The Case Against the Electoral College

Introduction
It appears that Hillary Clinton will win the popular vote by more than 2 million total votes, but she still lost the election to Donald Trump. This is the fourth time in history that the candidate winning the popular vote lost the election. This last happened in 2000, with Al Gore winning the popular vote but losing the election.
As I explained in the essay in support of the electoral college, we should not be crying over spilt milk; Trump won the election based on the rules that were established. Perhaps the rules should be different, but we have the rules we have and he won.
While it doesn’t make sense to continue to cry over the spilt milk of the electoral college in 2016, there is a reasonable case to be made that the electoral college system should be abandoned in favor of selecting President based on the popular vote.   In this essay I will explain the arguments in favor of this approach.
Arguments in favor of abolishing the electoral college (EC)
One person, one vote. The basic argument in favor of abolishing the EC is that a critical component of democracy is that each person’s’vote should count equally and that our leaders should be chosen through an election that reflects this principle.
Historically, there have only been four instances of a candidate losing the popular vote but winning in the electoral college. But until the year 2000, it hadn’t happened in more than 100 years, and now after the 2016 election, it has happened twice in 16 years, with the Democrats winning the popular vote in both 2000 and 2016 and losing the election.
Here’s a look at recent Presidential election outcomes.
1988 Bush senior won, won popular vote
1992 Clinton won, won popular vote
1996 Clinton won, won popular vote
2000 Bush, won, lost popular vote (lost by 500,00 popular votes)
2004, Bush, won, won popular vote
2008 Obama won, won popular vote
2012 Obama won, won popular vote
2016 Trump won, lost popular vote (by 2 million popular votes)
I publish all of this data to not only point out the two years of EC v. popular vote discrepencies but also to highlight how many times the Democrats have won the popular vote – they won the popular vote in 6 of the last 8 elections but they only won the Presidency in 4 of the last 8 elections. So guess which party prefers replacing the EC with the popular vote? J
And, well, they have a point – most elections are conducted based on the popular vote and there is an intuitive appeal to the basic thesis of one person, one vote, especially when two of the last four elections have selected winners inconsistent with such a vote.

Ky Fullerton, law student, Oregon, Oregon Law Review, Summer, 2000, Comment: Bush, Gore, and the 2000 Presidential Election: Time for the Electoral College to Go?, p. 746-7
Only a direct election system gives individual votes equal weight. In Gray v. Sanders, the Supreme Court stated: “The conception of political equality from the Declaration of Independence, to Lincoln’s Gettysburg Address, to the Fifteenth, Seventeenth, and Nineteenth Amendments can mean only one thing – one person, one vote.” Gray involved a challenge to Georgia’s indirect primary system, where candidates for statewide offices were elected by county unit vote rather than direct popular vote. Georgia attempted to analogize its unit voting system to the Electoral College, but the Court dismissed the analogy as “inapposite” and struck down the statute as unconstitutional. The Court’s only justification for this conclusion was that the text of the Constitution explicitly provides for the Electoral College. Reconciling the Court’s holding in Gray, which required an equal dispersal of votes among a candidate’s constituency, with the winner-take-all system of the Electoral College is difficult at best. Unless mere mention of the Electoral College in the text of the Constitution “exempts all of its various nontextual facets from scrutiny, the Court too quickly dismissed Georgia’s analogy.” The disparity in state voting power that results from the Electoral College significantly enhances the possibility that a single state will determine the outcome of an election, regardless of the reliability of its vote results or the methods used to select presidential electors. To witness this effect, one need go no farther than the 2000 presidential election between Bush and Gore. Despite the controversies surrounding inconsistent vote-counting standards, antiquated equipment, and confusing ballots, the results in Florida effectively decided the 2000 presidential election. Only direct election of the President, with each person having one vote, remedies these unequal effects of the winner-take-all system.

And now we don’t even have to go farther than the 2016 election.
All votes should count equally –

Rhonda Hooks, law student, Spring 2001, Thurgood Marshall Law Review, Has the Electoral College Outlived It’s Stay?, p. 220
The right to vote is a fundamental right for which Americans have given their lives. That vote should count – each and every one. In the unit-voting system, each and every vote does not count. This process is just another form of disenfranchisement of voters. To bring all states under one rule of uniformity has worked well in other areas dealing with fundamental rights. So why not address an equally important issue of the right to cast a vote and it count? The popular vote is a reflection of the hearts and souls of the American people; so, let all votes count.

The EC was designed to undermine popular democracy.

John B. Anderson is a professor of law at NOVA Southeastern University’s Shepard Broad Law Center in Fort Lauderdale, Florida, Spring 2005, Human Rights, The Electoral College Flunks the Test in an Age of Democracy, p. 17-18
There is another interpretation of the genesis of the Electoral College that is even less kind in explaining its inclusion in the Constitution: that the Electoral College grew out of the last-ditch efforts of the “states righters” of 1787 to preserve as much of the Articles of Confederation as possible. This group was intent on denying direct popular election of the president and preserving the power of the states. Just as they succeeded in a provision allowing state legislatures to elect the members of the Senate, they wanted the primary power to elect a president to be lodged in the states–not in a mass electorate comprised of individual voters. James Madison, James Wilson, and Gouveneur Morris preferred a vote by the people but fell back on the compromise of an Electoral College to appease the die-hard defenders of the Articles of Confederation and their exaltation of each state’s right to be its own principal governing force. The original design of the Electoral College was based on the notion that electors would be faithful agents of the people who were “men superior, discernment, virtue and information” and who acted “according to their own will.” GEORGE C. EDWARDS III, WHY THE ELECTORAL COLLEGE IS BAD FOR AMERICA 19 (2004). Fifty years later, Justice Joseph Bradley of the U.S. Supreme Court and a member of the 1877 Electoral Commission established to settle the disputed Hayes-Tilden Election of 1876, said, “Electors were mere instruments of party–party puppets–who are to carry out a function that an automaton without volition or intelligence might as well perform.” Id. Another commentator of the same period, Senator John J. Ingalls of Kansas, opined that “electors are like the marionettes in a Punch and Judy show.”

And if we got rid of the EC, there would be even greater democratic participation.

Bob Fredericks, November 25, 2016, New York Post, Electoral College significantly impacts voter turnout, https://nypost.com/2016/11/25/electoral-college-significantly-impacts-voter-turnout-study/
There’s another reason to get rid of the Electoral College: Voters in states that aren’t in contention are staying away from the polls in droves. Turnout was low in solidly Democratic or Republican states such as New York, California, Texas and Oklahoma compared to more competitive states. According to the Election Project, 52.4 percent of eligible voters cast ballots in New York, 53.8 percent in California, 51.1 percent in Texas and 52.1 percent in Oklahoma. Hillary Clinton was always a cinch to win the first two states, and Donald Trump the latter pair. In toss-up states, turnout was significantly higher: Florida, 65.1 percent; Ohio, 64.5 percent; and New Hampshire, 70.3 percent. Writing in The Hill, historian and educator Ross Rosenfeld said the apparent discrepancy was “a definite argument against the Electoral College,” which handed Trump his victory even though Clinton garnered at least 2 million more votes nationwide The numbers suggest that if not for the Electoral College, Clinton’s margin in the popular vote would have been far greater. “Given that Clinton’s lead in big blue states was often bigger than Trump’s in big red states, the overall likelihood is that a straight popular vote would’ve increased Clinton’s popular vote lead,” Rosenfeld wrote.

