"Election-rigging" and voting reform

[Jameson Quinn]

My name is Jameson Quinn; I'm a PhD student in statistics at Harvard and a board member of the Center for Election Science (electology.org). I appreciate your work at FiveThirtyEight and other venues explaining such issues as "rigged" elections (that is, how natural and artificial structural factors distort political outcomes) and primary reform. I'm writing you now to encourage you to discuss voting systems when you're doing so. In particular, I think you should mention the defects of first past the post, and highlight those defects by briefly explaining alternatives such as approval voting and/or mixed-member proportional representation.

Of course, I'm writing you as an activist, an advocate for voting reform. But I imagine that you don't see your own role as being an activist on these issues. So why should you help raise their profile? After all, while voting reform has had some localized successes in certain jurisdictions, it's not really high on the national media agenda, and there is no state that's particularly near to the kinds of statewide reforms I'd advocate.

But I'd argue that in order to get people to truly understand our system as it currently stands, you have to help them imagine how it could be. I think your readers might nod along as they read your articles, but without that imagination of what a solution would look like, the understanding they gain will be fleeting; if you quizzed them a day or two later, I suspect they would have returned to inchoate feelings that the system is rigged.

If this makes sense to you, allow me to pitch two solutions that I think are not just promising, but also illuminating. I'm sure that you're already somewhat aware of voting reform possibilities, but I'd like to explain some of the more recent thinking that you may not have heard yet. So I'll first explain these in the kind of detail I think you'd appreciate, then suggest the kind of simpler language that I think would be appropriate for the kind of articles you write.

First, as a single-winner voting reform, for electing presidents, governors, or mayors, I'd suggest approval voting. As I'm sure you know, that simply means that voters can approve as many candidates as they want, and whoever gets the most approvals wins. Although it's somewhat less well-known than instant runoff voting (IRV), it's not just simpler but actually a better solution to the problems of first-past-the-post. And it's not as if it's untested; it was used in the 19th and early 20th century in Greece, leading to Greek idioms around voting that are still used today.

There are two principal problems with FPTP: one, the potential for candidates to be spoilers, hurting their natural allies and helping their natural opponents; and two, the resulting incentive for strategic lesser-evil voting that ossifies two-party domination. But in approval voting, unlike FPTP and even IRV, there is never an incentive to betray your favorite candidate in order to support a lesser evil. There would still be two frontrunners in any given race, and in most places that would still be a Republican and a Democrat, but third parties would have a chance to grow organically. Thus, the party system would be much more dynamic and responsive to the voters, and less subject to corrupt equilibria or zero-sum, mudslinging campaigning.

Second, for multi-winner legislative elections, there's proportional representation (PR). I'm sure you're aware of such proposals as multi-member districts and single transferrable voting. As would any PR system, this would solve the problem of gerrymandering and loosen the two-party logjam. I would absolutely support something like this, but I recognize that there are valid criticisms of the idea too: that it requires complicated ballots, that it breaks the direct local connection between a voter and "their" representative, that it might give too much power to groups writing "suggested ballot orderings".

So I think it's worth underlining that modern voting theory allows designing a PR system without any of these downsides. The basic principles would be the following:

1. Existing districts would be unchanged.

2. Voters would choose a single candidate. The ballot would list the candidates from the local district first, in a larger font; but would also allow some means for more-engaged voters to vote for or write in candidates from anywhere in the state.

3. Winning a seat would require a Droop quota of voters; that is, one over the number of seats plus one. That means that in a state with 9 House seats, winning would take 90% of the average votes per district.

4. Obviously, it would be extremely rare for one candidate to get direct votes equivalent to 90% of a district. So the candidates with the fewest votes would be eliminated, and their votes transferred; first, spread equally over all other candidates in the same party, and then, if the whole party has been eliminated, to the candidate of their choice in some other party. Such transfers would continue until a full slate of winners each had a Droop quota of the remaining votes.

5. Each winning party would assign its winners to cover territories of one or more district, so that each district would be in the territory of one representative from each winning party. Thus, two neighbors from the same party would both be able to point to the same single representative as "theirs", although their neighbor from another party would point to a different representative.

The system I've just described (a slight simplification of PAL representation) uses the latest thinking in voting system design, so it's not actually used in any major democracy, though it has aspects of existing systems such as Germany's multi-member proportional system and the biproportional system used in Swiss municipal elections. It still tends favors parties which are among the two frontrunners in at least some local area, so, while it makes a far more level playing field for third-party or independent candidates, it still does not present an existential threat to the existing parties. For instance, in a state which happened to be getting a proportional result under FPTP, this system would elect the same incumbents if the same ballots were cast, so it's something incumbents might be able to support. And it achieves PR without complicated ballots, without party lists prepared in unaccountable backroom deals, and without breaking the local bond between a voter and their representative — the three biggest criticisms of other PR systems.

So, those are the kinds of reforms that we at electology.org advocate. But, you might respond, what does that have to do with you? Do I really expect you to get sidetracked on explaining obscure voting systems when you're writing articles about our current system? The answer is, of course I understand that you're not going to get too far into the weeds with this, but it's worthwhile to know that there are alternatives. If you simply draw the connection between the current-day problems that you discuss and the voting system, briefly mention that other systems are possible, and give a link or two for those who are interested in learning more, I'd be more than happy.

Thanks,

Jameson Quinn