The three fallacies of the popular vote

By Professor Edward Zelinsky

Via Oxford University Press Blog  In light of Secretary Clinton’s victory in the popular vote, prominent voices call for replacing the Electoral College with a direct, nationwide vote for President. Among the distinguished individuals now urging abolition of the Electoral College are former Attorney General Eric Holder and outgoing Senator Barbara Boxer. However, for three reasons, it is wrong to assume that the popular vote total in this or any other presidential election is the same as the result which would have occurred under a direct, nationwide election for President conducted using uniform national rules. Would Secretary Clinton or President-elect Trump have won in 2016 in a direct, nationwide election? We don’t know.

Abolishing the Electoral College would shift the campaigns’ allocation of their respective resources from close, toss-up states to party strongholds. The Trump campaign productively allotted its candidate’s time and its money to such closely-contested states such as Florida, Michigan, Pennsylvania, and Wisconsin. Without the Electoral College, the Trump campaign would have instead sent its candidate and its funds into Trump strongholds like Tennessee and Indiana to increase his already large vote totals in those states. The Trump campaign would also have devoted resources to upstate New York and inland California, Republican-leaning areas of no consequence under the winner-take-all allocation of their states’ electoral votes. The result would have been a vote pattern different from the one which actually occurred under the Electoral College system in 2016 when campaign resources were channeled to toss-up states.

Embracing this logic, the Trump campaign claims that the President-elect would have prevailed in a direct election since his campaign would have allocated its resources differently under such a system. Perhaps this claim is correct. Perhaps not. We simply do not know how a direct election for President would have come out in 2016 since we did not have such election.

There are two additional reasons that the popular vote under the Electoral College system may not reflect the outcome which would have occurred under a direct vote using uniform national rules: various states today use different voting rules. A direct nationwide election for president would change the incentives for many protest voters who cast their ballots for third-party candidates.

The United States today does not conduct a single presidential election under a nationally uniform set of rules. Rather, there are fifty-one separate elections each administered under particular state-established regulations. Comparing the different state tallies obtained under different rules may not quite be adding together oranges and apples. It is, however, at least adding together oranges and tangerines.

Consider, for example, the different approaches to mail ballots in Oregon and West Virginia. Oregon conducted its entire election by mail ballots and awarded its electoral votes to Secretary Clinton. West Virginia, in contrast, issued a mail ballot to a voter only under more traditional, restrictive circumstances reflecting the voter’s inability to get to the polls.

It is wrong to assume that the popular vote total in this or any other presidential election is the same as the result which would have occurred under a direct, nationwide election for President conducted using uniform national rules.

It is problematic to look at the votes cast under these different rules in Oregon and in West Virginia, and declare that these are the outcomes which would have occurred under a nationally-uniform direct election without the Electoral College. Suppose that West Virginia had this year taken Oregon’s liberal approach to mail ballots. Would this have increased Secretary Clinton’s West Virginia votes by stimulating Clinton supporters who were too discouraged to bother voting in an overwhelming pro-Trump state? Or would more liberal mail voting regulations have brought out relatively more of the pro-Trump electorate in West Virginia?

We don’t know. And since we’ll never know, it is fallacious to compare the actual vote totals of these two (and other) states which used different voting rules.

Moreover, the incentives under the Electoral College are different for many voters who cast protest ballots for third party candidates than the incentives such voters would confront under a direct national election of the President. Under the current system, a Republican unenthusiastic about Mr. Trump could in this fashion comfortably vote for Governor Johnson and the balance of the Republican ticket, confident that his presidential vote was symbolic since his state’s electoral votes were destined to go to Mr. Trump or Secretary Clinton. That same voter would have confronted a different calculation under a direct national vote since his vote might then have affected the outcome.

It is instructive in this context to compare the popular vote total received by President-elect Trump with the aggregate vote simultaneously received nationwide by Republican candidates for the US House of Representatives. While Mr. Trump received roughly two million fewer popular votes than did Secretary Clinton, Republican candidates for the House received in the aggregate approximately three million more votes nationwide than did their Democratic opponents. A substantial share of the additional votes cast for Republican House candidates likely came from individuals who simultaneously voted for Governor Johnson for president and their local Republican candidate for Congress.

Would this have altered the final result in 2016? We don’t know which, again, is why we cannot assume that the popular vote which actually occurred in 2016 is the same as the result which would have resulted under a direct election for President conducted under uniform national rules.

There are thoughtful critics and supporters of the Electoral College. Americans should debate about our Constitution and the institutions it establishes. In that debate, Americans should not succumb to the fallacious assumption that the popular vote under the Electoral College is the same vote which would necessarily occur in a direct national election for President.