Popular support. There is popular support to base the election on the popular vote.

Congressional Research Service, December 12, 2014, The National Popular Vote Initiative: Direct Election of the President By Interstate Compact, https://www.everycrsreport.com/files/20141212_R43823_4df94a11f0b45b91348c622a380b03da934c92a3.pdf
For the record, survey research findings have consistently identified broad support for direct popular election of the President and Vice President. In its most recent question on the issue, asked in 2013, the Gallup Poll reported that 63% of respondents favored an amendment providing for direct popular election, while 29% favored retention of the electoral college. This finding corresponds with similar levels of public approval recorded by Gallup since at least 1967.26 Further, National Popular Vote, Inc., reports that statewide surveys in 38 states found similar support levels for direct popular election among respondents in 238 state polls taken between 2007 and 2011.2

A simple system. One person, one vote makes the system transparent and easy to understand

Harvard Law Review, June, 2001, RETHINKING THE ELECTORAL COLLEGE DEBATE: THE FRAMERS, FEDERALISM, AND ONE PERSON, ONE VOTE, p. 2546-7
The one-person, one-vote principle promotes transparency by ensuring that electoral processes are comprehensible and justifiable: a process in which each person casts a vote and the canvassing board declares a winner based on a majority or plurality of the votes is simple to understand and normatively unimpeachable. Whether individuals are running for governor, senator, congressman, mayor, or dogcatcher, it is clear to everyone that the candidate with the most votes in the defined geographic area will win the office. In contrast, an electoral process that, for example, tallied votes but then flipped a coin when the tally indicated a margin of victory of five percent or less, would lack transparency. Although this method may be comprehensible, it is difficult to conceive of any publicly shared normative justification for the decision to randomize the vote. In light of this analysis, a central problem with the electoral college’s failure to conform to the one-person, one-vote principle is that it operates in a nontransparent way. The inconclusive empirical data regarding whom the electoral college benefits and the diverse and often conflicting rationales offered for its existence are so complex that it is unlikely that the public will ever be able to reach a secure understanding of the justification behind it.

Minority rights and interests. There are a number of ways the EC system operates to disenfranchise minorities.
One, since the majority vote is awarded to state electors, minority interests are inherently not reflected in the electors’ vote for the President.

Yale Law Journal, January 1996, The Illegitimate President: Minority Vote Dilution and the Electoral College,  p. 936-7
African-American voters in the United States have never held much sway in the election of the President. Although they often exert considerable influence in presidential primaries, their status as a racial minority ensures that, in the general election, their voices are frequently drowned out by the majority. In the electoral college, where the President is actually chosen, the preferences of minority voters count for almost nothing. The overwhelming majority of states provide that the presidential candidate who wins the popular vote in the state claims all of its electoral votes. Thus, so long as minority voters have different political preferences than the majority – a fact that is almost self-evident in many parts of the country – their votes will be virtually meaningless in the final selection of the President. This problem is perhaps most acute in the South, where the political disparities between African-American and white voters have historically been most pronounced.   As one pair of commentators has recently noted, African-American voters in the South “have had little more influence on most modern presidential general elections than Bulgarians. Their votes, although technically cast, have not usually counted.”

Two, it locks in a Republican advantage against minorities.

Yale Law Journal, January 1996, The Illegitimate President: Minority Vote Dilution and the Electoral College p. 959-61
In the four most recent presidential elections, the South has been solidly in the Republican camp, notwithstanding a dramatic surge in the number of African-Americans registered to vote.   But as the gradual realignment of white Southerners from 1948 to 1968 makes clear, the current Republican domination of the electoral college in the South is no accident. Rather, it is in large part the result of a conscious effort by white Southern politicians – first by segregationist Democrats, and later by racially conservative Republicans – to make race a focal point of presidential politics.   Today, race remains a polarizing force in presidential politics.   The realignment of the bloc of white racial conservatives from the Democratic party to the Republican party has altered the political picture, however. Because of this shift, racial conservatives no longer need to manipulate the electoral college in the manner that Thurmond and Wallace attempted. Instead, they rely on the discriminatory mathematics of the winner-take-all system, which ensures that racial minorities have no voice in determining the composition of the electoral college. Republicans today often refer to their party as having a “lock” on the electoral college by virtue of its dominance in the South.   Such terminology echoes – though perhaps unconsciously – the language used by Collins nearly fifty years ago. Racial appeals continue to be a staple of presidential politics. Rather than relying on overt racist imagery, as Thurmond did in 1948, modern politicians generally play the “race card” through subtle use of code words and careful manipulation of racial imagery.   For example, George Bush’s victory over Michael Dukakis in 1988 is frequently attributed to his campaign’s skillful handling of racial imagery – most notably the infamous “Willie Horton” episode.   Indeed, many of the rhetorical devices and subtle racial images employed by modern-day Republicans are essentially variations on themes developed by Wallace in his 1968 campaign.   Issues of race are frequently closely tied to a number of “social issues,” ranging from crime to welfare to affirmative action Race surfaces in other unexpected ways in our national politics. In the current presidential campaign, many members of the media enthusiastically touted General Colin Powell, a prominent African-American and the former chairman of the Joint Chiefs of Staff, as a possible Republican candidate for President. In many ways, this phenomenon was curious, because Powell had never sought public office or offered any hint of a political agenda. While the origins of the Powell noncandidacy are complex, it seems undeniable that one of the primary reasons for the media fascination with him was his race. As an African-American Republican (or at least, a presumed Republican), Powell did not fit into established categories. He was seen as someone who might be able to attract the support of both black Democrats and white Republicans. In this respect, Powell’s backers were tacitly acknowledging the stranglehold that race has on the American electorate. Indeed, Powell acknowledged as much in his speech announcing that he would not run for President, commenting that he wanted “”to help the party of Lincoln move, once again, close to the spirit of Lincoln.'”   Racial politics also play an important role on the other side of the political aisle. Democratic candidates frequently try to walk an almost impossibly fine line, desperately trying to distance themselves from black political interests without alienating their African-American political base.   In short, race has by no means been a trivial or incidental issue in presidential politics of the modern era. Far from it, race has been a central issue in the most important political trend of the last fifty years: the conversion of the South from a Democratic bastion to a Republican one, and the Republican party’s corresponding shift from a moderate stance on racial policies to a much more extreme position.   In effect, the battle between the parties for control of the South has led to their severe divergence with respect to racial issues – a change that has been felt not just in the South, but nationwide. For nearly five decades, politicians have been relying on the primacy of the winner-take-all scheme as a means of excluding African-American voters from the political process. The efforts of Strom Thurmond and George Wallace to inject race into the presidential contests in 1948 and 1968 help to illustrate the logic of George Bush’s attempts to do much the same thing in 1988. This recurring emphasis on race all but guarantees the continued occurrence of racially polarized voting, and consequently ensures that minority voters will not enjoy an equal opportunity to participate in the selection of their Chief Executive.

