s filibuster rule makes the inequities worse. It takes sixty senators to overcome a Senate filibuster and bring a bill to a vote (“cloture”). Thus, forty-one senators can thwart the combined votes of their fifty-nine Senate colleagues and the 435 House members, all of them elected by their respective constituents.

Moreover, the whole is worse than the sum of its parts. It is not just that a minority of states wield disproportionate power. The filibuster rule gives citizens of the low-population states yet another advantage, since senators who represent only a small number of people can obstruct legislation favored by far greater numbers. The combination of the Senate’s counter-majoritarian makeup and the filibuster enables senators from states that comprise only 13% of the national population to thwart the will of those who represent the other 87%.⁴³

The filibuster rule is the product of the US Senate itself rather than an action by the states. At least in theory, the Senate might have created, and can preserve, the filibuster with or without states. But given the smaller states’ incentive – and their disproportionate power – to create and maintain the filibuster, abolishing state government and replacing the states’ equal Senate representation with a direct vote of the people would at least diminish the life prospects of this counter-majoritarian add-on to an already counter-majoritarian chamber.

These inequalities are not merely a theoretical problem. Sanford Levinson summarizes some of their more pernicious effects: systematic net movement of taxpayer funds from the residents of large states to the residents of small states, without regard to their respective poverty levels; a disproportionate adverse impact on minority representation; and disproportionate say of small-state senators in the composition of the federal Judiciary.⁴⁴

The same inequalities have another source – the nearly total disenfranchisement of US citizens who live in the District of Columbia. Like their fellow citizens who live in the fifty states, they are subject to all federal taxes.⁴⁵ As of 2021, in fact, they “pay the highest per capita income tax rate in the nation.”⁴⁶ Yet they are denied any representation in the US Senate and have only a nonvoting representative in the US House.⁴⁷ (The Twenty-third Amendment now awards the District the minimum three electoral votes in presidential elections.)⁴⁸ Congress has broad authority over the DC budget and absolute authority to nullify any legislation passed by the local DC government.⁴⁹

The Constitution authorizes Congress to admit new states to the union,⁵⁰ but to date, Congress has declined to grant statehood to DC. Although some have argued that various provisions of the Constitution preclude DC statehood,⁵¹ that uncertainty is not the main roadblock today. Until recently a relatively nonpartisan issue, the principal barrier to DC statehood today is rock-solid Republican opposition. The political math is straightforward: DC residents vote overwhelmingly for Democrats. In 2020 and again in 2021 the House voted to grant statehood to DC (with every House Republican voting no), but in both instances the bill failed to pass the Senate.⁵² The systematic advantage that the two-senators-per-state rule gives the Republican party, aggravated by the filibuster, makes Senate passage of a DC statehood bill a near impossibility for the foreseeable future. Abolishing states would put DC residents on an equal footing with other US citizens in voting for the president and for members of both houses of Congress.

The discussion in this section hopefully conveys the major counter-majoritarian effects of the two-senators-per-state principle. This book does not propose eliminating the Senate. But in the stateless unitary republic posited here, the Senate, like the House, would consist of members elected from equipopulous districts (fewer in number than the House districts) throughout the country. The details appear in Chapter 6, Section B.

B The Electoral College System

Hundreds of books and articles have provided histories, descriptions, or critiques of the Electoral College.⁵³ It is another of those many impediments to democratic rule that can fairly be laid squarely at the feet of the states. This section assembles and responds to all the arguments that have been made in its defense. But first, a few words on how it works and how it started.

The president of the United States is not chosen directly by the people. Instead, the people vote for “electors” who in turn choose the president. This is by constitutional design: “[E]ach State shall appoint, in such Manner as the Legislature thereof may direct, a Number of Electors, equal to the whole Number of Senators and Representatives to which the State may be entitled in the Congress …”⁵⁴ Exercising that power, every state legislature today has directed that the electors of that state be chosen by a vote of its people.⁵⁵ Each candidate will have named a slate of electors, publicly pledged to him or her, for whom the people cast their votes.⁵⁶

As this provision says, the number of electors for a given state equals the number of its US House representatives plus its two Senators. (The District of Columbia, as just noted, is also awarded three electors.)⁵⁷ In every state except Nebraska and Maine, the presidential candidate who wins a plurality of the statewide vote receives all of that state’s electoral votes – or, technically, the entire slate of the electors who are pledged to that candidate. Nebraska and Maine each award one electoral vote to the plurality winner of each of the state’s congressional districts and two additional electoral votes to the candidate who wins a statewide plurality.⁵⁸

That brings the total number of electors to 538-435 based on the number of House members, 100 based on the number of Senators, and 3 for DC Together, the 538 electors are commonly referred to as the Electoral College. The Electoral College members meet in their respective states and cast their votes. Those votes are then officially tallied by the vice president, in his or her capacity as president of the Senate, in a joint session of Congress.⁵⁹ To win a presidential election, a candidate must receive an outright majority of the Electoral College. If no candidate wins such a majority, the House of Representatives chooses the president from among the three candidates with the highest numbers of Electoral College votes, as discussed in Section C. In that event, each state delegation – large or small – gets one vote. To win the presidency in such a House election, a candidate needs the votes of a majority of the state delegations, today twenty-six states.⁶⁰

Robert Dahl has described the unimpressive origins of the Electoral College. As he shows, the framers settled on the idea of the Electoral College out of a combination of weariness, desperation, and dissatisfaction with every alternative they could think of.⁶¹

Although the question of how to elect the president generated considerable anguish at the Constitutional Convention,⁶² the Federalist and Antifederalist Papers devote surprisingly little space to the subject. Two papers do address it head-on.

In Federalist 68, Hamilton thought it “desirable that the sense of the people should operate in the choice of the [president],” but not directly. They should instead vote for electors who “possess the information and discernment requisite to such complicated investigations.” James Ceaser expands on that theme. He argues that the framers’ dominant concern was that the president be a statesman – someone who would govern based on the best interests of the nation rather than a person who, out of personal ambition, would seek to curry favor with the masses. The framers wanted a strong leader rather than a follower. For that purpose, they did not trust popular sentiment. They feared it would be too easily swayed by either personal charisma or demagogic appeals on the issues.⁶³

Hamilton in Federalist 68 saw additional advantages. Voting for electors avoids “tumult and disorder,” for two reasons. First, he argued, the people will be voting for many electors, not just one president. Today, of course, electors are pledged to specific candidates, so as a practical matter people are still voting, albeit indirectly, for one candidate. Second, he wrote, the electors will be assembling in several places, not just one central location; thus, there is less likelihood of “heats and ferments.”

He also maintained that this system will guard against corruption, “chiefly from the desire in foreign powers to gain an improper ascendant in our councils,” because a foreign power won’t know who the electors are until after the election. Modern readers of this Federalist paper will note the irony. Compelling evidence of successful and unsuccessful Russian influence on US elections arose in 2016 and 2020, respectively. And, if anything, it is much easier for someone to corrupt a handful of electors and swing a state in a close national election than to corrupt millions of individual voters.

Finally, and certain to induce derision in modern-day Americans, is this passage, also from Federalist 68:

The [Electoral College system] affords a moral certainty, that the office of President will never fall to the lot of any man who is not in an eminent degree endowed with the requisite qualifications. … It will not be too strong to say, that there will be a constant probability of seeing the station filled by characters preeminent for ability and virtue.

At a time when George Washington was the consensus choice as the first president, that assumption might well have seemed reasonable. But in 2016 the Electoral College system awarded the presidency to Donald Trump, a man whom few would describe as “pre-eminent for ability and virtue” the national popular vote would have prevented that result.

At any rate, all of these arguments were merely for the proposition that the people should vote for “wise” electors who in turn would select the president, rather than vote for the president directly. Even if one subscribes to that viewpoint – highly unlikely in modern times – the specific counter-majoritarian effects built into the design of today’s Electoral College, discussed below, raise different questions.

One paper did object strongly to the whole concept of indirect voting through electors. In Antifederalist 72, “Republicus” offered these comments: “Is it necessary, is it rational, that the sacred rights of mankind should thus dwindle down to Electors of electors, and those again electors of other electors?”⁶⁴ Rather, he points out, “To conclude, I can think of but one source of right to government, or any branch of it – and that is THE PEOPLE. They, and only they, have a right to determine whether they will make laws, or execute them, or do both in a collective body, or by a delegated authority [emphasis in original].”⁶⁵

I note that that paper was published in the Kentucky Gazette. On the likely assumption that its anonymous author was from Kentucky, his preference for direct popular election of the president over the proposed Electoral College takes on extra moral force. The Electoral College system, in which every state receives the same two electoral votes from its senatorial representation, benefits small states like Kentucky by giving them a voice disproportionate to their populations. Yet he opposed it.

Importantly, however, even that writer’s objection to the Electoral College system was rooted solely in his concern for popular sovereignty – indeed, one of the three essential elements of a democracy. Missing from his argument was any reference to either of the other two elements – majority rule and political equality. It is those latter elements that lie at the core of the modern objections to the Electoral College.

All of this matters. At least five US presidents have been elected despite finishing second in the national popular vote.⁶⁶ The first case is distinctive. In a multicandidate race in 1824, Andrew Jackson won a plurality, but not a majority, of both the national popular vote⁶⁷ and the Electoral College. Yet, the House of Representatives, voting by individual state delegation as required by the Twelfth Amendment and discussed below, selected John Quincy Adams, who had finished second to Jackson in both the popular vote and the Electoral College.

In each of the last four of those instances, the winning candidate garnered an outright majority of the Electoral College despite losing the national popular vote. In 1876, Samuel Tilden won the national popular vote but lost to Rutherford B. Hayes by one vote in the Electoral College. In 1888 the Electoral College selected Benjamin Harrison, who had lost the popular vote to Grover Cleveland. In 2000, Al Gore won the popular vote but, after the Supreme Court’s controversial 5-4 decision in Bush v. Gore halting the Florida recount,⁶⁸ narrowly lost the State of Florida; that result enabled George W. Bush to win the Electoral College with 271 electoral votes – just one more than the required majority.⁶⁹ And in 2016, the people chose Hillary Clinton over Donald Trump by a margin of almost three million votes, but Trump prevailed in the Electoral College.⁷⁰

Those are the actual misses – where the American people chose candidate A over candidate B, but the Electoral College nonetheless delivered the presidency to candidate B. There have also been near misses – lots of them. These are cases in which the same candidate wins both the national popular vote and the Electoral College (so that the system didn’t do any harm), but where a switch of just a tiny number of popular votes, either nationally or in one or more close states, would have produced an actual miss. There are two kinds of near misses: In one scenario, a candidate wins the national popular vote but barely wins the Electoral College. A shift of a small number of popular votes, in close elections in states with enough combined electoral votes, would have awarded the Presidency to the candidate who lost the national popular vote. In the second scenario, the near mismatch is reversed. A candidate wins the Electoral College but barely wins the national popular vote. A shift of a minute percentage of the popular votes nationwide would have resulted in that candidate losing the national popular vote but still winning the Electoral College and therefore the presidency.

There are several modern examples of the first scenario. In 1976, Georgia Governor Jimmy Carter won the popular vote by almost 1.7 million votes, but “a switch of fewer than 4,000 votes in Hawaii and 6,000 in Ohio would have given [incumbent] President Gerald Ford an Electoral College victory.”⁷¹

In 2004, George W. Bush defeated John Kerry by more than three million popular votes. But Bush eked out a much narrower victory in the Electoral College, 286 electoral votes to 251, with 270 needed for a majority. (One faithless elector from Minnesota, a state that Kerry won, voted for Kerry’s running mate, John Edwards). So Bush made it through by just sixteen electoral votes. In three of the states that Bush won – Iowa, Nevada, and New Mexico – his combined popular vote margin was only 37,547 votes – approximately 1.2 percent of the total votes in those states, or fewer than one in 83 voters.⁷² So those states could easily have gone the other way. If they had, their seventeen combined electoral votes would have swung the election for Kerry, despite Bush’s decisive national popular vote margin.

