Speculations on the Legislative Council
Russell Degnan

From 2006 politics in Victoria could be very different. Under changes to the Constitution instituted in 2003 the Legislative Council will have proportional representation. Whether you think it is better depends on your notions of democracy and what you perceive as the outcomes, but it will certainly change a few things. One, the way elections for the Council are conducted. Electioneering for the council will no longer focus on local members - even if those members were often absent from the process. Preference deals, party politics and quotas will, like the Australian Senate decide the makeup of the Legislative Council. Two, the makeup of the Council will change. There is an expectation that minor parties will gain some representation and probably hold the balance of power. I'll discuss the likelihood of that below; suffice to say it may be true. Finally, in addition to the new districts the previous system of double (8 year) terms has been changed to single (4 year) terms. This may in fact be the biggest change, and the most dangerous, but I will get to that later.

Firstly, the boundary changes. There will be 8 new districts consisting of 5 members each. Three of those will be rural, five in metropolitan Melbourne. Because they have yet to be released any further comments are little more than speculation. In addition, because the final makeup of the seats will depend on things happening at the margin -- preferences, the vagaries of distribution, and small differences in the votes -- any prediction about the result of the Legislative Council elections will be at best, very, very rough. However, testing different results gives some interesting results. Not least, who will benefit from the changes to the electoral rules.

In order to make predictions, some sort of distribution is necessary. A good guide to what is likely can be garnered from the VEC website. Most submissions roughly conform with I would come up with personally, namely the three rural areas being: Western District, Riverina, and Gippsland, and the five metropolitan regions being West, North, Central, East and South-East. The splits between the current provinces can be seen in this spreadsheet along with other data pertaining to election results. Despite anomalies such as seats not being contested, and a relatively poor granularity for examining votes, I am using the Legislative Council results for voting data. The main reason is that minor parties record substantially higher votes already in the Council, indicating that people already conceive of it as a house of review and vote accordingly.

The summary table gives the expected votes in each new district based on the last election, and the quota that would afford them. As with the Senate, each full quota gives one member. Fractional quotas are redistributed according to preferences meaning they cannot be predicted with any great accuracy. However, we can make some general assumptions about preferences: namely, that the Liberals and Nationals will preference each other; as will the Greens and the ALP; that Family First will preference the Liberals; and the Democrats will preference the Greens then the ALP. Again though, I am less interested in predicting the end result than the likely effects, so it is not terribly important.

Because there are 5 seats available the quota is 16.6%. This is above the average vote of all but the two major parties, but affords the Greens and the Nationals full quotas in two districts: Central Melbourne for the Greens and Riverina for the Nationals, where their votes are around 20%. By contrast, the two major parties are almost always guaranteed the 33% they need for two seats. The chamber is rigged therefore, in their favour, with the others left to fight for the scraps. A summary of the way the council would look if the last election result was repeated can be seen below:

Summary TableSeatsWestDistRiverinaGippslandWestNorthCentralSouth-EastEast
ALP21177506.462.79133473.522.36154584.312.54213862.733.59210998.093.55139101.32.38171700.472.9173982.132.88
Liberals14140632.272.21112706.761.99133382.352.1995409.151.694577.031.59138468.822.37136993.752.32147307.882.44
Greens431045.380.4925018.280.4436366.160.639076.480.6642510.340.7266212.861.1338378.630.6536095.870.6
Democrats04752.330.0753900.14666.750.088986.060.1585940.146971.610.127895.650.134458.60.07
Nationals127460.560.4362872.441.11360860.590000000000

The ALP would have a clear majority, as it does already. And it could easily be bigger than the one indicated. Two of the Green seats -- Gippsland and West Melbourne -- are dependent on Democrat preferences. The house of review would be no more useful under the new system than it is already. Namely not at all. But is the typical? Looking at the likely results for each party, and for each seat gives a better idea. From here, you'll need to see the spreadsheet to see the calculations made. Note too, that using the spreadsheet you can easily change the assumptions and generate your own tables if you want to check my conclusions.


Party Summary

The number of seats a quotas affords a party seems to operate in a series of tiers. Getting just over or near the quota gives a party the seat. Getting below that range enhances the chances of that party's ideological allies. Hence, for the Greens an Nationals, who get around half to two thirds of a quota, their chances of gaining a seat can actually depend on whether the major parties are near the tier -- and utilising their vote -- or in the middle -- and therefore distributing their preferences across. This makes counting seats for the minor parties more problematic, but some general rules can be formulated.