Presidential legitimacy. Losing the popular vote threatens a President’s legitimacy

Ky Fullerton, law student, Oregon, Oregon Law Review, Summer, 200, Comment: Bush, Gore, and the 2000 Presidential Election: Time for the Electoral College to Go?, p. 753-4
Finally, and perhaps most importantly, losing the popular vote seriously threatens a President’s legitimacy. Allowing a President to win office despite losing the popular vote is contrary to America’s time-honored principles of majoritarian democracy. Akhil Amar predicted in 1995 that “one day, we will end up with a clear Loser President – clear beyond any quibbles about uncertain ballots. And the question is, will this Loser/Winner be seen as legitimate at home and abroad?” Others agree with Amar’s assessment. Instances where a popular vote loser prevails in the Electoral College are “fraught with danger. The legitimacy and governability of a president without a popular vote majority is prima facie suspect. A system that does not include direct expression of the voice of the people undermines the principle of a government with the consent of the governed.” Electing a President who loses the popular vote “undermines respect for the system and compromises the new president’s mandate to govern.” Indeed, following the Supreme Court decision that effectively handed George W. Bush the 2000 presidential election, questions immediately arose regarding Bush’s legitimacy as President. n Although time will tell just how Bush’s precarious electoral vote victory affects his ability to govern, history provides some guide. Prior to 2000, three candidates obtained the Presidency despite losing the popular vote: Adams in 1824, Hayes in 1876, and Harrison in 1888. The administrations of these three Presidents were largely unsuccessful. In 1828, Andrew Jackson avenged his prior loss and soundly defeated Adams in both the Electoral College and popular vote. By the end of his term, Hayes was so unpopular among members of his own party that he declined to run for reelection. Like Jackson before him, Grover Cleveland returned to deny Harrison a second term in 1892. Gerald Ford faced similar problems after assuming the Presidency in 1974. Earlier, Congress confirmed Ford as Nixon’s Vice President following the resignation of Nixon’s first Vice President, Spiro Agnew. Following Nixon’s resignation in August 1974, Ford faced the daunting prospect of governing a nation that had neither elected him to the Presidency nor the Vice Presidency. Without the popular mandate a President gains by virtue of winning an election, Ford found little support in Congress for his legislative proposals. In slightly more than two years in office, he vetoed a remarkable sixty-six bills. Although Ford’s circumstances are somewhat different from those of a popular vote loser who obtains the Presidency, they demonstrate the difficulties faced by a President who is viewed, at least by some, as lacking a popular mandate. The bottom line is that the Electoral College system has the potential to create a significant crisis with respect to the legitimacy and effectiveness of a President. Implementation of a direct popular vote system in presidential elections is the only way to fully address this concern.

Disproportionate representation. Some states have more representation than others.

Michael Herz, law professor and Co-Director, Floersheimer Center for Constitutional Democracy, Cardozo, Cardozo Law Review, May 2005, HOW DEMOCRATIC IS THE AMERICAN CONSTITUTION?: ROBERT DAHL’S HOW DEMOCRATIC IS THE AMERICAN CONSTITUTION?: AN INTRODUCTION, WITH NOTES ON THE ELECTORAL COLLEGE, p. 1194-5
On the second point, votes would only be of equal weight if electoral college votes were exactly apportioned among the states, and then on the basis of those actually casting ballots rather than population. But in fact electoral votes are allocated by populations, which are not perfect multiples of 435; the rates of registration and of voting vary from one state to another; each state, regardless of size, gets two electoral votes (corresponding to its two Senators) in addition to the electoral votes allocated by population (corresponding to its Representatives). The result, as has often been pointed out, is a wide variation in the weight of individual votes and a particular dilution of individual voting strength in large states. So in 2000 Vermont had 97,931 1/3 voters per elector; New York had 206,727 1/4 voters per elector. In this sense, a Vermonter’s vote counted twice as much as a New Yorker’s. This also looks odd at best, if not, in the words of Senator Durbin, “undemocratic and unfair.” So the electoral college conflicts with prevailing assumptions about elections.

Disproportionate individual representation. The EC system makes it possible for a single invididual to have a disproprtionate impact on the election.

Michael Herz, law professor and Co-Director, Floersheimer Center for Constitutional Democracy, Cardozo, Cardozo Law Review, May 2005, HOW DEMOCRATIC IS THE AMERICAN CONSTITUTION?: ROBERT DAHL’S HOW DEMOCRATIC IS THE AMERICAN CONSTITUTION?: AN INTRODUCTION, WITH NOTES ON THE ELECTORAL COLLEGE, p. 1211-12
This may not be a good thing. Critics of the electoral college decry how minute changes in votes cast could have changed many election outcomes (and produced a divergence between the popular and electoral vote winners). So, for example, if 1,983 voters in California had voted for Hughes rather than Wilson in 1916, Hughes would have won the election; a change of 11,424 votes, across five states, would have given Richard Nixon the White House in 1960; and a shift of 9,246 votes in Hawaii and Ohio would have sufficed for Gerald Ford to defeat Jimmy Carter in 1976. These possibilities are in the nature of a districted, state-by-state system, just as they are in the nature of a win-games system for the World Series. For electoral college critics, it seems self-evident that having election results hinge on such narrow differences is problematic. (In general, these arguments begin with the premise that it is unacceptable for the popular vote winner not to become President; since that has rarely happened, the argument is bolstered by pointing to the instances in which it almost happened.) But the World Series analogy draws attention to the fact that the salient fact may be not that the popular vote winner can lose the election, but that the election’s outcome can be determined by such small numbers of voters. The electoral college system increases the likelihood that a single individual, or a single group, will cast the deciding vote or votes. It increases the circumstances in which, like the Mazeroski home run, an individual can have a disproportionate impact. The 2000 election was a case in point. Never in recent memory, perhaps never at all, has the American electorate had a stronger sense that every vote mattered. It all seemed to come down to Florida, and there the election was excruciatingly close. And it only came down to Florida because of almost as close tallies in a number of other states, in any one of which a slight shift could also have changed the overall outcome. But the national election was not excruciatingly close, nor is it likely ever to be simply because the electorate is so large.

And some states have many more Electoral College reps per person than others

Michael Herz, law professor and Co-Director, Floersheimer Center for Constitutional Democracy, Cardozo, Cardozo Law Review, May 2005, HOW DEMOCRATIC IS THE AMERICAN CONSTITUTION?: ROBERT DAHL’S HOW DEMOCRATIC IS THE AMERICAN CONSTITUTION?: AN INTRODUCTION, WITH NOTES ON THE ELECTORAL COLLEGE, p. 1200
Second, the electoral votes are not allocated exactly in proportion to population. For one thing, every state gets two votes (corresponding to its two Senators) regardless of size. In those states with only a single representative, especially those with a population less than that of the theoretical ideal House district, an electoral college vote represents many fewer popular votes than in larger states. At the extremes, Wyoming has 165,101 people per electoral vote; California has 616,924. In addition, state populations are not all perfect multiples of 1/435th of the total number of U.S. citizens. Therefore the citizens per representative ratio (and so citizens per electoral vote) varies from state to state.

History of racism. One of the arguments used to support the creation of the EC was that it would protect the interests of the slave states.

Mark Weston is the author of “The Runner-Up Presidency – The Elections that Defied America’s Popular Will, The Hill, November 23, 2016, The electoral college has to go – and her’es how we do it, https://origin-nyi.thehill.com/blogs/pundits-blog/the-administration/307359-the-electoral-college-has-to-go-heres-how-we-do-it
The United States can’t pretend to be a democracy. First-place candidates have lost two presidential elections in the last 16 years: Al Gore lost despite beating George W. Bush by 544,000 votes in 2000, and Hillary Clinton was defeated even though she leads Donald Trump by more than 1.69 million votes this year. Most Americans realize that electoral votes have got to go.
Even James Madison wanted direct elections of the president. He only created electoral votes during the Constitutional Convention as a compromise. White Southerners were afraid of being outvoted by Northerners, and wanted their slaves, or at least 3/5 of them, included in the allocation of votes even though the slaves themselves could not vote. Electoral votes made this possible, direct elections did not. Our strange voting system is thus a relic of America’s original sin of slavery.

