IRV, not Fusion
Published in The Nation, November 7, 2005

Greg Dennis
gdennis at mit dot edu

Alyssa Katz claims fusion voting "sidesteps the Nader effect," but this is true only so long as third parties like the Working Families Party do not run their own candidates. Furthermore, fusion voting is only applicable to partisan elections and, therefore, does nothing to mitigate the spoiling of primaries and nonpartisan local elections.

Instant Runoff Voting overcomes these limitations. Under IRV, voters rank their candidates in order of preference (e.g., first: Nader; second: Kerry; third: Bush), and votes are counted so as to insure that the winner always has a majority of support. Thus IRV allows third-party candidates to run without causing the major-party candidate ideologically closest to them to lose. Plus, unlike fusion, IRV can be used for both partisan and nonpartisan elections. Recent IRV victories in San Francisco and Berkeley, California; Ferndale, Michigan; and Burlington, Vermont, demonstrate that Instant Runoff Voting has far more momentum than does fusion.

Instead of looking at recent elections in New York State, let's look at New York City, where Fernando Ferrer won a six-way Democratic mayoral primary with only a minority of the vote. His low percentage nearly forced a divisive runoff that could have cost the city up to $12 million. Although fusion would be no help here, IRV would consolidate the city's two-stage runoff process into a cheaper, fairer and less divisive instant runoff election. While fusion is an improvement over simple plurality, progressives should place their electoral reform hopes and energies in IRV (for more details see