For the ALP 18 seats is a typical number. They are practically guaranteed three seats in the West and North, and two seats everywhere else. Their vote has to drop precipitously to under 35% before they start to lose seats, but anything over 42% will see them gain, and be near a majority in the Council.

The Liberals strike a similar balance at 14 seats, missing out in the West and North where the ALP is strong. However, this assessment relies on ALP-Green preferencing. If a successful preference deal was struck between the Liberals and Greens several seats would end up in different hands. Like the ALP though, anything over 42% will put them close to a majority.

The Greens are cursed to sit in the middle. In Central Melbourne they are guaranteed one seat, but the others depend on where the other parties sit. If both are around 40% then the Greens can preference off Labor to get between 4 and 6 seats with no geographic predominance. Conversely, when one of the major parties is strong, the Greens run second, dropping to between 1 and 3 seats.

The Nationals are similarly cursed as well as being limited to winning in the three rural seats. They should get one seat regardless, but it is hard to tell. In addition they are difficult to model because their vote is not distributed and an increase of 2% will actually have a much bigger increase on their chance of gaining a seat. Finally, Family First may reduce the Nationals vote, or gain them valuable preferences. Either way, the number of seats they hold in the Legislative Council is set to halve, or worse.

At the very bottom minor parties and independents are being nailed. The 16.6% quota is well beyond their ability to collect, and the Democrats would need to treble their vote and get very good preferencing to get even one seat. Apparently, under a proportional system, all parties are equal, but some are more equal than others.


Seat Summary

So who gets what seats? First, let me repeat, without a proper redistribution, these are rough at best, and some subtle differences will give different results.

Riverina could be the last bastion of the National party in the upper house. Unless the Greens make big inroads it is almost certain to give 2 Labor, 2 Liberals and a National for some time to come.

Western District and Gippsland will also give 2 and 2, but the fifth seat is a lottery, with both major parties, the Greens and the Nationals a chance.

West and North Melbourne are Labor's strongholds. Even a huge Liberal victory will probably still see them win 3 seats, so the real question is whether the Labor surplus can get the Greens over the Liberals or not. Which given the same question applies in most other seats as well, raises an interesting question. Would the major parties be willing to stitch the Greens up by trading preferences with each other?

Central Melbourne is a done deal. Unless there is a big swing to the Liberals the solid Green vote will make this a 2-2-1. As with the lower house, my vote is only worth the money the party who gets it will cream off the government.

South East and East Melbourne are where elections are won or lost. The fifth seat is a lottery again, between the two major parties and the Greens.

For those counting, barring an unusually large victory, or a generational change in the fortunes of our political parties, the following are the base figures: Labor, 18; Liberals, 14; Greens, 1; Nationals, 1. And the most number of seats that could be won would be: Labor, 22; Liberals, 20; Greens, 7; Nationals, 3.


A House of Review?

There was a lovely contradiction in the approach the Labor party took to implementing Legislative Council reform. On the one hand they argued for the council to be a "House of Review", as neatly explained by John Lenders:

One view is that the principle of government mandate comes from the majority in the Legislative Assembly rather than those who win a majority in the Legislative Council, so that implies reviewing the deliberations of the house of government rather than attempting to be the house of government.

On the other, they argued against eight year terms by accusing them of giving a "stale mandate", because the honourable member may not have been required to stand for election during the previous cycle. But, if it is serving as a house of review it has no mandate, nor even the need of one. If it was really a review it would be broadly representative of the electorate and provide continuity from one term to the next. By removing the longer terms, the Bracks Government has removed that continuity. But worse, by setting the quota at 1/6 they have also made the Legislative Council the one thing they argued it shouldn't be: a copy of the Legislative Assembly.

If a government wins a big majority, once, not twice as with longer terms, it will -- possibly in coalition -- have a clear majority in both houses. It may not be a big majority, but unless the Greens become a serious force in Victorian politics -- for which they will need to raise their vote to 12 percent or more -- then the Legislative Council will provide little minor party representation, and a balanced house to be negotiated through only when the government is weak, and presumably cautious.

