Saturday 25 May 2013

In a Rich Man's World

So last week I made you all sit through a summary of the Labor budget, and threw in a few pretty pictures to keep you interested. Now we're going to do the fun bit -- or for those of you with no sense of edge-of-your-seat-adventure, the interpretation bit.

I mentioned that this kind of economic prediction is very difficult to do with any accuracy -- not least because the extent to which budget spending influences votes is not only very hard to calculate, but varies from year to year and community to community.

However, for obvious reasons, the role of the budget will be greater where all other factors are evenly balanced: in other words, in marginal seats. So what I'm going to do here is say boo to accurate predictions based on this data alone. And I'm going to say a slightly quieter boo to predictions factoring in existing margins, polling, voting history, traditional voting of demographics, targeted campaign strategies, ideology and promises and the like ... for now.

What I am going to do is work out roughly how people would vote if financial benefit were the only consideration in their vote. This will assume all financial benefit is equal, so a million dollars of funding is given equal weight regardless of how directly (or indirectly) it impacts the voters. Now obviously this is not how the real world works. One million dollars towards education is fantastic (a million dollars more fantastic than nothing, at least) but it will mean less to parents than a million dollars towards disability services would mean to the disabled and their carers (even if we assume there are as many parents as disabled people and carers).

So the calculations are cheap and nasty, but that is okay because they will not be a major plank in our raft of prediction tools - simply another factor to consider in close-run seats. Besides, we're still working with 2006 data anyway, so we're going to have to live with some inaccuracies.

So here are the maps:


Firstly, here is a map of seats gaining funding between the five initiatives outlined last week: (N.B. some averaging and approximation was necessary where electoral seats did not match up with ABS statistical divisions. Generalisations have been made, and urban seats are inaccurate due to the low resolution of ABS maps.)



For this map, each sheet was shaded with pure green (#00FF00) at an opacity level equal to the number of people affected by the funding gains. The intensity of the colour, therefore, shows the percentage of people won over by the pension, school and disability increases.

It is assumed that the benefits from the programs mentioned last week are independent of each other. In other words in an area with ~10% of the population aged 55-64 and ~5% living with a disability, the total people benefiting from the two initiatives related to these demographics is not 15% of the population but 14.5%, since .5% (10% of 5%, or 5% of 10%) fit into both categories and should not be counted twice. At first this seems illogical when we allow for a person between 55 and 64 to also vote according to the aged 4-and-under factor, but this is because people close to retirement may have children in that age bracket and act accordingly. None of the demographics are exclusive of any other - A 55 year old disabled man attending university and raising a school age child while his wife is pregnant would be rare, but would also be affected by all five budget measures.

As we can see the distribution is surprisingly even. There is a slight advantage to seats in southeastern states, but not one that is noticeable on this map.

Here is another map, showing places loosing funding:


Again, the burden is rather evenly spread. The concentration in urban areas is slightly more noticeable than the concentrations in the previous map. This is largely the result of the university cuts.

However, we need to remember that not all of this funding is equal. Different measures cost different amounts and benefit or disadvantage people to different extents. The above map demonstrates the population densities where people are aided or disadvantaged. The following map factors in how much their are aided or disadvantaged. The above map shows the government's target demographics, the one below shows the extent of the impact in average dollars per person.







Here, the intensity of the colour is relative to the amount of money gained per person on average. It is the benefit per targeted person multiplied by the proportion of people benefited. So 10% of the population gaining a $100 boost is valued here on par with 100% of the population gaining a $10 bonus. The problem with this model is that giving $5 billion to one person in Sydney is the same as giving $1000 to everyone in Sydney, although the latter will clearly get you more votes.

Here is a map where the amount of money is not a factor. votes can be bought for a fraction of a cent, and spending almost 20,000 per targeted person (e.g. the NDIS) is worth no more or less than handing out loose change on the street corner.



This makes the distinctions between winners and losers a little clearer. Again the metropolitan areas are feeling the university cuts, but over all the nation is in the green. And in practical terms this is probably a closer mirror to the thinking of people who vote according to personal benefit; they don't compare themselves to others along the lines of 'the money I got in my old age pension boost is only about 5% of what they get under the National Disability Insurance Scheme'. They merely want to know if they have more than they did before - and if the opposition will see and raise the government's promises.

The problem for the opposition is that they need to try to offer more money to more people while ensuring greater savings. Or do they?

How to Frame a Budget from the Other Side of the Floor:


The budget reply came as a big surprise to many. The opposition agreed with the bulk of Labor's proposed budget recommendations. There was none of the normal game playing, such as blaming Labor for the deficit and not supplying a surplus. Programs were not opposed simply because Labor proposed them. It was an odd day for many, however it was a fantastically well thought out move.