The fact that it is tainted by slavery is enough of a reason to disband it.

Darnell Weeden, Associate Dean and Roberson King Professor; Texas Southern University, 2008, Ohio Norther University Law Review, v. 34, Response to Professor Amar: Some Thoughts on the Electoral College’s Past, Present, and Futurep. 395
While sharing his thoughts regarding ten contemporary apologies for the Electoral College in his Ohio Northern commentary, Professor Amar utilizes the same Socratic technique that he used in his book, America’s Constitution: A Biography. Similar to his book, Professor Amar engages in what Professor Hall correctly describes as a Socratic narrative style by posing questions about whether America should continue its electoral system. Amar then offers plausible justifications for the systems which are articulated by other legal scholars; however, Amar summarily rejects these other systems as providing wrong answers, and then provides his view of the Electoral College as providing the correct answer. According to Professor Hall, Amar’s use of the Socratic method is “didactic and frustrating because the dialogue tends to be one-sided and the arguments often designed as straw men to set up Amar’s own conclusions. The technique also gives Amar the role of the grand ventriloquist who at the end sweeps all behind.” In my opinion, Amar masterfully uses the Socratic narrative to provide instructive and enlightening information about why an Electoral College system rooted in slavery should be abandoned as soon as possible. I think Amar has successfully set up the moral conclusion that an America that has rejected legalized slavery should also rid itself of the Electoral College as a tainted vestige of the evils of slavery. Although Amar concedes that it is possible in spite of its slave roots that the Electoral College has served a useful purpose unrelated to the slavery question. After reviewing Amar’s rationale for rejecting the top ten justifications for the continuing existence of the Electoral College, I endorse Amar’s conclusion that these arguments are not credible.   The section that follows is essentially my general agreement with Professor Amar’s skillful debunking of the top ten prevailing apologies for supporting the Electoral College.

Additional arguments. This card covers four arguments at one time, so I thought I’d just list it last

Luis Fuentes-Rohwer and Guy-Uriel Charles, law professors, Florida State University Law Review, 2001, v. 2, The Law of Presidential Elections: Issues in the Wake of Florida 2000: The Electoral College, the Right to Vote, and Our Federalism: A Comment on a Lasting Institution, p. 895-7

  1. Faithless Electors

We begin with one of the least compelling criticisms of the Electoral College, the “faithless elector” problem. A “faithless elector” is an elector who “takes it in his head to act independently” and votes for a presidential candidate other than the winner of their state’s presidential election. The argument, taken seriously, is this: if a “voter has a constitutional right to cast an effective vote for President, an elector who casts his ballot contrary to the voters’ mandate may be said to be acting under color of state law to deprive the voters of that constitutional right.”   This is another way of saying that, since 1796, electors have become “party dummies.” Their role is but limited to registering the preference of state voters at large, nothing more.
There are many reasons the faithless elector problem is one of the least captivating defects–if indeed it is a defect–of the Electoral College. As a point of departure, the faithless elector problem is often described as a “specter”; throughout the history of the College, “faithless electors” have not surfaced often. Longley and Peirce report that “since 1796 fully 21,291 electoral votes have been cast for president, but only nine votes in all those years were indisputably cast ‘against instructions.'” The statistical probability that a faithless elector would cast a decisive vote in a presidential election is extremely low.   Hence, the argument is about possibility, not probability.
Second, remedies for the “faithless elector” are readily available. As a matter of procedure, only those who are most committed to the party are selected as electors. Additionally, many states have passed statutes compelling electors to strictly vote for the presidential nominee of the party of their choice.   Further, if the faithless elector does become a significant problem, it can be fixed by the simplest amendment to the Constitution, the automatic vote. That is, instead of choosing electors as a bridge between the popular vote and the Electoral College, proposals have been made to assign electoral votes to the winning candidates automatically. Moreover, it is open to question whether the electors were intended to act as free agents, independent of the public’s opinion within their corresponding states. Finally, and as the Supreme Court’s opinion in Bush v. Gore   made vastly clear, a constitutional right to vote for President is simply nonexistent.

  1. Inspiring “Respect and Acquiescence”

For a second criticism, we look to none other than James Madison. “Next to the propriety of having a President the real choice of a majority of his Constituents,” he wrote to George Hay in 1823, “it is desirable that he should inspire respect & acquiescence by qualifications not suffering too much by comparison.”   To be sure, the popular vote is the simplest and most accepted way to chose a representative, and the chief executive is no exception. Yet, absent that, the chosen person must “inspire respect & acquiescence.”
Unlike Madison, we are not quite as worried with the qualifications of the chosen executive; perhaps our recent history leaves us a bit cynical to ask for too much. Instead, we worry more about the fact that, unsurprisingly, many Americans do not fully understand how the Electoral College works.   As an ABA commission on electoral reform concluded, the College is “archaic, undemocratic, complex, ambiguous, indirect, and dangerous.”   In this vein, others have gone as far as to call it “a deplorable political institution.” The fact that a majority of Americans do not understand how the Electoral College works is potentially problematic. After all, this is the process by which we select our President and nothing short of the legitimacy of the incoming President potentially hangs in its balance.
Defenders of the institution concede this point rather willingly, at least at first glance. Martin Diamond, for example, writes that:
 
Perhaps the fear is that voters are baffled by the complexity of the Electoral College and that their bafflement violates a democratic norm. It must be admitted that an opinion survey could easily be devised that shows the average voter to be shockingly ignorant of what the Electoral College is and how it operates.
And yet, these concessions lead him away from the expected conclusions, for, as he proceeds, “it all depends on what kind of knowledge the voter is expected to have. . . . However ignorant they may be of the details of the Electoral College, their ignorance does not seem to affect at all the intention and meaning of their vote, or their acceptance of the electoral outcome.”
Diamond raises a second objection, one grounded specifically in the republican tradition and our system of government. The Electoral College, he maintains, is “only one example of the complexity that characterizes our entire political system. Bicameralism is complex; federalism is complex; judicial review is complex; the suspensory executive veto is a complex arrangement; the Bill of Rights introduces a thousand complexities.” Thus, he asks, “if a kind of prissy intelligibility is to be made the standard for deciding what should remain and what should be simplified in American government, how much would be left in place?” Much can be said for Diamond’s position. Most Americans pay very little attention to politics. In general, the political knowledge of the average American is rather limited.   Taken together, these two premises lead to our present condition over the public’s understanding of the Electoral College. That is, the fact that many people misunderstand the workings of the Electoral College is undoubtedly a function of the attention that the average American devotes to politics and the minimal level of political knowledge of the electorate as a whole.
In response to this condition, one may argue that the fact that many Americans are unaware of how the College works undermines the legitimacy of the incoming President or of the electoral system as a whole.   We do not think so. An electoral institution may be said to undermine the legitimacy of the winning candidate in a case where political elites have deliberately conspired together for the purpose of coercing the electorate to accept an institution that it would not otherwise accept. Such is not the case here. Thus, the question is whether a “long-standing constitutional arrangement”   is undemocratic simply because it is complex. On this score, we agree with Diamond, who asserted that institutional simplicity is not the sine qua non of democracy.

