Preferential Voting

Getting another option when your first choice is not available

Minority Winners

Bill Clinton

In the 1992 American presidential election the three major candidates were Republican George H.W. Bush, Democrat Bill Clinton and third party contender Ross Perot. Because of the latter’s anti-Washington stance and concern for a balanced budget, the billionaire businessman appealed mostly to conservative voters who were disenchanted with George H.W. Bush. Winning only 43% of the popular vote, the liberal Bill Clinton won the election over the divided conservative candidates despite them totalling 56.3% of the vote.


George W. Bush   Ralph Nader

A mirror image situation happened to occur only eight years later in the 2000 presidential election where the left wing Al Gore was running against the right wing George W. Bush. There was an independent, the far left Ralph Nader, also running. Many Americans on the political left were not sure who to vote for: Gore or Nader.  Nader was estimated to have no chance by pundits and was viewed as nothing more than a spoiler. One of the often heard catch calls was “a vote for Nader is a vote for Bush”.
In the end Gore actually won more votes, but due to the gerrymander effect he unfortunately won fewer electoral votes (officials who actually decide who shall be President). Putting aside the controversy of the Florida vote recount, it is estimated that if the 2.74% of the voters who voted for Ralph Nader has instead voted for the other liberal candidate, Al Gore, then the result of the election would have been different, and the minority conservative candidate would not have won the election.

Jacques ChiracMarie Le Pen

French presidential elections are a two stage affair. The first election is for all aspirants vying to occupy the Élysée Palace and only the two candidates winning the highest number of votes go on to the second election to decide the winner. Jean-Marie Le Pen of the small, right-wing extremist National Front had the second largest individual vote in the first round of the 2002 elections, mainly because the leftist parties happened to split into smaller groups. The final stage of the election wasn’t between the two most popular candidates but between one of moderate popularity, Jacques Chirac, who received 20% of the first election votes, and one outsider. Whereas Chirac might have won the election under normal circumstances, the 2002 election definitely did not give an indication of who was the preferred candidate for the majority of French voters.



President of Singapore Tony Tan

Unlike most elections in presidential type democracies where there are only two major party candidates, three or more established parties or candidates enter the Singaporean presidential elections. In the 2011 election Tony Tan Keng Yan won over three other candidates, gaining 35.2% of the vote. The new president took office despite the fact more than 64% of the voters chose someone else.


Local Member elections

Below even a third

In the British general elections of 2010, the Liberal Democrats candidate, Simon Wright, won the constituency of Norwich South with a grand total of 29.4% of the vote. He thus became the MP to represent Norwich South in the House of Commons despite the fact that more than 70% of his constituents voted for someone else.lowest mandate ever!

Mr 24.5%

Poor Simon was not able to hold on to his representation ‘record’ for long. Five years later in the 2015 elections an Alasdair McDonnell, representing the Social Democratic and Labour Party, set a new benchmark by winning his seat of Belfast South with the approval of less than a quarter of the popular vote, leaving over 75% of his voting constituents desiring someone else.



All the above elections were under an electoral system known as first-past-the-post, aka, pluralist voting.

the first, who may be well outnumbered by the rest
The first in, the front runner in the race, is served with the laurels no matter how few votes he/she actually received and no matter if even a majority of the voters specifically dislike their new rep.




Why not the electorate’s least objectionable candidate?

An antidote to this fault in democracy are systems to allow voters to not only indicate their single preferred choice, but also further choices, thus designating candidates they specifically do, and do not, approve of.
In deciding the election winner all candidates’ primary (first preference) votes will be counted and if one candidate has a majority, then a winner is declared and the counting is finished. However if no candidate managed to prove himself popular with the majority of the electorate, then counting continues to find out who is least objectionable to the majority. This is managed by first eliminating from the election the candidate with the lowest vote count. His / her votes will now be distributed amongst the remaining candidates according to the preference indications on each ballot paper. This will continue to be done with the next least popular candidate until a candidate finally achieves a majority of votes.

