.')H QUOTA IN PROPORTIONAL REPRESENTATION, 



9. We have next to compare the two quotas in a con- 

 test between parties. Messrs. Douglas, Piesse, and Birch- 

 all, in their report on the Tasmanian General Election of 

 30th April, 1909, (^*), discuss this aspect as follows: — 



But these arguments do not decide the supeiioiitj' of one 

 quota over the other if an election is considered, not as |a 

 contest between candidates, but as a contest between parties. 

 For here we have to consider the possibility of one or more 

 candidates of a party securing election on less than a quota, 

 and so obtaining for their party an amount of representation 

 in excess of its proportional share. With the Hare quota it 

 is very easy for a party to secure excessive representation by 

 returning several candidates with less than the quota. With 

 the Droop quota this is impossible in a two-party contest 

 (except when papers become exhausted through the neglect 

 of voters to give a preference to each candidate of their 

 party), and in a contest between more than two parties dis- 

 proportional representation would probably occur much less 

 frequently with the Droop quota than with the Hare quota. 



Take the case of an election of six members by 210 voters, 

 63 of whom belong to party A, and 147 to party B, and 

 assume the Hare quota is used. Party A, having roughly 

 one-third of the voters, is entitled to two members, and party 

 B to four. When all candidates but seven have been excluded, 

 the state of the poll might be that the five remaining candi- 

 dates of party B had respectively 30, 30, 29, 29, 29 votes 

 each (total 147) ; and the two remaining candidates of party 

 A 35 and 28 each (total 63). The candidate lowest, on the 

 poll has now to be excluded ; that is, the A candidate with 

 28 votes is excluded, and there are left six candidates — five 

 of party B, and one of A, who are declared elected. That is, 

 party A^ instead of getting two members, has got only one ; 

 and party B, instead of four members, has got five. 



Now this has happened solely because the use of the Hare 

 quota (35) has wasted the four votes which the A candidate, 

 with 35 votes, had in excess of the Droop quota (31). If the 

 Droop quota had been used, this surplus of four would have 

 been distributed before the exclusion of the lowest candidate. 

 It would naturally have gone to the other candidate of the 

 party, whose votes would thus have been raised from 28 to 

 32 ; and the candidate excluded as lowest on the poll would 

 then have been one of the B candidates. Thus, the result 

 would have been the correct result — party A, two members ; 

 party B, four members. 



It is interesting to note in passing that if, in the election 

 for Franklin, the Hare quota had been used, and if there 

 had been no cross-voting between the candidates of the Labour 

 Party and other candidates by voters who gave their first 

 preferences to Non-labour candidates, and no exhaustion of 

 the papers of such voters, the Labour Party would have 

 secured only one member in place of the two to whom it was 

 entitled in proportion to the number of its supporters. 



There are also some cases in which the Droop quota 

 produces representation not proportionate to the strengths 



('*) See (TO), p. 5. 



