RSS Feeds

How the single transferable vote system works

HOW IT WORKS: Understanding the quota is key, writes Patrick Smyth

The correct description of our voting system is a bit of a mouthful - technically speaking we use the "single transferable vote system of proportional representation in multiple seat constituencies". Or STV for short.

"Single" is crucial - because we each really have only one vote, despite the fact we are asked on a ballot paper to vote 1,2,3... in the order we prefer the candidates. The preferences 2,3,4, etc are our instructions to the election counter, the returning officer, on what to do with our "single" vote if our first preference candidate is not in with a chance of election. Or if he/she, in being elected, has amassed more votes than he strictly needs.

In either case, the whole vote, or part of it, is "transfered" to another candidate.

Crucial to understanding how this is done is the "quota" - this is simply the number of votes needed to get elected and it varies with the number of seats in a constituency.

In a single-seat constituency - like the presidential election - the quota is 51 per cent of votes cast.

Clearly, only one candidate can reach this figure, so another way of defining the quota is that it is "the number of votes which can only be achieved by the number of candidates to be elected."

In a two-seat constituency, it would be possible for three candidates each to have 33 per cent of the vote (33 x 3= 99), but only two can possibly have as much as 34 per cent (34 x 3= 102). So 34 is then the winning line, or quota.

In a three-seat contest, using the same logic, it is theoretically possible for four candidates to each have 25 per cent of the vote. But we only want three elected - only three candidates can achieve 26 per cent, the quota in this case.

Applying this reasoning the general formula for the quota is as follows:

Quota = Total valid vote + 1 Seats + 1

(The total valid vote is the total vote less the spoiled vote)

If candidate A reaches the quota he/she is deeemed elected and the returning officer then subtracts the quota from A's vote to calculate the surplus.

That surplus can then be divided between the remaining candidates in proportion to A's second preferences - this ensures that votes for A are not wasted if there are more than enough to elect him.

Similarly, if there are no candidates elected during one of the many counts that the system usually requires, the returning officer counts the second preferences of the candidate with fewest votes and distributes them accordingly.

The two processes - distribution of surpluese and eliminations - are repeated, often many times, carrying on through the available preferences, until the required number of candidates is elected.

At the end of the election it often happens that the last elected is elected without reaching the quota - this happens because voters often do not use all their preferences up and so votes end up transfering nowhere.

This feature of the system has two effects - it lowers the effective quota at the end of the counting process allowing some candidates to be elcted without reaching the quota, and it means that those parties or individuals who are better at attracting lower preferences get a bonus from the system in terms of seats.

It's worth noting - contrary to a widely held misapprehension - that voting all the way down the list, even for candidates way down who you regard as objectionable, does not in any way advantage them over your higher preferences.

For the spectator in the grand sport of election watching it is possible at the early stages of an election to work out candidate X's chances of election by studying the early trends: firstly, how has X's party done? By adding up how many quotas X's party colleagues have collectively achieved on the first count, it is usually possible to estimate how many seats the party should take. Usually, but by no means allways, the party's candidates will then be elected in the order of their first preference vote.

And by looking at who is unlikely to make it, or where surpluses are likely to arise, it is possible to estimate where transfers will go - sometimes they will be to party colleagues, or coalition colleagues, or to candidates from the same town, or same sex ...

Party tallymen have such nuances down to a fine art. Endles speculation on the subject makes the election night the joyous political equivalent of the day at the races.