Electoral System Reform

Preamble
In a well functioning democracy, the outcome of an election should reflect the will of the people who voted. Interest groups, both big and small, should be able to look at the results and agree that the process was carried out with integrity and everyone is fairly represented. Further, the resulting government should have support from the majority of the population. Judged by that standard, Australian electoral systems are thoroughly broken.

Current systems fail voters
The number of seats a political party obtains in the Lower Houses of State and Federal Parliaments usually has little relation to the number of voters who support that party. Governments are regularly formed with nowhere near a majority of first-choice support. Sometimes 20% or more of the voting population goes without having their votes represented at all. In one particularly spectacular case in Queensland in 2012, an election produced a result where only a single party had any significant presence in Parliament. This disconnect becomes even worse when it is remembered that MPs almost always vote according to party lines.

The root cause of this problem is the use of many single-member electoral divisions. Voters who form a minority in many divisions fail to elect their preferred party's candidates and can easily go without any representation, even though they may be a large portion of the population overall. Conversely, voters who form small majorities in many divisions can easily become overrepresented. When this geographic bias is intentionally exploited it is called gerrymandering. But even when intentional gerrymandering isn't an issue, the problem remains unless the voting system is designed to consider the big picture as well as individual divisions.

Follow-on effects from this are numerous. Landslides happen whenever a party obtains many small majorities, producing overrepresentation. Safe seats become commonplace and become neglected in favour of swing seats. Minor parties face extreme difficulty obtaining seats, as they nearly always have thinly spread support. The system encourages excessive focus on local issues even in State and Federal parliaments, because concentrated regionalism is the only path to success. Focus for minor parties shifts towards where their chances are greater, which is usually the Upper Houses, and helps to produce overcrowded "tablecloth" ballot papers. The political landscape stagnates with a mere few adversarial parties playing tug-of-war, switching back and forth every few terms.

Details differ among the different levels of government in Australia, but this problem is nearly always present. Even in the ACT's Legislative Assembly and Tasmania's House of Assembly, where multi-seat divisions are used, some disconnect between voter preferences and party results can be observed. The only places where the problem and flow on effects are unobserved are the Tasmanian Legislative Council and a number of rural Local Councils, where independent candidates dominate.

Partly as a consequence of of the 'tablecloth' ballot papers, substantial nomination deposits are required to stand in many elections around the country. For all but a handful of candidates, these deposits are effectively fees, taking up limited financial resources that would be better spent on campaigning. Further, the continued size of the Federal Senate and NSW Legislative Council ballots indicate that such fees are insufficient as a barrier to entry anyway. Finally, it is common for parties to stand candidates in areas those candidates have no connection to - because for parties, the nomination deposit is the only external restriction.

Other more minor issues also plague Australia's electoral systems. Federal division, State district, and Local ward boundaries differ wildly. Optional vs full preferencing differs depending on State, as does the methods of filling unexpected mid-term vacancies. Term lengths differ, both in length and fixed vs varying, between different levels of government. Single Transferable Vote is counted by software processing of scanned ballots, yet the software used is not made available for scrutiny.

Solutions and non-solutions
Fixing the big problem and most of the follow-on effects requires a voting system that is designed to ensure that overall party support matches overall seats won. Voting systems that are designed to ensure this are called "proportional representation" systems. There are three well-tested voting systems that can provide this, or close enough to it.

Open Party List is a system where voters cast a vote for a party and a candidate within that party. Seats are then allocated to each party according to the proportion of votes that party received, and those seats are filled by the most popular candidates within each party. Implementations often involve several separate multi-seat divisions, but overall proportionality can still be maintained by reserving some seats to allocate according to the overall vote. Of the three systems mentioned here, this has the highest number of other countries already using it. However it is the most different to all systems currently in use in Australia, and it has been previously tried and rejected by voters in the ACT. It also does not handle independent candidates well, nor does it do a good job providing localised representatives.

Single Transferable Vote is currently in use for Local elections in some States, most Upper Houses, and the ACT Legislative Assembly and Tasmanian House of Assembly. Because political party is not taken into account, this system does not guarantee proportionality. It often gets close in practice but only when a single multi-seat division is used. This places practical limits on the overall number of seats and means no localised representation at all. The system does, however, handle independent candidates extremely well, and minimises potentially wasted votes when the available number of seats is small. These characteristics make it ideal for Local Council elections and the Upper Houses of Parliament, but not the larger Lower Houses that are generally intended to represent the population of the State/Territory/Country as a whole.

