Timothy Stafford is a Research Fellow with Pacific Forum-CSIS, and a former Parliamentary Researcher to Sir Malcolm Rifkind.

us-election-mapLast week, officials in Michigan certified Donald Trump’s victory in the state, handing him the last electoral college votes that had yet to be determined. The result underscores a central paradox of the 2016 election: a wide margin for Trump in the electoral college, despite a clear lead for Hillary Clinton in the national popular vote.

Split elections in the United States are not unprecedented. In 1876, Rutherford Hayes polled 250,000 fewer votes than Samuel Tilden, but won the Electoral College by a single vote. Twelve years later, Benjamin Harrison received 65 more Electoral College votes that Grover Cleveland, despite losing the popular vote by 90,000. More recently, George W. Bush won the 2000 election with five more votes than Al Gore, despite 530,000 more Americans voting for the Democratic ticket.

Yet what is unprecedented about 2016 is the scale of the divide. Trump is now over two million votes shy of Clinton, yet enjoys a 74 vote advantage in the Electoral College. In short, the level of correlation between the Electoral College count and the popular vote is as weak as it has ever been.

Commentators dismayed by the prospect of a Trump administration have responded with calls for the Electoral College to be jettisoned. Former Vice-President Gore has stressed that a popular vote system would increase participation, and Democrats on the Hill have already introduced legislation calling for electoral reform at the national level

Others have gone further, outlining and agitating for steps that could overturn the 2016 result itself. Noted Harvard Professor Lawrence Lessig has urged electors to simply ignore the statewide results, and vote for Clinton. Some liberal activists have initiated lawsuits designed to test the obligation of electors to vote as instructed, in an effort to shake loose those committed to Trump. In addition, a collection of pro-Clinton electors have formed a designated non-profit to campaign for an Electoral College coup d’etat. Their hope is that by throwing their support to a mainstream Republican candidate such as John Kasich, Trump electors might be persuaded to abandon Trump. Their efforts are beginning to show signs of success. Recently, a Trump elector from Texas announced his intention to become a ‘faithless elector’ by refusing to support him.

Meaningful change in the coming years is not an impossibility. Formal abolition of the Electoral College would require a gruelling campaign to secure passage of a constitutional amendment. Yet it could be brought about in de facto terms should states comprising an Electoral College majority simply decide to award their votes to the winner of the popular vote. This initiative, known as the National Popular Vote Interstate Compact or NPVIC, has already secured the backing of ten, primarily liberal, states.

However, calls to overturn or tamper with the workings of the existing system should be met with caution. Firstly, the wide gap between the Electoral College result and the popular vote is proof that the system designed by the framers of the Constitution is working as planned. The reason the Electoral College was designed in the first place was to preserve the health of the Union, by preventing a highly-localised bloc of voters from steering the affairs of America as a whole. This year, Clinton won just 20 states, fewer than any Presidential victor in the history of the US as a 50-state nation. Perhaps more significantly, she was defeated in the vast majority of individual counties. Whilst the health of the body politic isn’t well-served by having the winner of the popular vote lose the Presidency, it also isn’t bolstered with the election of a candidate who lacks broad support across the country as a whole, let alone one who wasn’t on the ballot in the first place.

In addition, those assuming that a national vote system would have resulted in a different outcome may well be kidding themselves. There is no guarantee that an alternative system would have led to a different result. The Electoral College forces candidates to focus their campaigns on voters in swing states, neglecting parts of the country where the statewide result is a foregone conclusion. Yet under a popular-vote based system, the 2016 campaign would have played out quite differently. For instance, Donald Trump would likely have spent much more time campaigning in the Republican-leaning areas of three of nation’s five most populous states: rural Illinois, upstate New York, and non-coastal California. He could also have have made a major play for urban-based voters in and around big cities in red states. For instance, two red states host five of America’s eleven largest cities: Arizona (Phoenix) and Texas: (Houston, San Antonio, Dallas and Austin). Under a popular vote system, this approach might have been enough to have secure a plurality of voters.

Therein lies the fundamental danger of major change. Moving away from the Electoral College would exacerbate the polarisation that already characterises the US political system. Freed of the need to appeal to moderate or centrist voters, national candidates would invariably turn to 51 per cent strategies that emphasise ‘running up the score’ amongst core supporters, while paying little attention to expanding their coalition through policy moderation. Given the divisiveness already wrought at the Congressional level due to primary elections and gerrymandered redistricting, would-be reformers should think twice before pushing for electoral reform at the national level.