The interpretation of election results in contests with third-party candidates is often difficult. Was the third-party candidate a positive force for change, or merely a spoiler? Did the third-party candidate swing the outcome and frustrate the will of the majority? Did the presence of a third party make it harder for the major party contenders to get their message out?
The reality is that a third-party candidate can blur the message voters intend to send. The most recent example comes from New York's 26th Congressional District. A Democratic candidate won a seat that Republicans comfortably held for the past 20 years. That candidate, Kathy Hochul, received 47 percent of the votes; the GOP candidate, Jane Corwin, 43 percent; and the third-party candidate, Jack Davis, 9 percent.
A reason given for the Democrat's victory was Corwin's early endorsement of the Republican Party plan to significantly shift the costs of Medicare to the elderly. However, Hochul's victory might actually have resulted from confusion created by the presence of the third-party candidate. Davis, who twice before had run as a Democrat, was the tea party candidate in this election, emphasizing tax cuts and smaller government. It is unclear which party's vote total was most affected by this candidacy. This kind of confusion occurs whenever there is a close race that includes a third-party candidate.
Fortunately, there is a low-cost way to avoid such postelection grief: instant runoff voting, or IRV. An IRV ballot allows the voter to indicate both a first and second choice for the contested office. When the polls close, the electronic machinery tallies the votes and identifies who came in first and second. Then it consults the ballots of those voters who did not prefer either of these candidates and allocates their second choices to the top two finishers. Only then is a winner declared. This process avoids a runoff election campaign. Virtually no additional costs are incurred.
In an election with three or more candidates, IRV increases the likelihood that the electorate receives the representation that the majority favored on Election Day. It enables people to vote both their preferred candidate and indicate their strength of preference. This procedure is used to elect the presidents of India and Ireland, the mayors of St. Paul, Minn., and London, England, and even for the Oscar for Best Picture.
Imagine how well this would have worked in the 2000 presidential election between Al Gore and George W. Bush. In that election, Ralph Nader received roughly 70,000 votes in Florida, while Bush led Gore by 535 votes — a virtual tie. Over the ensuing weeks, various recounts were attempted without changing this result. The Supreme Court ended the recount process, effectively declaring Bush the winner both in Florida and, in turn, the nation.
If a system like IRV were used, we likely would have had a winner on election night as the second choices of the Nader voters were reallocated between Bush and Gore. The nation would have been spared the angst caused by a month of recounts and legal challenges.
A similar important example is the 1992 presidential election among George H.W. Bush, Bill Clinton and Ross Perot. Neither Clinton nor Bush held a clear majority, but Clinton won with a plurality. What if the Perot voters' second choices had been heard?
With IRV balloting, candidates who have little chance of winning can still get their issues into the marketplace of ideas. Moreover, their issues should carry more weight since the major candidates will want the third-party voters to cast their second preference choice their way. For example, if Nader presented a great idea, then Bush and Gore would have had a political incentive to adopt some or all of that idea (and perhaps improve on it) to become the second choice of Nader voters. Without IRV, there is less incentive for a major candidate to include Nader's ideas since the only option is to persuade the Nader voters to change their first preference, a much more difficult task.
With IRV, there's no need to call third-party candidates "spoilers" since they cannot spoil the rule of the majority. It makes more sense to address the substance of their message than to question their right to run. Given the bitter division between the major parties on issues of social and economic policy, unclear election outcomes will only add to partisan rancor.
To adopt IRV would require a statutory change. In areas with computerized vote counting, this adoption would be easy and inexpensive. In those areas where more labor-intensive vote counting is used, the advantages of IRV may provide an impetus to employ computers.
Charles O. Kroncke is associate dean in the University of South Florida College of Business. William L. Holahan chairs the department of economics at the University of Wisconsin-Milwaukee.