Firstly, in the last budget Abbott did not promise the kind of immediate return to surplus many half-interested voters assume he intends to give. The Coalition has already done the groundwork on this. If a budget surplus is your top priority, the Lib-Nats have already framed themselves as the can-do candidates. And Abbott continued this theme by calling the situation an "emergency" which required such "objectionable" measures to be adopted. However, Mr Abbott has not promised unreasonable economic changes he cannot deliver - his reply is roughly equal in pragmatism to Labor's budget because it is a very similar deal.

More importantly, by agreeing on a lot of Labor's plans, he's taken the wind out of their sails. Any backlash - even post election and even if the Liberals come to power - will be against the Gillard government reforms. And by agreeing to Labor's terms on the baby bonus, the NDIS and most of the other budget plans, these are not going to win Labor much traction because they are not going to be debated. And this is the key to Abbott's budget reply. He has chosen where the two parties will differ, and he gets to frame the debate.

The key differences, and thus the main focus of this campaign, are the "carbon tax", the "mining tax" and "Gonski reforms". Importantly, only one of those is the term that Labor originally used to describe them. It is fair to say that many people do not fully understand the Carbon Tax or the Mining Tax. They sound like taxes on the common people, and sound like big ones. In one of the post-western-Sydney-trip reports on the ABC there was a shop owner concerned about how he needs to pay the carbon tax on top of the GST, income tax and the like. This is clearly one issue where the Coalition has clearly cut through more effectively than the Government.

The odd-one-out is the Gonski reforms, which are also poorly understood and somewhat nebulous as far as the average voter understands. There is increased funding for schools, sure, but how this funding is allocated (private vs public, infrastructure vs technology vs staffing) is all very confusing to anyone who doesn't have the time to sit down and read about the policy.

I could look at which areas will benefit the most from the scrapping of the "carbon pricing", the "mineral resource rent tax" and, to a lesser extent, the "Gonski reforms". But I suspect this is irrelevant, because the Coalition is selling this as a better deal for all Australians. And perhaps it is -- that is for the voters to decide -- but my point is this: Abbott's reply was not targeted at helping specific target demographics at the expense of the other. It was an exercise in downplaying Labor's strongest policies and framing the economic debate in terms he thinks he has the best chance of winning.

This is not necessarily a bad thing, and it is not necessarily a good thing. It is just a simplicity thing. Abbott might be hiding some further, more controversial measures he plans to bring in -- to do with "stopping the boats" perhaps -- by sweeping them under the carpet. Alternatively, these could simply be the core issues of his campaign, with nothing hidden by the time we reach the polls, which is why he has no declared plans to change industrial relations until the election after next.

Mr Abbott is offering a simple, easy to grasp choice. Perhaps this is the best response when Labor has trouble explaining their own policies so that the average voter can comprehend. Better the devil you understand...

Data Dump

This is a supplement to the post above. This does not count towards my limit of one post per week, and should be read in conjunction with the post to which it refers.

So, it's been a little while since I last did a data dump. This week's will be a little different to before, because the maps were largely derived from the ABS maps provided last week. Apart from these maps offering ranges rather than pure figures (and thus some inherent errors in this week's maps precision) this meant making assumptions wherever the boundaries did not match up - i.e. that each ABS division was homogenous and that its population was evenly distributed across the area.

Instead, I have the separate layers used to make the combined maps for you to ignore, and the raw data for working out how much was earned per person targeted by an initiative in the budget (and thus each layer's opacity for map 3).


Layers:

Superannuation – Benefits to Retirees

Demographic Distribution for Recipients of the Pension Increase

School Funding  Benefits to Parents of School-Aged Students



Demographic Distribution of School Aged Students

NDIS Benefits

Demographic Distribution of People Living with a Disability

University Indexation

Demographic Distribution of Tertiary Students

Baby Bonus Impacts

Demographic Distribution of the Parents of Young Children (as a Proxy for Expecting Parents)

 

Data:

 #  Pensioners will gain $35.50 per fortnight (source) which is around $925 per person annually. Note that couples pension increases by less (same source).

 # Schools are gaining $9.8 billion over 6 years (source), which is an average of $1.6 billion per year. With around 3.5 million school students (source) $1.6 billion a year is roughly $450 per student per year. Government funding schedules always seem to ramp up over time, so this is probably less initially.

 # The National Disability Insurance Scheme, which will support a projected 410,000 disabled people, carers and the like (source, p.6), costs $8 billion per year (source), or roughly 19,500 per person assisted.

