Every four years Americans wake up to the fact that a president can be elected despite receiving fewer votes than another candidate. Until 2000 the electorate couldn’t be blamed for being unaware of this possibility, because it hadn’t happened since 1888. But twenty years ago George W. Bush squeaked into office with a five vote majority in the electoral college even though Al Gore outpolled him by half a million votes. Then in 2016 Hillary Clinton received nearly three million more votes than Donald Trump but still lost by a substantial margin – 304 to 227 – among the electors. Ask a man or woman in the street why this system of electing a president was adopted and how it works and you will almost certainly draw a blank. It’s complicated, but the main point to bear in mind is that the president is elected indirectly. To be sure, on election day Americans think they’re casting a ballot for their preferred candidate. But, technically, what they’re doing is voting for electors pledged to support that candidate. The electors vote a month or so later and in almost all cases cast their ballots for the candidate who carried their state. No matter who won the national popular vote, they have the final say.
The United States prides itself on providing a global model of democratic government. But of the nearly two hundred sovereign states that make up the United Nations it is difficult to think of a single one that elects its chief executive as Americans do. Even countries with constitutions explicitly modelled on the US one have not thought the electoral college worthy of emulation. Liberia, established as a settlement for manumitted slaves, closely followed the American example, but opted for direct election of the president ‘by the people’. The post-World War Two constitutions of West Germany and Japan, their drafting strongly influenced by the American occupying authorities, did not adopt the system. The electoral college (an odd name for an institution whose members only assemble once every four years, in the fifty state capitals) certainly makes the US exceptional.
How the president should be elected was one of the most divisive issues to confront the constitutional convention of 1787. The delegates agreed that the new nation must be a republic, which ruled out a hereditary head of state. Some favoured selection by the legislature, the method used in parliamentary systems, but others feared this would make the president dependent on Congress. The most democratic option, of course, was election by the people (or at least the minority of the population eligible to vote in each state, generally white men with property), but most of the framers believed that unrestrained democracy was as dangerous as tyranny. Placing prominent men of ‘discernment’ between the electorate and the final outcome, Alexander Hamilton insisted, would hold popular passions in check and prevent a demagogue, perhaps beholden to a foreign government, rising to power. James Madison had a more self-interested objection to popular election. The political power of the South, where slaves made up 40 per cent or more of the population, had hugely increased, thanks to a clause adding three-fifths of the slave population to the number of free inhabitants when allocating on the basis of population the seats given to each state in the House of Representatives. Since the slave population would have no impact on the outcome, warned Madison, a Virginia slaveowner, a popular vote for president would deprive the South of ‘influence in the election on the score of the Negroes’.
The electoral college system was adopted shortly before the convention’s deliberations ended, and has remained almost unchanged ever since. Each state was given the right to choose electors by a method it determined (which ended up meaning either selection by the state legislature, or by popular vote). The number of electors in each state was equivalent to that state’s delegation in Congress – two senators plus however many members it had in the House of Representatives. The candidate who received a majority of the electoral vote would become president and the candidate who came second would become vice president. If no one won a majority, the House, with each state casting one vote, would select the president from among the top finishers. Thus, the electoral college imported into the election of the president two undemocratic features from elsewhere in the constitution – the allocation of two senators to each state regardless of population and the three-fifths clause – and added a third, the provision that in the event of a final election by the House, each state, large or small, would have the same influence on the outcome.
The constitution’s framers neither anticipated nor desired the rise of political parties, which they saw as divisive institutions that elevated factional interests above the common good. But parties quickly emerged anyway, and caused havoc in the electoral system. Ever since, instead of men of local prominence and independent judgment, each party has nominated as candidates for elector minor functionaries who can be relied on to vote for their chosen presidential candidates. The electors are not supposed to think for themselves. Not one voter in a thousand can name any of them past or present.
In 1788 and 1792, the state legislatures in most cases chose the electors, and they unanimously made George Washington president. After that, the trouble started. Initially, each elector cast two votes without differentiating between president and vice president, because it was assumed that candidates would compete as individuals, not as representatives of political parties and that the two most qualified would occupy the two highest offices. In 1796 this resulted in the winning candidate, John Adams of the Federalist party, ending up with Thomas Jefferson, leader of the opposition Republicans (not to be confused with today’s party), as vice president. Four years later, the Republican ticket consisted of Jefferson for president and Aaron Burr for vice. They outpolled Adams and his running mate, Charles Pinckney, but in order for Jefferson to become president, one or more Republican electors had to avoid voting for Burr. They failed to get the message. Jefferson and Burr both ended up with 73 electoral votes, sending the contest to the House of Representatives. Rather than withdrawing, Burr schemed to become president with Federalist assistance. Only after 35 indecisive ballots did Hamilton, who disliked Jefferson but thought Burr incorrigibly dishonest, convince enough Federalists to abstain, thus allowing Jefferson to be elected. This set in motion a train of events that culminated in the 1804 duel in which Burr took Hamilton’s life. It also led to the adoption of the Twelfth Amendment, requiring electors to vote separately for president and vice-president, in recognition of the fact that candidates were already running as party tickets and would continue to do so.
