Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Friday, 16 April 1948


Dr EVATT (Barton) (Attorney-General and Minister for External Affairs) .- by leave- 1move- (That the :bill .be now read a second .time. 3/he .purpose of the :.bill ;is to make provision for .the application of proportional representation .to the -election .of -senators. Prior »to '1918, -senators were .elected by what ds .usually 'termed the " -first :past -the post " ,method. Each voter was -required to place a cross in the -square, opposite the name-of each of the candidates for whom he desired to , vote. the .number of .crosses permitted being confined to the., exact num- ber.of senators to be elected. Thus where three -senators were .required the three candidates »with 'the greatest number of crosses were .chosen, and as was -usually the-case -,the .three- elected were (candidates of the same party.

In 1918, the "preferential block majority " system was introduced in respect df Senate elections. This system continued the principle of the old "first past the-post" system in that under it all seats. in a State generally go to the. party, or combination of .parties, 'favoured at the time. by a bare,- or simple, majority of the electors. It might 'be described as the "..all or .none'" system, either all or none of a party's candidates being elected. The system used since 1.918 may'be considered an improvement on the old system only to the degree (hat it ensures majority representation as against possible minority representation. The great defect, from the representation aspect, of both the old "'first past the post ".and the more recently used " block majority " system is that at an election, generally all seats in a 'State are won by candidates o'f 'the one party, leaving ,a minority of between 40 to '50 per cent, of the electors without :any representation at all in the ."Senate. "For many .years there has been .a demand 'that 'the Parliament should provide ,:a system of 'electing senators which would give more equitable results, and enable: the electorate to be more truly-represented in the Senate. The G.o- vernment has given careful consideration bo the matter [and has closely examined alternative methods. Jit h$s '.decided that, in relation £o the .election :of senators, ,where each .State 'votes ;as .one electorate, .the .fairest system -and .the one most .likely to enhance .the ^status .qf the Senate lis that of proportional representation.

The bill -sets out in. detail -the method, of counting ipr.oposed.to.be adopted tin respect of ,future elections of .senators. The method is .generally in r.accord with the practice laid down -by the 'Proportional Representation .'Society. AH I need say about the principle is that the marne of the system indicates the result -intended to be achieved,, that .the electors -will be represented, in the number of tho.se elected, approximately proportionately to their expression of opinion. -For example, if -five are to 'be -elected and one .party's candidates in the Proportional Representation Bill 1912 of' Great Britain and the system employed in respect of municipal elections in the United Kingdom and -South Africa. It is virtually identical with the method used in the election of the ^Parliament of :Eire and is similar to the -system which was employed in i respect o'f 'parliamentary elections for the Legislative Assembly in 'New South Wales from 192.0 to 11925. In principle, the method proposed lis 'the same .as that used in Tasmania although 'for reasons of workability .and simplicity it differs slightly in its practical application.

It .is not proposed to alter the - existing style of .the Senate ballot-paper or the provision 'that 'Candidates i may be grouped thereon with their mantes in such order within the' group. as they desire. .Nor is it intended to .vary -the .(requirement that voters must indicate -the order of their p reference for all the-candidates. (Whilst this latter requirement might have the effect of continuing to produce a -fairly high informal vote, .it definitely precludes -.the possibly greater evil of exhausted votes- that -is, votes which become exhausted in the process of transfer. If a voter were to indicate his preference for only three of, say, seven candidates, his vote would be effective up to the number of preferences shown on his ballot-paper and after that it would be effective no longer. At that stage, the vote would be said to be exhausted. In Tasmania, the elector need not vote for all candidates and therefore on occasions a fairly high percentage of votes becomes exhausted in the process of transferring the votes of a lower candidate or the surplus of a higher candidate to the next candidate in order of preference.


Mr Thompson - And the last candidate elected frequently does not get a quota.


Dr EVATT - That is so. One result of a system which does not require the electors to vote for all candidates whose names appear on the ballot-paper is that a candidate may be declared elected although the total number of votes credited to him falls short of the required quota. At the parliamentary elections in New South Wales in 1922 and 1925, the exhausted votes, which far outnumbered the informal votes, were the cause of niuch dissatisfaction and disputation.