  # I estimated the University Indexation to be around $3400 per student. Of course, the students aren't directly paying for all of this, that is just the financial cost of the services they would otherwise have had. (Note: this is not the cost of financial services lost, since university funding will still increase -- see last week's post for details -- nor is it the loss in value in real terms, since university numbers will increase sharing the burden but also straining resources.) This was based on a $2.8 billion dollar reduction (source) divided among roughly 830,000 domestic students in universities (source).

To be fair, not all of the tertiary education indexation is aimed at universities. Also, I could not find any data as to how many years this reduction was going to take place over, so I factored it all into a 12-month period from January. If this is a 5 year reduction, the per-student cost will drop per year. Further per-student drops in this impact may be calculated if you include foreign students or non-uni students who will share in some of the losses.

 # For the Baby Bonus, the cut is the whole $5,000 per child born (source). Also, this article suggests less than 30,000 families would have benefited in the next year. With a voting population of 16 million (source) last election, if each was in a relationship there would be 8 million couples. 30,000 of these expecting a child is around 0.4%. This is well below the minimum 5-10% of the population who actually had children born between 2002 and 2006 from last weeks census data. Part of this could be accounted for by twins, and part by the fact the census is data gives us the number of parents who had children over 4 years, as opposed to over a one year time span. And then there are parents who have two children within four years. And, of course, not everyone of voting age is in a relationship.

But all of this is irrelevant to some extent since the psychological effect (i.e. the effect which would play out during voting) is not to do with who will have a child -- and indeed anyone curently expecting will still receive the bonus -- but who might have a child. That includes parents trying for a child, or who plan to start a family in the near future. Whether or not this pans out in the long term these people will feel that they will lose out on the $5000 next electoion.

Both approximations (ABS data used and statistical number of actual births) are poor indicators of the voting demographic for hopeful parents, but in the absence of better figures this is the best I can offer. Treat the Baby Bonus data as a VERY rough guide. Considering how even the distribution on that map was, I don't think it had much of an overall effect, though.

Saturday 18 May 2013

Money Money Money

Responsible Spending:


It should not come as a great surprise to anyone that we will be turning our attention to the economic situation in a week when the budget was released and the opposition budget reply naturally followed. This was also a week that saw France officially describe its financial situation using the R-Word and Gina Rinehart once again bemoaned the state of Australia’s finances, despite having enough money to personally bail out a small nation (or at least share with her children). Tony Abbott described a Budget Emergency this week. We recently dropped below parity with the $USD. Bill Gates is once more the richest person in the world. It seems fitting to summarise the entire economic world with a phrase from the master of the dramatic arts: “it is a tale told by an idiot, full of sound and fury, signifying nothing.

I have written previously that I shall restrict myself to analysis and interpretation, not opinion. This is a little inaccurate if you take into account my admittedly subjective tirade against the current incarnation of our compulsory voting system. However I do strive to be nonpartisan and to rely on evidence wherever possible – within the loose shackles of a light, entertaining and occasionally jocular blog format. For this reason I will not be critiquing the contents of the budget or the Coalition’s reply. And, of course, factoring in the effect of the budget on the predicted voting patterns of the public can never be more than pure speculation, even with accurate and targeted polling data. I will, however, be looking at what the budgets can tell us about the strategy of both major parties for the upcoming election and how this may influence our election predictions.

This week we saw an austere and “responsible” plan from both [sic] sides. Responsible spending is one of those great political terms that seem to cluster around economics and finance, like “eliminating waste”. It is a great term because it is both simple and evocative. It is a great term because it is something every household can relate to. It is a great term because it automatically implies that your opponents are economically incompetent, and because it questions both the funding and necessity of their policies in general without calling on specifics. And it is a great term because it is something every Australian voter can agree to oppose in principle; who could possibly support irresponsible spending?

But most of all, from a political point of view at least, it is a great term because it can mean absolutely anything and therefore binds you to absolutely nothing. Normally it implies reduced spending – especially in the current economic climate – but this is not necessarily the only interpretation. Responsible spending can range from winding back funding to ‘unnecessary’ projects to throwing truckloads of cash at policies where it is ‘cheaper to fix the problem now than deal with it later’. It can mean not wastefully over funding schemes that could succeed on much smaller budgets, or it can mean providing enough resources to ensure targets are reached and previous investments were not wasted. One could argue that additional funding for childcare would be irresponsible if the budget were particularly strained. Conversely one could argue that any investment to allow many parents – particularly single parents – to get back to work, earn a decent salary and pay more income tax is necessary to boost government coffers.

