There are some legitimate concerns associated with IRV.  Unfortunately, Resolution #7 misses these legitimate concerns, relying instead on contrived and mostly unsupportable arguments. And the recommendation to stop using IRV without offering a better alternative takes us a step in the wrong direction.

First of all, some definitions are needed.  There are three common methods to select a winner from among three or more candidates.  (Note that although this discussion talks about individual winners, the same techniques can be used to select the top two finishers for a primary election.)  In plurality voting, all delegates vote one time for one candidate, and the candidate who receives the largest number of votes is the winner.  This method has the classic problem of allowing a minor candidate to draw votes away from the most popular mainstream candidate, allowing a less popular candidate to win.  Winners of these elections often do not receive a majority of the votes.

Runoff elections (also called the multiple ballot method) attempt to solve the problems of plurality voting.  Here, all delegates again cast one vote for one candidate, but only the candidate receiving the fewest votes is eliminated.  Another round of voting is used to eliminate another candidate, and this process is repeated as long as necessary to determine a winner.  This method allows delegates to more fully express their opinions by voting for a favored minor candidate in the early rounds.  Once their candidate is eliminated, they can switch their vote to their preferred mainstream candidate without having ‘wasted’ their vote.  And the eventual winner will have received a majority of the votes.  The major disadvantage of this system is the time involved to complete the many individual rounds of voting.  The lengthy process creates lots of opportunities for arm-twisting and back room deals to get delegates to change their votes along the way.  Some delegates may leave early, shifting the advantage to the more determined delegates, even though they might represent a minority position.  (The first round of a runoff election is sometimes used to eliminate more than one candidate, but this reintroduces problems inherent in plurality voting.)

Instant Runoff Voting (IRV) addresses the time issue by implementing a runoff election with a single ballot.  Instead of selecting only a single candidate, delegates rank all candidates in preferred order.  The candidate with the fewest first-place votes is eliminated, and the ballots that ranked him/her first now count for the candidate listed in second place.  This process of ballot counting, candidate elimination, and reallocation of the loser’s ballots is repeated until a winner is determined.  It is important to note that in the absence of delegates voting inconsistently or leaving early in a traditional runoff election, IRV will produce exactly the same results as the traditional runoff.

But IRV has some weaknesses.  Consider a scenario in which there are four candidates: A, B, C and D.  A, B and C are all somewhat controversial.  Approximately one third of the voters favor each one of them, and dislike the other two.  Candidate D is nobody’s favorite, but is widely respected and everyone’s second choice.  The ballots would look something like the following:

1/3: A D B C
1/3: B D C A
1/3: C D A B

Notice that candidate D is actually favored by a margin of 2:1 over each of the other candidates individually.  He is the obvious compromise candidate even though he received no first place votes.  But he will be eliminated in the first round, and the voters will have to settle on a candidate that two-thirds of them dislike.  (More detailed discussions and other examples of problems with IRV can be found using the links below.)

Another important point is that this problem is not unique to IRV.  Candidate D would also be eliminated in the first round of a traditional runoff, and would never win a plurality vote.  IRV improves on the traditional runoff method, which in turn improves on the plurality method.  But all three methods can produce a less than optimal outcome in some scenarios.

Resolution #7 is critical of IRV, but for mostly invalid reasons.  For example, the resolution and the author’s discussion argue that IRV would give minor parties or candidates a ‘disproportional amount of political power’ by allowing them to deliver their delegate’s second place votes to the mainstream candidate that offers them the best deal.  But this is no different than if the minor candidate endorsed a favored mainstream candidate after he was eliminated in a traditional runoff, or after withdrawing from the race shortly before a plurality vote.  Other related arguments that apply equally to the more traditional forms of voting include the ones labeled as ‘vote-brokering’ and ‘election manipulation’.  Some arguments criticize IRV for its improvements to the process, including ‘fast-food balloting’ and breaking from our ‘balloting heritage’.  Some arguments seem to simply be stretches to find something bad to say, including ‘unknown long-term political consequences’, ‘lackluster candidates’, and ‘no national mainstream party support’.  The argument that IRV ‘does not guarantee a majority winner in an election’ seems completely wrong.  And frankly, it is difficult to see how IRV ‘does not allow for a consensus winner in an election’ unless this argument refers to a scenario similar to the one described previously,  And this scenario also caused problems with the more traditional voting methods.

Resolution #7 instructs the party to discontinue the use of IRV, but fails to suggest a better alternative.  Yet going back to plurality or traditional runoff methods will throw away IRV’s improvements without solving any of its problems.

Resolution #7 should be voted down at the convention, not because IRV is flawless, but because the resolution’s reasoning is faulty and its recommendation would take us backwards, not forwards.

Some useful links:

• Resolution #7
• State Convention version
• Discussion of Resolution #7 by its author
• Information about voting methods, including potentially better ones