The following is an edited version of the address given to the Annual General Meeting of the Victoria-Tasmanian Branch of the Proportional Representation Society of Australia by Stephen Lesslie, President of Electoral Reform Australia.

Mr President,

We, Electoral Reform Australia, are looking for three things in any reform of an electoral system. They are:

1. The complete abolition of any form of above the line voting.
2. The rights of voters to have their vote counted.
3. Equality for voters.

The complete abolition of any form of above the line voting

Very briefly, for any reform to succeed it is essential that voters have returned to them the right to control and express their own preferences to the extent that they choose.

Without the above the line boxes, the two sets of instructions and the big distracting black line the ballot paper instantly becomes much smaller and less intimidating and more user friendly.

The rights of voters to have their vote counted

What sends my blood cold is the phrase, ‘we will require voters to…’. It then continues with various endings like ‘number all the candidates’, ‘number twice the number to be elected plus one’ or various arbitrary numbers like six, ten, twenty, etc.

Fully optional preferential voting is a fundamental right of the people. Anything else is a civil rights violation. You don’t have to be Margaret Thatcher with her poll tax or the Ku Klux Klan with their burning crosses to prevent voters from exercising their franchise – making the ballot paper more complex is just as effective.

The vote belongs to the voter. We laugh at the United Kingdom and Canada with their first past the post system but if a voter votes for a candidate, they count it!

We want fully optional preferential voting and we maintain that with fully optional preferential voting voter participation actually goes up – that both informal and exhausted votes drop. What is voter participation? Turnout minus the informal vote minus the exhausted vote. It is pleasing, but irrelevant, that voter participation goes up with fully optional preferential voting as we would still object to any form of compulsory numbering even if there was clear evidence that voter participation would fall without it.

It is their vote – we cannot require voters to do anything except give a single unambiguous first preference. Advise, recommend, encourage and educate voters that continuing preferences to further candidates can never harm the chances of their first choice: yes; but require: no.

While we are at it, let’s get rid of that other nonsense put forward by various commentators, including Kevin Bonham in Tasmania, who are worried about the problems posed to parties should they run, to quote Bonham, ‘strong candidates who have a cult-like popular appeal’. They appear to be worried that should such candidates receive too many single first preferences, the surplus might increase to above one and thereby reduce the vote available to the party as a whole.

My response to that is:

• STV elects candidates not parties.
• There have never been any such candidates. Can anybody name one?
• It does not happen. Can anyone cite an example of this happening when available candidates from the same party were still in the count?
• A full quota of votes must be just a single [1] before a ‘problem’ begins to manifest itself. Candidates are still grouped in party columns and it is very hard for voters to resist voting for other candidates within the party group. A candidate with 1.5 quotas will need to have two thirds of their votes cast as a single [1]. In the 2012 ACT election for the seven member electorate of Molonglo, Katy Gallagher received 23,996 votes – over two quotas. Only 124 of these were single [1] votes and none exhausted.
• Since this candidate is a hugely popular candidate they will, by definition, take more votes from opposition parties and this, despite the fears, will benefit their party.
• The voting instructions on the ballot paper, how to vote tickets, and the AEC’s advertising campaigns will encourage voters to express their preferences to the maximum.
• In the unlikely event that this occurs, counting the ballot by Meek will reduce the impact.
• It does seem an overreaction to have votes at every election declared informal, even those given to non-cult-like candidates, just to prevent the possibility of this happening.

Equality for Voters

Under the Australian Constitution, unlike local government and state governments, it is impossible to ensure that every multi-member electorate will have the same number of members. Each state has a predetermined number of members and prime numbers abound.

We should strive for an STV electoral system that does not have above the line voting and where the quotas are as even as possible. There is no reason why the district magnitudes within an individual state should vary by more than one.

Different district magnitudes are designed to prevent electorates returning an even numbered cohort of members. But even numbered electorates are not the problem that some people think. It is not necessary to achieve a two party majority vote in every electorate. The Liberal, National, Labor and any minor party members elected from a Western Victoria multi-member electorate will not be caucusing about how to solve the problems in Mildura.

Electorates are just the device by which members are elected to parliament. Only having uneven numbered electorates will not prevent the 150 member House of Representatives ending up as 75 Government and 75 Opposition members. Indeed, an undue insistence on uneven numbered electorates may distort the proportionality of the entire election. It certainly distorts the quotas.