More dramatic still was a near miss in 2020. In that election, Joe Biden defeated Donald Trump by more than 7 million popular votes but won the Electoral College with only 306 electoral votes – 36 more than he needed. In three of the states that Biden won – Arizona, Georgia, and Wisconsin – his combined popular vote margin was only 42,918 votes – approximately 0.37 percent of the total vote in those three states. Thus, if 21,459 of those who voted for Biden in those states – that is, just one out of every 269 of Biden’s 5,776,642 voters – had instead voted for Trump (and of course assuming they were distributed among the three states in the way most favorable to Trump), those states could easily have gone the other way. If they had, their thirty-seven combined electoral votes would have swung the election for Trump, despite Biden’s overwhelming national popular vote margin.⁷³

An example of the second kind of near miss occurred in 1968. Richard Nixon handily won the Electoral College over Hubert Humphrey. But his victory margin over Humphrey in the national popular vote was slim. Nixon garnered 31,783,783 votes to Humphrey’s 31,271,839, a difference of only 511,944 votes (0.7 percent of the total).⁷⁴ Therefore, if 255,972 of those who voted for Nixon (a large-sounding number but only one out of every 125 of his voters) had instead voted for Humphrey, the latter would have won the national popular vote but Nixon would still have been awarded the presidency.

These are just some of the modern examples of near misses. The aggregate probabilities illustrate that those elections were not quirks. One mathematically sophisticated paper finds that “[i]n elections within a one percentage point margin – about 1.3 million votes, based on 2016 turnout – the probability of [the winner of the popular vote losing the election] is around 40 percent. In historical fact, six presidential elections of the 46⁷⁵ since 1836 have yielded a popular vote margin within one percentage point.”⁷⁶

Jesse Wegman sums it up nicely:

[W]hat’s remarkable is not that a split between the Electoral College and the popular vote has happened twice in the past two decades, it’s that it hasn’t happened far more often. In sixteen other elections, a shift of 75,000 votes or fewer in key states … would have made the popular vote loser president. Six times, a shift of fewer than 10,000 votes would have done the trick.⁷⁷

Frequent future recurrences, in other words, are a statistical certainty.

What is it about the Electoral College that has produced these counter-majoritarian outcomes? Two of its features are to blame.

The principal culprit is the winner-take-all element. As noted earlier, except in Nebraska and Maine, whoever wins a plurality of the statewide popular vote receives all of that state’s electoral votes. It doesn’t matter how close the popular vote in that state was.

By way of illustration, in 2000 George W. Bush won the State of Colorado over Al Gore with only 51 percent of the popular vote. But his prize was all eight of Colorado’s electoral votes. Had Gore received a share of those electoral votes proportionate to his popular vote within the state – even just three of those votes, rounded down – he would have become president.⁷⁸ Attempts to save the Electoral College have in fact included proposals to mandate precisely such a proportional allocation system in every state. That idea is discussed below, though for the reasons given, it would be an incomplete solution and at any rate inapplicable to the stateless country imagined in this book.

Furthermore, the disconnect between how people vote and how the Electoral College votes is asymmetrical. Today, the likelihood of a Democrat winning the national popular vote but losing the Electoral College is much greater than vice versa.⁷⁹ That’s because, “in the Modern period, Democrats have tended to win large states by large margins and lose them by small margins.”⁸⁰ For example,

in 2016, Hillary Clinton won the electoral votes of three of the ten largest states – by 30 percentage points in California, 22.5 percentage points in New York, and 16.8 percentage points in Illinois. But in the seven large states that Donald Trump won, his largest margin of victory was 9 percentage points (in Texas), and in three states (Florida, Michigan, and Pennsylvania), his margin of victory was less than 1.3 percent. Overall, in the ten largest states, Hillary Clinton received 36,440,207 votes and Donald Trump received 31,295,308. But because of how their supporters were geographically distributed, Clinton garnered only 98 of that group of states’ electoral votes, while Trump garnered 138. These large states accounted for the majority of Trump’s electoral vote victory margin.⁸¹

There is a second source of the Electoral College’s counter-majoritarian outcomes. Every state (and DC) receives one electoral vote for every member of its US House delegation, plus two additional electoral votes for its two Senators. The size of each state’s House delegation is proportional to its population, but its Senate representation is not. As discussed in Section A, Wyoming gets two Senators, as does California. Those two electoral votes per state are freebies. They bear no relation to the size of the state’s population.

Several writers have suggested that, contrary to conventional wisdom, the two-extra-vote feature of the Electoral College system doesn’t really produce a net benefit for citizens of the small states. These writers acknowledge that citizens of small states receive a slight bump by getting the same two extra electoral votes as the large states. But, they point out, that advantage is far less mathematically significant than what the winner-take-all rule provides voters in large swing states. The latter have the potential to produce a much greater net electoral vote gain for their preferred candidates than their counterparts in small swing states have for theirs.⁸²

These writers’ comparative point is clearly correct, and it can help to explain why the various states’ stances on Electoral College reform have not correlated especially well with population size. But that doesn’t mean (and they don’t imply) that the two extra votes are fair. It merely means that both the winner-take-all feature and the two-extra-votes feature are counter-majoritarian. Neither the fact that the two effects work in opposite directions (with respect to the balance between large states and small states) nor the fact that one effect is usually greater than the other changes that inescapable bottom line.

Of more practical importance, while a disconnect between the popular vote winner and the Electoral College winner is more likely to result from the winner-take-all rule than from the two-extra-vote feature, the latter is still consequential. For one thing, it systematically tilts the playing field in favor of one major political party – Republicans, since they command clear majorities among the populations of the smaller states.⁸³ Given the impact that the choice of US President has on both the nation and the world, the two-vote bump would have enormous consequences if it were to make the difference in even one presidential election.

As it turns out, it already has. Those two “free” votes per state changed the outcome in 2000. Even after the Supreme Court awarded Florida to Bush, he ended up with only 271 electoral votes to Gore’s 266. (One elector, from DC and therefore pledged to Gore, abstained.) Bush won thirty states; Gore won twenty states plus DC.⁸⁴ If the two free electoral votes per state (and DC) were subtracted from each candidate’s total, Bush would have ended up with 60 fewer electoral votes, for a total of 211. Gore would have ended up with 42 fewer electoral votes, for a total of 224. Gore would thus have won the presidency by thirteen electoral votes.

We will never know how different the course of history would have been. It seems a safe bet that the United States would have pursued a more forceful climate change policy and that doing so would have spurred other world powers to agree to do the same. Perhaps the Iraq War would have been avoided. Whatever one’s views on those and other issues, the two extra votes changed the world.

Except for “state sovereignty is an end in itself” and “the framers were smart people who knew what they were doing,” are there today good reasons to use this complicated, counter-majoritarian institution to choose our presidents? To address that question, it is necessary to add “Compared to what?” To reframe the issue slightly, is there a better way to choose our presidents?

There has been no shortage of ideas. Two recurring proposals would eliminate actual human “electors” and therefore the Electoral College, but preserve electoral votes and the requirement of a majority of those electoral votes to win the presidency. More substantively, these proposals would abolish the winner-take-all feature as it now stands. As the preceding discussion shows, it is that feature, after all, that accounts for the lion’s share of the Electoral College’s counter-majoritarian effects.

One of those proposals is for a constitutional amendment that mandates the Nebraska/Maine model for every state. Two electoral votes would still be awarded to the candidate who wins a statewide plurality, but the rest of the state’s electoral votes would be allocated to the plurality winners of each of the state’s congressional districts. One variant of this proposal would be to allow each state legislature to create “presidential districts” in place of congressional districts for this purpose. A subvariant of the latter proposal would be to fold the two senatorial electoral votes into these presidential districts.⁸⁵

On its face, any of these district system variants might look like improvements, since they make it possible for a candidate who falls short of a statewide plurality to pick up at least some of the state’s electoral votes. In the end, however, each of these variants would still be a winner-take-all system. It’s just that the winner-take-all rule would be applied to each district rather than to each state. In the view of many, in fact, a district system of any kind would be even worse than the present system, because the combination of residential patterns, gerrymandering, and the retention of the single-member district feature makes the congressional districts poor proxies for proportional allocation of the electoral votes.⁸⁶ A district system, therefore, doesn’t eliminate the risk of the popular vote loser winning a majority of the Electoral College. In some elections, it might even increase that risk. By way of illustration, Jesse Wegman points out that “[i]f the district system had been in use nationwide in 2012, Mitt Romney would have become president, despite losing to Barack Obama by about five million votes.”⁸⁷

A second reform proposal would distribute each state’s electoral votes among the candidates in proportion to their popular votes in that state.⁸⁸ This proposal would be a clear improvement over the district proposal. It avoids not only the winner-take-all feature, but also the counter-majoritarian effects of intrastate residential patterns and gerrymandering, thus greatly reducing the chances of the popular vote loser becoming the electoral vote winner.

Still, while it would reduce those chances, in a reasonably close election it won’t always prevent the popular vote loser from walking away with the presidency. Moreover, as Wegman observes, it could result in no one getting a majority, since third-party and independent candidates could get significant numbers of votes in one or more states. In fact, that would have been the case in 2016. Hillary Clinton won the popular vote, but with a proportional allocation system the election would have been thrown into the House.⁸⁹ Anyway, if the goal of such a reform is to bring the electoral vote into close proximity to the popular vote, wouldn’t a direct national popular vote serve that purpose both more simply and more reliably? Why settle for an approximation when we could have the real thing? Put another way, why settle for “hopefully” when we could have “definitely”?

So neither of these reform proposals, in any of their variants, eliminates the counter-majoritarian effects of relying on electoral votes. That is reason enough to look for an alternative. Moreover, if state governments were abolished entirely, as this book recommends, there would be no practical way to implement an electoral vote system anyway, unless some kind of regional allocation of electoral votes were to replace the state-based system. And that would simply recreate the same counter-majoritarian problem. Thus, if we are to rid ourselves of the whole concept of electoral votes, the question is what would replace it. The most obvious alternative is a direct national popular vote. The next step, therefore, is to compare that to the status quo.

As the preceding discussion suggests, the major criticism of the Electoral College system, throughout its history, has been its counter-majoritarian structure and effects. Robert Hardaway, one of its leading defenders, mocks the many studies “devoted to proving by complex mathematical equations and formulae that voters in different states do not have precisely [my emphasis] the same voting power in electing a President.”⁹⁰ The implication is that the only problem with the Electoral College is that the outcomes don’t match the popular vote with scientific exactitude. With respect, no one has criticized the Electoral College system for not providing “precisely the same voting power.” Put aside the two free electoral votes that each of the fifty states receives just for being a state – whether it’s California or Wyoming, with the previously mentioned population ratio of approximately 69-1. The winner-take-all feature of the Electoral College system gives the voter in a swing state infinitely – not just slightly – more power than the voter in a safe state.

Hardaway’s ultimate response to demonstrations of unequal voting power is to point out that both the requirement of equal Senate representation (discussed in Section A) and the constitutional amendment process (discussed in Section E) could similarly be faulted for depriving citizens of equal voting parity.⁹¹ Michael Maibach offers the same argument: “The Electoral College is no more ‘undemocratic’ than is the Senate or the Supreme Court.”⁹²

Indeed. Those who believe that equality demands a national popular vote for the presidency should also object to the makeup of the Senate and the states’ roles in the constitutional amendment process – as I do. But simply pointing out that other features of the Constitution are similarly flawed isn’t much of a defense.