Clause 8 prescribes the manner in which it is proposed that the several vacancies shall be filled. As hitherto, the count will be carried out under the direction of the Commonwealth Electoral Officer for the State concerned in the offices of the respective Divisional Returning Officers. For instance, in New South Wales, where there will be 47 divisions, the count will be carried out at 47 points. This will ensure completion of the count with the greatest safety, the maximum of speed and the minimum of cost. When the Commonwealth Electoral Officer has received the final results of the count of first preference votes from all returning officers and has totalled them, he will determine a quota by dividing the total number of first preference votes by one more than the number of candidates required to be elected and by increasing the quotient so obtained by one. That appears to be complicated, but if in New South Wales, for example, there were 1,600,000 valid votes and seven senators had to be elected, the total number would be divided, not by seven, but by eight, giving a quotient of 200,000. This would give a quota of 200,000 plus one.

This formula for determining a quota, which, in effect, produces the lowest number which, when multiplied by the number of candidates to be elected, leaves a remainder of votes less than that lowest number, is the one recommended by the Proportional Representation Society and is used in Tasmania, in Eire, and generally in all places in British communities where proportional' representation has been applied. Any candidate who, either on the count of the first preference votes or at any subsequent stage, obtains a number of votes equal to or greater than the quota shall be elected, and until all vacancies have been filled, the surplus vote9 - that is, any number in excess of the quota - of each elected candidate will be transferred in the manner set out in the bill to the continuing candidates in strict proportion to the voters' next preferences.

The method of disposing of an elected candida te's surplus votes prescribed in the bill is the one recommended by the Proportional Representation Society and is precisely the same as that used in Eire and in municipal elections in Great Britain and South Africa, and it is similar to the method which was employed in connexion with the parliamentary elections in New South Wales in the early nineteen- twenties. It differs from the Tasmanian practice in that whereas in Tasmania all the votes of the elected candidate are transferred at a fractional value so that every single paper is looked at, under the proposed method only such number of votes as equals the surplus, taken in strict proportion to the preferences on the whole of the votes of the elected candidate, are transferred. No doubt that provision will require some consideration in committee.

It is thought that the Tasmanian system, while suitable where the number of votes is comparatively small and all such votes are concentrated at one centre, is> not readily capable of being efficiently worked by too remote a control. I'n order to employ that system at a Senate election, it would first be necessary to assemble the whole of the ballot-papers for the State at one centre. Not only would that delay completion, involve risks of loss in transit, and increase costs heavily, but also in the larger States the Commonwealth Electoral Officer would be faced with the almost insuperable task of securing for a period of several weeks the extensive accommodation, equipment and staff needed. Whilst it might be claimed that the Tasmanian system is mathematically more exact, tests that have been made reveal that a similar result is obtained by the employment of the method proposed. As an illustration, I shall state a hypothetical case in New South Wales and for this purpose I shall assume for the moment that a Labour candidate has received 900,000 votes of the total number of 1,600,000. The quota being 200,000, he has a surplus of 700,000 votes for distribution. This would usually result in three full quotas being obtained for other candidates on the list. The proposal embodied in the bill is that the officers will not look at all of the votes.. They will take the surplus available and will assume that the No. 2 vote, which would be the effective vote in the first instance, would reproduce a proportion in respect of the surplus of 700.000 that would be true of the total of 900,000. Tests have shown that the result would be the same as if all of the 900,000 votes were scrutinized.

Questioned as to the likelihood of difference, eminent mathematicians advised the Proportional Representation Society of England that " Whenever a considerable number of votes is in play the element of chance involved is so small as to be negligible ". Dealing with the point in its comprehensive report on electoral systems, the United Kingdom royal commission of 1908 said, " The chance of the result being affected is too small to be seriously considered. We agree with the Proportional Representation Society that the additional labour involved (in the Tasmanian system) is greater than it is worth ". In a report furnished in 1913, a committee consisting of Mr. H. E. Packer, Chief Electoral Officer for Tasmania, Mr. E. L. Piesse. LL.B.. and Mr. J. F. Daly stated inter alia, " We are justified in saying that in each district at each of the three elections - fifteen contests in all - the result would have been the same with the English rules as with the Tasmanian rules . . . We therefore recommend that if the form of the rules should again be considered by Parliament the English rules be adopted ". Thus it will be observed that a responsible local committee reported in favour of the adoption in Tasmania of the time-saving method of dealing with surplus votes proposed in this bill.

