Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
 Download Current HansardDownload Current Hansard   

Previous Fragment    Next Fragment
Wednesday, 18 February 1987
Page: 201

Senator VIGOR(6.16) —The report of the Joint Select Committee on Electoral Reform that is before us today deals specifically with the 1983-84 amendments to the Commonwealth Electoral Act. As such it provides only part of the picture for anybody who is a student of the electoral system. The Committee is to be commended for the detail in which it has examined the operation of the amendments made since the election of the Hawke Government. I believe that this type of technique of looking at the effect of legislation after it has been applied should be applied to more legislation coming through this place, which would probably lead to an improvement in standards.

Enormous amounts of very interesting statistical information have come out of the Australian Electoral Commission's work beyond just the scrutiny of the election. I believe that this in itself is a breath of fresh air, because previous conservative governments were not even prepared for basic information on the effects of compulsory making of references and so on to become public. However, in this process one of the Achilles heels of the Australian Electoral Commission has been exposed. This Achilles heel is the bureaucratic tendency to rearrange democracy for the convenience of its managers, instead of rearranging the procedures in the interests of democracy and of the people who are trying to elect a government. We find this in several places in the report.

Although I commend overall the findings of the report, I will concentrate on some of the aspects which I believe are problematic. For instance, I refer to the dangerous proposal that dates of birth and occupations be listed on electoral rolls and that this type of information be available as a salable commodity to whoever wants it, which I believe a bad practice. I strongly support my colleagues Senator Macklin and Senator Haines in their dissent on this particular recommendation. I believe that we should be strengthening the measures to protect privacy and not be supporting new steps along the road to corporate or fascist states. We have just been through the bureaucratically driven exercise of the identity card and we may have to deal with the concept in another form in the near future. My approach will not change when it comes to the protection of civil liberties.

A second area where the Australian Electoral Commission has shown its true colours is in its approach to suggestions that there are technical deficiencies in the procedures for Senate scrutinies. I believe that this attitude is most unfortunate in a body looking at its own procedures. Honourable senators will recall that the first attempt at a counting procedure in the 1983 legislation failed to guarantee even that the right number of senators would be elected. The problem was that those directing the drafting did not stick to basic principles. The confusion they created between `exhausted' and `non-transferable' votes led to the possibility of too many senators being elected under the unfortunate definition of what was called a reducing quota. This is highly technical stuff, but it is what we pay the Commission for. I believe another oversight led to the possibility of too few senators being elected because none of those remaining in the scrutiny could achieve a quota. These problems were patched up on the run after Senator Macklin and the late Senator Missen drew attention to the difficulties. The solution, however, suffers from a defect which can give voters more than one vote's influence in the scrutiny. It is quite a technical area so I seek leave to have incorporated in Hansard a document from the Proportional Representation Society of Australia.

Leave granted.

The document read as follows-



``Poor wording of the counting rules in Senate elections means that some voters can end up with more than one vote,'' said Jack Wright, President of the Proportional Representation Society of Australia. ``In some circumstances, voters can achieve more for a candidate they want to assist by giving a slightly earlier preference to someone from the major parties who is sure to be elected.''

This anomaly has been documented by the Society which is alarmed at the prospect of its use in the South Australian Legislative Council and other elections around the nation where proportional representation applies.

``The problem occurs in the way that successful candidates' surplus votes are apportioned. Each ballot paper which has gone to a successful candidate plays the same part, irrespective of whether it is a first preference or its transfer value is only 0.1. It is possible for the transfer value to increase, even though that should never happen because the transfer value represents the part of the vote that has not yet been used in electing candidates.''

The Proportional Representation Society is urging the South Australian Parliament not to let through a defective Bill which will require extensive amendment in the near future. It will draw to its attention ways in which the present problem can be overcome, and the whole scrutiny procedure streamlined and simplified without making voters' efforts less effective.


At an ordinary election for the South Australian Legislative Council, there are 11 vacancies.

Once the total number of formal votes is known, this is divided by 12 and the next highest whole number is called the quota. Any candidate whose progress total is at least as large as the quota is guaranteed election. The quota is the smallest number of votes that eleven different candidates could each get, but not twelve, in the case of the Legislative Council.

When a candidate is elected, votes in excess of the quota are distributed to continuing candidates in accordance with the wishes of those who supported the elected candidate. The transfer value for any ballot paper represents how much of it hasn't yet been used in electing candidates: it should therefore never go up in value.

To simplify calculations, it is supposed that all votes for candidates in a particular party have first preference for the candidate at the top of the ticket, and follow down that list before indicating anyone else.

