PR for the electoral college? No thanks

The following is a guest post by Nathan Batto

One of the proposals sometimes mooted (by disaffected Democrats) is that electoral votes should be allotted proportionally within each state according to the popular vote. Obviously, since Clinton won the popular vote, she would then win the election!

Not so fast. Let’s run the numbers. There are several different formulae to calculate proportional representation. D’Hondt is quite favorable to big parties; Ste. Laguë is quite favorable to small parties.

Ste. Laguë: Clinton 264, Trump 262, Johnson 10, Stein 1, McMullin 1.
D’Hondt: Clinton 267, Trump 267, Johnson 2, Stein 1, McMullin 1.

In both cases, no one gets a majority. The race would then be thrown into the House, where each state delegation would get one vote. Since Republicans hold majorities in 31 state delegations, Trump would almost certainly be elected president.

Of course, this assumes that no voters changed their votes, but of course small parties would almost certainly get more votes under this system. What that would do is make it very, very hard for either big party to get 270 EVs. Almost every election would be thrown into the House, where the Republicans hold a structural advantage in state delegations due to their popularity in rural America (read: small states). In other words, this reform would make it much harder for the Democrats to win the presidency.

[Nathan notes that the exact numbers could change based on updated vote totals. See comments for a point regarding possible thresholds. –MSS]

22 thoughts on “PR for the electoral college? No thanks

  1. I pointed out in discussion with Nathan that if states did adopt PR, it is likely that at least the larger ones would adopt thresholds, as most have for otherwise “PR” allocations in presidential primary delegations.

    Nathan responded: In the D’Hondt scenario, Johnson got one vote from CA and one from TX while Stein got one from CA. When I rerun CA and TX without Johnson and Stein, Trump gains 2 and Clinton gains 1. So the result with a threshold under D’Hondt is Trump 269, Clinton 268, McMullin 1. There is still no majority. If you further assume a monstrous threshold of 25% in Utah to eliminate McMullin, then Trump wins 270 to 268.

  2. Pingback: So, what about districts? | Fruits and Votes

  3. Surely, under this sort of scenario, the “third parties” would negotiate for concessions in exchange for passing their electoral college votes to another presidential candidate?

    So Johnson could be taking bids from both Clinton and Trump to see what compromise he’s offered in exchange for his EC votes – possibly the Vice-Presidency?

    Certainly Clinton would have a strong incentive to bid on the grounds that the House will presumably vote for Trump.

  4. Then I guess the question is whether such patterns of bargaining could ever become equilibrium. The opportunities for the president-elect to renege on deals would be high, and the opportunities for enforcement few and far between. Cabinet appointments are confirmed by the Senate, where the third parties would have no leverage. And once appointed, they can be fired by the president and there would not be much the third parties could do in retaliation other than hope for a close result in the next election and more electoral-college leverage.

  5. This post makes a big assumption: that we would have proportional voting in the Electoral College without changing how the House is elected. I find it hard to believe that we would have a proportional electoral college with a low threshold without a move to proportionality in the House.

    Also, as far as thresholds go, the reigning number in the states seems to be 15%. That’s the number to get delegates in the majority of proportional primaries as well as the percentage needed to qualify for general electionary debates.

    I think the most likely prospect for change to the Electoral College will come when/if Texas flips Democratic (the current demographic trends would suggest this happens around the middle of the next decade if the GOP does not begin to gain more of the Latino vote). That would give the Democrats such a built-in advantage that the GOP would almost certainly demand change.

    • The alternative is if National Popular Vote comes into effect there may be some push from the Republicanis, who would be somewhat disfavoured by NPV, to abolish the electoral college and push the election into the house or a runoff if no-one secures an absolute majority of the popular vote.

      • Still waiting for the inevitable three dozen eminent Oxford historians to pop up and explain to the US Republicans that the non-negotiable criteria for a democratic electoral system are that (b) it must be simple enough that a six-year-old can explain it and (b) it must, absolutely must, always ensure that the candidate with the most votes wins.
        Bonus points if one of the signatories has recently inherited a seat in the House of Lords.

      • Almost as persuasive, Tom, as the intensive pearl-clutching over the mere possibility that some Trump electors are being lobbied to vote for the popular vote winner.