Maibach acknowledges that the Electoral College system wastes votes because each state-by-state winner takes all the electors from that state. His response is that awarding the presidency to whoever wins the national popular vote would also be a winner-take-all system in which the votes for the losing candidate would be wasted.

That response misconceives the objection. The problem with the present system is not simply that it contains a winner-take-all feature. It is that it contains fifty different winner-take-all pieces and then adds them together without regard to the victory margins in each of those fifty pieces.

Faithless electors only amplify the counter-majoritarian essence of the Electoral College. These are electors who renege on their pledge to vote for a particular candidate. As of 2020, thirty-two states had laws that require the state’s electors to vote for the candidate who won a plurality of the popular vote in their respective states.⁹³ But the other eighteen states do not. In those states, a faithless elector may substitute his or her preferred candidate for the one chosen by the people. And even in the thirty-two states that require fidelity, there are often minimal consequences for those who stray.

To date, faithless electors have not altered the outcome of any US presidential election. But the potential is real. “[I]n 2016, 10 of the 538 electors cast ballots for someone other than their state’s popular vote winner, an unusually high number that could have changed the outcome of five of the previous 58 previous US presidential elections.”⁹⁴

Even after all the popular votes have been tabulated and the totals certified by the relevant officials in each state, the Electoral College system leaves several hurdles to clear. As will be seen, these additional steps create the potential – recently diminished but not eliminated – for still more counter-majoritarian effects. Each state’s governor (unless the state’s law designates another official for this purpose) officially submits a list of his or her state’s electors to Congress. Days later, the Electoral College formally votes. The vice president, presiding over a joint session of Congress, counts the electoral votes and announces the totals. Members of both Houses have an opportunity to object to the appointments of particular electors or to the votes those electors cast. To be sustained, an objection requires majority votes in both Houses of Congress. At the conclusion of the debate, both Houses vote on whether to certify the Electoral College results. Upon certification, the winner⁹⁵ of the election is sworn in as the president.⁹⁶

Until the 2020 presidential election, this part of the process was rarely controversial. But that year, President Trump ran for reelection against Democratic nominee Joe Biden. Even before the Electoral College returned a majority for Biden, Trump and many of his fellow Republicans had claimed repeatedly that Biden’s reported victories in several states were the product of widespread voter fraud. There was never even a kernel of truth to those claims, as detailed below,⁹⁷ but the Trump campaign tried nonetheless to persuade state election officials to revise the vote totals. Failing that, the campaign argued, Vice President Mike Pence (himself running for reelection as Trump’s running mate) legally could, and should, use his role as electoral vote-counter to refuse to declare Biden the winner. Pence, to his credit, made clear that he did not have that power. Nonetheless, all these claims aired nonstop on Fox News and other television and social media outlets favorable to Trump. Wild conspiracy theories flourished. In one December 2020 survey, “over 75% of Republican voters found merit in claims that millions of fraudulent ballots were cast, voting machines were manipulated, and thousands of votes were recorded for dead people.”⁹⁸

The ensuing hysteria triggered several events in rapid succession. First, approximately 14 Senate Republicans and 140 House Republicans announced their intentions to object, when the time came, to the certifications of Arizona’s or Pennsylvania’s electoral outcomes, or both.⁹⁹

On January 6, 2021, the time came. This was the appointed date for Congress to certify the final electoral vote. But before Congress could finish doing so, the false claims of election fraud spawned a second event – the storming of the Capitol building by heavily armed Trump supporters who had been led to believe that the Democrats had stolen the election. Amidst cries of “Hang Mike Pence,” the mob breached the Capitol walls, looted and ransacked the building, assaulted more than 100 Capitol police officers, and roamed the halls in search of terrified members of Congress hiding in bunkers and locked offices for several hours. The insurrection was eventually put down, but it delayed the required congressional certification until the wee hours of the following morning.¹⁰⁰

Even after all the violence (which included four deaths) and the toll it had taken on the members of Congress, police and other staff, the Capitol building, and democracy itself, and even as the nation watched the attack replayed on TV with a mixture of revulsion and horror, a third event occurred. When the police finally cleared the building and Congress belatedly reconvened, 6 Republican senators and 121 House members carried out their threats to object to Arizona’s electoral outcome; 7 Republican senators and 138 Republican House members voted to object to Pennsylvania’s. Some House Republicans, but no Senate Republicans, also objected to the results in Georgia, Michigan, and Nevada.¹⁰¹ Among those who earlier had pledged to object, only a handful changed their minds in light of the onslaught they had just experienced.¹⁰²

On December 29, 2022, President Biden signed into law the Electoral Count Reform and Presidential Transition Improvement Act (ECRA).¹⁰³ This statute amended the Electoral Count Act, which had governed the Electoral College and congressional certification processes since 1887. Among other things, the new law preserved the power of individual members of Congress to lodge objections but made it harder, in ways discussed below, for a small handful of such individuals to sabotage or delay the process. It also made explicit the existing understanding, which had been expressed only vaguely in the original law, that the vice president’s role in the counting of the electoral votes is purely ministerial. ECRA is in effect for the 2024 election.

ECRA, therefore, will now make it harder for congressional objectors to disrupt the timely selection of the President. Harder, but not impossible. Before ECRA, a single member of either House could object to the certification of a state’s electoral votes, thereby triggering debate and delay.¹⁰⁴ Under ECRA, an objection will now require a 20 percent vote in both Houses of Congress. And upon such a vote, it still takes majorities of both Houses to sustain the objection.

The 20 percent rule is a clear improvement. But if the 2020 election is a guide, the numbers related earlier show that even the new threshold is surmountable. Twenty percent means twenty Senators and eighty-seven Representatives. As noted, on January 6, 2021, even after the violent mob attack on the Capitol just hours earlier, 121 Representatives supported objections to Arizona’s slate of Biden electors and 138 supported objections to Pennsylvania’s – in both cases far in excess of the new 20 percent threshold. The Senate would have fallen short of the current threshold, as only six Senators supported objections to Arizona’s electors, seven to Pennsylvania’s. All those numbers, however, would surely have been greater but for some Members’ change of heart upon the violent attack that had interrupted their certification process and threatened their lives. The fourteen senators who before the insurrection had threatened to object got uncomfortably close to the current threshold of twenty, and the House objectors easily cleared the threshold.¹⁰⁵

So it is not hard to imagine a future scenario in which even ECRA’s 20 percent hurdle is cleared and the congressional certification delayed. Admittedly, there is one crucial safeguard: it will still take a majority vote in both Houses to ultimately reject a slate of electors. But even that check is less than ironclad in an age when those who fail to toe the party line face likely recriminations in the next round of their party’s primaries.

The antics of the congressional objectors in 2020 cannot be blamed – at least not directly – on the states. They did nothing wrong. But the existence of states is essential in multiple ways to the process that permits those congressional obstructions to occur. First, take away the states and there is no Electoral College system to begin with. Second, the congressional objectors themselves are elected through counter-majoritarian voting systems that account for at least some of the members of both chambers – in the House, because of intrastate residential patterns, single-member districts, gerrymandering, and voter suppression strategies (discussed in Sections A and B of Chapter 3), and in the Senate because of the equal state suffrage rule (discussed in Section A). Third, the congressional objections were only to the vote counts in selected states where the victory margins were relatively small, not to the national popular vote that Biden won handily.

Furthermore, the January 6, 2021 insurrection that Trump’s enraged followers staged, and the killings and destruction left in its wake, while still possible under a national popular vote system, would have been appreciably less likely. Since President Biden won the national popular vote by over seven million votes, getting people to swallow the myth of voter fraud on a scale that massive would have been a tougher sell. The events of 2020 thus provide a vivid illustration of the potential harms that could be avoided in a unitary republic, cleansed of state government, in which a majority of the voters choose the President through a process that is simpler, fairer, and more direct.

Of course, even with a national popular vote, the final tally would still have to be certified by somebody. That responsibility could be left with Congress and the vice president. For reasons considered in Chapter 6, however, this book instead proposes that the entire presidential election system be administered, and the votes ultimately certified, by a nonpartisan commission located within the national judicial branch, as is done with great success in Brazil.

Alexander Keyssar notes other, less publicized, features of the Electoral College system that have drawn criticism over the years. He points out that state legislatures could change the methodology from one election cycle to another whenever doing so achieves a partisan advantage.¹⁰⁶ As he also observes, the Constitution leaves the method for choosing the electors up to each state legislature. They arguably could, and in the early years of the republic did, choose the electors themselves rather than allow the people to do so.¹⁰⁷

In the current, radically polarized political climate, that scenario is not far-fetched. The recent attempts to resurrect the Independent State Legislature Theory (the “ISLT”), discussed in Chapter 3, Section C, make this danger quite real. The Supreme Court in 2023 rejected only the most extreme version of that theory, and only as it applied to congressional elections. But in the same case the Court opened the door wide for what I term “ISLT-lite,” and at any rate, the Court had no occasion to address its applicability to presidential elections at all. As a result, one cannot dismiss the real possibility of legislatures, particularly when the combination of counter-majoritarian forces discussed in this and Chapter 3 produces a Republican majority in a state where the voters lean Democratic, reclaiming a right to select their own presidential electors rather than leave the decision to the people.¹⁰⁸

An affirmative benefit of the Electoral College system, Maibach earnestly contends, is that it requires that “the winner build support across the nation, not in just a handful of large urban areas.” Hardaway similarly equates winning the Electoral College system with “winning the support of the people in States across the nation.”¹⁰⁹ In contrast, defenders of the Electoral College system contend, in a national popular vote a few big cities would control the election.¹¹⁰

To begin with, once a majority of the voters nationwide have made their choice known, it shouldn’t matter where in the country those who voted for or against particular candidates live. In a democracy, every person’s vote should count the same, whether they live in cities, small towns, or anywhere else. Yes, the aggregate vote totals in a big city will have a greater impact than the aggregate vote totals in a small town. And they should. More people live there.

At any rate, the Electoral College hardly requires the winner to “build support across the nation.” The winner need only build support in states that collectively comprise a majority of the Electoral College. Those states might well be concentrated in one or two specific regions, such as the west coast, the northeast, the south, or the midwest. As discussed below,¹¹¹ in today’s America very few states are competitive. Presidential elections are decided in a handful of swing states. Certain regions have become predictable. At this writing, the Northeast and the West Coast for the most part are solidly blue; the Deep South and the Central Midwest states are reliably red. Over time, that might well change. For the moment, though, whether our presidents are chosen by the Electoral College or a direct national vote, winning support “across the nation” is, sadly, a rarity.

Moreover, the idea that a few big cities will dominate the electorate isn’t even factually true. As Jesse Wegman points out, the fifty largest cities in the country together comprise only about 15 percent of the nation’s population – roughly the same population percentage as those who live in the rural areas. As he also notes, the aggregate vote split is about 60-40 for Democrats in those cities and about 60-40 for Republicans in the rural areas.¹¹² The feared urban domination of the national popular vote is as factually inaccurate as it is democratically irrelevant.

Related to the claim that urban voters would dominate the election is Maibach’s assertion, made by others as well, that in a direct popular election candidates would campaign only in densely populated urban areas. But where do they think the presidential candidates campaign now? Again, the vast majority of the states today are either solid red or solid blue. There is little incentive to spend precious campaign time and resources on those states. In practical terms, candidates today do almost all their campaigning in the few remaining battleground states. And if the worry is about rural voters being ignored in a national popular vote, does anyone imagine that in the present system the major party presidential nominees are spending their limited time giving speeches to small crowds in sparsely populated rural areas?

