Part Two: A National Popular Majority Rule for Presidential Elections

Guest Blogger

For the Balkinization Symposium on  Alexander Keyssar, Why Do We Still Have the Electoral College? (Harvard University Press, 2020), and Jesse Wegman, Let the People Pick the President: The Case for Abolishing the Electoral College (St. Martin's Press, 2020).

Edward Foley

What would be ideal

             Ideally, there would be a new constitutional amendment to eliminate the Electoral College and, instead, elect the president based on a majority of the popular vote nationwide. The amendment would empower Congress to choose the method of determining the majority winner of the national popular vote.  Congress could choose a traditional runoff system, in which there would be a second popular vote among the top two candidates in the event that none received a majority in the first popular vote. Alternatively, and perhaps preferably, Congress could enact a system of Ranked Choice Voting, in which a majority winner would be calculated from the rankings on the cast ballots.

            This constitutional amendment would achieve two main objectives. First, it would create a uniform national electorate for choosing the president. Voters in California and New York would each cast one ballot to be put in the same single national pile as the one vote cast by each voter in Iowa and Nebraska. No voter in any state would exercise any more clout over the outcome of the election than a voter in any other state.

            Second, this constitutional amendment would make sure that the winner not only received more votes overall than any other candidate but also was the candidate preferred by a majority of voters when compared directly against the second most popular candidate in the race.  The two criteria do not necessarily yield the same result. In a race with three or more candidates, Candidate A can receive more votes than any other candidate, but when all other candidates are eliminated and Candidate A is compared only with Candidate B, then Candidate B can be preferred by a majority of voters to Candidate A.  This constitutional amendment would insist that Congress adopt an electoral procedure for identifying a majority winner, not merely a plurality winner.

            In this way, the constitutional amendment would eliminate the risk of a spoiler situation. Candidates other than the two most preferred would be eliminated, either by a second round of voting or by the ranking of preferences in a single round of voting. Either way, there would be a head-to-head competition between the two most popular candidates, and the winner would be identified as the choice of the majority of voters without regard to the views of voters about other (now-eliminated) contenders.  Once eliminated, a candidate can have no effect on electorate’s preference for—and thus choice between—Candidates A and B.  Other candidates can run and be part of the race, but under this majority-guaranteed system these other candidates cannot “spoil” the selection of A or B. 

            If this system were in place this year, Kanye’s presence on the ballot could not affect the outcome.  Nor could any third-party or independent candidate.  It would be a head-to-head choice between Trump and Biden, and the one of these two candidates who was preferred by a majority of all voters nationwide would be elected president.

            It would be a great system. 

Why we can’t have the ideal

          But we can’t have it.  There are two reasons why. First, it is impossible to adopt a majority-winner version of a nationwide popular vote except by means of constitutional amendment.  Second, the difficulty of achieving any constitutional amendment to alter or abolish the Electoral College, as demonstrated by the history detailed in Alex Keyssar’s new book, makes it essentially a practical impossibility for the foreseeable future to secure the necessary constitutional amendment.

            Why Can’t We Do This Without a Constitutional Amendment? The National Popular Vote Interstate Compact (NVPIC) is a clever way to use the technical elements of the Electoral College to eviscerate its essential feature, namely the separation of the states for the purpose of casting and counting electoral votes. The one thing we know the Electoral College is not supposed to be: a uniform national popular vote. Yet, if enough states agree to pledge their electoral votes to the winner of the national popular vote, then it is possible by means of this agreement to convert the Electoral College system into a national popular vote.

            This is possible, of course, only because the states themselves have all chosen to use a popular vote as a method for appointing their own electors.  Therefore, when all the states conduct a popular vote on November 3, the states that have joined the NVPIC can use the popular vote tallies from all fifty states (and D.C.) to calculate a national popular vote winner and then appoint their own electors on that basis. If any state—say Florida—chose not to conduct a popular vote for the appointment of its own presidential electors, and reverted instead to the constitutionally permissible option of the state legislature appointing electors (as occurred early in U.S. history), then it would be impossible to calculate a national popular vote.  Florida, in this example, would be missing.  A national popular vote without Florida wouldn’t be a truly national popular vote.

             But the only type of national popular vote that is logistically possible under current conditions is a plurality-winner version of a national popular vote.  This is because the states themselves, apart from the single exception of Maine, only conduct a plurality-based version of a popular vote. States do not hold second-round runoffs in order to assure majority winners in the appointment of their presidential electors.  (Some states used to, but no longer.)  Nor, again with the single exception of Maine, do states currently used Ranked Choice Voting to appoint electors.

