Alaska’s lieutenant governor just killed a ballot initiative that would have mandated ranked choice voting (aka instant run-off voting, or IRV) in the state.
The official reason given was that the initiative combined separate subjects into a single initiative. This was doubtless frustrating to the initiative’s organizers, but avoiding a premature rush to implement instant run-off voting may be beneficial in the long term.
Instant run-off voting is wonderful in spirit but terrible in practice. The non-obvious problems include the following:
• Like our present system (plurality voting), IRV tends toward two-party duopoly and perpetuates it where it already exists. This happens mainly because, like plurality voting, IRV strongly rewards strategic voting (lying about your preferences).
• The above problem with IRV masks many of the problems below. These get downplayed because they don’t come up much where stable duopoly already exists.
• Ranking your favorite candidate first in IRV can make him lose. This is reasonably likely to happen in competitive three-plus-way elections.
• Ranking your least-favorite candidate last can make him win.
• Voting honestly in IRV is often worse for you than not voting at all.
• Competitive three-plus candidate IRV elections are both strategic and unstable. The vast majority (99%+) of voters will have no idea what strategy to use to get the candidate they want. Electing leaders by lottery isn’t a bad idea, but you wouldn’t want the lottery to include only politicians.
• IRV is harder to implement than other systems that work much better.
• IRV can’t be audited without giving up important aspects of voter privacy in elections with moderately large fields.
• Voters don’t like IRV. And where IRV has been implemented, they want to get rid of it.
Australia uses instant run-off voting, and it exhibits many of these problems. There isn’t room here to explain them, but MIT mathematician and election expert Warren Smith has published highly accessible explanations on rangevoting.org. He also explains the virtues of two systems that don’t have these problems: range voting and approval voting. These are much less widely known than IRV, but they aren’t complicated.
In range voting, voters give each candidate a score (e.g. between 0 and 10). Approval voting is just range voting with 0 and 1 being the only allowed scores. The candidate with the highest average score wins.
The problems with plurality voting are at the root of many other so-far intractable problems, and any prospective replacement must avoid the worst defects of the present system.
The worst defect is the well-known spoiler effect, in which third-party candidates end up hurting their policy-wise closest major party competitors. Media discussion of this effect treats it as a constant of the universe rather than a totally avoidable ghastly power-serving defect of the present system. When other systems get discussed at all, the discussion is usually about IRV as a supposed remedy for the spoiler effect.
Unfortunately, instant run-off voting fails to avoid the major consequence of the spoiler effect, which is two-party duopoly. Advocates for IRV usually wrongly assume that under IRV the field would look about like it does now (i.e. two major parties plus Greens and Libertarians). But because IRV gives small parties some limited real power via their capacity to deliver negotiated second-choice endorsements to major parties, it results in a profusion of small parties, all of which are usually unelectable due to strategic voting.
There’s a pernicious connection between mass media propaganda and two-party duopoly. It’s now widely accepted that by a process of selective emphasis and omission, the mainstream media ends up functioning as corporatist propaganda.
Duopoly plays a little-known but huge role in this. Under plurality or IRV in an entrenched duopoly, third parties aren’t interesting simply because they cannot win, and consequently they get little media coverage. This effectively reduces the propagandist’s job of controlling the overall narrative to the more manageable task of controlling the narratives in the major parties.
The founders of the Unites States understood the toxic effects of party politics. In his farewell address, Washington said, “The common and continual mischiefs of the spirit of party are sufficient to make it the interest and duty of a wise people to discourage and restrain it.” In practice this means preventing a small number of major parties from perpetually dominating the electoral landscape, and instant run-off voting fails to do this.
The intent of the recent ballot initiative effort is laudable, and its organizers should try again with a proposal for range or approval voting.
Britton Kerin lives in Fairbanks.