Formal Votes-779,998


Party A-350,000

Party B-300,000

Party C-57,998

Candidate D-70,000

Three candidates are elected on first preferences. When it comes to examine the papers that elected Candidate D, the first two candidates in both Party A and Party B have already been elected. The transfer value for these papers is 5,000/70,000 = 1/14.

Suppose that 28,000 papers have the third candidates in Party A and Party B next, and the other 14,000 go to the person heading Party C. 285,000 votes was the progress total of the second candidate in Party A, and 220,000 will go on to the third, taking his progress total to 222,000: altogether, 378,000 papers contribute to his election, so the Senate procedure makes the transfer value 157,000/378,000 = 157/378. This means that the 28,000 papers which were previously worth 2,000 votes are now worth 28,000 X 157/378 or roughly 11,630 votes, even though they have just helped someone else get elected, and should therefore have lost some of their value.

In other words, those who voted first for Candidate D, then for the third candidate in Party A and then for Party C assist Party C nearly six times as much as those who put Candidate D first and then went straight on to Party C. To put it bluntly, you can do more for a candidate by not voting for him until you've gone across to someone else you anticipate will be elected! This opens up possibilities for tactical voting in certain circumstances. No fair method of proportional representation gives advantage to a voter who falsifies his or her true wishes. It is relatively straightforward to remedy the situation by sticking more closely to first principles.

The problem has arisen because the papers of value 1/14 and the Party A votes which go to the third candidate at value 22/35 are treated as equals in the split-up of the surplus. If you take account of vote value, you don't run into the same problem. By amending other aspects of the scrutiny procedure you can also cut down the time taken to do the counting while dealing with surpluses in a more sophisticated way. In this particular case, the ticket votes would take up 220,000/222,000 of the surplus and the smaller parcel would determine what happens to the other 2,000/222,000: in other words the 28,000 papers would have total value 157,000 X 2/222 = 1,415 votes to the next highest integer, and their transfer value therefore be under 1/14. There is no excuse for persisting with a method that can distort the wishes of the voters, especially when it is easy to fix it.

Senator VIGOR —I thank the Senate. The document was drawn to my attention in 1985 when the South Australian Parliament was considering changes to its electoral legislation. I would like to summarise the points from the example. If a pile of ballot papers helps to elect someone we would expect its value to decrease as a result of the election of that person. This need not happen under the Senate procedures. That leaves the way open for tactical voting and other subterfuges which I believe should have no part in any fair electoral system. Voters should not have any second thoughts that they can achieve a better result for their favoured candidate by falsifying the expression of their wishes on the ballot paper. That is, people should be able to vote straight-forwardly and have their vote achieve the end which it appears to achieve. The problem can be rectified in a way which does not unduly prolong the scrutiny. The Committee has already recommended bulk exclusions and the deferral of transfers of surpluses where relevant. Transfer values still need to be tackled.

The attitude of the Australian Electoral Commission on this matter was most unsatisfactory. I shall deal with the technical points of this case at a later stage, perhaps in the adjournment debate. The Joint Select Committee has gone through the procedures fairly thoroughly for letting people have a fair vote. It has suggested quite a number of improvements. For instance, it has called for a simplification of the procedure for identifying voters on polling day before giving them ballot papers. It is well known that the formal questions that were supposed to be put were often not put by electoral officers. People whose English is not good can be intimidated if the procedures are rigorously followed.

The Joint Committee, in this report, was not able to examine ways in which the voters wishes could be implemented more effectively. In the last House of Representative selection nearly 4,000,000 voters were disenfranchised because of the single-member electorate system. Their votes were of no value at all. They may as well not have turned up to vote in the election and many may have chosen to do that to get away from such a travesty.

We need to adopt a proportional representational system, preferably a system such as the Tasmanian Hare-Clark model which allows voters to sort out the people whom they want as their representatives. The Australian Democrats' policy with which I fully concur is for optional preferential voting in multi-member electorates. I believe that this is the only fair way of voting. The Tasmanian Hare-Clark system comes closest to this ideal in Australia. The Tasmanian House of Assembly is the only lower House in which one vote one value is a matter of fact and not just an assertion. Not only does it treat parties and individual candidates fairly but it also leaves the real power in the hands of the voters-not just the string-pullers and apparatchiks in the Party machine. The rotation of candidates on the ballot papers and the fact that no by- elections are necessary also make this a less expensive option in the long run for people. We should be re-examining very closely the prospect of being able to fairly elect our House of Representatives in a proper way by using proportional representation and by adopting an optional preferential system in multi-member electorates. I commend such a process to the Senate. Although we do not have that, I believe that the proposals put forward in this report, in general, are excellent.