             Thus, even if some states wanted to form a compact in order to collectively appoint their own electors based on majority rather than plurality results, they would have no basis to calculate a majority rather than plurality winner beyond their own collective borders. To use Florida again as an example.  It does not calculate a majority, rather than a plurality, winner of its presidential popular vote—either by means of an actual runoff or a ranked-choice ballot. Thus, given the goal of calculating a nationwide majority winner, one is stymied by Florida as long as it does not change its own system for conducting a popular vote. It is simply impossible to calculate a nationwide majority winner if Florida is missing from the calculation.  There can be no determination of a national majority preference between the top two candidates if Florida does not collect this particular information, collecting instead only the plurality first-choice preference among all candidates.

             The only way to get the necessary data nationwide is to force all states to hold a runoff or use a ranked-choice ballot in order to identify a majority winner. But the only way to force all states to do this is by means of a constitutional amendment. A compact among states, even those who collectively have 270 electoral votes, won’t do the trick. Only a constitutional amendment (or voluntary opt-in by all states, which would be even harder to achieve) will work.

             There is an argument that Congress by statute could require states to employ ranked-choice ballots, just to facilitate a national calculation of a majority popular vote winner, even if those states remain free to appoint their own electors a plurality-win basis.  But this argument, while desirable as a policy matter, seems incorrect under current constitutional jurisprudence.  It would violate the “anti-commandeering” principle of the Tenth Amendment, as understood by the Supreme Court, to force a state government to use rank-choice ballots if the state government did not want to.  To be sure, Congress could compel the use of ranked-choice ballots for congressional elections, but not for presidential elections.  And even if Congress limited the grounds for this compulsion to facilitate the policy of other states to make a national ranked-choice calculation, Congress is constitutionally powerless to order sovereign states under the Tenth Amendment to obey that federal mandate.  The recalcitrant states would need to agree voluntarily to use ranked-choice ballots for it to be possible to make the necessary nationwide calculation. Congress of course could attempt to sweeten the deal by offering more federal funds to the recalcitrant states. Unless the financial inducement crossed the line of coerciveness, as in the Obamacare case, it does not violate state sovereignty under the Tenth Amendment for Congress to use its Spending Power to entice states to accept federal policy preferences.

             But if the goal remains nationwide compulsory use of ranked-choice ballots, a constitutional amendment—rather than a congressional statute—would be required for that.  

             Why Can’t We Get a Constitutional Amendment? During Thomas Jefferson’s first term as president, his supporters in Congress and the states had enough votes to clear the high hurdle for amending the Constitution to replace the original version of the Electoral College with a new one. It takes two-thirds in each house of Congress plus three-fourths of the states to adopt a constitutional amendment. The Jeffersonians barely had a vote to spare in Congress, but Jefferson’s popularity was strong enough for his party to gain seats in the midterm elections of 1802, and their Electoral College amendment squeaked through in time for Jefferson’s resounding reelection in 1804.

            As I explain in my own new book, Presidential Elections and Majority Rule, this Twelfth Amendment was not just a partisan ploy to facilitate Jefferson’s reelection. It also reflected a philosophically novel conception of presidential elections. While the original Electoral College was premised on the goal of electing a consensus candidate who would rise above the fray of partisan politics—George Washington was exactly what the authors of the Constitution had in mind—the new Electoral College adopted in the Twelfth Amendment had the goal of electing the candidate of the majority party, which inevitably was competing ferociously against an opposing party in a world where consensus was no longer possible.  When either the majority party or the minority party must prevail in their competition to control the nation’s presidency, the philosophical premise of the Twelfth Amendment is that republican government—rule by the people voting in elections—required the majority rather than the minority to win the highest office in the land.

            To be sure, the conception of majority rule embodied in the Twelfth Amendment was not the same as simple one suitable for electing a state’s governor.  Instead, because the United States of America was created as a federal republic, majority rule needed to be modified to reflect the distinctively federal character of the national government, including its president. Consequently, the Jeffersonians who crafted the new Electoral College in the Twelfth Amendment envisioned a compound form of majority rule: a winning president candidate was supposed to achieve a national majority of Electoral College votes by achieving majority support within enough states on a state-by-state basis. Jefferson himself, who was the model of the ideal winner for this new version of the Electoral College, achieved just this kind of compound majority-of-majorities in both his initial election of 1800, although he was almost deprived of his victory because of the faulty design of the original version of Electoral College, and in his 1804 reelection—which went exactly according to plan.