The actual numbers are jarring. During the 2020 general presidential election campaign, 96 percent of the campaign events (204 of 212) by the Democratic and Republican presidential and vice presidential nominees were confined to twelve states. One hundred percent of the events were in seventeen states. Neither of the two major party presidential nominees or their vice presidential running mates held a single campaign event in any of the other thirty-three states (or the District of Columbia). Two states – Pennsylvania and Florida – together received three-eighths of the nation’s campaign events for the major parties’ presidential and vice presidential nominees.¹¹³

That election was not an anomaly. Similar numbers describe the 2012 and 2016 general presidential election campaigns. In 2016, twelve states drew 94 percent of the events; six of those states drew two-thirds of them. And in 2012, 100 percent of the events were in only twelve states; four of those states received two-thirds of the nation’s total.¹¹⁴

Nor is it just a matter of in-person candidate appearances. As of October 13, 2020 (exactly three weeks before Election Day), almost 90 percent of all the money that the presidential campaigns had spent on TV advertisements nationwide had been channeled into six of the fifty states.¹¹⁵

Modern writers are not the first to object to a national popular vote on the ground that it would deprive some individuals of their disproportionate influence over the selection of the president. Alexander Keyssar highlights several early examples. Senator Robert Goodloe Harper, an influential US senator from Maryland, was a strong opponent of a constitutional amendment, introduced in 1816, that would have replaced the Electoral College with a national popular vote. He argued that the latter would violate the role of state sovereignty in the election of the president, thus obliterating the compromises that had been built into the Constitution. It would “destroy” the influence of the smaller states, he explained, because they would each lose the two electoral votes they receive just for being states. By way of example, he pointed out that Louisiana would lose disproportionately more than a large state like New York.¹¹⁶

Indeed it would. And it should. All Louisiana would be losing is the extra advantage the Electoral College gives it – an advantage that in a representative democracy it does not, and never did, deserve to have. In contrast, the national popular vote would give each citizen of Louisiana exactly the same influence that each citizen of New York and every other state has – namely, one vote. In the end, Harper’s arguments boiled down to nothing more than (i) state sovereignty is an end in itself; and (ii) we shouldn’t change the bargain that the framers had to strike to get nine ratifications.

Another Senator, William Wyatt Bibb of Georgia, shared Harper’s sovereignty concerns and added a new objection. The slave states, he lamented, would lose a constitutional advantage they then had. In tabulating the number of US House members the various states receive, and therefore their relative strength in the Electoral College, the slave states were allowed to count three-fifths of their slave populations even though they barred slaves from voting.¹¹⁷ Apart from both the obscenity of slavery itself and the compound injury of artificially granting the slave states both more representation in Congress and more Electoral College votes than their free populations would merit, the abolition of slavery mooted that specific argument.

Even still, none of the post-Civil War amendments have put African American voters on the same footing as white voters. As discussed below,¹¹⁸ the southern states in particular, and a broader range of Republican-controlled states today, have adopted increasingly sophisticated measures to depress the African American vote. Having done so, some of those states have nonetheless argued with a straight face that these very voting restrictions would unfairly disadvantage their residents in a national popular vote. Why? Because those voting restrictions would cause their citizens’ percentage of the national turnout to be lower than the state’s percentage of the national population, which in turn determines the size of their congressional delegation and therefore their percentage of the Electoral College vote. Now that is chutzpah! And it continues to play a prominent role in the Deep South’s opposition to a national popular vote.¹¹⁹

In the early twentieth century, the issue of women suffrage triggered a somewhat parallel objection. For a period of time leading up to the Nineteenth Amendment in 1919, some states allowed women to vote and others didn’t. That differential, the argument went, would give the states in which women could vote an unfair advantage in a national popular vote.¹²⁰ The obvious remedy – granting women the vote – apparently was not an acceptable option. At any rate, their argument was just as vulnerable as the earlier argument made by the slave states. Apart from the injustice of disenfranchising women in the first place, these states sought to have it both ways: They wanted to deny the vote to half the adults in their states, but still count disenfranchised women for purposes of maximizing the sizes of their states’ congressional delegations and, therefore, the amount of say they got in choosing the president.

Perhaps the most substantial defense of the Electoral College system is that, as Maibach argues, it almost always yields an outright majority. In fact, only twice in our history has the Electoral College failed to do so.¹²¹ That is because it is rare, at least today, for a third-party or independent candidate to win a plurality in a single state.¹²² The winner-take-all rule thus virtually ensures that all the electoral votes will go to the two major party candidates. In contrast, it is not at all unusual for independent or third-party candidates to prevent any one person from winning an outright majority of the national popular vote.

This is a fair concern. But I take exception to his fix. He says “The Electoral College creates a national majority for new presidents regardless of the popular vote margin. Reflecting the will of majorities in the fifty states, the College legitimizes the result”¹²³ [my emphasis].

It does no such thing, because the Electoral College is an artificial construct. The winner-take-all rule in forty-eight of the fifty states means that the majority Maibach extolls is a fiction, unless one believes that the president should be chosen not by the people collectively but by states – which, to add yet another layer of abstraction, are themselves creations aptly described by Hamilton as “artificial beings” and by Wilson as “imaginary.” Dressing up a mere plurality of the voters as a majority of politically constructed electoral votes doesn’t add any legitimacy to the process. Put another way, the Electoral College system does not salvage majority rule; it just masks its absence. Whatever the fiction, the bottom line remains: one can become president even when the majority of the voters choose other candidates.

That said, adoption of a national popular vote admittedly would require a decision as to what happens when no candidate wins a nationwide majority. One option, included in many of the constitutional amendments introduced in Congress in recent years,¹²⁴ would be simply to settle for presidents who win only pluralities of the national popular vote. In reality, we do this now, as just discussed. In fact, under the Electoral College system, we do this on the statewide level as well, because in every state except Nebraska and Maine, a plurality – not a majority – is all a candidate needs to win all the electoral votes of the state. In gubernatorial and other statewide elections as well, a plurality is typically all that is needed. Typically too, election to both houses of the US Congress and to both chambers of the various state legislatures requires only a plurality of the relevant popular vote.

But if the notion of a plurality president suddenly becomes unacceptable, a majority of the national popular vote could still be required. Options for accomplishing this include (i) a ranked-choice voting system; or (ii) a runoff election between the top two vote-getters when no majority emerges in the first round. The pros and cons of these options – not just for presidential elections, but in general – are discussed in Chapter 6.¹²⁵

A related defense of the Electoral College system is that it forces the voters to confine their focus to two candidates. Maibach’s argument here is that, as a practical matter, third-party or independent candidates have little to no chance of winning entire states. The winner-take-all feature of the Electoral College system therefore discourages them from running. In turn, any rule that discourages third-party and independent candidates from running increases the likelihood that some candidate will achieve an outright majority of the Electoral College. Maibach’s additional point here, though, is that discouraging third-party and independent candidates also helps avoid the often messy and unstable coalition governments that dominate many other western democracies.

But even assuming arguendo that limiting the voters to the choice between two political parties’ nominees is a good thing, the Electoral College system does not achieve that goal. People still run as third-party or independent candidates. Perhaps more would do so under a national popular vote system. Again, however, if it were decided that plurality presidents are no longer acceptable, either ranked-choice voting or a runoff election would solve the problem.

Hardaway also credits the Electoral College’s track record: It “has functioned far more successfully than was ever envisioned by the constitutional framers, and has, over the past 100 years, consistently produced clear-cut winners, all of whom received more popular votes than their opponents.”¹²⁶ Having published the book in 1994, Hardaway could not have anticipated that in two of the next six presidential elections, the Electoral College would hand the presidency to the candidate who lost the national popular vote. Still, his rosy assessment is surprising in the light of the many near misses that the country had already experienced.

Another claimed benefit of the Electoral College is what Hardaway calls the “immediate and decisive effect” of its “verdict.” As he puts it, “Amazingly, … advocates of direct election have chosen to find fault even with this undeniable feature of the Electoral College.”¹²⁷ Well, that feature, too, turns out not to be as “undeniable” as he thought. The ink was barely dry on his confident indictment of the critics when Al Gore in 2000 indisputably won the national popular vote, besting George W. Bush by more than half-a-million votes.¹²⁸ Had a national popular vote system been in effect, the outcome would have been clear and promptly known.¹²⁹ Instead, the Electoral College system triggered a weeks-long saga that ended only when the Supreme Court’s 5-4 decision halted the Florida recount and thereby decided the election.¹³⁰ The Court reasoned that a fair recount would not have been possible within the deadlines embedded in the Electoral College process.¹³¹ I am not suggesting either that a national popular vote will always yield clear and immediate results or that the Electoral College system never will. The point is that, in any given presidential election, either system could succeed or fail on that score. That is the reality in a nation as closely divided as ours.

During the early twentieth century, another objection to a national popular vote arose: If there were a national popular vote, the objectors said, elections would have to be managed by federal officials rather than state officials.¹³² Why that was a bad thing was never really explained, and Chapter 4, Section C.3 of this book considers the question of which level of government is best situated to manage federal elections. But even assuming arguendo that federal administration of presidential elections is an inherent evil, the argument is a non sequitur. Nothing about switching from the Electoral College to a national popular vote would prevent state officials from continuing to administer elections within the boundaries of their own states, just as they do now. For that matter, such a change doesn’t even have to alter the mechanics of the voting process. Of course, my view that state government should be abolished would make the question moot.

As with most defenses of the Electoral College system, Hardaway’s argument ultimately rests, more than anything else, on his deep-seated belief – reflected in the title of his book – in the value of American federalism and the Electoral College’s federalist roots. Maibach too defends the Electoral College as a necessary political compromise struck more than 200 years ago. It is, he says, just “part of American federalism.”

That observation is merely a description of its historical origins, not a normative justification on the merits – and certainly not a refutation of its antidemocratic effects. When the dust settles, it is hard for me to see any substance in the federalist defense of the Electoral College beyond “we’ve always done it this way” and “this and similar compromises accommodate the twin goals of popular democracy and state sovereignty” (my paraphrasing). Unless state sovereignty is seen – and somehow defended – as an end in itself, these arguments add no normative value.

Not surprisingly, public opinion surveys taken from the inception of scientific polling in the 1940s to the present day (with short-lived blips immediately following President Trump’s Electoral College win in 2016) have consistently demonstrated public discontent with the current system. Respondents in almost every poll have not only favored distributing electoral votes proportionally to the popular votes within each state (versus winner-take-all), but also abolishing the Electoral College entirely in favor of a national popular vote. And they have expressed both of those preferences by lopsided margins.¹³³

Despite the flimsiness of these defenses of the Electoral College, despite the longstanding and numerically overwhelming public preference for abolishing it, despite both the older and the more recent recurrences of its counter-majoritarian outcomes, despite the statistical certainty of many more to come, and despite the fact that “every other presidential democracy in the world did away with indirect elections during the twentieth century,”¹³⁴ the Electoral College has survived for more than 200 years. What is keeping it afloat?

Alexander Keyssar has written a (brilliant) 531-page book devoted to that one question. As he shows, two of the obstacles are fundamental and permanent. One of them, discussed in some detail in Section D, is that in general, the US Constitution is excruciatingly difficult to amend. The clash of diverse interests has impeded the degree of political consensus the amendment process requires. The other problem, already discussed, is states. Eliminating any state power is never easy, for states guard their authority jealously. Taking away the particular power, currently assigned to the state legislatures, to decide how to choose their presidential electors, is especially fraught.¹³⁵

The other main obstacles identified by Keyssar are more transient. They fluctuate with various states’ perceptions of whether abolition of the Electoral College would work to the short-term benefit or detriment of their specific partisan or other interests. As the preceding discussion suggested, at various times in our history states that believed rightly or wrongly that the Electoral College worked in their favor have included small states, slave states, and states that didn’t allow women to vote before the Nineteenth Amendment.¹³⁶ Today, states controlled by Republicans similarly tend to perceive that their partisan advantage is best served by keeping the Electoral College just as it is.