The framers had assumed that the House would decide most elections because in a vast, diverse nation it would be difficult for any candidate to win a majority of the electoral vote. But after presidential elections became party contests, nearly all produced a clear winner. After 1800, the only time the House chose the president was in 1824, when the party system was temporarily in disarray. John Quincy Adams, who had come second to Andrew Jackson in both the popular and electoral votes, struck a deal with Henry Clay, who came fourth, giving Adams a majority of the House votes. Adams then named Clay his secretary of state. What Jackson’s supporters called the ‘corrupt bargain’ was precisely the kind of political manoeuvring the framers had hoped to avoid.
Almost from the beginning, there were efforts to game the electoral college system. In the early republic, states switched back and forth between popular and legislative selection of electors, using whichever seemed to favour their preferred candidate. (When the legislature chose the electors, the majority party could simply assign the state’s electors to its preferred candidate.) In 1836, the Whig Party ran four regional candidates for president, in the hope that together they would prevent the Democratic candidate, Martin Van Buren, from winning a majority of the electoral votes (they didn’t succeed). In 1864 and 1876, Congress admitted a thinly populated territory (Nevada, then Colorado) as a state shortly before election day to bolster the Republican candidate’s electoral vote. The constitution, moreover, failed to explain what should happen if the result in a state was contested. This came about in 1876, when disputed returns from three Southern states made it impossible to know who had been elected president. After months of political crisis, Congress appointed a 15-member electoral commission to determine the outcome – a procedure with no basis in the constitution. Rutherford B. Hayes, a Republican, became president and as part of the ‘bargain of 1877’ his party agreed to recognise Democratic control of the disputed state governments. This marked the end of Reconstruction in the South.
By the 1830s, ‘democracy’ had lost its pejorative implications and every state bar South Carolina was choosing its electors by popular vote. Alongside this development came the tradition, not required by the constitution, that the candidate who carried a state received all of its electoral votes. Sometimes called the ‘general ticket’, the winner-takes-all system in the electoral college has been near ubiquitous for almost two centuries – today only Maine and Nebraska allocate some of their electors by results in congressional districts. (In 2008, Barack Obama carried a Nebraska district, winning one of that solidly Republican state’s five electoral votes.) Winner-takes-all maximises a state’s impact on the outcome, but also makes more likely a mismatch between the winner of the popular vote and the electoral vote. A candidate can carry a dozen or so large states by small margins and capture the presidency while trailing far behind in the popular vote. This is what happened in the momentous four-candidate election of 1860. Abraham Lincoln received virtually no popular votes in the slave states and only 40 per cent nationally. But by carrying the entire North, he captured an electoral vote majority. Indeed, if the popular votes of the other candidates had been combined and given to one of them, Lincoln would still have become president even though 60 per cent of the electorate opposed him.
Winner-takes-all discourages the emergence of third-party candidates unless they have a regional base: in 1992, Ross Perot, running as an independent, received nearly twenty million votes (19 per cent of the total), but no electoral votes since he failed to carry a state. It also has a powerful effect on the way that presidential campaigns are conducted. For reasons ranging from tradition to demography and ideology, the winner in most states is predictable well before election day. Neither candidate sees much point in campaigning in reliably ‘red’ or ‘blue’ states, since even a loss by a narrow margin translates into no electoral votes. As a result, the contest is confined to half a dozen or so ‘swing’ or ‘battleground’ states that both candidates have a realistic chance of carrying. In 2016, two-thirds of the campaign events held by Clinton and Trump took place in only six states. This year, the swing states include Arizona, Florida, Michigan, North Carolina and Pennsylvania. Voters who live in places like my home, the Democratic stronghold of New York, are essentially ignored. To be sure, I have the luxury of ‘throwing my vote away’ on a minority party candidate, knowing that this will not affect the electoral vote tally. (Things would be different if I lived in Florida.) Not surprisingly, voter turnout is higher in battleground states.