Preferential voting is also known throughout the democratic world as: Choice voting, the Alternative vote, Instant-runoff voting and Ranked voting. Four variations of preferential voting are:

  • Contingent voting:
    • When there is no majority winner after the first count, there is only one more vote count which uses all the preferential votes from the candidates except the top two.
  • Optional Preferential voting:
    • The voter has the option to mark off only as many candidates as one chooses. Hard to see the logic in doing this unless the voter values all the remaining candidates as equally bad.
  • Range voting:
    • Range voting alloRange Votingws voters to express preferences of varying strengths. Voters rate all candidates each with a number within a prescribed range (eg. 1-9). The virtue of this system is that not only do you rank the candidates in your order, but also by the degree you like or dislike them.  If a voter truly liked one candidate, could tolerate two others (by, coincidentally, equal amounts) disliked two more, and simply loathed the last one, she could mark her ballot paper in order: 10, 7,7,5,4 and 1. The results of the election are determined by the average each candidate receives.
  • Approval voting: 
    • You mark all the candidates with either a ‘yes’ or a ‘no’. Strictly speaking, not preferential as you don’t declare your order of preference for the candidate of the single-member electorate. However, you have the option to declare a group of candidates as equally acceptable over others, or just one as your choice.


An innovation universally embraced?

So why isn’t preferential voting utilized in most of the democratic world where the UK, USA, Canada, India, the Philippines, Singapore and Malaysia still embrace First-Past-the-Post?

The general response is that it is confusing and too complicated? This does not speak well for how certain governments view the intelligence of their average citizen.

Over the last 30 years Australia, with preferential voting utilized in both single member and proportional representation elections, has had an average rate of informal voting of only 4.3%. Considering Australia has compulsory voting where some people forced to the polls express their outrage by either not bothering to understand how to vote, or intentionally invalidating their vote, it would seem that introducing preferential voting has had little effect on the average citizen wishing to make a valid vote.



The 2013 Australian Senate Election


Suggested Reforms

Legitimate Reforms


As compared to the Australian House of Representatives, members of the Senate are elected by a hybrid proportional representation system utilising preferential voting. More than any other election in its history, Australia’s 2013 election created a broad range of parties to occupy Australia’s upper house and thus decrease the influence of the two traditional party groups, the Coalition and the Australian Labor Party. Almost one in four Australians (23%) voted for a minor party, causing members of these traditional party groups, as well as many in the media and academia, to respond, not by welcoming the increased multiparty democratic forum, but surprisingly, by calls for the electoral system to be “reformed” so as to return to a time when the number of parties at the nation’s forum was kept to a minimal few.
Mr Stephen Lesslie of Electoral Reform Australia has even gone so far as to declare “…50 registered parties is not a sign of a healthy democracy.”

The reasons given for this retroactive step are given below together with responses

Donkey-Vote Luck of the Draw

One of the winning, so called, micro parties, the Liberal Democrats, is said to have had significant help by being placed by the random draw at the #1 position on the ballot paper.  This indeed is probably true and even the party itself does not deny this. However this is the nature of practically all voting systems and also is nothing new with Australian elections. It does seem highly hypocritical that critics should complain of it now when the “wrong” party profits from it.


Voter confusion with party name similarity

It is claimed that one party profited from having a name similar to that of a major party, the Liberal Party of Australia. This is in all probability true and reform to aid political party identification would not inhibit democracy, and should not necessarily be opposed, however in this particular instance one thing needs to be noted.
In contrast to the Liberal Party, that particular party in question, the Liberal Democrats, has voluntary voting on its platform. Considering it would be very reasonable to believe that those forced to the polling booth and wishing to be in and out as quickly as possible, would be more likely to make a party identification mistake than a voluntary voter, is it not instead, rather a “chickens coming home to roost” problem for the Liberals?