            Since then, as also detailed in Presidential Elections and Majority Rule, the United States has strayed from the philosophical premises of the Twelfth Amendment. In retooling the Electoral College, the Jeffersonians left it to the states to choose the particular mechanism for appointing electors. Based on the experience of the first four presidential elections, the Jeffersonians assumed that states would choose mechanisms that were consistent with letting the majority party in the state prevail in appointing the state’s electors. The Jeffersonians did not anticipate that states would abandon majority rule and switch instead to a system that permitting a mere plurality of votes statewide, and thus a form of minority rule, to control the appointment of all the state’s electors.

            Over time, Americans also have become much less enamored of the state-based nature of the Electoral College.  As the nation has become accustomed to thinking of itself as a single country, rather than a collectivity of sovereign states (as the European Union, for example, currently does), Americans increasingly have come to believe that their president should be elected in a single nationwide vote in which each citizen’s ballot counted equally. The idea that Virginians should vote separately for president from New Yorkers, instead of voting all together as Americans, no longer seems to make sense as it once did.

            Although Americans have abandoned key premises of the Twelfth Amendment, they have never amended Constitution since then in order to alter or abolish the Electoral College.  But it is not for lack of trying.  Indeed, not long after the Twelfth Amendment was adopted, Congress was at it again considering various proposals for revising the system.  And ever since, Congress perennially has attempted to send the states a constitutional amendment that would revise or eliminate the Electoral College.  But these efforts all have failed, and Americans remain stuck with a system they do not embrace.

            It is the great achievement of Alex Keyssar’s magisterial new book Why Do We Still Have the Electoral College? to chronicle and explain this failure.  Every episode he analyzes is worth reading, starting with the early proposals both before and after the so-called “Corrupt Bargain” that ended the 1824 election, the only one since the Twelfth Amendment to use its contingent procedure of one vote for each state’s delegation in the House of Representative for when no candidate achieves an Electoral College majority.  Eliminating this contingent procedure, especially to jettison the equal voting power of each state regardless of population, has been an increasingly high priority over time.  Who thinks Wyoming’s one Representative and California’s 53 should have exactly the same weight in the election of the president? But even a constitutional amendment to eliminate this anachronism has proved impossible so far.

            Although each episode of failure is riveting in its own way (tragedies have a way of making for the most compelling drama, as Shakespeare surely showed), the pace of Keyssar’s narrative accelerates in the late 1960s, when Congress came closest to replacing the Electoral College with a national popular vote. Keyssar attributes failure to several factors, prominent among which is the hostility of segregationist Southerners still fighting to preserve as much of their regional power despite passage of the Voting Rights Act of 1965 and other civil rights reforms. One factor that I would highlight is the failure of advocates for a national popular vote to handle the problem of plurality victories. The leading version of the national popular vote proposal at the time would have set a threshold of a 40% plurality to win the presidency; anything less would have required a runoff. But other proponents of a national popular vote did not like the idea of a runoff, and the two groups could not figure out a way to overcome this obstacle.

            The lesson of this history for me is to avoid the plurality problem by proposing a constitutional amendment that requires a majority of the national popular vote to elect the president, and then to leave to Congress the details of how to implement the majority requirement. Keeping the constitutional amendment as simple as possible in this way would seem a sound strategy for maximizing the possibility of widespread public support.  Most Americans like the idea of majority rule and believe it to be an appropriate basis for conducting elections.  Deferring for congressional implementation the second-order questions of exactly how to conduct election procedures to identify a majority winner would also seem to reduce the chances that the constitutional amendment would get tripped up over debates over whether to hold an actual runoff or instead use Ranked Choice Voting or other divisions that might occur over exactly how best to achieve the basic agreed-upon idea of a nationwide majority winner.

            Still, it would be overly optimistic to expect even this streamlined amendment would be capable of reaching two-thirds supermajorities in both the Senate and the House. With Republicans seemingly hostile to any electoral reform that they perceive as threatening to the partisan self-interest, no matter how beneficial it might be to voters generally or to good government, and with Republicans currently believing that the state-based nature of the Electoral College benefits them relative to a national popular vote, there is little likelihood of Republican support for even this most simple form of a national popular vote amendment. Thus, unless the Democrats are capable of electoral successes that would put Republicans at less than one-third of each congressional chamber, the prospects of a constitutional amendment remain unpromising.

             Given this pessimistic prospect, Part Three addresses what’s the best alternative that’s actually feasible.

Edward B. Foley is Ebersold Chair in Constitutional Law and Director, Election Law, at Ohio State
Moritz College of Law. You can reach him by e-mail at foley.33 at osu.edu