For most of American history, the electoral college system has enhanced the political power of white Southerners. Without the votes of the extra electors that resulted from the addition of three-fifths of the South’s slaves to the population calculation, for example, Jefferson would not have defeated John Adams in 1800. In the late 19th century, the Southern states systematically stripped the right to vote from black citizens in flagrant violation of the Fifteenth Amendment, enacted during Reconstruction, which outlawed denial of the franchise on the grounds of race. But this did not affect these states’ representation in the House, since it is based on total population, not on voters or the number of Southern electors. Ironically, the abolition of slavery increased Southern political power because the entire black population, not just three-fifths of it, was now counted in the allocation of House seats and electoral votes. Like the three-fifths clause, disenfranchisement allowed the white South to benefit politically from the presence of the black population while denying it any semblance of democratic rights. For much of the 20th century, the Southern states resembled a series of rotten boroughs, whose tiny electorates wielded disproportionate power in Congress and in the election of the president. (According to the Fourteenth Amendment, states that deprive significant numbers of citizens of the right to vote are supposed to lose a portion of their congressional representation and electors. But this penalty has never been enforced.)
Given its undemocratic nature and long history of dysfunction and racial bias, it isn’t surprising that almost from the start proposals began to circulate about changing the way electors were chosen, or even doing away with the electoral college entirely. Over time, more than eight hundred such amendments have been introduced in Congress. Amending the constitution is a daunting task, requiring the approval of two-thirds of Congress and three-quarters of the states. But it has nevertheless been accomplished 27 times, effecting changes that have significantly democratised American politics: extending the right to vote to African Americans, women and 18-year-olds; shifting the election of senators from legislatures to voters; barring the imposition of poll taxes and allocating electoral votes to residents of Washington, D.C. But the stark fact is that with the exception of the Twelfth Amendment, which only tweaked the system, the strange way we elect the president has survived intact for over two centuries. These two new books try to explain why.
Alexander Keyssar’s Why Do We Still Have the Electoral College? examines efforts to change or abolish the system. Keyssar, who teaches at Harvard, is the author of The Right to Vote, which twenty years after publication remains the standard account of the history of suffrage in the United States. His new book is comprehensive and full of historical insight. Even specialists in political and constitutional history will encounter surprises. But in telling this story it’s impossible to avoid repetition. Madison described the debates about the presidency at the constitutional convention as ‘tedious and reiterated’, a comment that can be applied to the entire history of efforts at reform. The problem is exacerbated by the book’s partly chronological, partly thematic structure.
As Keyssar shows, the most common proposal has been for proportional allocation of each state’s electoral votes. Such proposals typically give one elector to the winner in each of a state’s congressional districts, as Maine and Nebraska currently do, with two chosen statewide. Election by congressional district would obviate the main problem of the winner-takes-all system, which is the effective disfranchisement of millions of voters whose ballots do not translate into electoral votes. It would undoubtedly increase the number of contested states and thus voter turnout. Allocating electors by congressional district, however, would introduce the problem of gerrymandering into the election of the president. In almost every state, state legislatures draw district lines. And ever since the early days of the republic, the dominant party has drawn them so as to maximise its electoral prospects. Today, thanks to sophisticated computer analysis of voting returns, politicians can effectively choose their voters rather than the other way round. Redistricting takes place every decade, when the census determines how many members of the House each state will be given. The 2010 elections gave Republicans control of a majority of state governments, and they proceeded radically to redraw district lines. In such circumstances, the district system would not eliminate the possibility of the loser of the popular vote becoming president. If electors had been allocated by congressional district in 2012, Mitt Romney would have been elected even though he trailed Obama by five million popular votes.
To avoid this problem, Henry Cabot Lodge, the Republican Senator for Massachusetts, proposed in the 1940s that each state’s electoral votes should be automatically distributed in proportion to the popular vote in that state, with a national run-off if no candidate received 40 per cent overall. Race played a major and somewhat paradoxical part in the debate. Lodge hoped that his plan would enable the Republicans to pick up electoral votes in the then solidly Democratic South, especially if black people regained the right to vote there. Nonetheless, some Southern Democrats initially supported the measure, believing it would weaken the power of black voters in the urban North. The massive migration of African Americans from the South to Northern industrial cities, where they enjoyed the right to vote, coupled with a continuing shift in their allegiance away from the party of Lincoln, had changed the political configuration of states like New York, Illinois and Michigan. ‘There are enough Negroes in New York City,’ proclaimed Ed Lee Gossett, a congressman from Texas who introduced Lodge’s measure in the House, to determine the electoral vote of the entire state. ‘With all due deference to our many fine Jewish citizens,’ Gossett added, the same was true of them. In February 1950, the Senate approved Lodge’s proposal, the first time in more than a century that either chamber had passed a constitutional amendment to change the way we elect the president. But the proposal died in the House. Liberals, as well as black and Jewish organisations, became convinced that it would weaken the power of Democratic urban enclaves in the North without affecting the South’s ability to continue to prevent black voting.