Effectively, responsible spending means funding the initiatives a party considers important, and cutting funding to those it opposes or simply considers ineffective. It means funding the policies they support, and not those that they oppose. It is stating the blatantly obvious, while making the other side look bad – and you don’t even have to make any new commitments.

Responsible spending is always a defensive position, a reaction to both the economic circumstances at home and abroad as well as the policies of others. After the aggressive and disenchanting election of 2010 where Labor and the Coalition both went on the offensive, it is interesting to see both [sic] sides of politics playing such a defensive game this year.

Both the ALP and the Coalition have proposed very similar plans with an emphasis on reducing spending – responsibly, of course – and delivering a budget surplus. The Coalition has to prove they can deliver a surplus because that is what the Liberal Party has been promising for quite some time now. The Labor Party needs to prove they can deliver a surplus because that is what the Liberal Party has been promising for quite some time now. However there is an argument to be made that a healthy economy is a means to an end, and not an end in itself. Having some money left over is useful in the event of a crisis. On the other hand, it is a bit redundant if, in order to obtain that surplus, the government needs to expose the nation to a crisis by cutting funding to education, health, defence and other vital services.

Normally a budget based on reducing spending is far more about perception and politics than about actual governance. Given the economic situation globally there is a genuine need to keep Australia in a position where we can avoid Eurozone-style recessions, so there is probably a mix of government and politics in this year’s budget. However, we are not experiencing the budget emergency Abbott suggests. Here is a graph representing debt among 19 of the G20 economies (the 20th economy is the European Union which is difficult to calculate and already includes several economies represented here).
 
Source for debt figures.

Now you can say that any debt is bad debt, but notice that all 19 of the most important economies in the world are in debt right now, because even bad debt is better than not being able to fund basic services. Australia’s 420-odd billion in debt does need to be paid back eventually, but it is insignificant compared to Japan or the USA, for example. Unemployment is at record lows and we have a triple A credit rating, both of which suggest we are in a good position to pay this debt back over the next decade or so. A more important factor is nominal GDP – the amount of money the country makes. This relates (through government revenue by taxation) to our ability to pay the money back. If debt is equal to nominal GDP then we are making as much each year as our debt. In other words if we put all of our profits towards paying off the debt it would take one year. Of course, we won’t put all of our profits towards that because most of that money stays in our pockets, and that which does go to the government also needs to be spent on important services that keep the country running. Nonetheless, a higher GDP puts us in better stead to pay back the debt.

In the next graph debt is still represented by the countries flags, which are now hoisted on flagpoles representing nominal GDP. The GDP can be measured against the scale on the Y axis, and the debt is to scale. The bottom of each banner, as a result, indicates on the Y axis’s scale the change each nation would get if it tried to pay off its debt with one years GDP. It’s the flag-to-flagpole ratio that illustrates how big the debt problem is. A small flag on a high mast is great news. A long banner on a short flagpole – I’m looking at you, Japan – is not so good. Compare our situation to, for example, Germany (the super-economy propping up most of Europe) or China (our local economic superpower) and decide for yourself how big Abbott’s budget emergency really is:

Source for debt figures; source for nominal GDP figures.

The Labor Budget:

During the WA election some of my predictions were thrown out by local spending, referred to by Coalition politicians as “sandbagging”. One of the lessons I took from that election was the importance of considering local issues – especially financial issues – in close-run seats. The following week I looked at the spending allocated to Western Sydney and we saw that Labor was not so much interested in winning seats here as they were in not losing them. I would like to do a similar, nationwide analysis of the budget spending.

One of my Archaeological colleagues recently suggested there are two types of voters – those who vote based on ideology and those who vote based on personal gain. I am not convinced that these are the only kinds of voters (e.g. there are people who vote based on perception and personalities). I am also not convinced that everyone falls neatly in one of these categories exclusively. Importantly, which type of voter someone is may depend on various demographic factors, so the budget spending’s effects might be downplayed or amplified in different seats depending on the average age, profession, education level etc. of the constituents. However, let us assume that all voters vote based purely on their own personal benefit from the budget. The question I then want to ask is ‘just who is Labor winning over, and who are they losing?’

Now obviously a 10 million dollar announcement will have less impact than a 10 billion dollar one, and all announcements will effect everybody to some extent (even if only through the non-funding of projects they would more directly benefit from). To simplify things, I am going to look at what I consider to be the top five announcements (as determined by the amount of media coverage, how much the ALP probably intends to trumpet the project and how much it costs) and then determine who are the people most likely to benefit or lose out as a result, and where they are most likely to live.*

Announcement 1: National Disability Insurance Scheme
The NDIS is a much celebrated scheme to ensure equal access to quality services for disabled people and an attempt to prevent as many cases ‘falling through the gaps’. The obvious group to be most affected by this are people living with disabilities.