  1. Ideological Purity, the Minority President, and Contingencies

A third argument against the Electoral College looks to the legitimacy of the institution and modern democratic understandings. To some commentators, the Electoral College should be abandoned in favor of a direct election system not simply because of voter confusion and ignorance but on the grounds that selection by popular vote is the more legitimate democratic alternative. n To be sure, when the electoral count and the popular vote count are convergent and elevate the same presidential candidate to the Presidency, concerns about electoral legitimacy are largely an academic exercise. The worry, however, is that at some point–such as the 2000 presidential election–the two counts will diverge and the legitimacy of the incoming President, if not the Presidency itself, will be implicated.
Not surprisingly, the critics have spilled most of their ink exactly here. There are two variations of this criticism. This first variation is primarily directed at the contingency plan–the established procedure were a candidate unable to garner the requisite electoral votes. As we know, such a scenario would send the election to the House of Representatives, where each state delegation would cast one vote.   To some, this is the feature of the electoral system “that has been most condemned.” At its most extreme, this procedure may allow one person to cast the deciding vote in an election–a decision that is clearly “distasteful in a democracy.” As such, “steps should be taken to prevent their recurrence.” n120 Even defenders of the College concede some ground here; Diamond, for example, has labeled the prospect of an election decided by the House of Representatives a “horror.”
The second variation–and more popular criticism–even has its own name: Senator Kefauver has labeled the possibility that the Electoral College votes and the popular vote do not match the “Loaded Pistol to Our Heads” problem.
Abbott and Levine have similarly likened the election of an executive who has not won a majority of the popular vote to the “Great San Andreas Earthquake.” Representative Emanuel Celler, while Chairman of the House Committee on the Judiciary, described the electoral process resulting in a minority President as “horrible,” “unsporting,” “dangerous,” and “downright uncivilized.” John Feerick forecasted that “resentment, unrest, public clamor for reform and an atmosphere of crisis would probably ensue”   “if the popular-vote winner were to lose a presidential election.” The advent of Baker v. Carr   and the “one person, one vote” revolution frames this particular criticism of the Electoral College. There are two ways understand this argument. First, and as Neal Peirce and Lawrence Longley write, “to lose their votes is the fate of all minorities, and it is their duty to submit; but this is not a case of votes lost, but of votes taken away, added to those of the majority, and given to a person to whom the minority is opposed.”   The second point is much simpler. As Alexander Bickel explained a generation ago: “It is time for the system to be ideologically pure. The Court has said that the Constitution commands equal apportionment. We should, therefore, reapportion the presidency.”   In light of the history of the College and specifically its precarious inception, this is an argument worth thinking about.
From these perspectives, the question presented is rather simple: is the Presidency less legitimate if the Electoral College process produces a wrong winner? Supporters of the College have offered a number of responses. We examine the three most persuasive defenses. First, defenders of the College point out that a “runner-up” Presidency would be mostly a “fluke.” As such, the possibility of a divergent vote is hardly a source of concern, for simply, “the American people would grin and bear it and throw him [the “runner-up” President] out the next time, if they did not like what he did. Flukes happen. . . . That is as close to Russian roulette as a pimple is to cancer.”
Second, commentators question the reflexive coupling of American democracy with majority rule. Hardaway argues, for example, that “[a] winner in the Electoral College who fails to win the most popular votes is no more a ‘wrong president’ than legislation passed by the Senate is the ‘wrong legislation,’ or an amendment passed by the States (and not by popular vote), is the ‘wrong amendment.'” Polsby and Wildavsky similarly argue that “there is no serious reason to quarrel with the major features of the present system, since in our form of government ‘majority rule’ does not operate in a vacuum but within a system of ‘checks and balances.'”
Third, one could question the criticism over the legitimacy of a minority Presidency on its own merits. Two such responses are particularly attractive. The first response argues that the institution of the Electoral College affects campaigning across the country and in so doing makes it difficult, perhaps impossible, to know how a candidate who does not win the popular vote would fare under a different system. On this view, it is unfair and ultimately inaccurate to separate the popular vote from its particularized electoral context. Plainly, the argument concludes, the popular vote is a distorted and ultimately useless measure. The second response looks to the function played by the Electoral College, particularly after a close election. Alexander Bickel notes, for example, that:
When some 70 million votes divide so closely, only an immensely dogmatic majoritarianism would insist that the so-called winner has the sole legitimate claim to office. In truth, there is a standoff, and all that is needed is a convenient device–any convenient device previously agreed upon–for letting one of the two men govern. That is all that is needed, and that is all that is possible.
These defenses are persuasive at least in one respect. They underscore the fact that legitimacy in presidential contests is not simply outcome-determinative. The supposition of illegitimacy presupposes an outcome-determinative conception of legitimacy: the presidential process is legitimate only if it produces the right outcome. In this case, “right” is defined as the outcome determined by the popular vote count. But that is precisely the question: should the popular vote count provide the normative baseline for judging whether the presidential election procedure identified in the Constitution n138 is sufficiently democratic?
In this context, democratic legitimacy is both outcome-dependent and process-dependent. If the rules of the game are described ex ante and the parties play by these rules, then any outcome is by definition legitimate. The fact that the Electoral College may produce a “wrong-winner” is part of the expected outcome as defined by the rules of the game. Put simply, the rule of decision is that the “person having the greatest number of votes . . . shall be the President, if such number be a majority of the whole number of Electors appointed.” The controversy surrounding the 2000 presidential election is useful to illustrate this point in a less abstract manner. The fact that George W. Bush lost the popular vote did not seem to terribly unnerve the general public. The public accepted the result of the electoral count once the legal challenges were exhausted because there was a sense that the selection of George W. Bush was part of the expected outcome.
In contrast, the reaction to the Bush Presidency in the African-American community differed markedly from that of many White Americans. Some commentators criticized African-American leaders for publicly announcing that they will not accept the result of the election on precisely the grounds mentioned above–that the result was democratic because it was part of a democratically expected outcome. However, African-American leaders responded that they were rejecting an outcome that they would have conceded ex ante was legitimate. Rather, they argued that the outcome was not legitimate because the rules of the game were not followed.
Thus, though there may not be agreement about whether the proper procedures were followed in the 2000 presidential elections, there is widespread agreement about the ground rules: When the proper procedures are followed, the outcome cannot be contested. In other words, in a society committed to democratic rule, systemic outcomes are by definition democratic when those outcomes are the product of preexisting commitments.
Even if one concedes this point to the defenders of the current system, however, their defense is nevertheless unsatisfying. The question posed is a normative one: whether electing the President by popular election is, as a normative matter, more democratic than the current system. The answer that the current system is democratic or that popular election is not the sine qua non of democracy does not resolve the normative inquiry. This answer does not speak to the question of comparative “democraticness.” We will return to this point in Part III. Before doing so, we discuss one last criticism of the Electoral College institution.

