Multi-member Parliamentary seats

Critics of Northern Territory self-government have sometimes argued that the NT has too small a population to generate the level of talent necessary to form a government and ministry. It’s certainly true that at just on 250,000 the Territory’s population is no larger than some local council areas “down south”.

But the next two smallest “states”,  ACT and Tasmania, only have populations of around 400,000 and a little over 500,000 respectively, and they don’t seem to have experienced the same levels of government chaos and incompetence as the Territory has recently experienced.

Of course one might argue that the recently deposed Giles Country Liberal government was a rare beacon of chaotic incompetence in a sea of mediocre but mostly vaguely competent governments.  To an extent that’s probably true, but focusing on the “Crocs in Cabinet” Giles fiasco obscures a more fundamental problem in the Territory’s electoral and broader governance system.

In my opinion the Territory’s single member seats with MLAs elected by preferential voting are simply inappropriate for a “state” with the NT’s population and demography. It seemed to work OK for the first 20 years or so of self-government, with the CLP continuously in government and able always to count on having enough experienced MLAs to step up and become Ministers as others retired (or very occasionally lost their seats).  In fact the stability was so evident that political pundits argued that the tiny size of each seat (around 5,000 voters) meant that incumbency was a massive advantage in the Territory because MLAs could form a personal relationship with just about every voter.

But all that changed after 1998.  MLAs from both parties found themselves being voted out with astonishing regularity, and governments changed frequently.  What happened to MLAs’ much-vaunted personal relationship with voters, you might well ask?  The answer is that if it ever existed it was only a weak factor.  NT seats’ voter numbers stay fairly steady, but the actual individual people making them up change by over 30% and sometimes significantly more during each 4 year term. The Territory’s population is a bit more stable than it once was but it remains highly transitory. Military personnel and mining and resource workers (e.g. on the huge Inpex project currently nearing completion) and their families are here only for the duration of their posting or contract.  Moreover, the extreme boom-bust cycle engendered by the Territory’s narrow, big project-based economy means that even tradies and others who move here hoping to stay are forced to move on when the economy goes into one if its periodic slumps on the completion of a big construction project.

In fact it is now evident that the remarkable stability of MLA incumbency over the first twenty years of self-government had little connection with tiny seat size but instead was mostly due to voter polarisation generated by a single major issue: Aboriginal land rights.  The great majority of Aboriginal people voted Labor because it supported land rights, and a clear majority of non-Aborigines voted CLP because it vehemently opposed land rights.  Because the great bulk of seats are urban and predominantly non-Aboriginal, the CLP didn’t have too much difficulty in retaining government and keeping most of its existing MLAs. Consequently it was also always able to count on having enough capable, experienced MLAs to form a Ministry.

When the era of land claims effectively ended in 1998 with a “sunset” clause on new claims, that single issue polarisation factor ceased to apply.  Labor won the very next election in 2001 and elections from then on have been very volatile with numerous MLAs from both parties frequently losing their seats.  Voters tend to be only loosely attached to either major party, if that.  Many are here for a good time not a long time, and to the extent that they think about politics at all they recognise that the ideological differences between ALP and CLP are quite small.  Although the small number of “tribal” partisans may bridle at the suggestion, both parties subscribe to a contemporary neoliberal model of pragmatic, professional government aimed at stimulating jobs, growth and prosperity.  Voters ruthlessly reject governments that don’t seem to be reliably delivering those things, albeit often with only a forlorn hope that the other mob will prove to be much better.

But with our current single MLA seats and preferential voting, this increasingly rapid turnover of politicians too frequently results in governments with few experienced MLAs and often none at all with any ministerial experience.  Labor came to government for the first time in 2001 having no MLAs with any ministerial experience (logically unavoidable in that circumstance).  However, the CLP government elected in 2012 also had no MLAs with any ministerial experience, and the same is true of the new Gunner Labor government elected in August 2016.

Fortunately this drastic shortage of experienced ministerial talent hasn’t resulted in complete chaos in day-to-day governance, even during the most chaotically divisive days of the Giles government, because we have always had the benefit of quite a few very experienced and capable departmental CEOs and both parties have tended to rely heavily on an equally capable cadre of senior party political advisers.  But the downside is that those veteran CEOs and long-time political advisers are unlikely to be especially self-critical or reflective about the correctness of their previous decisions, actions and policies during previous periods in government, while their new political “masters” lack the knowledge and experience to challenge accepted governmental wisdom in any major way.  Sir Humphrey Appleby is permanently alive and well and in charge up on the fifth floor of Parliament House.  Arguably that’s why the increasingly frequent changes of government since 1998 haven’t really led to significant change, despite voter dissatisfaction and frequent references to Labor as “CLP lite”.

Can anything be done about this? Might the cure be worse than the disease? I don’t know for sure but here are my ideas.

The solution? PR and multi-member seats

It is significant the other two tiny “states” ACT and Tasmania have opted for multi-member seats elected by proportional representation.  Both have slight variations on 5 member seats elected by the Hare-Clark proportional representation system.  It results in generally stable government, with one of the two major parties always elected, occasionally in coalition with a small number of minor party or Independent MLAs.