 Source.

Announcement 2: Raising the Super Guarantee
With an ageing population, many are concerned for their retirement. Younger voters will see many more changes to superannuation before they retire, and probably have more immediate concerns. This policy is complex, with gains and losses. There is less government contribution to superfunds and less tax concessions for the rich but a higher tax-free threshold and raised old age pensions. I am using people around retirement age as the most likely affected group.

Source
This data is a little old, so many of the people represented here will have already retired. However, this is still probably a good approximation of near-retiring population distributions.
 
Announcement 3: Funding Reduction for Universities
One of the most publicised losers of this budget, the government is winding back promised funding for universities. To be precise, universities will not be losing money, but their funding will not increase by as much as it has been recently. The problem for universities are firstly that some have already decided how to spend the money they now won't be getting, and secondly that the increase in enrolments may mean less money per student. The further commercialisation of tertiary education will most likely result. This is not something I will get into here since I try to avoid expressing personal opinions, but it is an issue that many students understandably have strong feelings on.


Announcement 4: Funding Increase for Schools
Although not a 1 for 1 correlation, there has been a pairing of cuts to unis as "paying for" schools. In reality, of course, savings from the universities are added to the savings columns, expenses on schools added to the expenses columns and the totals are tallied. You could equally argue university wind backs are funding our military and it is the scrapped baby bonus that will fund education reforms.


However you wrap it up, it is obvious that there are far more children attending (compulsory) primary and secondary education (as well as childcare facilities and other institutions gaining a funding boost) than there are uni attendees. Arguing that the government is disadvantaging the few to benefit the many, along with the point that without a strong school system the universities will struggle anyhow, will no doubt be a common focus in the election campaign.

Announcement 5: Scrapping the Baby Bonus
This is a hard one to model since the census does not specifically record the number of pregnant women or couples trying for a child. This is also the most likely category to be skewed by the age of this data, since couples with a baby in 2006 may or may not have another in 2013. It does however suggest that these areas are populated by demographics likely to have children (young couples with financial security to start a family, areas with high teen-pregnancy rates etc.) which may well still be the case today.

The best I could do was model the population of children under four years of age.
 

Further analysis of these maps to follow...

* Unfortunately the 2011 Census is still being processed for cartographic applications, so the map data dates from 2006 and may be inaccurate. Maps are based on place of usual residence.

Friday 10 May 2013

Long Posts and Small Fields

Small Fields:


One of the advantages of state elections from a reporter's point of view is that the number of voters is smaller, which means results are known sooner. With only three seats to count, this applies doubly in the TasLegCo.

My past predictive record is just above 90%. I only recorded 2 official predictions in the previous post:

  • Jim Wilkinson will retain Nelson
  • Vanessa Goodwin will retain Pembroke

You might expect us to get around 1.8 of those predictions right. The good news is that we got better than that, and so far have a 100% accuracy rating for Tasmanian state election predictions. This highlights the risk of working with small sample sizes, which I will no doubt end up doing at some point in the future of this blog. Meanwhile, our overall accuracy is at 90.3: up 0.2 percentage points.

I also made five unofficial predictions:
  • Leonie Hiscutt to win Montgomery
  • Cheryl Fuller to place second in Montgomery
  • Kevin Morgan to place last in Montgomery
  • Allison Ritchie to place second in Pembroke
  • Tom Baxter to place second in Nelson
Four of these were correct, including the first and most relevant prediction on that list. So had I gone out on a limb and predicted a Hiscutt victory, then I would have 100% off of three predictions (and raised our full record to 84/95 or 90.4%). However I was wary of that prediction and, as it turned out, rightly so.


Long Posts:


Normally I would skip over the correct predictions and focus on where I went wrong. As I didn't actually go wrong on any of the significant predictions, I'm going to look at why I was justified is playing cautiously with the Montgomery prediction.

Firstly, there was no reliable polling for Montgomery, so I was basing my results on generalisations and past results. Secondly, no-one in Montgomery had the kind of obvious advantage as in the other two seats. These combined make any prediction shaky to begin with. However, I set out my lengthy reasoning last week, beginning with the following lines:

I'm going to call Mr Morgan as the first to drop out, with most of his support flowing to Ms Fuller. The three way battle is hard to predict.

