Instant Runoff Voting: A New Way to Vote
A number of Democrats who might like to vote for Ralph Nader for president on the Green ticket are hesitant or unwilling to do so because they fear their defection from the Democratic ticket might allow George W. Bush to be elected on the Republican line.
Here in Vermont, some Democrats face the same problem in state politics -- they might prefer to vote for an independent candidate, Anthony Pollina, but they fear that their failure to vote for the re-election of Democratic Governor Howard Dean might result in victory for the Republican candidate, Ruth Dwyer.
This is an old American dilemma, resulting primarily from the nearly universal use of plurality, winner-take-all voting in U.S. elections. Suppose, for example, that either Al Gore or George Bush receives 46 percent of the vote next November, the other wins 39 percent, and Nader polls the other 15 percent (neglecting, for this hypothetical, all other candidates).
In the first place, the winner will be a minority president, as many American chief executives -- like Bill Clinton -- have been. In the second place, if that winner should be George Bush, Nader will be seen as a "spoiler" whose 15 percent probably cost Gore and the Democrats the presidential election -- even though 15 percent actually may underestimate vastly the number of people who preferred Nader to either major-party candidate.
Ross Perot was seen as just such a spoiler in 1992, after subtracting enough votes from George Bush the Elder to elect Clinton. Perot, George Wallace in 1968 and John Anderson in 1980, all started their third-party campaigns with strong poll showings -- but finished with relatively weak vote totals. That's because such candidacies, no matter how popular, are seen as inevitable losers -- and voters don't like to throw away their votes on losers.
Plurality voting is a strong reason why third parties and independent candidates so seldom win and why the two-party system is so entrenched in American politics. It explains why duly elected mayors, governors and presidents so often have to govern without an official, or at least a truly sympathetic, majority, and why so many Americans resent what they believe is the necessity to cast a "strategic" though insincere vote rather than one for a candidate they truly support.
The problems posed by plurality voting could be resolved, and far more democratic election results obtained, by a change -- unfortunately, one neither simple nor likely -- to a system of preference voting. In an explanatory acronym, this system is sometimes called IRV, for "instant runoff voting." Here's how it would work in November:
Gore, Bush, Nader and Patrick Buchanan, the Reform candidate, all would be listed on the ballot. Instead of putting an "X" by his or her choice (whether strategic or sincere), a voter would put a "1" by his or her first choice, a "2" by the second, a "3" by the third, a "4" by the fourth (and so on, if other choices were available). When the ballots were counted, if a candidate had a majority of first choice votes, he (or someday she) would be elected. But if no one had a majority of first choice votes, the candidate with the fewest such votes would be eliminated.
In an immediate second count -- in effect, an instant runoff -- the votes that had been cast by supporters of the eliminated candidate would be redistributed to their designated second choices. If this gave someone a majority -- even a second- or third-place finisher in the first round -- that candidate would be elected. If not, again the candidate with the lowest total would be eliminated. Votes for that candidate then would be redistributed to his or her supporters' designated second choices.
That would produce a winner with majority support, eliminate the idea of a "spoiler," make insincere strategic voting unnecessary, promote the chances of a third-party or independent (sure to be a loser in the usual plurality voting system), and give voters a wider range of choices. It would obviate the cost of the separate, runoff elections conducted in some constituencies in order to get a majority winner, as well as the drop-off in turnout such separate runoffs usually incur.
In North Carolina in 1950, for instance, Senator Frank P. Graham won the Democratic primary with the most votes ever cast for anyone in any election in the state's history. He did not win a majority, however, in a three-candidate race; in the run-off necessitated by state law, turn-out declined drastically, and Graham lost to the first round runner-up, Willis Smith (who was heavily backed by Jesse Helms, then a radio commentator, now the state's senior senator).
Turnout could not decline in IRV contests; and proponents believe IRV might even reduce negative campaigning. Theoretically, candidates eager to be the second or third choices of voters backing someone else would not wish to alienate those voters by attacking their preferred candidate, On the other hand, IRV might undermine the two-party system, an element of stability but also of the status quo in the American system. But enhancing the chances of some independents and third-parties, in some cases, would be useful.
IRV also might confuse voters. It would increase the complexity of vote-counting and the likelihood of cheating in that process. In many cases, it also would require constitutional or legal changes difficult to explain or effect. But it's not a radical or outlandish idea -- both the American Political Science Association and the American Psychological Association elect their officials by IRV.
The chief difficulty is the ingrained devotion of Americans both to plurality voting and the two-party system, which makes any variation from either seem radical or even subversive. The Republicans of 1860, with Abraham Lincoln as their candidate, were the last and only upstart party to win the presidency -- and at that in a four-party election. Even Theodore Roosevelt couldn't do it, in his "Bull Moose" campaign of 1912.
In the new millennium, election year 2000 is running true to form, with neither Nader nor Buchanan or any other two-party deviant to be included in those decisive televised debates between Bush and Gore, Republican and Democrat, mainstream versus mainstream.
A detailed explanation of IRV and other forms of non-plurality voting will be available, however, in time for this year's elections -- though it's not likely to affect them. Behind the Ballot Box, by Douglas J. Amy, a political scientist at Mount Holyoke College in South Hadley, Massachusetts, will be published by Praeger in October.
Tom Wicker was a columnist for the New York Times for twenty-five years before retiring in 1991. He is the author of more than a dozen books including, One of Us: Richard Nixon and the American Dream.