Given these difficulties, proponents of a national popular vote have turned to a clever alternative strategy. It is called the National Popular Vote Interstate Compact¹³⁷ (NPVIC). The idea is for individual states to enact legislation that awards all of their electoral votes to whichever presidential candidate wins a plurality of the national popular vote. Importantly, each such law will operate only in a year when the same law is in effect in states that collectively possess a majority of the electoral votes.¹³⁸ If enough such states pass this law, the winner of the national popular vote would gain a majority of the Electoral College and become president.

As of January 2024, the NPVIC has been enacted into law in sixteen states and the District of Columbia. Together they possess 205 electoral votes – 65 short of the 270 required for an Electoral College majority.¹³⁹

Getting to 270 will not be easy. As Wegman points out, “To date, all the states that have passed the compact did so under a Democratic-led [i.e., Democratic control of both houses] legislature and, with the exception of Hawaii, a Democratic governor.”¹⁴⁰ As a result of the 2022 midterm elections, Michigan now has a Democratic governor and Democratic majorities in both houses of the legislature. Still, even its additional sixteen electoral votes would leave the compact forty-nine electoral votes shy.

In addition, any interstate compact – especially one like this, which hinges on adoption and retention by states that together possess 270 electoral votes – is more fragile than a constitutional amendment that would enshrine a national popular vote permanently. Any state could withdraw from the compact, and, depending on the numbers, withdrawal by even a single state could take down the whole system. That risk will always be real, especially in battleground states, if a Republican Party trifecta suddenly displaces a Democratic Party trifecta.

Finally, there are constitutional concerns. First, there is always a danger that the Supreme Court will strike down the compact as an impermissible end run around the Constitution’s Electoral College provision. Second, the Constitution says “No State shall, without the Consent of Congress, … enter into any Agreement or Compact with another State.”¹⁴¹ Congressional approval will be highly unlikely unless and until the Democratic Party controls the White House and both Houses of Congress; even then, Senate approval will be a challenge unless the filibuster rule is repealed.

But as others have explained, the Supreme Court has rejected a literal reading of the congressional approval requirement. The National Center for Interstate Compacts, an entity established by the Council of State Governments, describes the Court’s case law as requiring congressional approval only when a compact “would increase state political power in a manner that would encroach upon federal authority.” It observes that only “[a]pproximately 40% of existing compacts required federal consent.”¹⁴² The Sightline Institute agrees, citing several Supreme Court decisions from 1893 to 1978. It goes on to point out that the Constitution requires each state to appoint electors “in such Manner as the Legislature thereof may direct.”¹⁴³ Thus, they conclude, the NPVIC does not encroach on any federal authority and therefore does not require congressional approval.¹⁴⁴ Others disagree,¹⁴⁵ and there is no certainty as to how the current Supreme Court would rule.

C House’s State Delegations Choosing the President

The same inequality of voting power is reflected in the Twelfth Amendment. If no presidential candidate gains a majority in the Electoral College, the House of Representatives chooses the president. Even if that were all there were to it, and even taking as a given both the Electoral College itself and the requirement of a majority of its electors, this wouldn’t be the best way to select the President when no candidate wins an outright Electoral College majority. As described in greater detail later,¹⁴⁶ the combination of intrastate residential patterns, single-member districts, and widespread gerrymandering makes the House anything but representative.

But it gets worse. When the election goes to the House, the president is chosen not by a majority of the House Members, but by a majority of state delegations. Those delegations that represent tens of millions of residents get no more say than those that represent only a small fraction of that number.

As Jesse Wegman has noted, several of the founders expected House selection of the president to become the norm. George Mason, in fact, predicted that nineteen out of every twenty presidential elections would be decided by the House.¹⁴⁷ Thankfully, as noted earlier, this process has been necessary only twice. In the 1800 presidential election, the Democratic-Republican Party ran Thomas Jefferson for president and Aaron Burr for vice president, against incumbent president John Adams and his vice presidential running mate, Charles Pinckney, both Federalists. The Jefferson–Burr ticket won just over 60 percent of the national popular vote. At the time, however, each elector cast two votes, with no distinction as to which vote was for president and which was for vice president. The result was that Jefferson and Burr ended up each receiving seventy-three electoral votes for president, to Adams’s sixty-five. Because no one candidate had a majority of the Electoral College, it fell to the House of Representatives, voting by state delegation, to choose the president. The contest came down to a battle between the two Democratic-Republican candidates, Jefferson and Burr. Jefferson was elected on the thirty-sixth ballot.¹⁴⁸ Amidst the political fireworks, at least on that occasion the outcome reasonably reflected the wishes of the majority of the voters.¹⁴⁹

Not so in 1824. That election featured four candidates, all members of the same Democratic-Republican Party. Andrew Jackson won 41 percent of the national popular vote. John Quincy Adams finished second with 31 percent. Henry Clay and William Crawford garnered 13% and 11%, respectively.¹⁵⁰

Jackson also won a plurality of the Electoral College with ninety-nine electoral votes to eighty-four for Adams.¹⁵¹ But when the election got to the House, Clay threw his support to Adams, whom the House, voting by individual state delegation as constitutionally required, ultimately selected. Adams ended up winning thirteen of the twenty-four state delegations; Jackson won seven and Crawford four.¹⁵²

To be fair, the final outcome in 1824 would have been no different had the House voted by membership rather than by state delegation. The results would have been closer, but either way, Adams would have won the election in the House, albeit just narrowly. He received the votes of 109 House members (51%) to Jackson’s 104 (49%).¹⁵³ But the basic takeaways remain the same: First, Jackson won a plurality of the national popular vote by a healthy margin over Adams; yet Adams won the presidency. And second, even if selection by the House were otherwise a fair way to resolve the lack of an Electoral College majority, the tightness of the House membership vote illustrates how close we came to the presidency, already having been tarnished once by the counter-majoritarian Electoral College, being tarnished again by the additional counter-majoritarian principle of one vote per House state delegation. And, of course, as public dissatisfaction with both of today’s major political parties grows, one cannot dismiss the possibility of a future third-party candidate winning a plurality of the popular votes in one or more states, thereby preventing an Electoral College majority.

D The Judicial Appointment Process

The appointment of a federal judge is a two-step process – nomination by the president and confirmation by the Senate.¹⁵⁴ To be clear, my counter-majoritarian objection to this process is not that federal judges are not elected by the people. As explained in Chapter 6, I wouldn’t want them to be.

Rather, as this section will illustrate, both steps in the process – nomination and confirmation – have been hijacked by counter-majoritarian forces that would not exist without the constitutionally assigned roles of the states. The first step becomes counter-majoritarian when the judge is nominated by a president whom the American people rejected but whom the state-centered Electoral College installed in the White House. The second step becomes counter-majoritarian when the Senate is under the control of a political party that the voters repudiated nationwide and that Senate, in turn, either votes to confirm the nominee of a popularly rejected president or, conversely, votes to block or delay the confirmation of a candidate nominated by a popularly elected president.

The impact of counter-majoritarian judicial partisanship has been the most visible at the Supreme Court level. At this writing (2024), conservative Justices appointed by Republican presidents hold six of the nine Supreme Court seats. Three of those Justices – Gorsuch, Kavanaugh, and Barrett – were appointed by President Trump,¹⁵⁵ who as noted earlier had lost the national popular vote to Hillary Clinton by a margin of almost three million. Had Clinton become president (and had the Republican Senate functioned in a majoritarian manner), it is a safe bet that it is Democratic appointees who would be holding a 6-3 majority.

That example illustrates the counter-majoritarian effect of the nomination component alone. The same three appointments reveal the additional counter-majoritarianism of the Senate confirmation process. President Obama, who had won both election and reelection by winning outright majorities of both the national popular vote and the Electoral College,¹⁵⁶ nominated Judge Merrick Garland for the Supreme Court. The nomination was made in March 2016, with approximately ten months remaining in his term. The Republicans, then in control of the Senate despite still collectively representing only a minority of the US population,¹⁵⁷ refused even to hold a hearing on Garland’s nomination.¹⁵⁸ Their stated reason was that the appointment was too close to the upcoming November election. It was necessary to see how the people would vote, they argued. Yet, when President Trump later nominated Amy Coney Barrett to the Court, the Republican-controlled Senate (with the Republicans still collectively representing only a minority of the US population) had no trouble confirming her only one week before Election Day and after tens of millions of Americans had already cast early ballots. The candidate who should have received a hearing did not; the candidate whose nomination truly was so close to the election that the Senate should have waited was rushed through.

Then-Senate majority leader Mitch McConnell’s defense was that “Americans re-elected our [Senate Republican] majority in 2016 and expanded it in 2018.”¹⁵⁹ As Pamela Karlan rightly points out, “a majority of Americans did no such thing.”¹⁶⁰ Per her cited sources, the Senate’s Republican majority that confirmed Justice Barrett represented only a minority of the US population.¹⁶¹ Population aside, as detailed earlier,¹⁶² Americans voted overwhelmingly in favor of Democratic senatorial candidates in both of the years that McConnell cited.

Justice Kavanaugh, meanwhile, after being nominated by a president who had lost the national popular vote, was confirmed by a Senate whose Republican majority not only collectively represented just a minority of the US population,¹⁶³ but had soundly lost the national senatorial popular vote in the then-most recent election, 53% to 42%, only to be rewarded with twenty-two of the thirty-four Senate seats up for election that year.¹⁶⁴ Had the Democrats come away from that election with even an equal split of the Senate seats that had been voted on – let alone a share proportionate to the votes of the people – they would have controlled the Senate by a comfortable margin.¹⁶⁵

It is not only the Trump nominees who owe their Supreme Court appointments to counter-majoritarianism. As Jack Balkin has observed, a majority of the current Supreme Court Justices – Justices Thomas, Alito, Gorsuch, Kavanaugh, and Barrett – were all confirmed by senators who collectively represented only a minority of the then-existing US populations.¹⁶⁶ Moreover, for them as well, it wasn’t just population. Democratic presidential nominees have won the national popular vote in a majority of the last fourteen elections (8-6). Yet, during the presidential terms resulting from those elections (1968–2024),¹⁶⁷ Republican Supreme Court appointments have outnumbered those by Democratic presidents 15 to 5.¹⁶⁸

That’s just the Supreme Court. During that same period, again despite losing the presidential popular vote in a majority of the elections, Republican presidents were able to appoint approximately 1,482 lower court judges; Democratic presidents managed only approximately 1,082.¹⁶⁹ That disparity can be traced to the hundreds of appointees by national popular vote losers President George W. Bush (in his first term) and President Trump.¹⁷⁰ In addition, however, the Republican-controlled Senate (whose members represented only a minority of the country’s people, in case I haven’t reminded you of this often enough), blocked and delayed the nominations of popular vote winner President Obama (during the last six years of his eight-year tenure) and then expedited the nominations of popular vote loser President Trump. As a result, President Obama was limited to an average of only forty-two federal judges per year¹⁷¹ – the lowest annual average of any President during this period (except for President Ford, a caretaker President who served the final two years of President Nixon’s second term after Nixon’s resignation). In contrast, the Senate’s actions enabled President Trump to average sixty-one appointments per year,¹⁷² by far the highest annual average of any president during this period (except for the special circumstance of President Carter, who benefitted from Congress’s huge expansion of the federal judiciary).¹⁷³

If both the composition of the US Senate and its actual practice (discussed below) make the judicial confirmation process counter-majoritarian, the Senate “blue slip” tradition adds insult to injury. A “blue slip” is the paper that any member of the Senate may submit to the chair of the Judiciary Committee to support a judicial nominee who would “represent” that Senator’s home state – meaning the nomination is for a district judgeship located in that state or for a court of appeals seat for which the particular state is regarded as deserving a “turn.” Depending on the policy of the Judiciary Committee chair, withholding a blue slip either automatically kills the nomination without so much as a committee hearing or implies that the nomination should be blocked as a matter of “Senate courtesy.” No reason for withholding a blue slip need be given.¹⁷⁴

As of this writing (2024), Senator Dick Durbin (D-Ill) chairs the Judiciary Committee. He has said he will follow the precedents established by his two immediate predecessors, Senators Charles Grassley (R-Iowa) and Lindsay Graham (R-So. Car.). Under that policy, a positive blue slip from the home state senator is required for district court nominees but not for court of appeals nominees.

