Charles Richardson in Crikey is interesting but deserving of additional comment. I’ll begin with points on which agree. (I had written that “the geography of Queensland differs from that of NSW. The latter state has more genuinely safe Labor seats. Consequently in its worst result for more than 100 years, Labor was still able to retain 20 seats”.)

Following the NSW election last year, I expected many articles from advocates of proportional representation. To the best of my memory there were none, and I wondered why. Labor is not significantly under-represented in the state parliament of New South Wales. At the time Labor was over-represented in every Australian lower house, in some cases significantly so, even where Labor had lost. For example, Labor lost in Victoria in November 2010 but won 48.9% of the Legislative Assembly seats with a primary vote of 36.3%, an over-representation of 12.6%. In September 2008, Labor had lost in Western Australia but had won 47.5% of the Legislative Assembly seats for a primary vote of 35.8%, an over-representation of 11.7%.

Labor emerged from the NSW election in March last year over-represented in the Legislative Council, due to the system of rotation of members, it having benefited from a good result in March 2007.

In the Legislative Assembly, Labor won 25.6% of the votes and 21.5% of the seats, an under-representation of a mere 4.1%. As Richardson, Antony Green and I have all noted, the geography of the Newcastle-Sydney-Wollongong conurbation meant that there were enough genuinely safe Labor seats to ensure that Labor would not be wiped out.

I publish pendulums before elections in The Weekend Australian. Regarding NSW, my most recent pre-election pendulum showed 14 Labor seats in which the Liberal Party needed a swing of more than 20% to win. Regarding Queensland, my most recent pendulum showed only three seats (Bundamba, Inala and Woodridge) in which the LNP needed a swing of more than 20% to win. Labor duly retained the 14 seats in NSW and the three in Queensland.

It added six weaker seats in NSW and four in Queensland. So, with roughly the same share of the vote in the two states, Labor retained 20 seats in NSW and seven in Queensland, in a Legislative Assembly of 93 and 89 members, respectively.

Several commentators  point to New Zealand’s mixed member proportional (MMP) system as the model for reform. I find that strange. Two Australian lower houses (Tasmania and the Australian Capital Territory) have the Hare-Clark system which is, in my opinion, much better than MMP. Under Hare-Clark every member is elected. Under MMP, by contrast, there are 70 elected members but these are supplemented by 51 party machine appointments to bring the total House of Representatives up to 121 members.

It is often said that MMP is proportional representation while Hare-Clark is described as “semi-proportional”. Quite so. In my opinion, that is one of the defects of MMP. It never allows the biggest single party to get a majority of the seats. For example, at the most recent New Zealand election in November last year, the National Party was by far the biggest party in votes. It secured 47.3% of the party vote and won 48.8% of the seats (59 out of 121). Translating that to Queensland, the LNP’s 49.7% of the vote would have won it 45 or 46 (certainly no more) of the 89 seats. Under Hare-Clark the LNP would have won, perhaps, 54 or 55 seats.

It really all depends on what one thinks best serves the interests of good government. I think that the interests of good government are best served by giving the single biggest party a bonus when votes are translated into seats. I also think it should not be made too difficult for that party to win a majority of the seats. That kind of bonus is given by all the electoral systems of the Anglosphere — except MMP in New Zealand.

*Malcolm Mackerras is visiting fellow in the Public Policy Institute, Australian Catholic University, Canberra Campus