A final criticism made against proportional systems of voting is that they do not secure the exact representation of all the electors in a country. Thus the Royal Commission on Electoral Systems, whilst admitting that the new method would generally produce more accurate results, mathematically at least, than the existing method, qualified their statement by saying that their success "in producing in Parliament the 'scale map of the country,' which they held up as the ideal, can be only partial"; and in another paragraph the Report contains this remarkable statement: "On the assumption, however, that proportional representation is desirable, can any system yet invented be guaranteed or reasonably expected to ensure it? In our opinion, only in a limited and generally unascertainable degree." No responsible advocate of proportional representation has contended that proportional systems, save when the country is treated as one constituency, will result in a mathematically accurate representation of opinion. But the close approximation to accuracy obtained in the practical application of proportional systems is so pronounced that the statement of the Commission is wholly misleading. The following figures of the Belgian election of 1910 will show to what extent accuracy is obtained by a proportional system, even when, as in this case, the mechanism slightly favours the larger party:
BELGIAN ELECTION, 1910
Seats Seats in
Parties Votes. Actually Proportion
Obtained. to Votes.
Catholics . . . . . 676,939 49 47.0
Liberals and Socialists . 561,052 36 37.5
Christian Democrats . . 16,170 —— 1.0
Independents . . . . 20,428 —— 1.5
In Finland, where another system of proportional representation is in operation, the result of the election of 1909 was as follows:—
FINLAND ELECTION, 1909
Seats Seats in
Parties Votes. Actually Proportion
Obtained. to Votes.
Social Democrat . . . . 337,685 84 80
Old Finn . . . . . . 199,920 48 47
Young Finn . . . . . 122,770 28 29
Swedish . . . . . . 104,191 25 25
Agrarian . . . . . . 56,943 14 13
Christian Labourers . . 23,259 1 6
The single transferable vote has yielded results which are remarkably accurate. It has been used in Tasmania, with adult suffrage, in the Transvaal, with the municipal franchise, and in the election of the Senate for United South Africa, by members of Parliament. Each of the five constituencies in Tasmania returned six members, and the total result was as follows:—
TASMANIAN ELECTION, 1909
Seats Seats in
Parties Votes. Actually Proportion
Obtained. to Votes.
Labour . . . . . . 19,067 12 11.7
Non-Labour . . . . . 29,893 18 18.3
These figures speak for themselves. In the municipal elections in the Transvaal each of the parties obtained its fair share of representation. In Johannesburg the elections were fought by a commercial ticket of ten candidates, a Labour ticket of three candidates, and ten Independent candidates; the number of valid votes was 11,788, and the quota—that is, the proportion of votes which would ensure the election of a representative—amounted to 1072. The ticket of "ten good men" polled in all some 6185 votes, or 247 votes short of six quotas, and the ticket succeeded in returning six members. This result was strictly fair, for the deficiency in votes was made up by those supporters of independent candidates who, having failed to return their first choice, had indicated members of this ticket as their next choice. The three Labour candidates polled in all 2126 votes, or 18 votes short of two full quotas, and the Labour Party was successful in securing two representatives. The remaining two seats fell to two Independent candidates, each of whom had a considerable personal following. In the third test, the election of South African Senators, each of the parties obtained representation in proportion to their force in the Parliaments of the respective colonies. The details of the voting have not been published,[10] but the returning officers have all borne testimony to the satisfactory working of the system and absolute fairness of the results.