Mixed Member Proportional is a system where approximately half of the available seats are filled with candidates elected from single seat divisions. Then the remaining seats are allocated to each party to ensure that the overall proportion of votes each party received corresponds to the number of seats each party obtains. If preferences are used for the single seat divisions, the required ballots closely resemble current Instant Runoff ballots. Independent candidates and localised representation are both handled about as well as with current single seat divisions. This system works best when the overall number of seats to be allocated is not too small, and most candidates are affiliated with a party. These characteristics make it ideal for the Lower Houses of Parliament. This system was chosen by New Zealand in the early 1990's when conducting reform to their electoral system.

Electronic voting, as in where votes are cast and counted entirely electronically, is not a solution. The most it could offer is an easier time handling excessively large ballot papers. But, as pointed out, that is only a minor symptom of the real problem, and in return for convenience electronic voting introduces a laundry list of issues that compromise trust in any election. Electronically assisted voting, as in where votes are selected on a computer, printed out, then handled the same as other paper ballots, does not address the underlying problems. It too will only offer an easier time handling large ballot papers, and currently would add significant unnecessary expense to elections.

While it is distasteful to impose barriers to democratic participation, it is acceptable to require prospective candidates to demonstrate some level of community support. This support should not be measured in dollars (as it currently is), but rather in people: nomination signatures from electors of that district, region or state.

The importance of a Royal Commission
Any political party pushing directly to change the electoral system is subject to a conflict of interest. Political rivals will rightly have suspicions, and the entire issue will likely fail to gain traction. The only way to build a consensus and ensure a fair outcome is to empower an independent body to investigate the issue and come up with solutions. For that reason, the goal of this policy is to have a Royal Commission appointed which will determine a fair course of action.

Policy Text
Pirate Party Australia advocates for a Royal Commission on the electoral systems used in all levels of government in Australia, to decide on how to best make our elections fairer, more democratic, and more accessible to the voting population.

To that Royal Commission, Pirate Party Australia will make the following recommendations:

Voting Systems

 * Mixed Member Proportional is the best option for use in all State and Federal Lower Houses
 * This would ensure both accurate overall representation of the votes, and that all areas have an MP tasked with representing their concerns
 * Optional preferential is the best option for electing the division seats, to allow voters full control and for consistency with Single Transferable Vote
 * No thresholds should be used, as they distort the results away from how people voted
 * The Largest Remainder Method with Droop quota is the best option for allocating top-up seats to parties, for consistency with Single Transferable Vote
 * Best Near Winner should be used for filling top-up seats to ensure all MPs are directly elected by the people
 * Single Transferable Vote is the best option for use in all Local Councils and State and Federal Upper Houses
 * This works well with both the large number of Independent Local Councillors and the small number of seats available in most State and Federal Upper House elections
 * Above The Line voting should be abolished, as reform to the Lower Houses should ease the candidate crowding that originally necessitated Above The Line
 * The number of Wards for Local Councils should be minimised, ideally using only one whenever practical to avoid geographic bias
 * Robson Rotation should be adopted for all elections in order to eliminate the effects of donkey voting

Nominations

 * Abolish nomination deposits entirely, or else reduce them to a nominal level
 * Instead, require nomination signatures, from residents of the relevant electorate
 * The number of signatures should be set at a level reflecting the minimum number of campaigners required to interact with a majority of the electorate
 * For the House of Representatives, 10 nominators per candidate, as the top 10 polling places in each Division serve about half the population there
 * For the Senate, 10 nominators for every House of Representatives Division in the State or Territory, per candidate
 * To ease the volume of election-time paperwork:
 * Prospective candidates would be permitted to lodge nominations throughout the year prior to the election
 * Parties undergoing membership audits in the year prior to the election, who demonstrate membership numbers in any electorate equal to or exceeding the nomination requirements, would be pre-approved to stand candidate(s) in those electorates
 * Members of Parliament who are re-contesting their seat would be exempt from these requirements

Consistency

 * State and Federal geographic division boundaries should be aligned, but Local Council areas should be left to reflect natural community boundaries
 * This would be made practical through the flexible number of single division vs topup seats involved in Mixed-Member Proportional
 * All government elections should be standardised to fixed four year term lengths

Integrity

 * Paper ballots should remain the only method used for casting votes at all elections, with voters encouraged to cast votes in person at dedicated polling places
 * Any and all software used in the scanning and counting of paper ballots should have its source code made available to the public for scrutiny

= References =