Voters, not parties, should declare where preferences go

     “People have got to understand who they are voting for and this system doesn’t allow that.”
                                                                      Nationals Senator Ron Boswell

When commentators complain that the average punter doesn't know who their preferences go to when voting above the line, the obvious implication is that the voter has in their own head a list of their preferred parties from one to 110. Really? How many voters know the difference in policy platforms between, for example, the Christian Democrats and Family First? This does seem to be a furphy put up to disguise the intention of simply denying electoral success to minor parties. Voters’ knowledge of the political scene is generally limited to three or four parties they do like and three or four they detest. Basically they have their one desired preference and when they give that party their tick above the line they also hand over their long list of preference votes to the trust of that party for one of two reasons: either because they deduce that if they like the policies of that party they assume that that party’s preferences should thus be somewhat similar to their own, or because they realise that whatever the preference deals that have been done, have been done to maximize chances of success and as they want success for their party, they have no objection to their own vote supporting that deal.
Also it must be remembered that a voter has what is called constructive knowledge (knew or easily could have known) of the preferences he was nominating because all preference lists are published prior to the election.clandestine dealing

The basis of the whole criticism of preference arrangements between parties is that there is something sinister and immoral about this “behind-closed-doors-trade”, to quote erstwhile Senator Bob Brown, as though some party apparatchiks are being paid a back-hander to preference certain parties. Parties do arrangements simply to maximise their chances of electoral success, and all such “grubby preference deals”, to quote Senator Nick Xenophon, done in allegedly smoke-filled back rooms, are made public before elections anyway, so voters have fair warning of who they will ultimately be supporting when giving their above the line vote to any party.


Winning with only a miniscule percentage of the primary vote

One of the most spurious objections to the validity of preferential voting is that some candidates win a seat with only 2% or even less, of the primary vote. The rationale being that a second, and beyond, preference vote is considerably inferior in quality to a first preference vote. The alleged response to this “problem” is that the law should change to deny victory to any candidate without at least 4 or even 5 percent of primary votes.

[More than one commentator has even gone as far as to omit the descriptor and stated that some micro party candidate has won on only a few thousand votes. The logic used is quite challenging. Apparently  as the elector only voted above the line, then his second and further preferences were not in fact his, but only that of his party’s. Thus the preference votes, if needed, simply  float around in the ether and no candidate at all should have the right to claim them.]

The argument for discounting further preference votes seems to be that, for some reason, a voter’s submitted ballot paper loses value the more often it tries to be part of a full quota in preferential counting. The implication in this criticism of the preferential system is that even though you are allowed a second preference, in effect a second vote, it is somehow not as good a first preference vote, thus the reason critics talk about ‘primary votes’ but ‘only preference votes’.

But if someone votes for candidate X, is then, so to say, told that X has no chance, so then subsequently votes for candidate Y, that second vote (or any subsequent vote)  is still their legitimate single vote. How can it not be? Your vote will bounce off a number of other failed candidates first, but ultimately it goes in the bank of only one candidate. How can the misfortune of your primary candidate not being sufficiently popular mean that you are subsequently denied the full value of a vote to expend somewhere else?

To his credit, Senator Nick Xenaphon, albeit while still calling for reform, has declared that a prerequisite of having any minimum quota of primary votes would be improper.


Forced to select, in order, up to 110 candidates

Those voters not wishing to be tied to the preference order of their chosen parties have the option under the system to mark off their own preference of parties. However the complaint is that this is unnecessarily arduous in that one has to go to the trouble of marking off all of up to 110 names (if living in NSW). The change suggested is to offer what is called option preferential voting.
This is where the voter completes the below the line preference list, not according to his party’s call but as to his or her own preferences, and only voting as far as one wishes.table cloth ballot paper
If this reform were enacted it would, in fact, do no substantial harm to the electoral system. Perhaps it would decrease informal votes, however the reason it is not as yet allowed is that voters may be unaware that if they stop voting too soon, and their preference list becomes exhausted by their last vote (not making up a quota on any of their chosen candidates), then their vote is a null and in effect wasted.