In addition, it is a system that effectively locks in party seats. Effectively, only 6 seats will be genuine contests in the new Council. The spoils will go to party hacks and the real fight will be the factional brawls leading up to the election. The vast majority of the Council will have little to worry about from electors; good if you want them to get on with the job of "reviewing"; bad if you think they will be little more than drunk time-servers milking the public purse while they divide their time between publicly ridiculing their opponents and ensuring the continuing support of their own party.

Even better, we now need a referendum to change large slabs of the constitution. In short, the new system probably sucks. And we are stuck with it.

Sterner Matters 27th May, 2005 03:56:15   [#] 

Comments

NQR
Russ, your figures aren't quite right - there are some things that aren't accounted for - I'll give you a fuller analysis when I have some time. One of teh main points is the gains that will be made where party's didn't field candidates last time....
Pearcey  27th May, 2005 11:29:06  

You're right
Mate, that would be appreciated.

Check the spreadsheet for where most of my speculations are coming from. One thing, the ALP, Liberals and Greens did field candidates in all provinces in the last election. The Nationals didn't but their vote is pretty small in the areas they didn't (except Ballarat which probably plays down the Western District vote).

A lot of things were not taken into account, and a lot more won't be known until after the election. A shortlist for you to speculate on if you have time:
- The Democrats fielded candidates in most places, but their vote is in free-fall, so I wonder what influence they will have?
- What will other minor parties do, if anything? Particularly Family First, Liberals for Forests and the DLP. Their preferences might decide a few seats, although I can't see them accumulating enough votes to get over the line (again, see the spreadsheet regarding the Dems)
- Will people vote differently for proportional representation?
- Will the minor parties preference the Greens to tip them in?
- Even if they had candidates the major parties didn't campaign much. How will a larger campaign make a difference?
- Is it possible to run a localised campaign for a small group of candidates?
- Is it possible for a high profile independent to get near a quota in a district? Even half a quota -- 8% -- would make things interesting.
- And the biggest question of all, what will the major parties' relative vote be?
Russ  27th May, 2005 13:34:54  

Expect minor party votes to improve
I would expect minor party votes to improve next time around - simply because it is now far more likely for minor parties (especially the Greens) to at least get one quota in each of the upper house regions.
Aaron Hewett  28th May, 2005 18:19:32  

Riverina very difficult for independants
As we were discussing, it's going to be very difficult for an independent to get up in Riverina. If you look at the major population centre groups you've got Bendigo, Albury-Wodonga/Wangaratta, Shepparton/Echuca, and Swan Hill/Mildura, who all have independent media (so the Mayor of Bendigo is a complete unknown in the other regions, for instance), are almost as close to Melbourne as they are to each other, and have little commonality of interest on things like water, roads or rail any more than they have commonality with other regions of rural Victoria.

The only "common interest" if you like is the Murray-Darling Basin, but even there it's limited as Mildura and Wodonga are in conflict over water usage more often than not.

Not to mention that it's such a distance from end to end you'd need a plane to campaign effectively (otherwise you'd spend your days travelling from one end of the electorate to the other and have no time to actually campaign).

Rob  1st June, 2005 16:23:21  

The minor players
Rob, on Riverina I agree. But I also think that is almost certainly going to go 2-2 Lab-Lib and 1 to the Nats, so it may be a moot point. Western District would be the most likely candidate for mind, because Geelong makes up almost half the electorate, and Ballarat another quarter. Gipplsand would have a similar distribution with respect to the Latrobe Valley. So, independents might muddy the waters there. Certainly I think they could get a big enough vote to decide the 5th seat, and that gives them an outside chance of taking it if the preferences fall the right way.

Aaron, most minor parties haven't been running so their vote will naturally increase - by how much is another question. Your assumption is that people understand the electoral process. As I said, the LC vote was much higher in the past, which would indicate (some) people already considered it a house of review and voted accordingly. Family First gets 5% of the (Federal HoR) vote in some places. So potentially they could cause problems. But many of the parties may not even run at all. The parties that have previously run are the Democrats and the Greens. I suspect the Democrat vote will decline, although the Greens should improve, albeit marginally. As indicated, they will get at least one LC seat though, and be well placed for several more. Overall, the total minor party vote will probably increase, and their preferences will probably decide the fifth seat in 3/4 of the districts (meaning those 6 seats are completely unpredictable).
Russ  2nd June, 2005 01:13:31