Australia’s voting system gets it right – or does it?

One of the key themes of this blog throughout its six and a half years has been that electoral systems matter: that election results and political developments in general are partly determined by the mechanics of how votes are aggregated.

As I put it in the very first post,

[P]olitics matters: not (just) the sweep of great historical forces, but the detail of how political systems translate people’s preferences into actual decision-making. Without some understanding of that, important aspects of how the world works will remain mysterious.

Last Saturday’s election in Australia provides a good lesson in what I meant.

Many pundits over the last couple of years have remarked on how Australia seems to have escaped the rise of a far-right populist movement of the sort that produced the Brexit vote, the election of Donald Trump, and the success of Marine Le Pen and Matteo Salvini. Pauline Hanson and Clive Palmer seem like poor substitutes.

And there are useful things you can say about cultural factors that might explain that. But a key thing, usually neglected, is the electoral system.

The United States, uniquely for the democratic world, has a legally entrenched two-party system. Since it’s not practically possible for a third force to break into the system, a major change in direction involves insurgents capturing control of one of the existing parties – as the conservatives once did to the Republican Party, and the Trumpists have now done in their turn.

At the other end of the spectrum, most other democracies, particularly in Europe, operate one form or another of proportional representation. There it is relatively easy for minor parties to establish themselves, and shifts in public sentiment can be reflected in a shifting balance among a number of parties in parliament.

The rise of the far right in Europe has accordingly taken the form of actual far-right parties returning substantial contingents of MPs, and being potentially able to bargain with other parties for office or for policy concessions.

Australia is somewhere in between. We have no proportional representation in our lower house, so it is almost impossible for a new party to elect more than the occasional lone member.

But we do have minor parties. The existence of some proportionality in the Senate, the ability to bargain with their preferences in the lower house and our liberal rules on ballot access all give them a place in the system and some hope of reaching a position of influence.

So with that in mind, let’s look at Saturday’s results (counting is still continuing, but the figures for the lower house at least won’t change very much). The Liberal-National coalition* government was returned with 41.4% of the vote and probably 78 of the 151 seats. Its Labor opposition won 33.9% and a likely 67 seats.

Three minor parties won a seat each – the Greens, Katter’s Australian Party and Centre Alliance – and three independents won seats.

On a proportional basis, however, the result would have looked very different. A Sainte-Laguë calculation on the current figures (leaving out the independents) gives me Coalition 67, Labor 54, Greens 16, United Australia Party and One Nation five each, and one each for the Animal Justice Party, the Christian Democrats, Fraser Anning’s Conservative National Party and Katter’s Australian Party.

Interestingly, the total left-right balance comes out about the same. The proportional result is 80 right to 71 left; the actual total was 79 right to 69 left, plus three (Centre Alliance and two of the independents) difficult to classify.

In a very close election, our system risks getting the wrong result, giving the minority party a majority of seats. If it’s not close, the system produces lopsided majorities, giving the minority party far fewer seats than match its share of the vote. This is just the sort of election, close but not too close, that it’s most likely to get right.

But the system of single-member districts gives a massive advantage to the two major parties at the expense of the rest, and particularly of the Greens and the assorted far-right parties.

Many people clearly see that as a good thing – favoring the mainstream rather than the extremes (although outside of New South Wales it’s very hard to see the Greens as extreme). But the problem is that it also changes political behavior.

If a free-standing far-right party was more likely to be viable, it would attract more competent politicians; it would be less the sort of amateur hour that One Nation and the UAP now are. In particular, it would attract some people who now end up in the Liberal Party even though their politics are far from mainstream.

That might seem like a problem. Perhaps it’s better to keep the extremists within the tent, at least as long as they’re a small minority. But there comes a point where they reach a critical mass and start to exert serious influence, as the far right has been doing over the Liberal Party in recent years.

Each system has its pros and cons. If we had a European system, we would probably have a substantial far-right party with around ten per cent of the seats in the House of Representatives. Other parties would have to deal with that situation, and the voters would judge them on how they did.

But we would also probably have major parties that kept more to the centre, having divested themselves of their extreme elements. And without a sane, stable centre-right party, the whole system is in peril, as the United States has clearly demonstrated.

Which is better, to have the fascists concealed within a supposedly mainstream party, or out in the open where they can be seen?

* The relationship between the coalition partners is complex. In Queensland and the Northern Territory they constitute a single party; elsewhere the Nationals run as a separate party, except that they don’t exist in the ACT and are negligible in South Australia and Tasmania. To avoid difficult decisions about disaggregating them I am treating the whole Coalition as if it were a single party.