      • Indeed ironic. Reversing a plurality lead because of the bonus for winning smaller states = good. Reversing a plurality lead by flipping individual Electors = bad. And yet the latter is arguably closer to the Founders’ intent, in the sense that the population disparity between the biggest and smallest Colonies/ States in 1789 was 12:1 whereas in 2016 it is closer to, I think, 80 to 1.

  6. Actually, Tom, the ratio is *only* 66 to 1. The House is malaportioned enough to drop it to 53 to 1 and in the Electoral College just 18.3 to 1.

    • Sorry, then, stand corrected. My larger point remains – people might tolerate some population loading between units with an N to 1 disparity in population, but not with a much larger disparity. (I suspect that population disparity is also more palatable to many when it’s combined with huge disparities in area. Here in Australia, I’ve lost count of the number of people who assume that the Northern Territory, which is huge in area, has more people than the Australian Capital Territory: whereas the city-state of Canberra, though small in area, has just under twice as many).
      While population is, in legal theory, irrelevant to the status of Statehood, it does seem to play some political role. I believe that Alaska and Hawaii were only admitted when their populations seemed respectably large and even have some recollection of reading that Congress in the 19th century by statute required a particular population threshold – forty thousand, it may have been? – before a Territory’s request for admission would be entertained.
      Tangentially, I do find it odd that anyone (looking here at the Bush v Gore majority and its defenders, eg Posner J) could read the US Constitution’s Electoral College procedures (either the Twelfth Amendment or the original clause) and come away with the understanding that the Founders thought it constitutionally imperative that a president must be elected without delay, however tenuous his (or her) claim to majority support…

      • jd

        Both entireties are treated as if they were states without the minimum representation of 5. The ACT is just under 2.5 quotas (2.4392) and the NT is just over 1.5 (1.5572) quotas.

      • The District of Columbia is the only federal district that is denied representation in the legislature. Vienna and Berlin are states in their own right.

      • The ACT has 2.499 quotas of population and the NT has 1.501 quotas or thereabouts, so both get 2 MHRs. Canberrans can join Utahns in the queue to lodge a complaint… The Howard government also did some fiddling with the Electoral Act to help the NT keep its two seats a few years back, which the Parliament is allowed to do for Territories even though anything but strict proportionality to population (on a “nearest whole number greater than 4 seats” basis) is verboten in relation to States. The Australian Constitution has a lot of front doors that are heavily barred, locked, guarded 24/7, and protected by passwords and retinal scans, while out the back it has flyscreen alumin[i]um doors blowing open in the breeze.

  7. Isn’t the electoral college elected by a bloc vote system? Isn’t this quite unusual with a bloc vote system that it would lead to a reverse plurality?

    No wonder why we are stuck with the electoral college, it is too hard to figure out what to replace it with.

    What about increasing the House of Representatives? What would the election result had been? Would a reverse plurality be less likely with a bigger House? Then also re-run it with PR for the electoral college.

    If the next election produces another reverse plurality, would this lead to efforts to abolish the electoral college, what would our options be? I would want a preferential vote system, using the Irish STV for President. The problem with such a system is that too many exhausted ballots may lead to a candidate not winning an overall majority. I doubt most American would want to rank all candidates like they do in Australia.

    • The most common method is, of course, two round majority. I doubt anyone would want that, either, given that the presidential election process is already unbearably long and unbelievably expensive. Still, you have to admire Le Republicans’ primary for its speed and efficiency. Perhaps if Maine’s adoption of AV sticks and expands to other states, Americans will develop a taste for it.

    • If the idea of a completely per-capita national popular vote won’t fly, maybe something like;
      1. One electoral vote for the presidential ticket with the most votes in each State.
      2. Another block of electoral votes (four times the number of States, plus 19 – to preserve the existing weighting) allocated proportionately to the nationwide votes.
      This would solve the problem of wasted votes in the “safe” States while still ensuring that you can’t get a majority of Electoral Votes just by winning the ten largest States out of 50 (as is possible with winner-take-all voting).

      • I cannot see a constituency for this proposal. Those who object to the electoral college are unlikely to support adding another level of complexity to the system. Those who do not object are unlikely to support any move towards ensuring the victory of the *popular vote winner.

        *A more correct term would be winner, but we are forced by the contortions of the system to use the qualification.

      • Sometimes compromises fall between two stools (eg, AV in the UK). Other times they give half a loaf to both sides (eg, AV in Australia).

      • When Megalexandros came to Gordion, he carried a sword, not several lengths of rope and a primer on knot theory.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s