The U.S. is not alone in suffering from low voter turnout. Some countries have reacted by requiring citizens to vote; that does not seem unreasonable, for a democracy to work, a government needs to determine what its citizens want. Australia imposes a fine for failing to vote; and reports are that Australians are more politically engaged than Americans are. And we might consider whether mandatory voting might help dampen efforts at vote suppression?
But taking this approach does seem to be treating a symptom, rather than the underlying problem. More emphasis on civic education might help. But surely, it would be important to improve elections enough so that citizens would more appreciate the value of voting.
Increasing the number of candidates would be one way to do this. Additional candidates competing in our elections would increase the prospects for voters to enthusiastically favor one or more of those candidates; surely that would motivate voting. Likewise, with more candidates, the opportunities would increase for a voter to strongly oppose at least one. That too would motivate voting (particularly if the election allowed voters to explicitly vote in opposition to those candidates).
Instant Runoff Voting (IRV) recently changed its name and it is now mostly called, ranked-choice voting. Whatever you call it, IRV is surely the most widely known alternative to plurality voting. Many people know of IRV as the only alternative to BAV. And with only that one tool in their toolbox, it no seems to be the solution to every problem; if your only tool is a hammer, the world looks to be full of nails.
It seems possible that IRV was invented in the hope it might encourage more candidates to compete, but experience with IRV gives little support for that hope. Earlier articles in this series have illustrated several serious deficiencies of IRV and at the end of this article we examine yet another example.
Variations of IRV have been considered as well, such as holding two or more elections (actual rather than simulated as is done with IRV). This idea of runoff voting has proponents, typically using a series of only two elections. A preliminary election serves to reduce the number of candidates, typically to only two, and after some time to allow for campaigning, there would be a final runoff election. Why two? It is argued that this would ensure that the winner could boast of having the support of a majority of voters. Such a claim of majority support seems quite feeble, however. But many voters would probably have preferred one or even any of several candidates who had been eliminated in the preliminary election.
Moreover, just because a voter expresses support on the ballot for a candidate, we cannot be certain that the support is real. In fact, the motivation quite frequently comes from opposition to the other candidate. This even happens when there are three or more candidates, inarticulate and imprecise as that may be. With many voting systems this is the only way to a voter for even hinting at being opposed to a candidate. And with ranked-choice voting or plurality voting, it is quite possible for there to be an optimal consensus candidate who is rejected.
Most people would probably agree that artificial barriers for blocking candidates from competing in an election should be avoided when possible; ideally, no one should be arbitrarily excluded. However, it clearly is also important to avoid allowing an excessive number of candidates to compete; that is just too confusing for voters. And this is particularly a concern for the final decisive election.
Even in a twenty-candidate election, many voters would likely confuse one candidate with another or in some other way be unable to properly recall important details about the candidates while voting. Concern about such difficulties would be much less if the election had only five or six candidates. So, some method for limiting the number of candidates to a manageable level seems to be unavoidable, particularly for the final election. Elsewhere, I've suggested keeping the number of candidates in the final election to be between five and eight, but psychological studies might show a different range would work better; my suggestion is, in truth, just my guess.
As necessary as a preliminary election may be, it would be a mistake to overly constrain the number of candidates in the final election. Allowing only two candidates is clearly a mistake, particularly if the purpose is merely to accommodate our longstanding attachment to plurality voting. Experience cries out to us that a limit of only two candidates gives voters too few options. That choice leads to a two-party duopoly and eventually, a toxically polarized electorate. It surely would be preferable to encourage wider competition with the number of candidates being constrained only out of concern for over-taxing the abilities of voters. But for that to happen we must, at a minimum, adopt a voting system that sensibly accommodates a larger plurality of candidates. Despite its name, plurality voting falls short of that objective.
As the number of viable parties increases, gerrymandering will become increasingly difficult and possibly even counter-productive. Providing more candidates would also increase the opportunity for a voter to feel enthusiastic about supporting one (or more) of them; and finding a candidate to enthusiastically support would surely motivate a person to vote. Conversely, a long history of failure to discover such a candidate would surely dampen any voter's enthusiasm.
Clearly, we would try to find a voting system that can work well with multiple candidates. Because IRV has enjoyed active, enthusiastic and generously-funded promotion, many people will point to IRV as the obvious answer. But, as popular an idea as IRV has become, that unfortunate voting system has numerous and quite serious deficiencies; deficiencies that would make IRV a quite poor choice. Approval Voting would be distinctly better, if only because it avoids forcing voters to make arbitrary decisions. But Balanced Approval Voting (BAV) would be an even better choice. New ideas are always possible so in time, it is possible we might discover an even better alternative.
But BAV would work well. It retains the advantages of Approval Voting, while avoiding a bias that unfairly favors famous candidates. The bias stems from failing to distinguish indifference from opposition, a fault that is common to many voting systems, including plurality voting, ranked-choice voting and even approval voting. But an even more important advantage of BAV is that it directly discourages polarization, actually making it difficult to maintain a duopoly. This allows electoral victory by a new or small party to become a realistic possibility. BAV will improve competition in elections and it promises eventually to usher in a truly multi-party politics that more closely approximates an ideal democracy. At a minimum, BAV would cause our politicians to be more attentive to the needs of voters.
When competition in a preliminary election grows excessively large (perhaps 30 or more candidates) other voting systems may seem more practical than BAV. One alternative, which applies statistical methods, was proposed in an earlier in this series. Alternative systems, BV-1, BV2, BV-3, were described in one of the earliest articles of this series, and one of these may seem preferable for such elections.
(Note: You can view every article as one long page if you sign up as an Advocate Member, or higher).
Post Article Comment and Rate This Article