Those outcomes aren’t all that different from the NT, you might think. But the difference is that they (almost) always have enough experienced MLAs to form a competent Ministry, even after a dramatic change of government of the sort the NT has experienced at its last two elections in 2012 and 2016.  Let’s try a rough comparison of the outcome of the votes cast at the 2016 NT election on the existing single MLA seat system with an estimate of what would have occurred if we had (say) 8 seats each with 5 MLAs elected on a proportional representation voting system.  That’s more than the number of seats in either ACT or Tasmania, but the NT’s huge geographical size probably means that more seats are needed to provide a reasonable reflection of  regional community of interest.

At the August 2016 election Labor gained 42.2% of first preference votes, CLP 31.8% and “others” (independents) 19.6%.  The current system resulted in Labor winning 18 out of the 25 seats, CLP just 2 and Independents 5.  Labor is massively over-represented now based on its percentage vote and the CLP is massively under-represented, to the extent that it cannot provide the effective Opposition that Westminster democracy requires to deliver open, accountable government. Just as importantly, there are hardly any Opposition MLAs in Parliament to learn their political craft so they can step into government and become effective Ministers when the electoral wheel of fortune turns.

Now let’s imagine a system involving 8 separate 5 MLA seats (four in greater Darwin, one for greater Alice Springs and three for remote and regional NT). That is, a 40 member Parliament. Assuming uniform vote percentages in each seat exactly as actually occurred, what would be the outcome? As this excellent analysis  of Tasmania’s system by ABC political analyst Antony Green explains, the “quota” for election as a MLA in a 5 member PR system is 16.7% (to a single decimal place). Don’t forget that preferences of eliminated candidates are still distributed in such a system, so that the more successful candidates will still get more than just their first preference votes (although to what extent depends on the order in which candidates are eliminated).  In that system Labor would have had at least 2 MLAs elected in every one of the 8 seats and no doubt would have achieved 3 in at least half of the seats. Labor would probably have ended up with 20-22 MLAs in this 40 member Legislative Assembly i.e. a likely clear but not overwhelming majority.  CLP with 31.8% of first preference votes fell just short of two quotas in its own right, but would almost certainly have achieved two quotas in all eight seats after distribution of preferences of eliminated candidates.  That is, CLP would have ended up with 16 MLAs, enough to provide an effective Opposition and retain enough MLAs to gain experience over time so they could make effective Ministers when their time in government came.  The Independents would likely have ended up with a similar number of MLAs as they actually now have under the current system (3-5).

With a less dramatic victory for one of the major parties than occurred in 2016 (which you would expect to be the norm), it is likely that the major party with the most seats would need to enter into a coalition agreement (however informal) with one or two Independent or minor party MLAs. That might not sound attractive to the major parties at first glance, but the fact is that the Henderson Labor government was forced into just such an arrangement with Gerry Wood to stay in government between 2010 and 2012, while the Giles CLP regime had a more covert arrangement with Larisa Lee for its last year or so in government.  Political realists should accept that the days of the vast majority of voters being rusted on supporters of one or other of the two major parties are over. Independent MPs are a permanent part of Australia’s political landscape even in a system based on single member seats, and periods when a major party is forced to form a coalition of necessity with one or more Independents will increasingly become the norm.  A system of multi-member seats at least makes a virtue of that necessity, and more importantly ensures that the losing major party will retain enough seats not only to form an effective Opposition but to live to fight another day. Given that both parties will inevitably be on the wrong side of an election result from time to time, it’s a system worthy of serious consideration.

Leave a Reply

11 Comments on "Multi-member Parliamentary seats"

Notify of
Sort by:   newest | oldest | most voted
Geoffrey Goode

The case for replacing the Northern Territory’s system of single-member electoral districts with a system of multi-member districts using the Hare-Clark electoral system that applies in Tasmania and the Australian Capital Territory is very effectively made.

It would be important, for the best representation of voters, to include two other aspects of that system, Robson Rotation and optional preferential voting, as is explained at – NT

Norm Ellis

Sounds good, & well reasoned.
However, selling an increase in the number of elected politicians to the NT electorate may be a big problem.

Jeremy Lawrence

Dear Ken, why did you propose a 40 member parliament? Why not retain 25 members, using 5 districts that each elect 5 members?

Jeremy Lawrence

Further information on multi-member parliamentary seats is available at:

Henk van Leeuwen
Thank you Ken Parish for this wisely and perceptive presentation of the practical case for going with PR Quota Preferential on the NT. This is indeed the way it works in representative and sophisticated parliamentary democracies in Europe – I’m closely watching the 15 March national elections in The Netherlands where I’m from (1965) but still have a vote in since I have dual nationality. What you are advocating the NT Assembly should also be the system of voting for the Federal House of Reps, and we would have a much more representative and perusing and debating government. One vote… Read more »
Geoffrey Goode

Dear Ken,

Could you please change the hyperlink, which the software automatically, but incorrectly, uploaded by substituting a hyphen for the required hash, in my comment to

John Pyke
Yes, Ken, in principle the only sensible approach to the true representation of the people. But NO, not an even number of districts! What if 2 like-minded groupings each get 20 members? Then it becomes a competition to see who can bribe someone from the other “side” to switch or at least become Speaker. Odd number of members per district, and odd number of districts should be the formula. As for your 5×5 + 3×3 alternative, I can appreciate the argument for keeping rural areas more compact, but it does set a higher bar for minor rural parties to get… Read more »
Jeremy Lawrence

Hi Ken, on behalf of the Proportional Representation Society of Australia, I extend an invitation to join. We advocate for PR-STV, such as the 5-member electorates used in the ACT and Tasmanian Legislative Assemblies. You can find out more at and a membership application form is at