While our elections for the President of the United States are a bit more complicated because they involve the Electoral College, they too are based on FCV. Who could even imagine using anything but this tried-and true system of casting votes? FCV is such a familiar system to us we may have trouble even imagining an alternative. Is there any other way? Doesn't FCV work, if not perfectly, at least better than anything else we could possibly do?
In fact, there are actually many other possibilities and we do need to consider alternatives and in this series of articles we introduce several of them. But in this article, we focus on the familiar FCV approach with a view towards pointing out its flaws.
With FCV, first-preference votes are counted and whichever candidate receives the most votes, even by the very slimmest of margins, wins the election. This system guarantees that the most voters possible (even if that is only a minority of voters) will be very pleased to find out that their first preference won the election; however it also makes it likely that there will be a large number of voters, perhaps even a majority, who are disappointed, perhaps very, very disappointed, with the outcome. So FCV can and does at times lead to elections in which many more voters are seriously disappointed with the outcome than are happy with that outcome. Is this what we want from our elections? Could we do better?
FCV is a simple system that works marvelously in elections between just two candidates for an office. But when there are three (or more), a potential problem arises, and a famous law of statistics has much to say about potential problems. Murphy's law says that if something can go wrong it will (eventually), so the very possibility of a problem is not something to be taken lightly when doing something over and over again as we do with elections.
Two candidates in a three-way race might seem very similar to the voters. The majority of voters might prefer either of them to the third candidate, but not have a strong preference between these similar candidates. In a two-way race between this third candidate and either of the two similar candidates, the third candidate would lose (perhaps only getting 36% of the vote) but in the three-way race, the majority of the voters are forced to choose between the other two. In this situation, using FCV, the most unpopular candidate could easily win (say, in a 36%, 35%, 29% vote). This is an example of the spoiler effect, but it is only an example. Could such a thing actually happen? Results in the most recent election for Governor of Maine closely resembled this with results in what was actually a four-way race with the votes distributed 38%, 36%, 19% and 5%.
Aside from the extreme case of the least popular of the candidates (judged by first preferences) being elected the introduction of a third candidate, even a relatively unpopular one, in an FCV election has the potential for changing the outcome of an election. For example, two candidates in a statistical dead-heat with 49.4% and 49.6% in a two-way race could easily be changed when a third candidate is added to the mix and that third candidate gets only 0.5% of the vote provided that small vote draws unevenly from the two major candidates. While such an event in the abstract may seem unimportant, only changing the outcome between two candidates with nearly the same vote totals, we can be sure that two of the candidates and nearly all of the voters would feel strongly that it was actually a hugely important matter.
The potential for a minor party candidate entering to become a spoiler in U.S. Elections is rarely far from the minds of politicians and political strategists. From time to time there are rumors of major parties helping to fund third-party efforts that are apt to spoil the chances for the other major party. Understandably, promoters of third parties and of third-party candidates soft-pedal the notion that there is any such thing as the spoiler effect. They try to assure voters not to worry about it and sometimes they are successful. But most voters do worry about it and and many do take it into account in deciding how to vote.
The United States currently has only two major political parties and except for brief periods in its history this has always been the case. It is true that neither of the two dominate parties want additional competition and and not simply because they don't like the uncertainty raised by the spoiler effect. They are not apt to favor any changes that will encourage the establishment of additional parties. But however politicians feel about it, would democracy not be better served if we had a voting system could tolerate more than two candidates with less danger of producing outcomes that are in conflict with the aggregate will of the voters?
Politicians of the two major parties can be expected to oppose serious minor-party competition, but the media seems to as well. For reasons about which one can speculate, the media denies third-party any coverage in newspapers or much time on radio or television. As a consequence, voters have become accustomed, when going to the polls, to finding candidates on the ballot that they know absolutely nothing about. Not surprisingly, voters rarely vote for these candidates.
Given FCV voting, the very size of the two major parties militates against the rise of any new party to become a significant contender in elections. A new party will surely start being very much smaller than either the Democratic Party or the Republican Party. A a small party is not apt to win an FCV election that depends only on voters' first choices; at best, a small party may get a chance to make its policy views known to more people but it is unlikely to gain significant power; without any power it has no record of achievement and that, in turn, makes it difficult for a small party to grow larger. In time, many supporters of that small party who have policy objectives they want enacted are apt to drift away to one of the major parties in order, hoping to have gain some influence there for their objectives.
With FCV, a voter must decide which one of the candidates to vote for. For some this is easy, but there are people who feel uncomfortable about choosing to vote for a candidate who they consider to be only the lesser of two evils; casting such a vote seems too much like casting a lie. Potential voters sometimes justify not voting as a way of expressing to the world know that they truly don't like any of the candidates. The world often fails to interpret their non-participation that way, but they feel that to vote for the least bad one would simply be dishonest; it is something they just cannot do. For this reason some have called for a "None of the Above" option to be on the ballot.
Arguably not enough, but many voters do show up to cast a vote for the candidate they consider best (though perhaps too often, the least worst candidate). When there are only two candidates to choose from this choice is is not so difficult, but when there are three or more then things can get more messy under FCV. For one thing, a voter may find several candidates acceptable but one seems, to that voter, not at all acceptable; the only option under FCV is to cast a vote for one of the acceptable ones -- but deciding which one can be a dilemma for a voter who does not have a clear favorite. The best strategic choice is to cast the vote for the one most likely to win, but which candidate is most likely to win is not always clear and in any event, some voters find the very concept of strategic voting to be quite offensive.
Another difficult situation is for a voter who really does have a first choice but that first-choice candidate seems very unlikely to win. Voting for the true first-choice candidate means not voting for a second-choice candidate who might actually have a chance of winning, thereby helping yet a third candidate that this voter opposes. This is truly the hold-your-nose-and-vote predicament. With FCV, the voter is all too often faced with casting vote for on the basis of strategy rather than an honest preference for a candidate. It can be a difficult choice for someone who really cares about the election and many will decide not to actually vote for their first choice. The issue of honesty (to the principle of voting for one's first choice) arises once more and this apparently bothers some voters intensely. They may conclude that the only honest thing to do is to not vote at all; this is not necessarily just an excuse. And even those who do vote often find the strategic considerations distasteful.
As a practical matter, in the United States, voters only rarely have much choice beyond the two major parties, even though there may be other parties with names on the ballot. Only a small minority of voters may even recognize the names of these minor-party candidates and those who do realize that these candidates have but a negligible chance of winning. American voters generally understand that the election will serve to choose between just the two candidates that are running under the banner of the two major parties. But with (effectively) only two choices (and our propensity for campaigns dominated by negative advertizing) there is a strong possibility that a given voter will not favor either of the two alternatives. Our FCV system ignores all of these quite possible subtleties of opinion and insists that each voter indicate which candidate he or she likes the very best.
The two-party system that we have chosen, or perhaps more accurately that has been imposed on us by our government's structure and our voting traditions, has consequences for our society at large. The example that comes to mind is the widespread aversion we have as a society to freely discuss politics; we are all too aware of how easy it is for heated arguments to begin so. to maintain the peace, we too often restrict our political conversations to situations where we are with people of a like mind. This is a sad commentary on a society that, at least at some level, claims democratic values and that, at least in theory, puts a high value on discussing issues to achieve understanding if not consensus.