The Electoral College, Alternative Voting Methods, & Interstate Compact Issues

[Aaron Hamlin]

I'd like to start a discussion about presidential elections using a workaround with an interstate compact.

Background: An interstate compact is an agreement between states. In this context, we're talking about states agreeing to allocate their electoral votes according to the national popular vote upon the occurrence of some event. The current approach is to have the compact trigger upon enough states signing on so that their electoral votes total 270.

Issues:

1. When is a system like an electoral college useful, if at all?
2. If an alternative voting method is used, is anything besides approval voting compatible given lag with states using plurality voting and precinct summability issues?
3. Given the presence of an interstate compact, how would it trigger and what are the issues of the current approach for triggering?

[Aaron Hamlin]

Also, Warren has done some analysis on rangevoting.org:

http://rangevoting.org/NPVtrainwreck.html

http://rangevoting.org/DeepComGrowth.html

[Toby Pereira]

It seems to me that it's crazy to have an electoral collage rather than just counting people's votes, and it's also crazy to weight anything by state rather than having a single nationwide election.

[Andy Jennings]

On Tue, Sep 27, 2016 at 2:11 PM, Aaron Hamlin <aaron...@electology.org> wrote:

When is a system like an electoral college useful, if at all?

I have little faith in the integrity of an election in, say, Chicago (which has a reputation of dead people voting).  With the electoral college, I know that Illinois is a safe blue state.  It's 20 electoral college votes will go to the Democrat in any case, so I don't have to worry much about ballot stuffing there.

[Aaron Hamlin]

Below is Warren's issue and solution, which I like.

The big issue is being able to freeride off other states and the cost of joining in early.

Warren's solution is nice and simple as well. Have the compact effective immediately, but only use the popular vote of compact states.

He has some analysis on alternative voting methods. But with that, I have a hard time seeing the practical implementation of a voting method that doesn't have a precint summability feature. This would be an obstacle for Condorcet and IRV. Systems like Borda, approval, and score voting would be fine. The best would be approval voting because it provides the easiest flexibility for mismached voting methods of states within the compact.

Excerpt of Warren's analysis:

Problem 1: Suppose the countrywide popular vote is for A, but nearly all (or anyhow some) of the states in the Compact prefer B. In that case those Compact-states would, by obeying their own rules to make A win, be acting against their own interests!

Problem 2: The compact will have no effect until and unless enough states join it to get a clinching majority.

Problem 3: Why should a state S join such a Compact?

• If it joins, it may be forced to act against its own interests!
• But if it does not join, its popular vote will still be taken into account by the Compact and will still act to swing the Compact's bloc-vote.

So there is no motivation for any state to join this compact – joining can hurt it but can never help it, whereas by not joining it still gets to "free ride" and gain all the advantages it could have gained by joining! 

We suggest to the reader, that if states have no motivation to join, then they won't join! And then (by problem 2) the Compact will never have any effect! Result: NPV's whole proposal is a dead dog.

Don't believe this? OK, here's why CA governor Schwarzennegger said he vetoed CA's inclusion in the Compact: "I appreciate the intent of this measure to make California more relevant in the presidential campaign, but I cannot support doing it by giving all our electoral votes to the candidate that a majority of Californians did not support." (You'd think after the biggest state in the USA dodged the compact for precisely the reason we said, the NPVers might listen to us and adopt the simple fix we propose. Just in the interests of their own survival. You'd think.)

Problem 4: If a state outside the compact adopts a new and weird voting system (not in the list above), it can still unilaterally destroy the compact!

Simple Solution to all 4 problems: Make the compact cast a bloc vote for the popular vote winner among the compact's member-states' combined population only, and make it do so no matter how many (or few) states have joined the compact. This way, nonmembers cannot free-ride. Hence states are motivated to join the compact. Further, the compact, even well before it has a clinching majority will often be able to throw the presidential election due to the power of its solid bloc vote. Even with only a few states joining the Initial Compact, it will be an 800-pound gorilla that presidential candidates will pay immense attention to. Nonmember "shirker states" who fail to unite with that gorilla, will do so at their peril; presidents are not likely to pay much attention to the shirkers. Hence the motivation to join, will be major.

On Tuesday, September 27, 2016 at 5:11:14 PM UTC-4, Aaron Hamlin wrote:

[Jameson Quinn]

Warren's solution is too clever by half. It relies on entire states following exactly the logic that Warren sets out. States will supposedly join the compact essentially because they want to take it over; but they won't worry about another later state then taking them over.

Here's how I'd rephrase and respond to Warren's 4 problems:

1: Warren: "A deal is always a bad deal because if it were a good one you'd have done it anyway." Me: the whole point of a deal is that you give up some of your own freedom in order to increase the expected value of the result. In particular, "safe states" would benefit from NPV, because they'd get more attention.

2: Warren: "But I want it now!" Me: the fact it's safe until it is decisive is a selling point. It makes it clearly fair, and less of a risk to join early.

3: Warren: "Free-riders." Me: Who cares? A state that doesn't join doesn't matter. The fact that it "gets to" still cast its EVs for the statewide rather than the countrywide winner is inconsequential, not a benefit.

4: Warren: "Voting systems." Me: This is a valid issue, and the compact should ideally be written with this in mind. But pretty much any arbitration mechanism in which the deciding votes are cast by experts who aren't partisan hacks is good enough. Warren's proposed systems for counting various systems are decent (though his proposal for Condorcet should give half a vote to each of the top two for ballots which don't distinguish between them). But trying to ensure that that stuff gets written into the compact itself is just a way to make it never happen. Lawmakers can understand arbitration boards and arbitration boards can understand voting theory; but that chain breaks without the middleman.

...

Aside from Warren, here's what I think:

-Andy is right that the EC makes a kind of "firewall" that stops a corrupt state election from having more than a certain fixed effect on the national election; an NPVIC would break that firewall. So, hand-in-hand with NPVIC, you'd need strong national election integrity standards, with some possibility for a temporary federal takeover of elections in states where corruption can't be stopped otherwise. These standards wouldn't have to be perfect, but they'd have to be good enough to keep corruption to a small scale vote differential. Paper ballots; mandated risk-limiting audits of those ballots; a minimum national standard for voter registration; a clear and non-excessive standard of what constitutes sufficient voter proof of ID; and all equipment and procedures must be certified for minimum security and maximum error rate. (Note that the risk-limiting audits immensely help the security standards become attainable.) States could still vary within those parameters, but some basic minimum standards would ensure that Chicago can't suddenly say that it got 110% votes for candidate X.

-I think that an NPVIC with approval voting should be our goal. The current NPVIC can be read to work with approval, but it would be good if that were clarified somehow. It probably would not work with anything beyond approval, though I don't share Warren's worry that this could be used as a sneaky sabotage.

Since I've been pushing MUMA (formerly U/P) voting, let me spend a minute considering whether I can make this work with NPV. Remember, MUMA stands for Most Upvoted Majority Acceptable, which basically describes how it works.

Say states have a mix of plurality, approval, and MUMA. First, tally the totals including all the MUMA non-downvotes. If any of the candidates gets a majority in this tally, then any who did not get such a majority will be "crossed out" and get no votes from the MUMA states. Now, retally using only the upvotes from the MUMA states, except for those who are "crossed out".

This does reduce to MUMA if all states use MUMA, and can give meaningful outcomes in cases where there's a mix of states. But ugh... I dread trying to write that into a law, or even to write a clearly-understandable NPVIC which could accommodate such complications.

So I think that for presidential elections, approval should be the goal, and it's not worth shooting for anything beyond that, even as a long-term second step.

--
You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electionscience+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.