Ending up with a representative you had never heard of

As any student of civics is taught, preferential voting essentially means that the voters do not have to put up with a truly objectionable candidate merely because he or she won more votes while still being short of a majority. The Australian system, in contrast to most other democracies, allows the compromise choice. It is deemed better for the majority to be represented by a somewhat acceptable candidate who, in the voter’s estimation, will support relatively benign policies, than by a totally unacceptable candidate supporting distasteful policies. The virtue of the compromise candidate is not that he or she is a repository of wisdom on a broad range of policies. In most cases it is simply that he is not any of the other choices. He is the “begrudgingly will accept” middle-of-the-road candidate who can unite the majority of opposing political sentiments.
 The fact that few, or even absolutely no one, might prefer the ultimate winner as their first choice candidate is totally irrelevant. In fact, the reason he would be successful as a compromise candidate, being bland, anodyne and harmless, is probably the same reason few would want him as their primary preference. To deny him or her that Senate seat is to deny those 400,000 people (or whatever the quota size is) their chance not to be stuck with an objectionable representative.

Lottery Aspect to Senate Voting

Noted election analyst Antony Green has commented that winning a seat has become a lottery for the micro parties. Reform “…would remove the lottery aspect of the present Senate race, and force parties and candidates to do what they are supposed to do - campaign for votes, not negotiate outcomes through preferences.”

He is correct in that there is an element of luck in being elected: with the same number of accrued votes, your chance of success still depends on the votes other parties you exchanged preferences with, get. However he is perhaps a little disingenuous to imply that parties should instead do the right thing, so to speak, and get out there and campaign for votes to be rightly deserving of a Senate seat. 

Australia has 76 members in its democratically appointed Senate and thus, technically, each Senator represents 1.315% percent of Australian voters. Not that it did them any good but, accumulated throughout Australia, the Sex Party at the 2013 election, and Family First and the Liberal Democrats (where they were not in a privileged spot on the ballot paper) at the previous 2010 election did in fact attain that percentage.  So if we have allocated 76 members’ seats for the federal Senate, is it really too much of an extreme position to take, to ask that a party that wins one seventy-sixth of the vote should have a right to claim one of those 76 seats?

Those parties have already done “what they are supposed to do” to sell their message to one segment of Australian voters. Why should they have to campaign further because the electoral system is arranged such that the natural quota of less than two percent eventually becomes over 14% for various highly questionable reasons?

The irony is that despite the micro parties being accused of “manipulating” and “gaming” the system, it appears that over the last century it is the larger established law-making parties who have taken advantage of, and done nothing to modify, a system, allegedly proportional, which prevents 1.3% of the voters from electing 1.3% of the seats in the Senate.


Denying the government its mandate

The strangest comments made regarding the success of the new independent Senators were about their threat to the new Prime Minister’s “mandate”.
Compere Leigh Sales of the ABC’s 7.30 Report echoed many when she addressed the following question to some of the Senators-elect, “Tony Abbott has been elected with a really substantial majority. If you found yourself in this position would it really be fair to use your Senate slot to block any of his legislative agenda?”
It is a similar sentiment to that expressed by University of New South Wales Law Professor George Williams, who referred to the new members of the Senate who would be in the position “to hold the government to ransom”. (Will all the Coalition MPs be held in the Parliament house basement chained to water pipes and fed take-away pizzas throughout their ordeal?)plebiscite
It is difficult to understand the premise these statements must be based upon.
The fact those Senators-elect got voted in, even if by preferential votes, means that their voters chose them over Coalition Senators and thus were attempting to give a mandate to someone else.
On analysis, the question is even stranger. In 1917 Australian Prime Minister Billy Hughes held a plebiscite to gauge public opinion on conscription. He didn't need to, to pass legislation and make it law, but he wanted to be assured he had a mandate. So why are we to believe today that any and all of Abbott’s policies have a mandate merely because he won more seats in the lower house? Many have said that it was Rudd who lost the election rather than Abbott’s policies that won it. Also, is it not reasonable to believe that the Coalition’s support may be divided up into a fraction who wanted the boats stopped, a fraction who wanted abolition of the carbon tax, and perhaps a fraction wanting diminishing foreign aid? Are we to assume that the majority of Australians want all of his policies adopted, including the billion dollar paid parental leave scheme?