  1. Unit Voting

A common criticism often offered in favor of the eradication of the Electoral College is that the College diminishes the voting power of various political minorities, including both voters of color and voters who support third-party candidates. The criticism is essentially that the winner-take-all system or unit rule unnecessarily “wastes” votes, particularly when compared to a districted-vote system or a direct-vote system.
The argument is disarmingly simple. The unit-voting system employed by most states–which is not constitutionally required–“wastes” votes whenever the unit winner wins by a greater margin than necessary to carry that unit. The votes are considered wasted not simply because they represent an unnecessary marginal excess, but because the excess margins could make up the difference in another unit that the candidate lost. Similarly, the unit vote rule also wastes the votes of the unit loser. Even though the unit vote loser was able to win some votes, because the losing candidate did not carry the unit, he or she does not get any electoral votes. Consequently, the voters who voted for that candidate “wasted” their vote, especially if their candidate was from a third party.
The Electoral College and Third Parties.–Let us first think about the Electoral College’s impact on third parties before we examine the College’s impact on voters of color. The unit vote system, as opposed to a proportional system, does in fact make it more difficult for political minorities to win electoral votes. When the unit is defined as the “state,” which it is in every jurisdiction with the exception of Maine and Nebraska, political minorities will find it more difficult to carry the unit. The unit rule is one reason that third parties have found it nearly impossible to break the hold that the Republican and Democratic parties have had on presidential selection.

Appendix Table 1 shows how Ross Perot’s third-party candidacy fared in 1992. As is evident from Table 1, 19,741,657 Americans voted for Perot. Unfortunately for the Perot supporters, Perot carried nary an electoral vote. From this perspective, 19,741,657 Perot voters wasted their votes.
Appendix Table 2 demonstrates how a Perot candidacy would have fared under a proportional distribution system. Under a proportional distribution system, Perot would have garnered 102 electoral votes. Even though Perot could not have acquired sufficient electoral votes to win the Presidency, he and his party could have played a significant role in who would eventually become President. In many respects, the current system minimizes the role that third parties can play as brokers and kingmakers in presidential politics.
In making this claim, we must underscore the fact that the solution here is not achieved by changing the size of the unit but by getting rid of the winner-take-all feature of the electoral system. Thus, in this circumstance the debate is not really about the unit rule as much as it is about the benefits of proportional representation over a first-past-the-post system.
Further, note that the unit role provides a certain amount of stability to the system and underscores the importance of two-party politics to the constancy of the current system. Without the unit rule, a third party or third-party candidate of reasonable political strength would almost always prevent either of the two major parties from getting an Electoral College majority.
A third-party candidate’s impact does not necessarily depend upon capturing 20 percent of the electorate’s imagination as Perot did in 1992. Appendix Table 3 illustrates that even if a third-party candidate secures 10 percent of the popular vote as Perot did in 1996, the third-party candidate can prevent either major party candidate from attaining an Electoral College majority. Thus, under a proportional distribution system, Perot could have prevented or delayed Clinton’s Presidency in 1996.
In sum, the unit rule does have a disproportionately negative impact on supporters of third-party candidates. Compared to a proportional distribution system, the unit rule minimizes the voting power of third parties. In exchange, the unit rule provides some stability to the two-party system. This is a feature that, while many find virtuous, others find accursed.

Answering Objections to the Popular Vote
There are a numb of objections to abandoning the EC that I wil address in this section.
A nationwide election is too difficult. This may have been true in the 1700s and the 1800s, but technology makes a nation-wide recount feasible.

Ky Fullerton, law student, Oregon, Oregon Law Review, Summer, 200, Comment: Bush, Gore, and the 2000 Presidential Election: Time for the Electoral College to Go?, p. 743-4
Critics also maintain that the direct vote alternative is fraught with administrative difficulties. One such difficulty is the potential for a nationwide recount in a close popular vote contest. The election of 1960 provided such a scenario. Obtaining a nationwide popular vote figure would have been nearly impossible given Alabama’s eccentric method of counting popular votes and the fraud allegations present in other states. With direct election, the nation would have undoubtedly been subject to numerous recounts and legal battles to determine the 1960 popular vote winner. With technological advances and the possibility of congressional funding for a national voting system, conducting an efficient and timely nationwide recount is far from an impossible task today. Individual states are also currently taking steps to modernize obsolete voting systems. For example, Florida legislators recently approved a bill that eliminates punch-card voting machines and provides new uniform standards for vote recounts. Such improvements will certainly make future recounts faster and more accurate. Furthermore, state election officials are able to complete statewide recounts in a sufficiently expeditious manner. The 2000 U.S. Senate race in Washington state between Slade Gorton and Maria Cantwell illustrates this point. Washington, like Florida, requires a statewide recount when a candidate’s margin of victory is less than one-half of one percent. Washington officials completed the mandatory recount in twenty-four days, eleven days less than the time needed to resolve the Florida vote controversy. The Gorton-Cantwell recount was also prolonged by the fact that Washington absentee ballots need only be postmarked by election day to be valid. Assuming that Congress followed most states and required a national recount only when one percentage point or less separated the top two presidential candidates, the 1960, 1968, and 2000 elections would have been subject to recounts.

The EC protects small states. While this is theoretically true, there is no proof that small states end up being protected.

Michael Herz, law professor and Co-Director, Floersheimer Center for Constitutional Democracy, Cardozo, Cardozo Law Review, May 2005, HOW DEMOCRATIC IS THE AMERICAN CONSTITUTION?: ROBERT DAHL’S HOW DEMOCRATIC IS THE AMERICAN CONSTITUTION?: AN INTRODUCTION, WITH NOTES ON THE ELECTORAL COLLEGE, p. 2522
The second standard explanation is that the Electoral College was designed to protect small states. Professor Dahl does not mention this explanation, and rightly so. The explanation is tied to the smaller states’ disproportionately high number of electors, which flows from apportioning electors according to the number of Representatives and Senators. As a description of the Framers’ goals, this account is extraordinarily weak. For one thing, no one actually said anything to this effect at the convention. Rather, this seems an instance of reasoning from effect to intent. Moreover, as a mechanism for protecting small states it is pretty pathetic, as I shall briefly discuss below. There is little or no scholarly support for this explanation.

And if there was a direct election, their interests would also be protected.

Ky Fullerton, law student, Oregon, Oregon Law Review, Summer, 200, Comment: Bush, Gore, and the 2000 Presidential Election: Time for the Electoral College to Go?, p. 742-3
Supporters of the Electoral College also maintain that the system preserves the concept of federalism that was so important to the framers. Opponents of direct election claim that presidential candidates would focus on states like California, New York, and Texas at the expense of less populous states. They maintain that the Electoral College requires candidates to campaign in nearly every state and address regional issues. This notion that small states would suffer neglect under a direct election plan lacks merit. Common sense dictates that presidential candidates will expend limited amounts of time and effort in states that are dominated by either Republicans or Democrats. For example, neither Bush nor Gore visited Idaho during the 2000 general campaign, an overwhelmingly Republican state with only four electoral votes. In similar fashion, Bush made one brief campaign stop in the Republican stronghold of Utah, while Gore bypassed the state entirely. Even small swing states received little attention in 2000. In Nevada, a poll taken one week before the election showed that Bush’s lead over Gore was within the poll’s margin of error. But during the campaign, Bush visited Nevada only once, and Gore stopped in Las Vegas twice. These data suggest that direct election would actually encourage presidential candidates to visit states where the popular vote outcome is a foregone conclusion. If every vote for President receives the same weight, candidates would campaign wherever additional votes might be obtained. Under a direct election system, states like Idaho and Utah would actually see more of the presidential candidates than under the Electoral College system.

The electoral college promotes federalism. While true in theory, there is no statistical support for the claim that the electoral college promotes federalism.