Only in the mid 20th century, Keyssar shows, did a national popular vote become the preferred alternative for the electoral college’s detractors. Thanks to the civil rights revolution of the 1960s, Southern blacks finally regained suffrage, weakening the advantage the electoral system gave to white voters there. Then in 1968, George Wallace, a pioneer of white backlash politics, won 46 electoral votes as an independent candidate. Wallace didn’t succeed in throwing the election into the House, where he hoped to influence the outcome, but the prospect of this happening in future led to an upsurge of support for replacing the electoral college with a popular vote for president. The leading proponent in Congress was Senator Birch Bayh, a liberal Democrat from Indiana. But support for his proposal transcended party and ideological lines. President Nixon endorsed it, along with the US Chamber of Commerce, the League of Women Voters and the American Bar Association. A Gallup poll found that 81 per cent of respondents favoured the change.
Bayh’s amendment passed the House in 1969. More than half of the votes against it came from Southern Democrats who hoped that the enfranchisement of blacks could somehow be reversed. In the Senate, three segregationists – Strom Thurmond of South Carolina, James Eastland of Mississippi and Sam Ervin of North Carolina – mobilised opposition. (Ervin’s role may surprise those who remember him only for his principled part in the Nixon impeachment investigation. But before that he was mostly known as an outspoken opponent of racial integration.) As with the Lodge amendment, black leaders outside the South joined the campaign against reform, fearing a diminution of their political influence in the Northern industrial states. The Senate failed to break a filibuster and the amendment died.
Keyssar opens his book with a warning not to expect an analysis of current debates or a clear prescription for change. For these, one can turn to Let the People Pick the President by Jesse Wegman. A member of the New York Times editorial board, Wegman is a fluid writer who manages to make constitutional debates and the history of political parties lively, even amusing. His account skims much of the history related by Keyssar but offers a sustained argument in favour of the latest proposal to replace the electoral college, the National Popular Vote Interstate Compact. States that sign up to the compact pledge that their electors will cast their ballots for the winner of the national popular vote, regardless of whether they carry their state or not. The compact will go into effect when states with a combined 270 electoral votes – a majority of the electoral college – have joined. The advantage of this plan is that it avoids the laborious process of constitutional amendment. As of today, 15 states and the District of Columbia, representing 196 electoral votes, have joined, so there is still a long way to go.
Wegman devotes considerable space to debunking misconceptions about the electoral college. The most common is that the system benefits small states, because, since they start with the two votes representing their senators, they have more influence on the outcome in proportion to their population than if a president were elected directly. This belief is a major obstacle to winning over the three-quarters of the states required to change the constitution. Wegman argues persuasively that the winner-takes-all system negates the small state advantage. It makes much more sense for candidates to focus on populous states when winning them by even a narrow margin yields an electoral vote bonanza. In 2000, Bush’s majority of 537 out of six million votes cast in Florida gave him all of the state’s 25 electoral votes.
What is to be done? Keyssar refers briefly to the compact but does not seem to consider it a viable alternative. It carries a whiff of duplicity – many citizens would be outraged if a candidate who failed to carry their state nonetheless received its electoral votes. Whatever the plan’s shortcomings, however, it would encourage both parties to maximise turnout in every state and ensure that whoever wins the popular vote becomes president. But the hyper-partisanship of current US politics makes agreement on any proposal for change unlikely. All the states that have endorsed the compact are Democratic strongholds, not surprisingly, since the party’s candidates have won the popular vote in six of the last seven presidential elections. For the same reason, Republicans are convinced that the current system favours them. The two most recent Republican national platforms oppose any change in the electoral college system.
As another presidential election looms, these books deserve a wide readership. But the electoral college is only one symptom of a far deeper problem. American democracy is sick in ways that go well beyond the way the president is chosen. The symptoms include widespread efforts in Republican states to suppress the right to vote and to rig elections, employing such tactics as onerous identification requirements, partisan gerrymandering and the removal of many thousands of citizens from the voting rolls for trivial reasons. A partisan Supreme Court, in addition, has allowed unlimited corporate spending on campaigns and abrogated key parts of the Voting Rights Act of 1965, which restored black suffrage in the South. These problems will not be solved by allowing the people to elect the president, but that would be a valuable first step. Rooted in distrust of ordinary citizens and, like so many other features of American life, in the institution of slavery, the electoral college is a relic of a past the United States should have abandoned long ago.