Ballot Boxing

It never fails. Whenever I attend a political gathering – such as the ever-delightful FreedomFest – I am approached by someone pitching a third political party. I tell him that I don’t vote for third parties, as they almost never win, whereupon he tells me I’m a fool for voting for the lesser of two evils, and I reply that it’s better than voting for the greater of two evils. For my efforts, I get a hostile stare.

Let me see if I can clear this up a bit. I am by no means opposed to third parties. But we live in a democracy that uses the plurality method of voting. In this system, you can only vote for one candidate, and the candidate who gets the most votes wins. (On the presidential level, of course, the system is modified by the Electoral College mechanism). This is almost invariably a candidate of one of two parties, which tend to have major policy differences (with variation among individual members). In the plurality method of voting, you get only one vote, so to vote for anything but the lesser of two evils in the case where only two candidates have a chance of winning is in practice to vote for the greater evil.

This is not merely an abstract concern. I can recall several races in which if the few votes that went to the Libertarian Party candidate had gone instead to the conservative one, the conservative would have won, and would have voted for a number of policies I favor, such as lower taxes, tighter restrictions on welfare, fewer regulations, and increased school choice.

Moreover, if I am not supposed to vote for the person closest to my belief set who has a real chance of winning, but instead vote for the person closest to my belief set regardless of his chances of winning, even then, why would I vote for a third-party candidate? Faced with a mainstream candidate who shares maybe 70% of my views, or a third party candidate who shares maybe 80% , my best choice under the theory we are entertaining would be to write in my own name – thus voting for a candidate who shares 100% of my views!

So it is that I prefer to work within a large, traditional party, which is a coalition of disparate subgroups, funneling my money and efforts through the libertarian subgroup with the intention of moving the coalition more in that direction.

This choice is based upon the reality of our political system. I am certainly open to changing our electoral system in ways that would make me more inclined to vote for a smaller, more ideologically pure party. Such systems exist. Indeed, there are a surprisingly large number of other voting schemes, some merely proposed, others actually in use. While I certainly don’t want to get into grisly, statistical political science, perhaps a sketch of the alternative voting regimes might be useful.

Let’s focus on single-winner elections, i.e., those in which only one person is to be elected. And let’s focus on single-round elections, in which voters go to the polls just once per election. What are some alternatives?

Perhaps the biggest class of non-plurality, single-round voting schemes is called “ranked choice” or “preferential voting” systems. Probably the best known among preferential voting methods are Borda counts, Bucklin voting, Condorcet methods, and Instant Runoff voting. In all these methods, the voter begins by ranking the candidates (for the given office) in order of preference. So if Barack, Gary, John, Ralph and Ron are the candidates running for president, I might put “I” for Gary, “2” for Ron, “3” for John, “4” for Barack, and “5” for Ralph.

In Borda counts, for each ballot counted, any candidate ranked 1 would get 5 points added to his total (assuming in our imaginary case that there are five candidates), any ranked 2 would get 4 points, and so on down. The candidate with the highest point total (as opposed to the one with the highest number of l’s) wins the race. This system is used to some degree in a few countries – Kiribati, Nauru, and Slovenia come to mind.

In Bucklin voting, first-choice votes are tallied to begin with. If any candidate wins a majority, that ends the election. If none wins a majority, then the second choices are added in, and so on till a winner emerges. Bucklin voting was once rather widely used in the United States in the first part of the 20th century, but was eventually repealed or declared unconstitutional in all of them.

The Condorcet “method” actually refers to a group of similar counting techniques, and is somewhat more complicated. The aim is to select that candidate (if there is one) who would beat every other candidate in a one-to-one election. For each possible pairing, you count the number of ballots that rank the first candidate higher than the second. If one candidate wins all the possible pairings, he is the winner. Otherwise, some tie-breaker method is employed.

This is what one might expect from the mind of a great French philosopher and mathematician: sophisticated but depressingly intricate. Perhaps that is why no country I know of uses this method, although many private organizations do.

Finally, in Instant Runoff voting, the first count is of the candidates ranked first. If one of the candidates wins a majority, he is elected. Otherwise, the candidate with the lowest total of the first preference ballots is eliminated, and those ballots on which he was ranked first are recounted with the second ranked candidate now counted as first. Again, if a candidate now has a majority, the election ends. Otherwise, dump the lowest scoring candidate and repeat. (In a sense, this method simulates a multiple-round election; hence the name “Instant Runoff method.”)

This system is used to some degree in a fairly large number of jurisdictions, including Australia, Canada, Fiji, Great Britain, Ireland, Malta, Papua New Guinea, New Zealand, Northern Ireland, and Scotland. It is also used in a few places in the United States.

I don’t want to leave you with the idea that the Borda, Bucklin, Condorcet, and Instant Runoff methods are the only types of preference or ranked choice methods. There are others, but this is not a comprehensive study of voting methods. Nor do I want to leave you with the impression that the only kinds of single-winner, single-round voting schemes are either plurality or preferential voting ones (i.e., just schemes that ask the voter to pick one candidate or rank them all).

For example, in Approval Voting, the voter doesn’t rank the candidates, but simply marks the ones he approves (or at least can stomach). So, on the ballot consisting of Barack, Gary, John, Ralph, and Ron, I might mark “Gary,” “John,” and “Ron.” All the ballots are tallied, and the candidate with the most votes wins. I don’t believe this is used in any political venues, but it is used by a number of private associations.

Now, I am not quite as reverential toward our Constitution as many conservatives are. While I certainly view it as a creation of genius, and amending it as a matter to be undertaken with the greatest care, I certainly do not view it as perfect, now or ever. Nor did the Framers, I suspect, which is why they built in a perfectly useful mechanism for amending it.

So among the half-dozen amendments worth adding, I might include one permitting the adoption of ranked preference voting. I especially favor the Instant Runoff system. But as I don’t see that happening any time soon, I’ll keep voting against the greater of two evils, if you don’t mind.

Leave a Reply

Your email address will not be published. Required fields are marked *