Harvard Law Review, June, 2001, RETHINKING THE ELECTORAL COLLEGE DEBATE: THE FRAMERS, FEDERALISM, AND ONE PERSON, ONE VOTE, p. 2526-7
Rather than revisiting each of these arguments, n6 this Note focuses on federalism justifications for the electoral college and asserts that the empirical biases of the institution in favor of small or large states are inherently unpredictable in a way that renders any such rationale untenable. Although there are other reasons to support the electoral college, this Note argues that the lack of a viable federalism justification makes the present system more difficult to defend. Part I introduces the Framers’ rationales for the electoral college and explains that unforeseen developments since the Founding have rendered many of their concerns inapplicable to today’s electoral system. Part II then examines the continuing validity of the principal reasoning that survives: the desire to balance the interests of large and small states in a federal system. To perform this task, Part II enlists the assistance of statisticians and political scientists who have tried to discern large-state or small-state bias in the electoral college, and concludes that the system is unpredictable so as to preclude any federalism justification. Part III considers other justifications for the electoral college and concludes that without a clear, original basis in federalism, the overall case for the electoral college becomes more difficult to balance against a popular-vote alternative predicated on the equality and transparency values of the one-person, one-vote principle.

We should defer to/respect the framers. There are three reasons not to defer to the framers on the question of the electoral college.

Harvard Law Review, June, 2001, RETHINKING THE ELECTORAL COLLEGE DEBATE: THE FRAMERS, FEDERALISM, AND ONE PERSON, ONE VOTEp.2529-30
Since the ratification of the Constitution in 1789, political and social trends that the Framers failed to foresee – both in the early days of the Republic and more recently – have rendered many of their concerns about the electoral process inapposite today. First, to the extent that the Framers’ original intent was to benefit the South through a compromise that incorporated and gave constitutional recognition to a regime of slavery, their rationale is illegitimate in modern times. In fact, one commentator has argued that the shadow of slavery taints the entire electoral college system.   Second, the Framers’ concerns about allowing a provincial and inaccessible population to vote are also invalid today. The nation no longer has a primitive communications infrastructure, an uninformed electorate, or transportation problems that impede candidates from traveling the country. The gradual democratization of the country and the increasing importance of popular voting reflect a fundamental change in the nation’s political philosophy that has been effectuated through the numerous amendments to the Constitution in the 200-plus years since the Founding. In fact, the emergence of a national two-party system has eviscerated most of the features of the electoral college that the Framers created to suppress popular voting and to carve out a role for Congress in presidential selection. The two-party system has made the election of a president with an electoral college majority quite easy: if all the nation’s electors vote for one of two candidates, one candidate will nearly always receive the necessary majority. Only the elections of 1800 and 1824 have gone to the House of Representatives – all others have been decided in the electoral college. In addition, the two-party system has eliminated the role of electors as an intermediate body of contemplative actors to filter the preferences of the uninformed electorate. By recruiting slates of electors in every state pledged to vote for their candidate, the two national parties allow voters (or state legislators) to choose one or another candidate directly. Unlike the prior two rationales, however, the Framers’ third justification for the electoral college – the desire to strike an appropriate balance between large-state and small-state representation in presidential selection – remains valid and important. The breakdown of the congressional selection procedure, however, has fundamentally altered the specific large-state/small-state balance that the Framers envisioned. What remains today of the Framers’ vision is a much narrower small-state advantage than originally intended, deriving entirely from the Senate add-on provision of the electoral college. As Part II argues, however, the emergence of the popular-vote, winner-take-all system for state selection of electors, which the Framers did not anticipate, has rendered the notion of any consistent small-state, regional, or federalist protection in the electoral college highly tenuous. Today, all states except Maine and Nebraska (which use the district system) use the winner-take-all system, an important development that many commentators believe produces a significant overall large-state advantage.

The election could end up in the House. There is no real chance that a third party candidate will get enough votes to throw the election to the House.

Ky Fullerton, law student, Oregon, Oregon Law Review, Summer, 200, Comment: Bush, Gore, and the 2000 Presidential Election: Time for the Electoral College to Go?, p. 750-1
In contrast to the current domination of the presidential electoral process by the two major political parties, the framers envisioned presidential elections with numerous candidates. They believed that because an elector would favor a presidential candidate from his own state, a single candidate would rarely obtain a majority of the electoral votes cast. Consequently, the framers expected the House of Representatives to decide most elections. Only twice, however, in 1800 and 1824, has a presidential election been thrown into the House. In fact, the only recent election where resort to the House contingent procedure was even a remote possibility occurred in 1968, when George Wallace’s candidacy threatened to deprive the major party candidates of an electoral vote majority. But the possibility of a third-party candidate receiving enough electoral votes to throw an election to the House of Representatives today is remote. Despite the recent independent candidacies of Ralph Nader, Ross Perot, and Pat Buchanan, the two major parties have a virtual lock on the presidential election process. One substantial factor in this domination is the federal matching campaign funds program, enacted by Congress in 1974 in response to President Nixon’s perceived fundraising abuses in the 1972 campaign. To qualify for matching funds during the presidential primary, a candidate must raise a minimum of $ 5,000 in each of twenty states, and the contributions must also be of $ 250 or less and from individual donors.   Candidates who accept matching funds are subject to a spending limit during the primary election. The advantage to the two major parties is even greater in the general election, where a party that obtains twenty-five percent or greater of the popular vote in the previous presidential election is considered a “major party” and qualifies for substantial federal matching funds. Minor parties qualify for a smaller amount if they garner more than five percent of the popular vote in the previous election. In light of the two major parties’ enormous edge in fundraising, the current system strongly favors Republicans and Democrats. Matching funds provided by the government only enhance the financial gulf between the major parties and smaller parties. In addition to enormous amounts of money, parties seeking matching funds must also have a widespread, national organization. The requirements needed to obtain matching funds are a likely reason why no third-party candidate after the advent of federal matching funds has come even remotely close to depriving a candidate of an electoral vote majority. As a result of Perot’s performance in the 1996 election, the Reform Party qualified for federal matching funds in 2000. Perot’s relative success in 1992 and 1996 was in large part self-financed: he contributed millions to both campaigns and limited outside contributions to very small amounts. Further, Perot provided the bulk of the resources necessary for the Reform Party to build an organization large and widespread enough to qualify for matching funds. In short, to obtain federal funds for a presidential campaign, a candidate must be the nominee of a major party or be independently wealthy. Under the current system, the possibility of the House deciding an election between multiple presidential candidates is highly unlikely. Consequently, the federal matching funds system, in conjunction with the Electoral College, perpetuates the domination of presidential elections by the two major parties. This notion alleviates, to some extent, concerns that direct election of the President by popular vote will encourage splinter candidacies and promote faction.