The blue slip process is rife with partisan counter-majoritarian abuse. First, empowering a single senator to stop the appointment of a federal judge is inherently counter-majoritarian. Even lower federal court decisions frequently have interstate and even nationwide impact. As noted below, recent years have witnessed a spate of federal district courts entering nationwide injunctions that prohibit the federal government from carrying out its announced policies. And the precedential decisions of the US courts of appeals (except for the DC Circuit and the specialized “Federal Circuit”) always have at least multistate, and often nationwide, effects. As a matter of principle, the notion that a single state – much less, just one of the state’s two senators – deserves a veto power over the President’s selection of a federal judge should be a nonstarter for that reason alone.

Actual experience, especially in recent years, attests to the partisan abuse that the blue slip process invites. Senator Pat Toomey (R-Pa.) withheld his blue slip for President Biden’s nomination of Arianna Freeman for a seat on the US Court of Appeals for the Third Circuit because her area of practice was a “niche” area.¹⁷⁵ She was a public defender. I could find no record of his having withheld blue slips for nominees who were criminal prosecutors. When President Biden nominated Andre Mathis to a seat on the US Court of Appeals for the Sixth Circuit, Senator Marsha Blackburn (R-Tenn.) withheld her blue slip, describing as a “rap sheet” what the President of the NAACP described as “less than a handful of speeding tickets.”¹⁷⁶

Does this really matter much? Oh my, yes, as the following discussion will show. By shaping the composition of the federal courts, these powerful counter-majoritarian structures have fundamentally transformed US law. The impact has been especially pronounced in the election law cases themselves. Here is just a sample of the affected outcomes:¹⁷⁷

In 1986, a six-Justice majority of the Supreme Court in Davis v. Bandemer¹⁷⁸ had held that federal courts could review claims of partisan gerrymandering of state legislative districts. In 2004 the Court returned to the issue in Vieth v. Jubelirer,¹⁷⁹ this time in the context of alleged partisan gerrymandering of congressional districts. By then the composition of the Court had become decidedly more conservative. Four Justices concluded that such claims were nonjusticiable and wanted to overrule Bandemer; the other five disagreed (one of them holding out the possibility that the Court might one day declare them nonjusticiable if a manageable standard could not be agreed on). But by 2019 new appointments had moved the Court even further to the right. In its decision that year in Rucho v. Common Cause,¹⁸⁰ the Court overruled Bandemer, holding that partisan gerrymandering claims (in that case, relating to congressional districts) were nonjusticiable. Even in the face of what the Court admitted were “blatant examples of partisanship driving districting decisions,”¹⁸¹ the Court held that partisan gerrymandering claims present political questions beyond the reach of the federal courts.¹⁸² Essential to that 5-4 decision were Justices Gorsuch and Kavanaugh. Both had been nominated by President Trump, who had lost the national popular vote, and then confirmed by Senators who together had represented only a minority of the national population.

In Abbott v. Perez,¹⁸³ the issue was whether Texas maps updating both congressional and state legislative districts had been drawn with a racially (not just partisan) discriminatory intent, in violation of Section 2 of the Voting Rights Act. Again the Justices divided along party lines. The five Republican appointees held that the challengers had failed to prove a racially discriminatory intent; the four Democratic appointees felt otherwise and dissented. Justice Gorsuch, filling a spot that as noted earlier should have been filled by President Obama (and, failing that, would have been filled by Hillary Clinton had the national popular vote been honored), voted with the majority. As in Rucho, his vote made the difference.

In Husted v. A. Philip Randolph Institute,¹⁸⁴ the Supreme Court upheld an Ohio law that required the purging of eligible voters’ names from the registration lists for having failed to vote in recent prior elections. This was another 5-4 decision in which the Justices divided along partisan lines. And, again, the vote of Justice Gorsuch was essential to the outcome.

In Brnovich v. Democratic National Committee,¹⁸⁵ the Supreme Court addressed two of Arizona’s voting restriction laws. One of those laws required election officials to throw out any votes cast on election day unless they were cast in the precinct in which the voter lived. The other challenged law made it a criminal offense for anyone other than the voter or his or her family member, household member, or caregiver, or a postal worker or elections official, to collect or return the person’s ballot – practices that no one denied were most prevalent in African American communities. The question was whether either of these laws violated the race discrimination provisions of the Voting Rights Act.

Splitting 6-3, once again along strictly partisan lines, the Court upheld both Arizona laws. The three Trump appointees – Justices Gorsuch, Kavanaugh, and Barrett – all voted with the majority. As noted earlier, the Gorsuch spot should have been President Obama’s to fill. And if majoritarian principles had been followed, the Kavanaugh and Barrett spots would have been filled by national popular vote winner Hillary Clinton – or, in the case of Justice Barrett, arguably held over for President Biden. As in the preceding cases, counter-majoritarian judicial appointments made the difference.

Those are just the election law cases. They barely scratch the surface, as there are countless other subject areas in which counter-majoritarian Supreme Court appointees have made the difference in cases of huge national import. In 2022 alone, the votes of one or more of the three Trump appointees were frequently outcome-determinative. The blockbuster, of course, was Dobbs v. Jackson Women’s Health Organization,¹⁸⁶ where a 5-4 majority that included all three Trump appointees overruled the Court’s landmark decision in Roe v. Wade¹⁸⁷ and its constitutional protection of abortion rights.

There were so many more. In that same year, the Court declared a Second Amendment right to carry handguns outside the home for self-defense; banned the EPA from regulating the carbon emissions of existing power plants; prohibited the federal government from mandating that large employers require their employees to either vaccinate or self-test for COVID; barred states from selectively denying public grants to religious schools; and allowed the football coach of a public high school to lead prayer sessions on the fifty-yard line of the school’s field.¹⁸⁸ In 2023, in Sackett v. EPA,¹⁸⁹ the Court voted 5-4 to strip the EPA of its power to regulate huge areas of wetlands and other waterways, jeopardizing important sources of clean water. And on the last two days of June 2023, the Supreme Court, by identical 6-3 votes, struck down colleges’ race-based affirmative action programs¹⁹⁰ and the Biden Administration’s student loan forgiveness program.¹⁹¹ In every one of those cases, Justices appointed by President Trump, whom voters nationwide had rejected, made the difference on matters of huge public importance.

Nor is it just the Supreme Court. In recent years, the ideological and partisan preferences of federal lower court judges have assumed increased importance. That, of course, means that the selections of both the Presidents who nominated them and the Senates that confirmed them – and therefore the methods by which those presidents and senators are elected – have taken on greater significance as well.

One empirical study, analyzing 650,000 US Court of Appeals cases, found that judges’ political affiliations help predict the outcomes in categories of cases that represent 90 percent of all court of appeals decisions. In particular, this study demonstrated, judges appointed by Democratic presidents are significantly more likely than those appointed by Republican presidents to rule in favor of the less powerful party.¹⁹²

Two particular features of the US legal system, acting in concert, are especially notable at the district court level. One feature is that, since actions of the federal government ordinarily apply nationwide, lawsuits challenging the legality of those actions may usually be filed in any federal district court in the country. That fact gives the challenging parties considerable leeway to choose the forum they believe will be most favorable to their positions. Often they can find a judicial district in which there is only one active judge, whose views are known or easily discerned to favor the challengers’ positions. Second, the judges they select have shown not only a marked tendency to issue injunctions blocking the government’s actions – not surprising, since that’s why the challengers chose those judges in the first place – but also a sharply increased willingness to extend those injunctions nationwide rather than confine them to the territory covered by the particular district court or the state in which the challengers reside.

These realities have armed state politicians with a powerful weapon that they have begun to use to shut down federal executive actions to which they object on either policy or purely political grounds. The resulting power that these developments offer to each individual state would be of concern in almost any political era. But the present era is one in which three trends have combined to form a now-familiar sequence. First, extreme polarization and the US model of divided government virtually paralyze Congress. Second, the executive branch steps in to fill the void, addressing urgent national issues through major policy initiatives of its own. And third, the willingness and the ease with which state officials can now use their favored judges to frustrate the Administration’s policy decisions does more than create a vacuum in which pressing problems go unaddressed; they also enable officials who at best reflect the view of their local constituents to thwart the policy decisions of nationally elected presidents and their chosen appointees.

So investing a single state with these powers has unusually far-reaching effects. Whichever political party controls the executive branch of the US government, there will always be at least one state with the incentive – borne of either sincere policy concerns or crass political advantage – to use its newfound power to cancel controversial federal policies.

There have always been examples of these judicial interventions, but state lawsuits against the federal government began popping up with dizzying regularity during the Obama years. Texas has been an especially zealous plaintiff, having frequently teamed up with other Republican-controlled states to challenge actions of the Obama Administration and more recently the Biden Administration. States controlled by Democrats returned the favor during the Trump Administration.

Many of the lawsuits took aim at a series of President Obama’s executive decisions on immigration. Those policy decisions had offered a form of temporary relief, called “deferred action,” to undocumented immigrants who met certain specific criteria and were found to merit the favorable exercise of prosecutorial discretion. One such policy, relating to certain individuals who had been brought to the United States as children and had lived here continuously ever since arrival and ever since specified past dates, was “Deferred Action for Childhood Arrivals,” better known by its acronym, DACA.¹⁹³ The other, relating to certain parents of US citizens and certain parents of lawful permanent residents, was “Deferred Action for Parents of Americans” (DAPA).¹⁹⁴

Shortly after the 2014 announcement of DAPA, a group of twenty-six states led by Texas and represented by Republican governors or attorneys general brought a lawsuit seeking to block it. Although one might have expected the Texas attorneys to file the lawsuit in the state capital of Austin, where the state’s legal operations were based, they elected instead to file it in the federal district court some 351 miles away in Brownsville, Texas. At the time, the only active judge in that courthouse was Andrew Hanen, who was already on record as a staunch critic of President Obama’s immigration policies, particularly on matters of prosecutorial discretion. As other legal scholars have noted, this judge in several previous cases had excoriated President Obama’s immigration policies in vitriolic, emotional, ad hominem language rarely seen in judicial opinions.¹⁹⁵

The judge did not disappoint. In an opinion that this writer would respectfully describe as long in words but embarrassingly weak in legal reasoning, the judge issued a preliminary injunction prohibiting the Administration from implementing the challenged policies anywhere in the country.¹⁹⁶ Relevant here, Judge Hanen was a first-term appointee of President George W. Bush,¹⁹⁷ who as noted earlier had lost the national popular vote.

Although the Texas litigation and the history of the particular judge furnish an admittedly extreme example, there were many similar episodes during the same period. Several of them involved related immigration issues,¹⁹⁸ but President Obama’s executive actions in other subject areas also fell prey to lawsuits brought by Texas and other adversary states.

In a case similarly named Texas v. United States,¹⁹⁹ a group of thirteen states, again led by Texas, went to court to enjoin an Obama Administration interpretation of the federal civil rights laws. The federal interpretation required schools to grant transgendered individuals equal access to restrooms and similar facilities based on their gender identity rather than on the gender assigned at birth. Once again Texas’s attorneys found an exceptionally conservative judge, who promptly issued a preliminary injunction nullifying the federal government’s interpretation. And once again, the judge elected to extend the injunction nationwide. In another case, also decided by a federal district judge in Texas,²⁰⁰ twenty-one states sued to prevent the Obama Administration from implementing a Labor Department regulation that would have expanded the number of employees who qualify for overtime pay. The judge granted the injunction and applied it nationwide. In none of these cases was there any convincing explanation of how a narrower injunction limited to the plaintiff states would have failed to serve their interests. Nor could these judges explain how the broader injunction could avoid burdening the federal government or the other twenty-nine states (who opposed it) more than was necessary to protect the interests of the plaintiff states.