The bill further provides that if after the count of the first preference votes or after the transfer of the surplus votes of an elected candidate at any stage, no candidate or less than the number required to be elected has or have obtained the quota, then the candidate with the fewest votes shall be excluded and the whole of his ballot-papers transferred to the continuing candidates; and if thereupon no candidate has yet reached the quota, the process of excluding the candidate with the fewest votes and the transferring of his ballot-papers will continue until a continuing candidate has received a number of votes equal' to the quota or in respect of the last vacancy a majority of the votes.

Where candidates are elected at the same time, the order of their election shall be determined by the extent of their surplus votes. The candidate with the largest surplus shall be the first elected and so on. The same principle will apply in relation to the transfer of surplus votes. The largest surplus will be transferred first and so on. If on any count two or more candidates have an equal number of votes and one of them has to be excluded the Commonwealth Electoral Officer shall decide the candidate to be excluded, or if two or more candidates are elected with an equal number of votes the Commonwealth Electoral Officer will decide the order of their election and the transfer of their votes, or if in the final count for the filling of the last vacancy two candidates have an equal number of votes, the Commonwealth Electoral Officer will decide which shall be elected but that except as so provided the Commonwealth Electoral Officer shall not vote at the election. That is really the last resort. The chance's against its happening in a Senate election are enormous.


Mr Barnard - About a minion to one.


Dr EVATT - That is a gross understatement. Probably millions of millions would be nearer the mark. The result of the system is clear. There will be two major groups of political parties, and seven Senators will be elected, in the absence of any casual vacancies, in each State. The party which secures 53 per cent, or 55 per cent, of the aggregate votes will, provided the party ticket is followed, have four of the seven candidates to be returned. The three remaining seats will be filled by the candidates from the other party which secured 47 per cent of the votes. That is certain.


Mr Archie Cameron - I desire to ask the Attorney-General one important question. Will he inform me whether any othercountry has an electoral law under which votes are taken from a pool to fill a vacancy at random?


Dr EVATT - I have already mentioned the countries in which that system operates.


Mr Archie Cameron - What are they ?


Dr EVATT - It operates in municipal elections in the United Kingdom and South Africa, and in the national elections in Eire It also operated in elections held in New South Wales between 1920 and 1925. However, this matter is somewhat complex and difficult, and I do not desire to elaborate it now. In Tasmania, every vote is examined, and the proportion is attached to the No. 2 preference in each case. I shall explain the reason why the alternative method is proposed for Senate elections. Let us suppose that the leading candidate in New South Wales has 900,000 No. i votes. Then 700,000 votes would beavailable for distribution on the assumption which I mentioned.


Mr Archie Cameron - I understood that part.


Dr EVATT -.- That is the only part in which the selection will be taken. Instead of selecting all the surplus ballotpapers and applying to each a fractional transfer value, the electoral officers will select at random that numberof surplus ballot-papers which isascertained by multiplying the total number of surplus ballot-papers by the fraction. In 98 per cent, or 99 per cent, of instances the electors will vote the party ticket. The honorable member is aware of that.


Mr Archie Cameron - I thought that the Attorney-General was trying to popularize the lottery.


Dr EVATT - It is not a lottery in that sense at all. The lottery factor has been excluded.


Mr Archie Cameron - When a lottery is drawn, marbles are taken out of the barrel at random.


Dr EVATT - It is hot a matter of chance in that respect. The ballot-paper will already have been marked, so the element of chance will have been excluded by the elector. That is the only sensein which there is any element of chance.


Mr McEwen - The position is quite clear.


Dr EVATT -I am sure that it is clear to the honorable member for Indi (Mr. McEwen). Like most other people, he will probably be more interested in the result than in the method by which the result is reached.

Debate (on motion by Mr. MEnzies) adjourned.







Suggest corrections