One outcome of the 2016 election has been renewed interest in reforming the way we elect the President of the US.  It is now fairly commonplace for progressives to call for abolition of the electoral college (EC) so as to award the presidency to the winner of the popular vote. There is good reason to get rid of the EC.  Because every state is guaranteed at least two electoral votes regardless of population, the college over-represents more rural, less populous states, violating the basic democratic principle of one person/one vote.  (The US Senate is even worse in this regard, since it represents all states equally.)

Abolition of the EC would require a constitutional amendment, which in turn requires a two-thirds vote in both houses of Congress and ratification by three quarters of the states.  The actual chances of overcoming these formidable hurdles in any foreseeable future are close to nil, for two overlapping reasons: First, there are enough less populous states to block a change that would dilute their political power.  Second, the EC favors the Republican Party, which has generally not been inclined to advance democratic principles over political expediency. But the practical unlikelihood of achieving a desired objective isn’t necessarily a reason not to advocate for that objective.  Advocacy can make an important political point about the serious flaws in our democracy, flaws that tend to be defended most predictably by one of our two major political parties.

Still, I’m uncomfortable with the idea of electing the POTUS by a simple plurality of the popular vote.  Consider the 2000 presidential election. George W. Bush won the electoral vote even while narrowly losing the popular vote to Al Gore. That is widely and reasonably considered to have been an unjust outcome.  But just suppose that instead of losing the popular vote by almost 600,000, Bush had won it by that same margin.  The method of election, it would seem, would have properly reflected the popular will. But wait: Bush and Gore weren’t the only two candidates running.  There was a third party candidate—leftist Ralph Nader—who won over 2 million votes.  It is reasonable to suppose that most of Nader’s voters would have preferred the centrist Gore to the rightist Bush.  So, in this hypothetical vote breakout, the rightist would have won the election fairly and squarely even though a majority of the electorate preferred a candidate to his left.  The popular vote plurality method of election would not have reflected the popular will in this particular hypothetical case, which could easily have occurred.  A similar situation could easily recur: In only 3 of our last 7 presidential elections did any candidate win a popular vote majority.

The election of the POTUS by popular vote would pose no problem in a straight tw0-party system in which one candidate was always certain to win a majority.  The problem arises when a third party entrant keeps both major party candidates below 50%. The simple abolition of the EC would effectively ignore the votes of 3rd party candidates.  This, to my mind, would be a serious flaw from the standpoint of democratic principle. We need to rectify this flaw in our proposal.  There are various proposals for and experiments with ways of dealing with the third party issue, like asking voters to indicate their second choice, and then factoring those second choices into the election decision, but I think the complexity of such procedures could impede the legitimacy of the outcome.   The simplest and therefore best solution to the problem of reflecting the will of third party voters would be to call for a run-off election between the two leading candidates. Voters for third party candidates would then have an opportunity to switch to their preferred major party candidate.  Run-off elections have a substantial history at the state level in the US, so the idea of a run-off presidential election shouldn’t seem alien to our political tradition.  The process for election of the POTUS should be simple and transparent.  A runoff election meets that bill.


NOTE: A movement to get around the constitutional problem described in the second paragraph above is gaining ground at the state level.  Delaware’s governor recently made his state the13th to sign on to a pledge to award all the state’s electoral votes to the winner of the popular vote. Participating states now represent 184 of the 270 votes needed to elect the president, which is also the threshold at which the compact would take effect. Once it takes effect, the plurality winner would be assured election to the presidency.  I’m all for advancing this compact, which would effect a significant improvement in the way we elect the POTUS. It represents a practical though imperfect solution to the problem of presidential elections that don’t represent the popular will. But  if we are going to advocate a constitutional amendment to solve this problem, we should do it right.




  1. Donald Campbell April 30, 2019 at 9:47 am

    I like your ideas in the context of the existing system, but isn’t it like building a house on an unstable foundation? Wouldn’t a parliamentary system make the expression of political ideas through third parties more effective? And wouldn’t that expression lead to a more sophisticated and responsive political system?

  2. Daniel April 30, 2019 at 10:53 am

    A few thoughts.

    I’m not convinced that it makes sense to tie discussion of the electoral college to discussion of third parties/runoffs. The electoral college has its set of problems. As you point out, it magnifies the power of small states. Less often noted, but perhaps just as bad in my mind, is that it magnifies the power of swing states. With a nationwide popular vote, both democrats and republicans would have reason to care about the difference between a 70/30 and 55/45 even in states (big or small) that are solidly red (or blue).

    The fact that we don’t have any kind of runoff system has its own, quite separate set of problems, which you nicely bring out in reference to the 2000 election.

    Why think that proposals for addressing the first set of problems must be bundled with proposals for addressing the second? Is the idea that we’ll only get one constitutional amendment concerning federal elections, so we’d better pack everything into it that we need? Historically, there have been multiple amendments addressing voting rights (12, 14, 15, 19, 25, 26), so if we’re imagining constitutional amendments, it’s not crazy to imagine two separate ones, one addressing the problems with the EC, the other addressing runoffs. But as you point out, moving to a de facto national popular vote wouldn’t require a constitutional amendment, as it could be achieved via an agreement between states that comprise a majority of the electoral college votes.

    The problems with the EC and the problems with third party preference are analytically separate. And as a matter of political strategy, the most realistic method for addressing the former is via a pact that couldn’t be used to address the latter. So while I agree that some kind of runoff system is a good idea, I’m not convinced that there’s anything wrong with discussing and advocating for moving to a national vote independently of runoffs.

    Also, about the issue of doing a separate runoff election, versus something like ranked preference, it’s worth noting that the latter is used at the state level in Maine, and in a number of localities throughout the US. I don’t know anything about how it’s been received there, but it seems to me too quick to write it off as opaque:

    • tonygreco April 30, 2019 at 12:44 pm

      “The problems with the EC and the problems with third party preference are analytically separate.” Well, yes and no. You raise one problem with the EC–over-attention to swing states–that I didn’t deal with and that is indeed analytically separate from the issue of 3rd party preferences. But if we define the main democratic problem as reflecting the will of the popular majority, then third party voter preferences is an intrinsic part of the problem. My broad point is that a constitutional amendment is a very big deal. If we are going to advocate such an ideal long-term solution with the purpose of better reflecting majority preferences, then we shouldn’t do it knowing that we would figuratively set in stone a change that we consider defective. Again, I am fully supportive of the pact as the more practical approach to reform.

Have a comment?

Required fields are marked (*)