The polls indicate that if President Donald Trump wins reelection Tuesday, it’ll probably be because he manages another narrow victory in the Electoral College while losing the popular vote, just as he did four years ago. This would surely trigger calls for changing the peculiar manner in which we select the leader of this country.
But despite the many problems with the Electoral College, there’s a good reason to keep it, even at the cost of sometimes having the president elected by a minority of voters: The system prevents utter electoral chaos every time the popular vote is close nationwide.
By partitioning the popular vote among 51 constituencies (50 states plus the District of Columbia), the Electoral College structure serves to restrict vote recounts in close elections to a very few states — just the ones that are closely divided in the popular tally and have sufficient electoral votes to affect the outcome.
The consequence of this vote partitioning can be appreciated in the standoff between George W. Bush and Al Gore in 2000. The electoral vote was closely divided, but the outcome was in dispute only in Florida — which had 25 electoral votes then — and the recount was limited to that state. At the national level, the popular vote was also close: Gore led Bush by a mere .5% of the vote. Had the president been determined by the national popular vote, a nationwide recount would have been likely, requiring tabulation of the 101 million votes cast in the country, along with a consideration of the rejects, with their hanging chads, questionable signatures and issues of voter identity; all this would have taken place against a background of 51 different sets of rules on electoral matters.
Nor was the 2000 election an anomaly. The 1960 contest was even closer — a 0.2% difference separated John F. Kennedy and Richard M. Nixon — and in 1968, the popular vote difference between Nixon and Hubert Humphrey was also less than 1%. A close national popular vote in a presidential election is not a rarity, and the inevitable recount in this circumstance would be convulsive, if not chaotic, to the country.
Yes, the failings of the current system are well known: It distorts the popular vote, giving an advantage to rural states with fewer residents, and has led to two instances in the past 20 years of the election of a president with fewer popular votes than his opponent. It compels candidates to concentrate their efforts on the few states that are competitive, ignoring the interests of large regions of the country. And the very origins of this institution are tainted — it was created as a concession to slaveholding states at the formation of the republic.
Given these shortcomings, it’s not surprising that there has been interest in replacing this troubling anachronism with the eminently more equitable popular vote count. Such a reform would, however, entail a constitutional amendment, a difficult undertaking since passage requires a willing reduction in electoral influence by small-population states, to which they are unlikely to acquiesce. An end-run around the amendment route has been proposed in the form of the National Popular Vote Interstate Compact, a multistate agreement in which each state’s electoral votes would be cast for the winner of the national popular vote. Already endorsed by 15 states, this agreement is slated to come into effect once the participating states account for 270 electoral votes, the minimum for electing a president.
But abandoning the Electoral College might not improve things. It could even make them worse. A change in election rules will stimulate behavioral adaptations by the political parties to the new electoral environment. At present, in one-party states where the winner of the presidential contest is not in doubt, there is little motivation for the dominant party to inflate its vote count. But if the electoral decision were determined by the national popular vote, the calculus would change — since it would no longer matter where a vote is cast.
Indeed, this would heighten the attractiveness of odious, if marginally legal, strategies that have sprung up in recent years with the objective of diminishing voter turnout — such as the exclusion of felons, burdensome identification laws, challenges to mail-in voting. Up to now, manipulations of this sort have affected the presidential contest mainly through their presence in the states that are “in competition” and can influence the election outcome. However, if the electoral decision were to be based on the national popular vote, we can expect such laws, along with outright vote tampering, to be instituted more extensively, especially in states dominated by a single political party, since the institutional arrangements for ensuring a fair contest are weakest in such settings.
The Electoral College is a flawed anachronism, and its failings are manifest. Understandably, people tend to focus attention on the vexing deficiencies of the current system, since we are familiar with — and exasperated by — them. But any change will involve a trade-off. And we should consider the benefits of the current, imperfect system as well as the potential downside of any suggested replacement.
We must ask how much it matters to be certain to have a president selected by the popular vote majority — surely, the gold standard in a democracy — if the rules also bring the potential for a chaotic and perhaps indecisive outcome in close national votes, in which even a recount might remain inconclusive and the recount process would probably be fraught with simultaneous legal challenges and court fights in many states. Alternatively, we can keep the Electoral College as a buffering institution, which would limit the scope and havoc of a recount, but at a cost of sometimes producing a minority president. This is the trade-off we must weigh.