Tea Parties, GOP primaries, and elections

The Daily Camera printed my response to the following question in its September 18 edition:

Looking to fall, the Tea Party — though loosely defined and with no official structure or leadership — will play a major role in the November elections, having sealed up several important primaries. But some Democrats have chosen to see Tea Party victories as their own, saying that the Republican Party will be the one that suffers as the newly excited, more conservative base siphons votes from the establishment. What do you think?

My response:

Who cares if elections hurt political parties? What matters is how well winning candidates represent voters’ preferences.  But Democrats and Republicans cooperate to support our lousy voting system at our expense. Known as “simple plurality voting,” it shields the two major parties from third-party competition. Like all government barriers to competition, this degrades product quality. In elections, the degraded products are the candidates.

In our elections the winning candidate need only be the proverbial “lesser of two evils.” Because of vote-splitting, he need not best represent the preferences of the voters. For example, even if more voters prefer either Dan Maes or Tom Tancredo to John Hickenlooper, Hickenlooper’s victory is ensured if the other two split the vote.

The “Tea Party victories” in primaries also show problems with simple plurality voting. Voters who prefer either the “Tea Party candidate” or the establishment Republican may combine to exceed those who prefer the Democrat. But the Democrat need only defeat the Tea Party candidate.

Here’s a simple way to run elections that both avoids these problems and chooses candidates that better represent voters. First, repeal or liberalize ballot access restrictions. Government should not keep candidates off ballots. Second, use “approval voting,” which means you can vote for as many candidates as you want. Whoever gets the most votes wins.

For example, you could vote for both Tancredo and Maes. No spoiler effect. No vote splitting. No in-fighting.  And compared to instant runoff voting, approval voting is simpler and does not punish popular second-choice candidates.

* * *

There’s some debate about which is system is better: approval voting or instant-runoff voting (IRV).  Last month I endorsed instant runoff, thinking that the range voting was the only better, but prohibitively complicated, alternative. But approval voting is the simplest form of range voting: you give a candidate either a “1” or “0” instead of more discretized range, e.g., 1, 2, or 3.

A couple of articles about IRV vs. approval voting are:

Approval Voting vs. Instant Runoff Voting, FairVote.org (groups favors IRV)
Range and approval voting vs. IRV, by “Maikeru,” which links to an article at RangeVoting.org about how instant-runoff can promote 2-party domination.

I haven’t read this book, but have heard good things about it: Gaming the Vote:  Why Elections Aren’t Fair (and What We Can Do About It), by William Poundstone.



Filed under public policy, published

3 responses to “Tea Parties, GOP primaries, and elections

  1. But as you note, it’s precisely because the two major parties could lose their grip on power that they’d unite to oppose any system that could favor a third party. In which case, voters may need to stand up en masse to demand a change — otherwise it won’t happen.

    To that end, I’m glad you’re bringing some of these seemingly-arcane procedural issues to public attention!

  2. Anonymous

    One difference between instant runoff voting and approval voting: one has a history in thousands of major elections around the world (IRV) and the other is mostly a theoretical system with no track record in the USA of earning support (approval)

  3. Score Voting is FAR SIMPLER than Instant Runoff Voting, not “prohibitively complicated”. This is mathematically quantifiable in terms of Kolmogorov complexity and ballot spoilage rates. And it is one of a litany of things that makes Score Voting (aka Range Voting) superior to IRV.

    Jaboy is right that IRV has been used in thousands of major elections. But that has proved that criticisms of IRV are legitimate. E.g. in Australia it has maintained two-party domination, and appears to result in massive tactical exaggeration. Hundreds of years of Approval Voting show that it has fared far better in REAL LIFE.

    Also, in complex computerized election simulations (which allow you to peer into voter’s minds, in a way that you CANNOT in real elections), Score and Approval Voting massively outperform IRV as measured by Bayesian regret:

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 )

Google+ photo

You are commenting using your Google+ 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 )


Connecting to %s