In 2017 Donald Trump became president. He immediately rescinded DACA and issued a long series of other executive actions on immigration and other subjects. Suddenly the tables were turned. Now it was Democratic-controlled states (and private plaintiffs) doing their own forum-shopping and obtaining nationwide injunctions in the vast majority of the cases.²⁰¹

Upon the election of President Biden, the tables turned yet again. As a partial response to the limited capacity of US detention facilities, the Biden Administration adopted a policy of using supervisory alternatives to the detention of some asylum seekers. In Florida v. United States,²⁰² the State of Florida sued to shut down that policy. Florida might have been expected to file the lawsuit in the state capital of Tallahassee, where its legal operations were based. Instead, their lawyers chose to file it 196 miles away in Pensacola, where all four active judges had been appointed by Republican presidents.²⁰³ They landed a judge appointed by President Trump,²⁰⁴ and the judge accommodated the state’s request to terminate the Biden Administration’s policy nationwide.

Trump-appointed District Judge Drew Tipton is the only active federal judge in Victoria, Texas. He had already blocked two of President Biden’s immigration policies. In one of those cases, he had ruled not only that states had standing to challenge the federal government’s immigration enforcement priorities, but also that those priorities were illegal.²⁰⁵ That ruling was in such clear violation of established Supreme Court precedent that even the current Supreme Court had to reject it by a vote of 8-1 (Justice Alito being the lone dissenter).²⁰⁶ So Texas filed a lawsuit in his court challenging yet another Biden Administration immigration policy. That policy had allowed limited numbers of nationals of a few specified countries that are in turmoil to apply for a statutory remedy called “parole,” which authorizes the Secretary of Homeland Security to permit temporary entry into the United States. Calling out Texas’s forum-shopping, and observing that the case had no connection to Victoria, the Administration asked Judge Tipton to transfer the case to either the state capital or DC. He refused. The Administration had previously filed similar requests, also in vain, to other Trump-appointed judges in small-city courthouses with only one active judge and no apparent connection to their respective cases.²⁰⁷

As the only active judge in the Lubbock Division of the Northern District of Texas, Trump-appointed District Judge James Wesley Hendrix²⁰⁸ is assigned two-thirds of the civil cases filed in that court.²⁰⁹ He too has a history of blocking President Biden’s policies. Not surprisingly, therefore, Lubbock is where the Texas state lawyers elected to file their lawsuit challenging Congress’s Biden-supported 1.7 trillion dollar government spending plan for fiscal year 2023.²¹⁰ In 2024, he held that Congress’s enactment of that law had been unconstitutional because members had been allowed to vote by proxy under COVID-era rules. In particular, he struck down the challenged provision that had strengthened workplace protections for pregnant women.²¹¹

Another Trump appointee is Chief Judge Terry A. Doughty of the Western District of Louisiana. (Unlike the others discussed here, Judge Doughty is not the only available judge in his division.) At the height of the COVID pandemic, President Biden had required COVID vaccinations for workers in the federal Head Start program. Judge Doughty issued a twenty-four-state injunction blocking President Biden’s order. His opinion has been sharply criticized for its series of false statements about vaccinations – including a bizarre declaration that these vaccines are useless because boosters would eventually be required.²¹² In a subsequent case, Judge Doughty entered an injunction prohibiting the Biden Administration from asking (not requiring, since the federal government doesn’t have that power) social media companies to remove misinformation endangering public health.²¹³ The same judge has also enjoined the Biden Administration’s ban on new leases of federal land for oil and gas drilling.²¹⁴

Yet another Trump appointee, District Judge Matthew Kacsmaryk in Amarillo, Texas, is well-known as “a favorite judge for litigants opposing Biden administration policies.”²¹⁵ He has been easy to access, especially since September 2022, when the District Court for the Northern District of Texas decided that “any civil case filed in Amarillo would be assigned to Judge Kacsmaryk.”²¹⁶ As journalist Kate Riga notes, “Of the couple dozen lawsuits that Texas Attorney General Ken Paxton (R) has filed against the Biden administration, over a third have been funneled through the relatively small city [Amarillo], despite its distance [485 miles] from the state capital.”²¹⁷

In one case, Judge Kacsmaryk issued an injunction prohibiting President Biden from ending President Trump’s “Remain in Mexico” policy. Under that policy, people applying for asylum at the southern border were required to wait in Mexico, under extremely dangerous conditions, during the months or years before hearings on their claims could be scheduled. Among other things, continuation of the policy would require the US government to enter into negotiations with the government of Mexico – and with all the leverage on Mexico’s side. Even for the current Supreme Court, this was a bit much. Holding both that Judge Kacsmaryk had badly misread the relevant law and that at any rate a judge had no business effectively ordering the federal government to enter into negotiations with a foreign country, the Court sent the case back to Judge Kacsmaryk for a redo.²¹⁸ He promptly found another way to at least temporarily block the Biden Administration from rescinding Trump’s Remain in Mexico policy.

Judge Kacsmaryk is best known for legal interpretations that appear to be driven by his publicly expressed, deeply held, personal religious beliefs. In one such case, despite a Supreme Court decision that had clearly held to the contrary, he interpreted the Affordable Care Act’s prohibition on sex discrimination as not covering discrimination based on sexual orientation or gender identity. To distinguish a binding Supreme Court decision, he had to reason that “discrimination because of sex” is somehow different from “discrimination based on sex.”²¹⁹ And in Deanda v. Becerra,²²⁰ Judge Kacsmaryk was faced with a federal law that “encouraged” family participation in minors’ family planning decisions but pointedly did not require parental consent. He nonetheless interpreted that law as allowing the State of Texas to require parental consent for minors who wish to obtain contraception.

But Judge Kaczmaryk’s best-known ruling came in 2023, when he struck down the FDA’s approval of mifepristone, an abortion medication that had been in use for twenty-four years. The judge rejected the longstanding findings of both the FDA and the health care profession that the drug is safe. Coming less than a year after the Supreme Court’s overruling of Roe v. Wade, the litigation assumed outsized importance, because medication abortion then accounted for roughly one-half of all abortions in the country. Even the Republican-controlled US Court of Appeals for the Fifth Circuit reversed Kaczmaryk’s ruling in part, but it let stand his injunction on obtaining mifepristone by mail.²²¹

Judge Kacsmaryk and the other judges described here are just part of the bonanza for Texas Republicans. As journalist Kate Riga observes, “Texas works out particularly well for judge shoppers – they can get a case into Kacsmaryk’s hands in Amarillo, or maybe into Reed O’Connor’s in Wichita Falls or Drew Tipton’s in Victoria – resting easy in the knowledge that the state is controlled by the ultraconservative Fifth Circuit Court of Appeals. That leaves as liberals’ greatest hope for intervention … the Supreme Court.”²²²

Riga was certainly right to describe the Fifth Circuit as “ultra-conservative.” More to the point here, it is also one of the most grotesquely counter-majoritarian courts in the country. At the risk of overkill, I will note that, as of December 16, 2022, twelve of the sixteen active Fifth Circuit judges (there was one vacancy) were appointed by Republican presidents, including six by national popular vote loser President Trump. All of those judges were appointed between 1985 and 2022,²²³ a period in which the Democratic presidential nominees had won the national popular vote in seven of the ten elections. In fact, fourteen of the sixteen judges were appointed between 1994 and 2022, during which period the Democratic presidential nominees had won the national popular vote in seven out of eight elections.²²⁴ So when rogue federal district judges based anywhere in the states that the Fifth Circuit covers – Texas, Louisiana, and Mississippi – issue nationwide injunctions against policies of Democratic presidential administration, appellate remedies typically prove illusory.

Some will applaud this state of affairs; others will bemoan it. But whatever one’s normative views on the specific issues presented in those cases,²²⁵ there should be deep concern. Individual states whose political predilections do not align with those of the national government have now amassed an unhealthy power to impose their political views on the entire country. That power has come at the expense of both individual states with contrary policy preferences and the nation as a whole.

My objections to the constitutional process for appointing federal judges, then, are twofold: First, because of the powers that the Constitution confers on the states in the first place – specifically concerning the Electoral College and the equal suffrage of states in the Senate – both the nomination and the confirmation of federal judges are by inherently counter-majoritarian actors. Second, I worry about the combination of (i) individual states’ abilities to file their lawsuits with whichever judges they see as their ideological and political soulmates and (ii) politically radical judges who are all too happy to accommodate those states by shutting down actions of the federal government nationwide. That combination has been a gift to states, as well as to judges who either were nominated by a president whom the people had rejected or were confirmed by a nationally unrepresentative Senate (or both). Those state officials, and selected judges, now possess the frightening ability to impose their own policy preferences on the entire nation in place of the policy judgments of nationally elected presidents.

While court-shopping is not new, it has become more lethal in recent years in at least one way. Most federal district courts have multiple judges. This means would-be forum shoppers, at least in the past, could not guarantee getting the judge they wanted; they had to settle for playing the odds. But as Riga points out, today “the surgical specificity of targeting divisions of district courts overseen by one or two judges is newer.”²²⁶

None of this should be surprising, except perhaps as to the increased brazenness of some of the legal contortions these judges have had to perform. At his Senate confirmation hearing, now-Chief Justice Roberts famously disclaimed the influence of ideology on the decisions of judges; all he does, he testified, is “call balls and strikes.”²²⁷ Then-nominee Neil Gorsuch added a slightly different twist, looking the Senators in the eye as he told them that “[t]here’s no such thing as a Republican judge or a Democratic judge.”²²⁸

I don’t blame them for saying those things. No one who is nominated for a federal judgeship and who wants the job would dare acknowledge candidly, before the US Senate, that they intend to decide cases based on their personal policy preferences, their religious views, or their perceptions of the best electoral interests of the Republican Party. But both of these men had to know that what they were saying was not true. For decades, empirical studies have consistently exposed the extremely high positive correlations between judges’ decisions and (i) their personal ideologies and (ii) in the case of federal judges, the political parties of the presidents who nominated them.²²⁹

If there has been a shift – and I believe there has – it has been in the subtle transformation from ideologically driven judicial decisions to those driven by naked partisanship. The Supreme Court’s decision in Bush v. Gore, discussed earlier for the light it sheds on other issues, seems a likely foundation for this evolution as well.

In that case, the American people in 2000 chose Democratic nominee Al Gore over Republican nominee George W. Bush. But in the Electoral College, the outcome hinged on the electoral votes of one state – Florida. The voting machine tabulations in that state showed Bush holding a razor-thin lead. But there were widespread mechanical problems with the paper ballots in several counties with Democratic-leaning populations. Chief among the problems were the famous “hanging chads” and “dimples.” On those ballots, the voter had punched an indentation or hole for a particular candidate, but not all the way through, with the result that the machines did not record their votes. Florida law required every ballot to be counted as long as the intent of the voter could be ascertained. So the Florida Supreme Court, interpreting the Florida election laws, ordered a manual recount. But different counties employed different standards in judging the voters’ intentions. On that basis, a US Supreme Court majority of five Republican Justices rejected the Florida Supreme Court’s interpretation of Florida’s election law, found a likely denial of equal protection, and therefore temporarily halted the ongoing recount.