Four scenarios followed - two won by Mr Vincent, one by Ms Hiscutt and one tossup between Ms Hiscutt and Ms Fuller. None of these scenarios unfolded, because the one unofficial prediction I did not get right last week was that Mr Morgan would drop out first.

Ed Vincent polled the lowest, with only 9.73% of the primary vote (ignoring informal votes), presumably as he and Ms Hiscutt had a very similar voter base, and Ms Hiscutt got the bulk of them. This highlights the problems with first past the post polling that I have been discussing over the last few weeks; similarly placed candidates can undercut each other and let a less popular contender come out on top.

Tasmania, though, being a decently constructed system, is a preferential polling system. As such, Mr Vincent's supporters' votes were transferred to their next preferences. Surprisingly Ms Hiscutt received the least of these, while the most ideologically opposite candidate (Mr Morgan) received the most. This may represent a preference for Independents in the Legislative Council (the other independent, Ms Fuller, didn't do too badly out of the preferences.)

Whatever the reason, the fact remains that my predictions were based on a flawed premise. True, Ms Hiscutt was a strong contender either way. However, given that my predictive methodology went astray in the first round, it is understandable, I feel, that I was not confident enough to put my prediction for Montgomery outside the zone of a tossup.

Coming Up:

Tuesday next week is the second Tuesday in May, and therefore Budget Night. That may or may not steer my content for next week. We'll all have to wait and see.

Friday 3 May 2013

Girt by Sea

Before I get to the main body of this post and my next set of predictions I should take this opportunity to reflect on my previous predictions. Counting has well and truly finished in WA now, and it turns out that the Shooters and Fishers party did not win the Mining and Pastoral seat they were in line for.

I noted that this may well happen in my review of that election, but I can now confirm that I did in fact get my predictions for the Mining and Pastoral region correct. This brings my accuracy rating up to 90.1% which is above my arbitrary "mediocre" threshold of 90% and below my "good" threshold of 95%.

TasLegCo:


Tasmania does elections differently to (and in many ways better than) the rest of us. I'm not going to go into Hare quota vs Droop/Hagenbach-Bischoff quotas here, but there are some other, less well known differences. For example, federally and in all other states (except Queensland*) the upper house is elected state-wide and the lower house is elected by electoral divisions or "seats". In Tasmania the lower house is elected state wide (which is why the Greens so regularly play such a big role - they garner enough support state-wide rather than out-muscling the Liberal and Labor candidates in an expensive seat-by-seat campaign). The upper house - or Legislative Council (or LegCo) - is therefore elected across seats.

Another difference is that ballots do not need to be fully filled in to be valid. You can always leave your last preference blank, and you never have to list more than your first three preferences. Also, a process called 'Robson rotation' is employed, where the candidate order is randomised between different ballots in the hopes of reducing the influence of the dreaded Donkey Vote.

Like other Legislative Councils and the Federal Senate (and other upper houses around the world, such as the U.S. Senate) candidates are not elected all at once. For example, in the Federal Senate, half of the senators are elected every three years (assuming the Senate is not overhauled by a double dissolution or the like) and serve for six-year terms as a result. In the U.S. one third (roughly) of the Senate is replaced every two years, again setting six year terms. This means there is continuity in the Senate  minimising the disruption of an election to ongoing processes  and ensures that the party makeup is more stable than the lower house, so a landslide in one election does not necessarily give one party both houses.

The Tasmanian LegCo has six year terms as well, but there is a LegCo election every year. The numbers of councillors has changed a fair bit in recent years, but the current system has three seats up for election in odd numbered years and two in even numbered years for a total of 15 seats. (One previous incarnation had 18 seats, and thus three seats up for election every year.)

Being an odd numbered year, three seats are up for election on Saturday: Montgomery, Nelson and Pembroke.

Because the state of Tasmania is divided into seats for the upper house elections, is is not uncommon for safe seats to emerge, and incumbency is reasonably stable. In most cases you expect the same person to be re-elected. Interestingly, only two upper house members are party affiliated - one Liberal and one ALP. One of these is up for re-election.

There is little valid polling available for the LegCo. Many seats are not contested by all major parties (Labor is not officially contesting any this year). When seats are contested, the major parties do not necessarily do well. This means margins and predicted swings - the bread and butter of the psephologists predictive sandwich - are not applicable. Instead, we're going on past voting trends (but without me having the time to put together any variable-dependent transparency arrays...) and other, less reliable factors. This round of predictions is going to be a bit rough as a result.