Two small wins for the same side of politics in two five member electorates may give a 6:4 result, where the true proportional result may actually be 5:5. The socio-economic divide in Australia is sufficiently wide and diverse that a 54.55% two party win, and therefore a genuine 6:4 split in a ten member electorate, would not be unusual.

We have come to ‘electoral stasis’, so a quick definition would be appropriate.

Electoral stasis occurs when an electorate cannot realistically change its political composition regardless of the swing occurring in a general election. In an STV proportional representation ballot, electoral stasis is the equivalent of a safe seat in a single member electoral system.

Electorates in electoral stasis are also so small that it is generally not possible to change the composition of members within the same party. The problem for small, even numbered electorates is the same problem that we get with small, uneven numbered electorates: they are too small and there is therefore a high probability that they will be in electoral stasis.

When formulating policies or allocating campaign resources, party strategists would not need to consider seats that are in electoral stasis. The major parties would not seriously campaign in these electorates or run extra candidates, so voters would not be given a choice of candidates even within the same party.

South Australia is the problem when trying to devise an STV electoral system for the House of Representatives.

With eleven members (a prime number), South Australia cannot be divided into sensible electorates. Any division has one or more of the following problems:

• Quota difference is too great. One electorate returning 5 members and two electorates returning 3 members gives an 8.3% quota difference. Why should a candidate need 25% to be elected in one electorate but only 16.7% in an adjoining electorate?
• Electorates that are too small and are therefore, potentially, in electoral stasis.
• Electorates that are too big geographically to service, or more to the point, that are perceived as too big.
• Gerrymandering. Even a 5/6 split can be gerrymandered by the simple decision of having the Division of Grey in the five or six member electorate.

Dividing South Australia into small electorates is unsatisfactory, but having South Australia as one electorate with eleven members achieves many positives:

• The same quota of 8.34% for every voter and every candidate.
• The electorate will not be in electoral stasis and the likelihood that political swings will change the political representation is high.
• Parties will be forced to run sufficient candidates, enabling voters to choose candidates from within party groups.
• Gerrymandering will be impossible.
• Redistributions will be unnecessary. Should the State gain or lose a Member then the electorate just returns one more or one less Member.

Once it is decided that South Australia should be one electorate, the other mainland states can be divided to make electorates as close as possible in size to South Australia’s eleven. The states that have to be divided are divided so that the quota difference between the electorates is at a minimum.

State	Members	Electorates
QLD	30	3 x 10 member
NSW	47	3 x 12 member; 1 x 11 member
Vic	37	2 x 12 member; 1 x 13 member
WA	16	1 x 16 member

Australia’s Constitution makes it impossible to include Tasmania within the proposed model. But Tasmania, with over a hundred years’ experience, is better able than most to utilize STV to its maximum, even though it is limited to a single electorate returning five members. (Tasmanians may also take comfort from the fact that constitutionally they are entitled to five members but mathematically to only four members.)

The quota in the mainland States varies between 5.55% and 9.09% – a difference of 3.54%

Compare this to the differences between 3 and 5 member electorates (8.33%); 5 and 7 (4.16%); 5 to 9 (6.66%); and 3 to 9 (15%).

The advantages of the Electoral Reform Australia model are many.

Electorates returning between ten and sixteen members incorporate all those ideals that we in the proportional representation movement hold so dear, and then most of us ignore.

 

• Equality

 

The maximum quota variation within individual States is 0.65% and between States is 3.54%

 

• Proportionality

 

Apart from Tasmania and the Territories, every electorate has a quota that is less than 10% and more than 5%. There appears to be a consensus around the world that, where thresholds apply, they should be set at 5%.

While I don’t agree with the arbitrary nature of thresholds, I do believe that any candidate unable to poll at least 5% should struggle to be elected. These quotas will achieve that. I also believe that candidates who achieve 10% of the vote have met the proportionality requirement and deserve to be guaranteed election via a quota. These electorates also achieve that.

 

• Choice of Candidates

 

Electorates of ten or more will force parties hoping for a favourable result to stand more candidates than they might expect to be elected. This will give supporters of these parties more choice. To avoid looking like a bunch of clones, the parties will also have to offer a diversity of candidates.

 

• Electoral Stasis

 

There will be no safe seats. The quotas for all these electorates are sufficiently small that even moderate political swings are likely to change the political composition of the electorate, meaning Parliament as a whole will be responsive to the mind of the electorate.

