Hi, I’m IRV

by Danny Muchoki

The underlying assumption of elections is that they capture voters’ preferences. Voters go into a booth, push a button/punch a card/pop a chad and when they’re all counted up we know that the person who wins over 50% of the votes is the winner. It’s obvious, right?

Not necessarily. In 1992, Bill Clinton became President with 43 percent of the vote. In 1998, Jesse “The Body” Ventura became governor of Minnesota while winning about 37 percent of the vote. In 2010 (again in Minnesota) Mark Dayton became governor with 43.6 percent of the vote. The runner- up was behind by just .4 (point four) percentage points – 43.2 percent.

A plurality system is simple, but some argue it is fundamentally unfair to let a candidate win with a plurality, let alone a plurality that is far short of a bare majority.

This potential eventuality has led some to push for electoral reform in how votes are counted. One of the most popular proposals is the Alternative Vote, but it’s also known as Instant Runoff Voting or IRV. According to Dr. Benjamin Reilly of the Australian National University, IRV is already in use in Australia, Estonia, Fiji, and Papua New Guinea with a few jurisdictions using it here in the US as well (such as San Francisco).

Dr. Reilly was involved in the (ultimately unsuccessful) campaign to implement IRV in Alaska in 2002 and in his paper, The Global Spread of Preferential Voting, he outlined a simple IRV election:

1. Voters rank candidates on the ballot paper in order of their choice, by marking a ‘1’ for their most favored candidate, a ‘2’ for their second choice, ‘3’ for their third choice and so on.
2. A candidate who gains an absolute majority of [first preference] votes—as happens in roughly half of all cases—is immediately elected.
3. If no-one has a majority, the candidate with the lowest vote total is ‘eliminated’ and his or her ballots re-examined for [the voters’ listed] second preferences, which are assigned to the remaining candidates in the order as marked on the ballot.
4. This process is repeated until one candidate has an absolute majority or until there are no votes left in the count.

“My involvement in the Alaska campaign in 2002 was mostly fortuitous,” Dr. Reilly said.  “I knew some of the guys from Fairvote, the main lobby group that has been pushing for electoral reform in general and IRV in particular – and they invited me to come to Alaska, so I did! It was a lot of fun even though I don’t think we were very successful in convincing the electorate, as the outcome suggests.”

IRV does two things at once – it ensures that the winner of an election has the support of a majority of voters, and it simulates several elections at once. Voters don’t have to keep going back to the polls – they can list their favored candidates, drop the ballot, and be on their way.

How would this work in Alaska? Let’s take a look at a local primary election in Alaska’s Senate District P . For the sake of this example, we have more or less invented the voters’ preferences (since that data was unavailable). We also assume that the candidates’ supporters are voting in blocs – it makes the example easier to follow.

So let’s track what happens on IRV election day.

1. After the votes are counted the results are:
   1. Moronell – 28 percent of the vote
   2. Giessel – 46 percent of the vote
   3. Johnston – 25 percent of the vote
      1. Nobody has an absolute majority here, so there’s no immediate winner. Johnston is eliminated as the lowest vote-getter, but we don’t stop there. Johnston’s voters listed Moronell as their second preference, so their votes are transferred to their second preference, Moronell.
2. Round 2 of the vote
   1. Moronell – 28 + 25 = 53 percent of the vote.
   2. Giessel – 46 percent of the vote
      1. Moronnell wins the primary. How come? Because Johnston’s supporters prefer Moronell over Giessel, and IRV allows this preference be accounted for. It identifies the true majoritarian winner.

Admittedly these preferences are completely made up, but the example illustrates how IRV works and how it can lead to completely different outcomes from the usual first-past-the-post system.

Curiously, the race for Senate District P was a completely open one after the incumbent chose to retire. The general election was a three way race as well, with Giessel winning 50.29 percent of the vote. With an IRV system, Giessel would automatically win the election as the majority winner and there would be no second round. But given that Giessel might not have won the nomination at all during the primary, there’s no real way to tell how the general election for Senate District P could have unfolded.

Even though its implementation in Alaska was unsuccessful, IRV has a growing presence in American elections. Time will yet tell whether it’s really here to stay and what impact it may have. But there are other voting systems that take the voters’ preferences into account – is IRV the right one?

Danny Muchoki is a third-year student at William & Mary Law School.

Permalink: http://stateofelections.pages.wm.edu/2012/11/14/4702/