Nelson:

Incumbent: Jim Wilkinson (Indepenent)

Mr Wilkinson was elected for the seat of Queenborough in '95. In the 1998-99 redistribution caused by the above mentioned alterations to LegCo seats, he was shifted to the newly created seat of Nelson. Wilkinson won the 2001 elections with 48.99% of the primary vote in a 4-way contest, almost double his nearest competitor. In 2007 he won again with 61.60% of the vote in a two-way race against a Greens candidate. Boundary redistribution since then has removed some of the strongest Greens sections from this seat.

Mr Wilkinson will be running against three other candidates on Saturday. Two are independents unlikely to garner enough support to topple the current MLC. The third is a Greens candidate which is likely to be Mr Wilkinson's main competitor. However, given Mr Wilkinson's comfortable victory over the Greens in 2007 and the subsequent redistribution, the incumbent is in a strong position.

Prediction: Jim Wilkinson Returned

Pembroke:

Incumbent: Vanessa Goodwin (Liberal Party)

Ms Goodwin is the only Liberal party member in the LegCo, and one of two party affiliated MLCs (the other being Labor's Craig Farrell in the Seat of Derwent, up for election in 2015). After the withdrawal of Labor-associated Allison Ritchie - an active advocate for women in politics and the youngest person ever elected to the Tasmanian LegCo - Ms Goodwin won the 2009 by-election. In this eight-way contest she collected 38.55% of the primary vote - more than three times her nearest competitor (and the only other party affiliated candidate, who stood for the Greens).

Ms Goodwin faces two contestants on Saturday - the Greens candidate she beat in 2009 and Allison Ritchie, who is running again. General opinion supports Ms Goodwin, however Ms Ritchie may yet benefit from her 2001** and 2007*** victories to get her message out and is likely to be the main competitor.

With the exception of Ms Ritchie, however, Pembroke has been repeatedly strong performer for Liberal or Liberal-leaning politicians. I suspect this trend is the reason for the socially conservative campaigns run be all candidates (even the Greens candidate suggesting moderation to her own party's proposed abortion bill).

Prediction: Vanessa Goodwin Returned

Montgomery:

Incumbent: Sue Smith (Independent)

Ms Smith actually began as the Councillor for Leven after the 1997 election. She was shifted into the newly formed seat of Montgomery when it was created in the same redistribution as Nelson. She won the seat in 2002 with 51.71% of the primary vote in a 4-way contest - more than twice the votes of her nearest competitor - and again in 2007 where she was unopposed. From 2003 until 2005 she was President of the Commonwealth Parliamentary Association and in 2008 she became President of the Tasmanian Legislative Council.

Normally this would be a classic example of an incumbent that would be nearly impossible to dislodge. However, I am predicting that Ms Smith will not hold her seat because of one specific factor: she is retiring.

Because it is so hard to remove a sitting MLC, an opening like this attracts a lot of speculation and a few more candidates than normal. This will be the main seat to watch, hence I will go into more detail than for Nelson or Pembroke.

Montgomery, like much of Tasmania's northwest coast, is rather conservative. Liberal candidates would be expected to outperform Labor or Greens, especially in the current federal political climate. There are, however, no Labor or Greens candidates in Montgomery, so in the absence of polling it becomes a question of who the "others" are.

Candidate 1: Kevin Morgan (Independent) has past links to Labor. He has made a campaign around the insistence of his independence which is probably necessary to give him any real chance, yet also distracts from getting any policy across. When he does get down to his guiding values (paragraph 13 of this media release provided on his website) he admits:

"I do have Labor values in equality and social justice, I strongly support Education, Health and Law Enforcement and believe that all these systems must be sustainably supported and run efficiently to attain high levels of service to the state and the community."
I suspect he's the right candidate in the wrong seat, and expect him to place last.

Candidate 2: Leonie Hiscutt (Liberal) is the only officially party-endorsed candidate in this race, and playing the exact opposite strategy. Although the prevalence of independents in the LegCo is generally believed to be the result of conservative voters preferring conservative independents to Lib/Nat coalition members (and non-conservative voters supporting non-conservative independents over Labor or the Greens) Ms Hiscutt is flying the Liberal flag high.

This certainly allows her to bring in more political muscle (although there is a $14,000 funding cap for campaigns in Tasmania, so she won't be dipping into a bottomless Liberal Party slushfund). This political muscle is evidenced by her updates on her campaign Facebook page (which, according to ABC is her main election website). Her formulaic short-post-and-a-photo updates are punctuated by such distractions as:
"My baby was playing for the Two Blues today at Penguin for the seconds. They won by one point!"
Or:
"I am just opening the mail and have come across a letter from the Fox Baiting Task force. The letter tells me where the baits have been laid in our area. Would you believe the name of the Operations Manager of the Fox Eradication Branch is Nick Bates!!"
Feigned folksy amateurism, or the result of cost-cutting at every turn? Ms Hiscutt was the Liberal candidate for Braddon at the 2010 Tasmanian Election, and may be missing her $20,000 warchest.