“Flyover states.” There are three reasons that the popular vote won’t cause “fly over” states to be ignored.
– Campaigns would spread resources equally and respond to “fly over” states

Congressional Research Service, December 12, 2014, The National Popular Vote Initiative: Direct Election of the President By Interstate Compact, https://www.everycrsreport.com/files/20141212_R43823_4df94a11f0b45b91348c622a380b03da934c92a3.pdf
NPV advocates also assert the compact would provide a practical benefit to non-battleground “flyover” states. With “every vote equal,” NPV maintains that presidential and vice presidential nominees and their organizations would need to spread their presence and resources more evenly as they campaigned for every vote nationwide, rather than concentrate on winning key “battleground” states. They assert that, under the present system, … candidates have no reason to poll, visit, organize, campaign, or worry about the concerns of voters of states that they cannot possibly win or lose. This means that voters in two thirds of the states are effectively disenfranchised in presidential elections because candidates concentrate their attention on a small handful of “battleground” states. In 2004, candidates concentrated over two-thirds of their money and campaign visits in just five states; over 80% in nine states, and over 99% of their money in just 16 states.71

– Campaigns would pay attention to large states they don’t pay attention to now

Congressional Research Service, December 12, 2014, The National Popular Vote Initiative: Direct Election of the President By Interstate Compact, https://www.everycrsreport.com/files/20141212_R43823_4df94a11f0b45b91348c622a380b03da934c92a3.pdf
For instance, they note that California voters seldom see the presidential or vice presidential nominees or benefit from campaign spending because the Golden State, having voted Democratic since 1988, is considered to be reliably “blue,” and Democratic Party candidates are said to take its 55 electoral votes for granted. They also note that Republican candidates make few California appearances, but for the opposite reason: why spend time and resources in support of an apparently hopeless cause? Similar arguments on the Republican side apply to Texas, a state that has voted for Republican presidential nominees since 1980. In 2012, for instance, neither presidential candidate, nor his vice presidential running mate, appeared in California or Texas during the general election campaign. By comparison, Ohio and Florida, closely contested battleground states, received, respectively, 73 and 40 candidate appearances.72 According to Every Vote Equal’s analysis of campaign appearances, the 2012 major party candidates for President and Vice President appeared at a total of 253 campaign events during the general election campaign, but they only visited 12 states; 38 states and the District of Columbia were bypassed during the campaign.73

— The current system depresses turnout in “fly over” states

Congressional Research Service, December 12, 2014, The National Popular Vote Initiative: Direct Election of the President By Interstate Compact, https://www.everycrsreport.com/files/20141212_R43823_4df94a11f0b45b91348c622a380b03da934c92a3.pdf
NPV advocates also maintain that the concentration of campaign resources, advertising, and candidate appearances in battleground states depresses turnout in “flyover” states. Every Vote Equal notes that in 2012, voter participation in battleground states, as identified by political commentator Charles Cook, averaged 67%, while nationwide turnout averaged 59.4% in the nation as a whole.74 The NPV manifesto further cites a Brookings Institution study of the 2004 presidential election in support of its argument, “Because the electoral college has effectively narrowed elections like the last one to a quadrennial contest for the votes of a relatively small number of states, people elsewhere are likely to feel that their votes don’t matter.”75 It should be noted, however, that a range of other political, social, cultural, and economic factors may also contribute to the disparity in turnout between battleground and non-battleground states.

The public is too stupid. Today most of the public is literate.

Paul Boudreaux, law professor, Stetson, Marquette Law Review, Fall, 2004, LEAD ARTICLE: THE ELECTORAL COLLEGE AND ITS MEAGER FEDERALISM, p. 206-7
In assessing the work of the constitutional convention, hindsight reveals a number of changes in society and politics that the delegates did not anticipate. First, the assumption that the people are too ignorant to directly elect the President strikes modern sensibilities as un-American. But of course even relative democrats in the eighteenth century were skeptical of too much democracy. This was an age when the historical experiments with republics, such as ancient Greece, seemed unstable, and this was an age when many Americans were illiterate and knew little of national politics. Despite the original prominence of the “ignorance” argument, however, today’s federalist supporters of the electoral college do not focus on this rationale, for obvious reasons. Fears of foreign intrigue or undue influence by groups such as the Society of the Cincinnati likewise seem quaint. The delegates of 1787 did not know that the United States would rise from being a breakaway outpost of European civilization to a great world power, or that the American political system would prove to be among the best in history in avoiding military influence over the government.

And citizens have easy access to information on candidates from other states.

Ky Fullerton, law student, Oregon, Oregon Law Review, Summer, 200, Comment: Bush, Gore, and the 2000 Presidential Election: Time for the Electoral College to Go?, p. 748-9
Moreover, due to the limited means of communication present at the time of the founding, the framers believed that citizens would have limited access to information about candidates from other states. Because the top two vote-getters became President and Vice President, the obvious solution to this problem was to require a voter to vote for two different candidates from different states. This proposition presented the framers with a dilemma. An individual would undoubtedly have the necessary information about a candidate from his state, but how could a voter in Georgia learn about a presidential candidate from New York in 1787? The establishment of the Electoral College was undoubtedly, at least in part, an effort to address this problem. A small number of electors, prominent in government and chosen by the state legislatures, were much more likely to be familiar with the various presidential candidates than the average citizen. The delegates alleviated their fear that electors would blindly support home-state candidates by requiring electors to vote for at least one candidate who was not an inhabitant of their state. America is today, of course, a markedly different nation than it was in 1787. National broadcast and print media provide extensive coverage of presidential campaigns from start to finish. From the moment a presidential candidate declares his candidacy, his every move, speech, or comment is transmitted to all fifty states and around the world by newspapers and television. The television networks beam presidential debates into millions of homes nationwide. Americans certainly do not lack information about a candidate’s views, positions, or personal life. Particular mediums, such as the cable network C-SPAN, provide continuous programming exclusively related to government and politics, including substantial coverage of presidential elections. Current voting trends also suggest that residents of a particular state are unlikely to vote for a candidate solely because he or she is a resident of the voter’s home state. Indeed, Al Gore did not carry his home state of Tennessee in 2000, and, ironically, his failure to capture the state’s eleven electoral votes cost him the election.

Electors are not the “wise people” of the community

Ky Fullerton, law student, Oregon, Oregon Law Review, Summer, 200, Comment: Bush, Gore, and the 2000 Presidential Election: Time for the Electoral College to Go?, p. 747-8
Another factor supporting direct election of the President is that the role of the presidential elector has changed substantially since 1787. The Constitution’s only restriction on the eligibility of electors is that “no Senator or Representative, or Person holding an Office of Trust or Profit under the United States, shall be appointed an Elector.” By entrusting the selection of electors to the state legislatures, the framers undoubtedly believed that, the “wise men of the community” would select the President. As early as 1826, however, a Senate committee noted that electors were “‘usually selected for their devotion to party, their popular manners, and a supposed talent for electioneering.'” Today, the criteria for selecting presidential electors have changed dramatically. The most common traits shared by current electors are service to and financial support of a political party. Longley describes today’s electors as “little more than a motley state-by-state collection of political hacks and fat cats usually selected because of their past loyalty and support for their party.” Noted author James Michener, a Pennsylvania presidential elector in 1968, agreed with Longley’s assessment:
In my case I was chosen to be an elector because I had worked hard for my party. I was passed upon by no public hearing, no primary vote, no board of qualifications, no review of my prior public service. My finest credentials were that each year I contributed what money I could to the party. The notion that the nation somehow relies on the special knowledge and expertise of presidential electors in choosing a President is weak at best. Further, although electors are not bound by the Constitution to vote for a particular candidate, the reality is that electors rarely exercise discretion and almost always vote for their party’s nominee. Between 1796 and 1988, only nine presidential electors voted for a candidate not supported by their party, and no faithless elector has ever changed the outcome of a presidential election. Nevertheless, the danger remains that in a close election, an elector seeking publicity for a particular cause or personal notoriety might switch or withhold his or her vote, causing a popular vote winner to lose or throwing the election into the House of Representatives.