Suggested reforms for these “problems”

Above-the-Line Preferential Voting

Commentators such as Senator Nick Xenophon, erstwhile Senator Bob Brown and current Attorney-General George Brandis have suggested that so called ‘above the line preferential voting’ should be introduced to Senate voting (while leaving ‘below-the-line’ as is). This would mean that instead of voting above the line for just one party with its attendant full list of preferences, its group voting ticket (GVT), one would vote in sequential order for as many parties as one wishes but each vote would be for only the candidates of that party, and no further. The alleged virtue of this system is that voters would not be giving support to parties they were not aware of (parties they may, or may not, express any particular sentiment towards), which might otherwise be in their preferred party’s list of preferences.

This is indeed true. However there is one thing the supporters of this suggestion fail to mention. Micro party supporters are not the ones complaining about their preferences electing unknown candidates.

For years the approximate 20% of the electorate who don’t support the major parties have been voting for their individual small parties and ending up after elections with approximately 0% of representatives in the Senate. This was because traditionally after voting for their first, second or third preferred micro party, they then went back to an established major party for their third or fourth preference. A practice which always got them nowhere but that established major party, their final successful candidate. Finally they have awoken to the fact that the only way to get ones preferred party elected is to give all preferences to small parties then hope, perhaps by the luck of the drawer, that your chosen party may win that final seat. For the very small percentage of micro party supporters who still might happen to complain, they still have the option to vote below the line anyway.

The ‘reformers’ in this situation are offering a solution to a problem which in fact doesn’t exist. Apart from members of larger parties attempting to beat up an issue, they don’t know anyone of smaller parties who complain about unintended voting results. However what they do know is that introducing above-the-line preferential will definitely increase the exhausted and informal vote, and thus decrease the vote of micro parties who depend to a much greater extent upon preference votes.

While supporters of the larger parties will only have to vote, in order, for up to three parties above the line, if even that, micro party supporters will definitely have to fill in, in correct sequential order without mistakes, almost the complete list to guarantee that their vote does not become exhausted (discarded before ever being used) before it would be effective. It does appear a spurious claim to suggest reform, which will undoubtedly seriously impact upon micro parties, for the sake of a small minority of voters who could  alleviate their problem anyway by simply voting below the line. 

A Minimum Threshold of Primary Votes to be Elected

  • A quota of 4% is “a solid base of support in the community”
    Professor George Williams, Sydney Morning Herald 10th Sep 2013
  • “No doubt the first response will be to clamp down on micro-parties to be elected. This could include tightening party registration rules and introducing threshold quotas a party must pass to win election.”
     Antony Green, Sydney Morning Herald 10th Sep  2013
  • “…not allowing candidates/parties with less than a certain percentage of the [primary] vote to collect preferences” Amanda  Vanstone, The Age, 23rd Sep 2013 

The problem with this argument is that it is very feasible that not all of six Senate candidates in a state can be guaranteed to accrue even 4% of the primary vote. Coalition politicians Julian McGauran, Judith Troeth and Bridget McKenzie as well as Labor’s Stephen Conroy all, in at least one of their elections, won their Senate seats on primary votes of 0.04% or less

It would be rather ridiculous to create new electoral laws which might actually prevent some states sending their full allotment of upper house Members of Parliament to Canberra. How should the legislation then be written? “A candidate must first gain 4% of the primary vote, unless he or she is a member of an established larger party…”?