When the Court returned to the case for a final decision, the same majority decided that equal protection did indeed require the same evidentiary standard in every county. As the four dissenters pointed out, even then the Court could have given the Florida election officials a chance to restart the recount using the required uniform standard. By then, however, it was getting close to what the majority regarded as a firm deadline for Florida to submit its slate of electors to Congress. The majority thus decided on its own that the Florida election officials would not be able to get the job done in time. So the recount was permanently enjoined. With Bush holding a lead of just 537 votes out of more than 100 million cast, he was declared the winner of Florida’s electoral votes and therefore the presidency.²³⁰

The majority had to dodge several bullets to reach that result. These federal Justices – all five of them states’ rights conservatives – had to overturn a state court’s interpretation of the state’s own law. They also had to hold that equal protection required all the counties in the state to employ the same criteria when assessing the intent of the voter in cases where the machines do not record the votes. And once it did that, the Court weeks later had to deny the Florida election officials any opportunity to try to complete the recount in the required manner before a deadline that the dissenters pointed out was not inflexible anyway.

For me, the most interesting part of the obstacle course that the majority was willing to run involved the equal protection issue. Five conservative Republican Justices who until this case had barely noticed the equal protection clause suddenly not only discovered it, but, having done so, placed on it an interpretation far more expansive than what any of their more progressive colleagues could ever have dreamed of. And that epiphany just happened to occur in a case where their robust interpretation of the equal protection clause enabled Republican nominee George W. Bush to win the presidency over his Democratic rival.

Without entirely dismissing the possibility of mere coincidence, one cannot help but be struck by the typical lineups in the election law cases in particular – both those just discussed in this section and the many others that will be discussed in Chapter 3. Time and again, it is Democratic-appointed judges who vote to rein in both gerrymandering efforts and voter suppression strategies. It is Republican-appointed judges who strain to uphold them.

The New York Times editorial board did not mince words: “Over the past several years, the court has been transformed into a judicial arm of the Republican Party. … In cases involving money in politics, partisan gerrymandering, and multiple suits challenging the Voting Rights Act, the court has ruled in ways that make it easier for Republicans and harder for Democrats to win elections.”²³¹

The public has also noticed. In the same editorial, the Times said this:

In a Gallup poll taken in June, [even] before [my emphasis] the court overturned Roe v. Wade with Dobbs v. Jackson Women’s Health Organization, only 25 percent of respondents said they had a high degree of confidence in the institution. That number is down from 50 percent in 2001 – just months after the court’s [already] hugely controversial 5-4 ruling in Bush v. Gore, in which a majority consisting only of Republican appointees effectively decided the result of the 2000 election in favor of the Republicans.²³²

The Presidential Commission on the Supreme Court of the United States, in a passage relating the arguments in favor of Supreme Court expansion, said:

[C]ritics maintain that the Supreme Court has been complicit in and partially responsible for the “degradation of American democracy” writ large. On this view, the Court has whittled away the Voting Rights Act and other cornerstones of democracy, and affirmed state laws and practices that restrict voting and disenfranchise certain constituencies, such as people of color, the poor, and the young. This has contributed to circumstances that threaten to give outsized power over the future of the presidency and therefore the Court to entrench that power.²³³

In addition, the report might have added, the nation suffers when the vast majority of its citizens lose faith in the basic institutions of government.

All this is disturbing enough. But as this section has sought to illustrate, so many of the most egregious abuses of judicial power have been by judges who would not even have been on the bench but for a federal judicial appointment process riddled with systematic counter-majoritarian biases. At multiple levels, those biases, in turn, would not exist but for the constitutionally entrenched roles of the states.

E The Constitutional Amendment Process

The process for amending the US Constitution is yet another example of elevating the equality of the states over the equality of the citizens. And this one takes counter-majoritarianism to an extreme. Article V of the Constitution reads:

The Congress, whenever two thirds of both Houses shall deem it necessary, shall propose Amendments to this Constitution, or, on the Application of the Legislatures of two thirds of the several States, shall call a Convention for proposing Amendments, which, in either Case, shall be valid to all Intents and Purposes, as Part of this Constitution, when ratified by the Legislatures of three fourths of the several States, or by Conventions in three fourths thereof, as the one or the other Mode of Ratification may be proposed by the Congress …

As the text indicates, there are two ways to begin the process of amending the Constitution. Two-thirds of the House membership, already distorted by the residential patterns, single-member districts, and gerrymandering and voter suppression practices discussed below,²³⁴ together with two-thirds of the Senate, in which the citizens of high-population states are grossly underrepresented, may propose a constitutional amendment. Alternatively, two-thirds of the state legislatures – similarly both individually unrepresentative of their respective citizens’ voting preferences²³⁵ and representing dramatically different state populations to start with – may file an “application” for a constitutional convention that could then propose amendments. Thus, even the origination of a constitutional amendment is layered with counter-majoritarian hurdles.

Once that first set of hurdles has been cleared, additional supermajority requirements govern ratification. Three-fourths of the state legislatures – not states that collectively represent three-fourths of the American people, but three-fourths of the state legislatures, no matter how large or small the populations of the states they represent might be – must ratify the proposed amendment. And since the state legislatures themselves are counter-majoritarian for all the reasons just mentioned, one might think of this step in the process as a counter-majoritarian trifecta: Amending the Constitution requires ratification by (i) three fourths (ii) of already counter-majoritarian state legislatures (iii) that represent states of any population size. And all this is after a 2/3 majority vote by both of the counter-majoritarian chambers of the United States Congress. One book advocates eliminating the need for ratification entirely,²³⁶ and this book considers the alternative option of ratification by nationwide referendum.²³⁷

As many others have observed, this process makes the US Constitution painfully hard to amend. Donald Lutz, in his thorough 1995 comparative study, found the US Constitution to be the hardest to amend of any constitution in the world.²³⁸

To Madison, that was a good thing.²³⁹ To Jefferson, it was a problem. He felt that new generations should be able to alter the US Constitution far more easily.²⁴⁰ State constitutions, in contrast, are quite easy to amend. Typically, all they require is a simple majority vote of the people²⁴¹ – by definition, pure majoritarianism.

The resulting data are not surprising. The US Constitution has been amended only twenty-seven times – and only seventeen times since the final ratification of the Bill of Rights in 1791. In contrast, the average state constitution has been amended 150 times,²⁴² even though 37 of the 50 states have been around for far fewer years than the US Constitution.

I won’t wade deeply into that part of the debate. I do not object to the requirement of a supermajority for amendment of the Constitution. Our supreme law, which guarantees fundamental rights and protects other essential institutions, requires some measure of durability. Too easy an amendment process would leave the Constitution, and especially unpopular minorities, dangerously vulnerable to rapidly shifting political winds. But if the state constitutions are too fluid to rely on, the US Constitution seems to me to veer too far in the opposite direction.

The question being one of degree, my objection is to the endless layers of super-majoritarian requirements piled on top of one another. So much of that (not all) traces back to the outsized roles assigned to the states. The requirement of a 2/3 vote in the Senate accentuates the small-state favoritism already built into that chamber, and the requirement of subsequent ratification by three-fourths of the states (small versus large doesn’t matter) adds another dose of double counter-majoritarianism – a triple dose, actually, when one considers the counter-majoritarian problems that afflict so many state legislatures to start with. As in other places, the Constitution sacrifices the equality of citizens for the equality of states.

The supermajority requirements matter. Take the case of the Equal Rights Amendment. Proposed by the requisite two-thirds of both Houses of Congress, it provided, in Section 1, that “Equality of rights under the law shall not be denied or abridged by the United States or by any State on account of sex.” By the 1979 expiration date specified in Congress’s proposal, the amendment had garnered the ratifications of thirty-five states, falling just short of the required thirty-eight. In six additional states, one of the two Houses of the state legislature had voted to ratify, but the other had not. To estimate the percentage of the US population represented by the opposing groups of legislatures, I added the populations of the ratifying states to one-half of the populations of the six states where half the legislature had voted to ratify. (I did not add in the populations of three other states that voted to ratify the amendment after the congressional deadline had passed.) Under those assumptions, pro-ratification legislatures accounted for 78.4 percent of the national population. Legislatures that declined to ratify, even in combination with those legislatures that ratified it too late, accounted for only 21.6 percent of the national population. Despite the lopsided score, the latter won out.²⁴³

Apparently feeling that the dizzying array of counter-majoritarian requirements did not erect a high enough hurdle, the framers attempted to make one provision of the Constitution unamendable entirely. The last line of the amendment provision creates the following exception: “No State, without its Consent, shall be deprived of its equal Suffrage in the Senate.”²⁴⁴ This insistence on not only constitutionalizing the principle of equal Senate suffrage, but then taking the further step of insulating that principle from the already backbreaking constitutional amendment process, showcases how zealous the Antifederalists were about state sovereignty. Madison understood this. In Federalist 43, he explains this exception by saying that it “was probably insisted on by the States particularly attached to that equality.” It was, in other words, just a concession necessary to get all the states’ agreement, not a normative argument for the equal suffrage provision – much less an argument for perpetuating it.

This Senate exception to the amendment process does present an interesting, if inconsequential, conundrum. If it had been a freestanding provision, rather than a proviso to the amendment process, it might be easier to change. If there were enough support for making each state’s Senate representation proportional to its population, an amendment would have been able to simply do that. Even an amendment abolishing states could either expressly delete the requirement of equal state Senate suffrage or simply ignore that requirement, as the issue would become moot. There simply wouldn’t be any states with unequal Senate suffrage, because there wouldn’t be any states at all.

But the provision is not freestanding. It is cast as an exception to the constitutional amendment process itself. How could the Constitution be amended to eliminate equal state Senate suffrage when the very process that would have to be followed to amend the Constitution contains an express prohibition on the elimination of equal Senate suffrage?

One possible way around the exception is to argue that no legal document, and no provision of any legal document, should ever be interpreted to permanently prohibit the parties from agreeing to any future changes. The counterargument would be that, while such a principle makes sense when there is some elasticity in the relevant text, this language is crystal clear. It expressly prohibits depriving a state of its equal suffrage in the Senate without its consent. Besides, if this exception were interpreted as merely a statement that states must have equal state suffrage in the Senate, and not as a bar on using the constitutional amendment process to end that practice, the provision would be superfluous. The principle of equal state suffrage in the Senate is already guaranteed elsewhere in the Constitution.²⁴⁵

Alternatively, perhaps one could argue that a constitutional amendment abolishing states would not violate the equal suffrage requirement. If there were no states, then there would be no unequal state Senate suffrage. Or if, as proposed here, states remain as geographic areas and as sources of identity, affiliation, and pride, and state government is all that is eliminated, then it would seem that the states’ suffrage in the Senate would remain equal: Every state would have zero senators.

Perhaps the most clearly valid solution – assuming for the sake of discussion that there were the political will to abolish state government in the first place – would be to proceed sequentially. First, amend Article V itself, deleting the Senate proviso. That amendment would not violate the exception because it would not eliminate equal state suffrage; it would merely eliminate the exception to the constitutional amendment process. Then, once that amendment has been ratified, add a second amendment, either to make Senate districts equipopulous or to abolish states entirely. Perhaps even a single constitutional amendment with two clauses, rather than a sequence of two amendments, could effect both changes. As long as the clause abolishing states takes effect only after the clause that deletes the exception to the amendment process – and perhaps even if they take effect simultaneously – the amendment should be valid.

At any rate, the Senate suffrage exception to the amendment process, whether or not interpreted literally, is also one illustration of how counter-majoritarianism feeds on itself. The states, whose counter-majoritarian impact on our democracy has manifested itself in the many ways described in this chapter, insisted on that exception. By doing so, they have at least attempted, very possibly successfully, to inoculate themselves from termination even by a supermajority, thus preserving themselves for all eternity. As will be discussed in Chapter 3, Section A, state legislatures have achieved an analogous self-reinforcing effect through partisan gerrymandering.