 

• Redistributions

 

Redistributions are easy. Should a State gain entitlement to a seat, all that is required is for the most populous electorate to gain another member. If a State loses a seat all that is required is for the least populous electorate to lose a member. However, if even-numbered electorates are forbidden, redistributions would entail major boundary and district magnitude changes.

 

• Members Of Parliament servicing electorates

 

There is a mistaken concern that with large multi-member electorates, Members of Parliament will not be able to adequately represent the voters. Parties that elect more than one candidate will be able to share responsibilities. Elected members will live and have past careers in different parts of the State or electorate and, because of the rotation of candidates on the ballot paper, will probably have campaigned in different parts of the State or electorate. It would be electorally disadvantageous to the party if they did not. These more popular parties will delegate the appropriate member or members to deal with an area or issue. Voters will understand this – their concern is being dealt with by a member of the party for whom they voted.

Elected members who are the sole representative of their party in a State or electorate are under a disadvantage and will have to decide how best to service their constituents. Senators from a party with only a single member have the same problem. This is an inevitable downside of belonging to an unpopular party. The upside is that without these large multi-member electorates with their small quotas, they would not have been elected in the first place.

Senate Reform

Electoral Reform Australia’s views on Senate reform are very simple.

 

• No above the line voting of any kind.

 

Voters must be given back the right to express their own preferences to the extent that they choose.

 

• Fully optional preferential voting.

 

Just a reminder: voter participation goes up with fully optional preferential voting because both the informal and exhausted votes go down.

 

• A simple linear rotation of candidates.

 

In a modern STV ballot the rotation of candidates is essential.

• It spreads the vote of the popular parties amongst all their candidates.
• It allows voters to choose their favourite candidate with a reasonable chance of success .
• It helps keep the occasional single [1] vote from exhausting.
• It helps ensure that votes for micro, or less popular parties, eventually help elect candidates from the more popular parties instead of the other way around.
• It helps ensure that quotas have some meaning in determining which candidates are elected.

The Robson rotation, however, especially as it has been modified in the ACT, has taken this concept too far.

There are too many variations and, while there would be no problem with six candidates on a Senate ballot paper, it would be impossible with twelve candidates in a double dissolution. There will always be the possibility that some billionaire will think that running twelve candidates will be beneficial.

Senate elections are largely anonymous with regard to individual candidates. Who can name two of the four Labor Senators for Victoria and how many Victorian Labor voters would be able to name any? Yet even a New South Welshman can name the Victorian minor party Senators that no one voted for.

With the Robson rotation it is likely that where second and subsequent candidates are elected they may beat their fellow party candidates not on merit but by luck, or worse because there is just enough avoidance of candidates because of their name or sex to change the result.

Preselectors, the members of political parties, are not evil people who have to be thwarted at every opportunity. Rotating candidates is essential for them, the party members, as they are likely to get more candidates elected, but their input into who is elected should be respected. Allowing political parties to present their list of candidates in the order of their choice will respect the parties’ choice. Preferences are more likely to flow along the order of the parties’ choice. If the voters believe the party has made an error, as they may with the recent Tasmanian Senate preselection, then they can still correct the decision. If the voters are supportive or neutral about the preselection then the party’s choice has a higher chance of success.

All that is needed is a simple linear rotation. Three candidates, three variations; six candidates, six variations; and so on.

 

• Electoral Deposits

 

Substantial electoral deposits should be required: $20,000 at a minimum.

A Senator gets a base salary of $195,000 a year plus perks. If an independent wants to be elected they have to be prepared to do what Nick Xenophon did – spend ten or twenty years building a profile and get the support of hundreds of friends. If they do this, they have a chance of being elected and having their deposit returned – if they don’t, they are a joke candidate and deserve to lose their deposit.

We have to stop thinking that election to Parliament is available to anyone who wants it. 110 Senate candidates ran in NSW and 97 in Victoria, but only six get elected each time. (‘What! I didn’t get elected? Oh well, only two thousand bucks down the drain. But I did have fun.’)

Fifty parties and one hundred candidates is not necessarily a sign of a healthy democracy.

 

• Meek Counting

 

The vote should be counted using Meek. Should a vote exhaust, it is as if the voter did not vote at all and the quota is adjusted down accordingly. At the end of the count, every candidate is elected with a quota. No candidate is elected with the largest remainder.

Conclusion

We need to trust the voters.

Incremental change is not an option.

Before this parliamentary term is over there will be some reform. This will happen regardless of what we do. Our job should be to keep pushing for genuine reform. We might even succeed – at worst we can laugh at the politicians and say, ‘We told you so!’