Candidate 3: Ed Vincent (Independent) is a former Liberal candidate, differing from Ms Hiscutt chiefly on the Forestry legislation. Competing for similar voters, Mr Vincent will chip away some of Ms Hiscutt's support; the question is "how much?"

The "forest peace deal" is broadly opposed by Liberal voters in the only polling I am aware of (actually more applicable to Nelson voters). It may be that these people identify as Liberal voters because of their stance on the forestry bill, or they may have adopted this position because it is the Liberal position and they are long-time Liberal supporters. In reality there is probably a bit of both going on. As a result, this polling suggests Mr Vincent is on the losing side of his only point of difference.

Candidate 4: Cheryl Fuller (Independent) is an active and outspoken candidate with two strikes against her. Firstly she is Deputy Mayor in her Local Council. If this sounds like a good start for a politician, I should point out that Ms Ritchie came to power in Pembroke - the lone Labor leaning candidate in a considerable Liberal history - after defeating Liberal Cathy Edwards who was also a Mayor; that Ivan Dean lost his Mayoral role after winning the seat of Launceston; and that the main campaign against Mr Wilkinson in Nelson is that sitting on the LegCo should be his full-time job - the same argument used to thwart Edwards and Dean.

Secondly, Ms Fuller is a moderate, even a centrist, describing her political stance as "somewhere in the middle" and certainly not a conservative. She will probably pick up a lot of the Left-wing vote, though, especially if Mr Morgan drops out in the first round.

Scenarios: I'm going to call Mr Morgan as the first to drop out, with most of his support flowing to Ms Fuller. The three way battle is hard to predict.

1) If Mr Vincent is leading at this point then we will probably see either:
Ms Hiscutt eliminated and (being the most right-wing of the candidates) the preferences going to centre-right candidate Mr Vincent over Ms Fuller
or
Ms Fuller eliminated with preferences going to Mr Vincent (or rather, not going to the Liberals) for a comfortable win.
Outcome: Mr Vincent elected.

2a) If Ms Fuller leads and Ms Hiscutt is eliminated, right wing preferences will flow to centre-right Mr Vincent before moderate Ms Fuller.
Outcome: Mr Vincent elected.

2b) If Ms Fuller leads and Mr Vincent is eliminated, preferences may flow to Ms Hiscutt with similar policies, or away from the far right and towards another Independent (Ms Fuller)(it seems Tasmanians like Independents in their independent house of review...)

Assuming a roughly even split, Ms Fuller would maintain her lead. In reality a pro-Hiscutt skew would be expected.
Outcome: Tossup.

3) If Ms Hiscutt leads, it is probably at the expense of Mr Vincent. Vincent voters would split preferences as above: perhaps 50-50, but probably towards Hiscutt.
Outcome: Ms Hiscutt elected.

Of these, the general feeling I'm getting is Scenario 3 as the most likely. However, while I will predict Ms Hiscutt elected as a best guess, I am not prepared to stake anything on this.

Prediction: Tossup.

Conclusion:

Although the Legislative Council's low news rating means little polling and rough approximations in our predictions, I'm suggesting:

  • Jim Wilkinson will retain Nelson, and
  • Vanessa Goodwin will retain Pembroke for the Liberals

I'm also making some informal predictions, not counted in my record tallies (for the reasons mentioned in parentheses):
  • Leonie Hiscutt to win Montgomery for the Liberals (tentative)
  • Cheryl Fuller to place second in Montgomery (tentative, prediction has no practical value)
  • Kevin Morgan to place last in Montgomery (prediction has no practical value)
  • Allison Ritchie to place second in Pembroke (prediction has no practical value)
  • Tom Baxter to place second in Nelson (prediction has no practical value)

While it may seem a little cheap, and even pointless to call the perhaps obvious seats of Nelson and Pembroke while shying away from the tough case of Montgomery, there really isn't enough data to judge this case and I have been burnt too many times by my hesitation to use the 'tossup' category. However, I have still made my best guess, which is all you would have got out of me anyhow. I'm just stating that I'm not confident enough to risk reducing my success rate on it.

Check back next week for a review of these predictions!

* Queensland only has one house in its parliament
** 2-Way contest, won with 57.09% of the vote
*** 6-way contest, won from 42.87% of the primary vote