  • “Professors of politics Dean Jaensch and Clem Macintyre … reject minimum thresholds as anti-democratic” Senator Nick Xenophon (who received 25% of the primary vote in the 2013 election).  Canberra Times, 12th Sep 2013
  • ‘I can understand why people might proffer the idea of some sort of threshold. However, every democratic inclination that I have suggests that that is an unfair bias. It is a disincentive for people to put their hand up for office, and it also discriminates against smaller parties, be they minor parties or micro parties”  Erstwhile Senator Natasha Stott-Despoja,  Radio National Breakfast 10th Sep 2013

Legitimate Reforms to the System

Luck of the Ballot Drawer

Party positions on the ballot paper are decided by ‘drawing out of a hat’ random choice. This is certainly an improvement on earlier versions where candidates’ names were listed alphabetically.  However there still remains the problem of the so called ‘donkey vote’, especially prevalent with compulsory voting, where some people simply tick the first box in sight so they can be out of the voting booth as soon as possible. Thus the candidate / party lucky enough to draw the first position gets an unequivocal advantage.

A solution to this, as practiced in Tasmanian elections, is to introduce what is called Robson Rotation. With this system there is no need for a random draw and instead multiple sets of voting papers are printed each with a different party listing.

Name Confusion

It is without doubt that in the 2013 Australian Senate election the Liberal Democrats gained an advantage due to the similarity of their name to that of the established Liberal Party of Australia. In fact a decade earlier the Liberal Party took the party ‘liberals for forests’ to the Administrative Appeals Tribunal claiming they had no right to use their name. In its ruling the AAT ruled that in politics the word ‘liberal’ is quite a  generic term due to the fact it is used  widely in all manner of political and social discourse, and it would be wrong for one political party to claim ownership and sole proprietal rights.

However what was perhaps remiss of the Australian Electoral Commission was to fail to make any concerted effort to warn voters of possible confusion. This would not have been that difficult. Published print ads and even posters in the voting booth could list those parties with similar sounding names together with identifying their dissimilar attributes.

Reforms the Major Parties won’t be Supporting

Voluntary Voting

Two of the three oft repeated criticisms of the 2013 Senate election was the advantages one successful minor party had due to the luck of the voting paper draw and the similarity of its name to that of an established major party. Some of their support came from voters who either took little time to check which party they were giving their support to or in fact took no time at all and simply ticked the first box in sight. It would be quite fair to say that without those counted votes, the result would have more correctly reflected the wishes of those Australians concerned about their political future. It does not take a genius to work out that those voters careless with their vote are those who don’t wish to be there in the first place and are simply performing “their duty” as expeditiously as possible.

Thus an effective solution would be to do away with that rarity in the democratic world, compulsory voting, and return to the almost universal system, whereby only those truly concerned about their country’s future should be the ones to direct it.

Remove Artificially High Quotas

The third main criticism of the election related to the preferential system of voting utilised. Where a person’s vote did not immediately become part of a quota to elect a candidate, it subsequently went into, what may be called a holding pattern of floating preference votes, which after various tries, would eventually become part of a quota to elect someone, a candidate often way down on the list and not that well known, if at all, with the voting public.   

While suggestions of dubious democratic legitimacy have been made to remedy this perceived problem, possibly the simplest of all have been ignored.

Remainder preferential votes have become such a big issue in contemporary times because there are becoming so many of them. Many of the political parties engaged in the preference swap negotiations garner less than 2% of popular support, and yet by current law, to win a seat in the Senate, a quota of 14.3% of the vote must be accrued. At the last election 21 of the 36 states’ Senate vacancies were won on only primary vote quotas, meaning that the remaining 15 vacancies, over 40% of the vote, went into this preference-exchange, holding pattern maze.

But why does this quota have to be so high? With 76 seats in the Australian Senate the natural quota is only 1.35%, a fraction of the existing artificial level, and small enough to ensure that almost all primary votes could in fact elect a candidate. Yes there would still be some preference votes left over, but far fewer than the current 41% of the total.

The arguments made to defend high quotas tend to ignore not only the loss of political representation for minorities but also the clarity and confidence gained from elections where preference dealing would play such a little effect.


Why Not Proportional Representation?

[Home] [Glossary] [Chartists] [Malapportion] [Voting] [Info] [About Us]