Condorcet: A Better Election Method

The reason we have a two-party system in the United States is widely misunderstood. It is not because the Democrats and Republicans consistently have the best ideas, nor is it because the media or the debate commision shut out the other parties. We have a two-party system because voters are not allowed to fully specify their preferences. To vote for minor parties, voters must effectively withdraw from the races between the two major parties, even though they may have a strong opinion on those races too. Voters who vote for minor parties essentially ``waste'' their votes and fail to oppose political movements they strongly disagree with. Protest votes may send a ``signal,'' but the unfortunate reality is that they have virtually no direct effect on the actual outcome of elections--and the indirect effect is usually contrary to the voter's intention.

The right to vote is the foundation of democracy, and a simple expansion of voting rights could improve the democratic process dramatically. Instead of allowing voters to select only a single candidate for each office, they should be allowed to rank the candidates according to preference. Such an expansion of voting rights would allow voters to fully specify their preferences. Although ranked voting may not at first seem important, it could end the two-party system as we know it. It will not make the voters wiser, nor will it guarantee that they will elect better candidates, but it can give them the kind of leaders they really want--and that is the essence of democracy.

When voters cast their votes in our current plurality system, they are allowed to select only a single candidate for each office. That is far better than no choice at all, of course, but it is nowhere near as good as also being allowed to specify a second and third choice, or beyond. Current voting rights are therefore incomplete. Complete voting rights would allow voters to vote according to their convictions and principles without wasting their vote on a candidate with little or no chance of winning. The rules for determining the winner would be slightly more complicated than they are now, but they would be based on elementary mathematics and should be understandable by virtually anyone old enough to vote.

The proper method of counting ranked votes is called the Condorcet election method, named after the French mathematician who conceived it a couple of centuries ago. The main idea is that each race is conceptually broken down into separate pairwise races between each possible pairing of the candidates. Each ranked ballot is then interpreted as a vote in each of those one-on-one races. If candidate A is ranked above candidate B by a particular voter, that is interpreted as a vote for A over B. If one candidates beats each of the other candidates in their one-on-one races, that candidate wins. Otherwise, the result is ambiguous and a simple procedure is used to resolve the ambiguity. Condorcet voting is explained in more detail elsewhere at this web site.

The Condorcet election method allows voters to vote their true preferences without worrying about wasting their vote on a candidate with little or no chance of winning. It frees voters from the possibility that, by voting their true preference, they will neglect to oppose a candidate they strongly disagree with and who could actually win. That is, the Condorcet system eliminates the ``horserace'' effect, which forces voters to consider not only which candidate they prefer, but also what each candidate's chances are of winning. It allows voters to vote for the candidate they agree with most rather than against the major-party candidate they disagree with most. In other words, it eliminates the need for defensive or strategic voting.

Distortions such as occurred in the 1992 presidential election could not happen under the Condorcet system. In that election, Ross Perot was a strong third-party candidate and may have taken enough votes from George Bush to allow Bill Clinton to win, despite the fact that most Perot voters may have preferred Bush over Clinton. Clinton received substantially less than a majority of the popular vote, and Bush might have defeated him in a one-on-one race (neglect the electoral college for now, which should be abolished). In a Condorcet voting system, the outcome might have been quite different. With no worry about splitting the vote, more Republicans might have voted for Perot, and Perot might have won, for example. On the other hand, most voters who voted for Perot might have selected Bush as their second choice, and Bush might have won re-election.

The Democratic and Republican parties will probably not like the Condorcet election method, at least not in general elections, because they could eventually lose their effective duopoly. Competition would be spurred dramatically, and some of the parties now considered minor would become stronger. Voters who believe in laissez faire government could vote Libertarian, for example, and still register their preference for Republicans over Democrats. Similarly, those who believe that the government should provide economic security could vote for their preferred parties and still register their preference for Democrats over Republicans. Although Condorcet voting removes the artificial advantage of the two major parties in general elections, however, it would help them tremendously in their own primary elections.

Our current plurality voting system is particularly inadequate in primary elections with many candidates. Suppose, for purposes of illustration, that the Republican party is predominantly conservative, and their field of candidates consists of six conservatives and one moderate. The conservative candidates are likely to split the conservative vote, but the moderate would presumably get the entire moderate vote and could possibly win with a small plurality much less than a majority of the votes. That outcome would be unfair to the conservative majority. The issue here is completely non-ideological, however, and labels are used as examples only. The same phenomenon could happen to the Democratic party, for example, if the field of candidates consisted of six liberals and one moderate--or, for that matter, six moderates and one liberal.

Although the examples given for purposes of illustration are simplistic, the basic principle applies in every election. Popular political trends tend to attract more candidates, and candidates with similar platforms tend to split the vote with each other, which is unfair to the voters who wish to support those platforms. This fundamental flaw in our current incomplete plurality voting system makes it very difficult for the majority to consistently get the kind of leaders they really prefer. The Condorcet system is much more likely to give them what they want and is therefore more democratic.

The current system has other deficiencies too. Consider the 1996 Republican presidential primary election as an example. Suppose that a particular voter preferred Alan Keyes and strongly disagreed with Pat Buchanan. He could have voted for Keyes, but if he were rational he would have realized that Keyes had virtually no chance of getting nominated, and that if he voted for Keyes he would be taking the chance of letting Buchanan (who started out strong) win the nomination. Rather than voting for Keyes, his rational vote in the current system would be to vote defensively against Buchanan by voting for Bob Dole, the ``highest ranking'' Republican. In a Condorcet system, on the other hand, he could have voted exactly as he wished: for Keyes first and Dole second (or at least ahead of Buchanan).

The preceeding example shows how the current system strongly favors prominent ``insider'' candidates and magnifies the inherent disadvantage of less prominent candidates. It is no wonder that the last three Republican presidential nominees were George Bush (then vice-president), Bob Dole (Senate majority leader), and George W. Bush (son of a former president). Nor is it any wonder that sitting presidents rarely, if ever, lose a bid for renomination by their own party. Many voters are willing to settle for familiar but mediocre candidates rather than wasting their votes on ``long shots'' and taking the chance of letting the nomination go to someone they strongly disagree with. In the Condorcet system, such compromises are unnecessary.

The Condorcet election method is fundamentally different than our current plurality method and obviously cannot be implemented overnight. The public needs to be educated on the inadequacy of our current system and the major advantages of the Condorcet system. Although the rules of the Condorcet system are simple enough, their profound implications are not immediately apparent, and many will mistakenly dismiss the whole idea as unnecessarily complicated. Also, the Condorcet system can easily be tested in a non-binding mode before actually being phased in for real public elections. Such testing can be started immediately by private polling organizations.

Aside from irrational fear of change, the only reason to oppose complete voting rights is to protect the current political duopoly from outside competition. Such protection obviously cannot be good for democracy. Unfortunately, it is those very ``Republicrats'' who currently have the power to change the election method, so democracy will undoubtedly have some difficulty evolving. Rest assured that they will ridicule Condorcet voting as too complicated, but they will only be insulting the intelligence of the American public. They will also characterize it as radical, but so was the concept of voting itself when first introduced. With a proper understanding of what is at stake, however, rationality can ultimately prevail.