The Electoral College, initially designed as a check on popular will and a guardian of elite prerogative in American government, has delivered the presidency to Donald Trump, a populist outsider who campaigned on the promise to burn the bipartisan political establishment to the ground. Though he is losing the popular vote by a 500,000 vote (and growing) margin, Trump was able to obliterate Hillary Clinton’s “blue wall” in the industrial Midwest, securing a decisive advantage among the electors who will cast the ballots that actually select the President of the United States on December 19.
Democrats are seething at having lost the presidency while winning the popular vote for the second time in the twenty-first century. Liberal media outlets are publishing a flurry of pieces attacking America’s allegedly antiquated system for electing its presidents, and social media is abuzz with petitions to replace it with a national popular vote system. This reaction is understandable: Our democratic instincts tells us that all votes should count exactly equally, and that the person who gets the most total votes should win.
But the case against James Madison’s original design isn’t quite that simple. Here are four points that liberals upset with the outcome might weigh against the visceral indignation of an electoral-popular vote split.
First, there is no way of knowing, in 2016 as in 2000, what the vote margin would have looked like if winning the presidency required winning the most votes rather than winning the most electors. Donald Trump might have held rallies and run advertisements in the predominantly white and economically stagnant regions of California east of San Joaquin and north of Sacramento, or similar areas in the southern half of Illinois and upstate New York, driving up turnout and narrowing Hillary Clinton’s massive victories in these deep blue states. For her part, Hillary Clinton might have done more to boost her performance in demographically favorable regions of solid red states, like the Houston and Atlanta metropolitan areas. The point is that this election was decided by a razor thin vote margin under one set of rules, and it is impossible to speculate whether it would have tilted in one way or the other if those rules had been changed in advance.
Second, scrapping the Electoral College, however desirable it might be in theory, is such a remote possibility (as long as the Republic survives) that liberal attacks on it merely distract resources and energy away from more realistic paths to Democratic recovery. The first route to replacing the Electoral College—through a Constitutional amendment—is hardly worthy of consideration. Even in the extraordinary event that two-thirds of both houses of Congress voted to repeal it (this has never occurred, despite hundreds of efforts, although the Congress elected in 1968 came close) there is simply no possibility of getting buy-in from 38 state legislatures. Between swing states that relish the attention presidential elections bestow on them, small states looking to maintain their outsize influence, and states that feel the current Electoral College map favors their dominant party, it’s hard to see how there will be ever be fewer than 13 states with an interest in maintaining the existing system.
A marginally more realistic path to reform is the “National Popular Vote Interstate Compact,” an agreement between states to allocate all of their electors to whichever candidate wins a plurality of the national popular vote. Thus far, this Constitutional workaround has been approved by ten deep blue states and Washington, D.C., which together control 165 electoral votes. Once states controlling 270 electoral votes sign the compact, it is theoretically put in place, and whoever wins the popular vote will automatically also win an electoral majority. The first problem with this scheme is that, as Nate Silver has shown, the creators of the compact have already scored the easy victories in high electoral vote states like California and New York. It will be extremely difficult, if not impossible, for them to reach the 270 electoral vote threshold needed to activate the compact.
And even if the compact were activated, as the law professor William G. Ross has argued, it might not survive constitutional scrutiny. For one thing, the Constitution requires congressional approval for any interstate compact that infringes on federal power or the power of other states. While the advocates of the NPVIC claim this compact would be exempt from this command, it would certainly be challenged, meaning that unless Congress approved the scheme, the election outcome would likely be decided by the United States Supreme Court the next time the electoral and popular votes are split (hardly a more “democratic” outcome). Finally, there is no enforcement mechanism for the compact, meaning that individual state governments could pull out of it as soon as it became clear that it no longer served their partisan interest in a given election. They could even pull out after the votes were counted and before the electors were allocated. This will likely be fatal to the whole project.
Third, given that the Electoral College is not going away in the foreseeable future, it’s worth highlighting the good that geographic electoral systems might do for our politics. As the axis of political conflict across the West shifts from right versus left to nationalism versus cosmopolitanism, political coalitions are becoming increasingly sorted by geography, with Republicans dominating less-populated rural and exurban counties and Democrats packed into dense urban areas. Geographic polarization makes political conflict more intense and compromise more elusive, because it means that people with divergent views are less likely to know or even interact with people who disagree with them. Early reports suggest that this deeply divisive election had the highest level of geographic polarization on record, and that this polarization tilted the electoral map in Donald Trump’s favor: Hillary Clinton’s supermajorities in big blue state metropolitan areas ran up her popular vote margin but didn’t add to her electoral vote total.
Partisan agendas don’t exist in isolation; they are responses to political institutions, like the Electoral College, that set the rules for attaining political power. Going forward, the Electoral College might moderate the violence of the nationalist-cosmopolitan clash by forcing the Democratic Party to court blue collar states more aggressively, instead of doubling down on their already-sizable advantage in metropolitan areas. In other words, it gives the Democrats an incentive to build a more geographically diffuse political coalition (and increased competition from Democrats in rural exurban areas might in turn force Republicans to think more about how to compete in cities). In this election, Trump’s America and Clinton’s America were unrecognizable to one another. In a popular vote system, there would be no obstacles against the parties engaging in an arms race to widen this chasm for political gain. It’s possible that because of the Electoral College verdict this time around, the next Democrat to run against Trump will work to build more appeal outside of metropolitan areas, blunting the trend toward the mutual ghettoization of today’s red and blue communities.
Fourth, in addition to the usual concerns about a disastrous national-recount scenario, a national popular vote system could produce a number of unanticipated consequences. The winner-take-all Electoral College system creates major obstacles to third-party presidential candidacies. Scrapping it would lead to stronger third-parties vying for the presidency, as these parties wouldn’t need to win any states to register on the electoral scoreboard. As a result, it’s possible that no candidate would come close to getting a majority of the popular vote. America could then regularly end up with plurality presidents with support from thirty percent (or less) of the voting public. Parliamentary systems manage this problem by requiring coalitions to form a government. The party that wins the most votes in the first round doesn’t immediately win power; it must create a coalition with other parties so that together, they represent a majority. (In a number of European countries, far-right parties are kept out of power despite having a plurality of popular support because the governing coalition excludes them). In America, there is no such mechanism. Popular vote champions looking to avoid minor plurality presidencies (the legitimacy of which might also be challenged on democratic grounds) would need to also seek to implement a runoff election or else scrap the entire U.S. Constitutional structure.
In 1951, the economist Kenneth Arrow demonstrated that any voting system for selecting one option from among three or more will generate an outcome that could be described as “unfair” from a small-d democratic perspective. The most obvious purpose of representative democracy is nonetheless to ensure that the government responds to the preferences of its citizens, to the extent that these preferences can be captured by elections. But another purpose, as Benjamin Zycher has noted, is to encourage a stable and effective political system—to facilitate the peaceful transition of power, and to encourage coalitions to work together while remaining accountable to the people. The Electoral College—by demanding geographically dispersed coalitions and converting close races into decisive electoral victories—seeks to accomplish all of these things. It may not be the single best system for balancing the competing priorities of republican government, but it’s not clear that a pure national popular vote system would always do much better. And even those of us who are deeply shaken by Tuesday’s result should not forget the subtle ways our system has served us well in the past, and might even continue to do so in the future.