Delayed versus Instant runoff

In the usual runoff system ("delayed runoff") the election is carried out in two stages. The first stage is a regular plurality election. If nobody gets over 50% of the votes, then there is a second ("runoff") election, which is a 2-contender election between the two candidates who had the top two vote counts in the first round.

The instant runoff system (IRV) attempts to compress those two rounds into only one round, but at the cost of increased complexity, and also at the cost that the resulting system is really not equivalent to the old one. The voters supply rank order ballots as their votes, e.g. "Bush>Gore>Nader" could be a vote. The candidates with the fewest voters top-ranking them, are eliminated (both from the election and from all votes). For example, if Gore were eliminated, that vote would become "Bush>Nader." We continue doing these eliminations until only one candidate (the winner) remains.

Where are they used?

Instant runoff is currently (2006) used in Australia to elect House members, Ireland to elect the President, and in Malta in perhaps some situations. (A recent addition to this list was Fiji but it ceased to be democratic after a 2006 coup.) Also some bastardized forms of IRV are currently used to elect London's mayor and in San Francisco. IRV was used to elect the mayor of Burlington, Vermont, but it was repealed after a bad experience. It was also used (or similar systems used) in numerous USA cities during the 1900-2000 century but repealed in almost all of them.

Delayed runoff is used in France to elect the president. (It also is used for many city elections in the USA, e.g. New York, Chicago, and Denver mayoral; city council, etc.) It is also used for presidential elections in numerous other countries, e.g: Argentina, Brazil, Central African Republic, Chile, Colombia, Congo, Cyprus, Finland, Gabon, Ghana, Iran, Liberia, Macedonia, Mali, Mauritania, Moldova, Monaco, Nicaragua (but runoff only used if winner got below 35% in first round), Niger, Peru, Poland, Portugal, Romania, Senegal, Slovakia, and Uruguay. Of these 27 countries, 21-23 have broken free of 2-party domination according to my reckoning in 2006.

Which is better?

Proponents of IRV seem to have the idea that IRV is "obviously" superior to delayed runoff, because one round must be better (i.e. cheaper, higher typical voter turnouts) than two.

We agree one round is better than two if you just look at that factor in isolation. But there nevertheless are good reasons to prefer delayed runoff over instant runoff, and we dispute any assertion of "obvious" IRV superiority. (Indeed, we regard the repeated efforts of a certain US pro-IRV group to replace delayed runoff by IRV elections wherever they can – as opposed to the course, which would be a far larger and genuine improvement, of trying to replace single-round plurality elections by IRV or delayed runoff – as insane. Why are they wasting their time on such an incredibly minor issue, whose payoff may actually be negative?)

The main reasons we know to prefer delayed runoff over IRV are

  1. IRV leads to stifling 2-party domination, whereas delayed runoff encourages the formation of many stable political parties, offering voters more choices. Why does that happen? Regardless of why that is, it is hard to dispute: all the IRV countries listed above are 2-party dominated in their IRV seats, whereas 21-23 of the delayed runoff countries listed above have multiparties. And this is true despite the fact most of these delayed-runoff countries have strong presidents (unlike the IRV countries), a factor that normally would enhance 2-party-domination. Is the goal of that pro-IRV group we alluded to, to destroy USA's third parties?
  2. For the statistics nerds in the audience who are wondering how statistically significant that was: consider a total of 31 countries, 27 TTR and 4 IRV, and ask: "what are the chances that the 4 IRV ones, by pure chance, would just so happen to coincide with the 10-12 that are (≤2)-party dominated, if there were no actual political forces causing IRV to yield 2-party domination." The chances are (X choose 4)/(31 choose 4) where X=10-12. This is 0.66% to 1.57%. In other words with about 99% statistical confidence we can assert that IRV leads to 2-party domination and TTR does not – or at least that the two differ significantly in that respect... While that is not 99.999999%, still, it is pretty convincing. Also, the TTR countries here usually have strong presidents (and the IRV countries not), which is a factor that tends to encourage 2-party domination, and they still avoid it – a fact which strengthens our case that IRV leads to 2-party domination while TTR avoids it.
  3. Yes, one round is cheaper and easier than two, but with IRV, that one round is more complicated and it cannot be done on ordinary "dumb totalizing" voting machines, whereas both rounds in delayed runoff can be done with such machines; and IRV is non-additive (no such thing as "precinct subtotals") and non-monotonic; and the second round in delayed runoff often does not happen. (Top-two runoff also is non-monotonic, but each of its two rounds, in isolation, of course is monotonic.) In view of those facts, it is not at all clear to us that IRV actually saves money. And in any event, the money spent on elections is negligible compared to other government expenditures, so it is more important to get quality in elections, than to save money. Thus "saving" money would be a false economy that surely would actually cost more in bad government than it saved in election expenditures. For example (2006), a pro-IRV group was recently arguing that Oakland California should switch to IRV because each runoff election under the old delayed-runoff scheme cost Oakland "hundreds of thousands of dollars," which was their way of saying $200,000. However, they did not mention that Oakland's annual budget is over $1 billion so that the "cost savings" they were lobbying for was of order 0.02% fractionally. Surely there are superior ways to save Oakland's money! Also they did not mention that it cost (neighboring, comparable size) San Francisco $1,600,000 to upgrade its voting machines to run IRV two years before. So the payback time required to justify this cost "savings," as you can see, would be very large, perhaps 30-40 years assuming elections every 2 years and runoffs required half the time. Quite probably Oakland would be re-replacing its machines before that time, in which case the costs never would be repaid.
  4. IRV is more complicated for both voters and talliers. Thus when San Francisco adopted IRV in 2004, they experienced 7-times-higher voter error rates leading to invalid ballots, than in plurality races.

On the other hand, Chris Benham points out that instant runoff would have shown superior behavior to delayed runoff in France's 2002 presidential election. Also (Benham continues) IRV enjoys certain logical criteria which top-2-runoff does not enjoy, such as "dominant mutual third" (which says "if more than a third of the voters rank [in any order] the members of a subset S of candidates above all others, and all the members of S pairwise-beat all the non-members; then the winner must come from S"). So we certainly cannot argue that one system is better than the other under all circumstances.


Return to main page