What’s wrong with single-member electorates - Part 2

This is part two of Bogey Musidlak's commentary. In part one he discussed the problems with Australia's current system of representative democracy.

As I argued previously, the winner-take-all nature of single-member electorates encourages a 'whatever it takes' mentality in the relatively small number of marginal electorates that will determine government, following earlier internal scrambles for preselection in safe seats.

The preponderance of safe seats and narrow concentration of competitive political effort elsewhere are inherently unhealthy for democracy. Conducting primary elections, extending optional preferential voting or making voting voluntary would not get to the heart of the problem of geographic determinism.

So what can be done in multi-member electorates and what ought to be done? The key is to encourage voter involvement through effective voting that strongly shapes the composition of the parliament, rather than participation in largely foregone conclusions. This brings the levels of individual and collective accountability where ongoing sympathetic contact with each electorate is essential for political survival.

While Australia has extensive experience of good practice, especially in the Hare-Clark system that has served Tasmania continuously since 1909, this is not always recognised or followed in the discussion or design of electoral systems involving multi-member electorates.

One approach that seeks to remedy gross imbalances between votes and seats in single-member electorates is to use a top-up mechanism to produce fairer outcomes.

This may involve having regional consolation prizes for parties that have most missed out, as set out in the Jenkins Commission recommendations in the United Kingdom. At best, there will be a slight softening of the worst distortions.

Alternatively, as happens in the hybrid systems of Germany and New Zealand, there can be two separate votes for individuals, one in a single-member electorate and the other among regional or national party lists.

Overall results are determined by some form of proportional representation applied to the party-list votes. Single-member victories are usually subtracted from the 'correct' result to determine how many candidates are elected from the various party lists: however, if there are lopsided single-member regional results, as regularly occurs in Germany, additional seats may have to be created after an election to deal with these major departures from proportionality.

Hybrid systems don’t actually fix the problems of unbalanced single-member outcomes that are known in advance. Voters usually have little or no say about the order in which list candidates get elected. In addition, because the methods used to determine the overall allocation of seats are based on comparisons of (sometimes weighted) votes-per-seat ratios for different parties, large numbers of these party-list votes can be completely wasted.

This can occur either because of excessive splintering of votes among numerous small parties or through the deliberate imposition of qualification thresholds (often around 5 per cent) for representation to discourage such behaviour.