Possums Pollytics

Politics, elections and piffle plinking

ALP preference deals: The Greens vs Family First.

Posted by Possum Comitatus on January 28, 2008

In the last post we had a quick look at the interesting relationship between the two party preferred swing to the ALP and the Greens primary vote. The most interesting feature of that relationship was how, on average, a higher Greens primary vote correlated with a smaller ALP swing.

But what is also interesting is that the higher the level of the Greens primary vote, on average, the higher is the size of ALP two party preferred vote. This combination of realities might sound somewhat paradoxical, but they finally put some numerical reality to the seemingly endless question of which of the two minor parties the ALP should make preference deals with, especially if they seek to maximise their TPP swings.

Firstly let’s recap the relationship between the Greens primary vote and both the ALP two party preferred vote and the ALP two party preferred swing.

alptppswingvsgreensprim.jpg alptppvsgreensprim.jpg

What we’ve got here is a scatter plot where the Greens primary vote at the election is plotted against either the ALP two party preferred vote or the swing to the ALP in two party preferred terms for all 150 electorates.

As we can see, on average, as the Greens vote gets higher, so too does the ALP TPP. However, as we can also see, the seats which had a large TPP movement to the ALP at the election also tended, on average, to have a lower Greens vote – similarly the seats which had the smallest ALP swings tended to have a high Greens vote.

Now let’s compare this to the relationship between the Family First primary vote and the ALP swing by electorate.


Here we find the opposite occurred – the bigger the ALP swing, on average, the bigger the Family First vote.

There’s lots of complicated demographics behind these relationships, but one of the key pre-election arguments this data helps to resolve is whether the ALP were justified in trying to make preference deals with Family First at the expense of making similar deals with the Greens- especially from a purely hardball perspective of maximising the ALP TPP vote in the seats that mattered.

Another question of course remains; “Is that political hardball perspective always the right prism to view preference deals through “?

I’d be interested in your thoughts on these.

To add a further bit of raw data to assist the impending debate, if we make a scatter plot that combines both the Greens and the FFP primary votes against the ALP swing to put the whole lot in perspective we get:


Which is probably the most important diagram of the lot.

And finally, to add some fuel to the argument of whether the ALP should make a preference deal with the Greens or the FFP at the next election, we can plot the Greens and FFP primary votes against the current ALP TPP results:


Another piece of food for thought is the extent to which either minor party’s voters would actually follow the HTV cards should such a deal be made. How many Greens voters would not preference the ALP above the Coalition if the ALP were to do a preference deal with FFP, and similarly, how many FFP voters would actually preference the ALP when they ordinarily would not – simply on the basis of the FFP HTV card saying they should?

On another note, Simon Jackman had a squiz at the informal vote as well. Simon wrote for the Bulletin over the election – and now the Bulletin is dead. Personally, I’m blaming Simon for it! :mrgreen:

AddThis Social Bookmark Button

add to kwoff


20 Responses to “ALP preference deals: The Greens vs Family First.”

  1. kwoff.com said

    ALP preference deals: The Greens vs Family First. « Possums Pollytics

    In the last post we had a quick look at the interesting relationship between the two party preferred swing to the ALP and the Greens primary vote. The most interesting feature of that relationship was how, on average, a higher Greens primary vote corre…

  2. Zaf said

    My completely uninformed first reaction:

    Family First voters seem to be concentrated in the same electorates where the aspirational ‘Labor Howard supporters’ who became ‘Liberal Rudd supporters’ can be found in numbers. It would be interesting to see whether this is true – were these the specific seats that went from Labor to Liberal, bringing Howard to power, and then the other way, bringing in Rudd? Is there a demographic overlap between aspirational voters and family first supporters – young families worried about bills and bringing up their kids but finding different political/cultural answers to common questions/concerns?

    Are Green voters concentrated in the rusted on Left leaning suburbs? (Except for Wentworth, which has now gotten interesting, this ‘sounds’ right.) In which case there was little swing because the voters were already ‘there’, in terms of whom they were likely to vote for.

    To acquire more seats, Labor might well try and cut more preference deals with Family First. But the question is, how will these deals play in the ‘progressive’ (??) suburbs which already have a high Green vote – will it move some votes from Labor to the Greens, and will this movement be enough to lose the seat for Labor and gain it for the Greens?

    Even if the number seats gained and lost to Labor in this manner were the same (ie, balanced out), would Labor prefer to deal with more Greens in parliament (which would allow them to take measures which would be placate their own left wing vote while maintaining their street cred with business ‘we HAD to, the Greens made us do it, but we’re pro business, really’) or with more right leaning independents, small parties or Coalition members in Parliament?

    My guess is that Labor would prefer to claim the large middle ground – which would include swapping preferences with Family First – thereby marginalising the Green and Conservative ‘rumps’.

  3. dawson said

    Poss, I’m touched. Thank you.

  4. Don said

    (Poss) plot the Greens and FFP primary votes against the current ALP TPP results:

    So far as I can see, the FFP primary vote seems to be unrelated to the ALP TPP, but there is a slight positive correlation between the Greens primary vote and the ALP TPP.

  5. Possum Comitatus said

    No worries Dawson, it was a great set of questions.

  6. David Richards said

    Stuff the FFP – 2% max compared to 7-8% Greens – more bang for your dead deer skin by doing a deal with the Greens and let the godbotherers go jump.

  7. barney said

    I’m confused. I am a non mathematical person so your scatter sheets fall somewhere between “interesting” and gobbledeygook for me.

    The link between Higher green vote and Higher ALP 2pp makes sense given the overwhelming majority of Green preferences go to the ALP . Why doesn’t it follow that this would result in a higher swing? That it doesn’t is counter-intuitive. Given the low base 2pp that the ALP was on prior to November, an increased 2pp can only come about via an increased swing. Or have i missed something?

  8. Possum Comitatus said

    Those scatter plots just show each of the 150 electorates plotted using the value of the Greens (or Family First) primary vote and the ALP TPP vote (or the ALP swing)for each of those 150 electorates (depending on the chart you are looking at).

    The difference between swings and TPP is a bit counterintuitive. But what it shows is that even though the Greens deliver high preference flows generally to the ALP, when it comes down to a seat by seat basis, the higher the swing that the ALP achieved (on average) in the 150 electorates, the more the ALP picked up non-Greens preferences to deliver them those high swings. So while the ALP improved their margins in a lot of their own seats off the back of Greens preferences, in the seats that counted – the seats held by the Coalition, the bigger the swing was to the ALP in those seats, the less (on average) Greens preferences were responsible for delivering those big swings.

    The best way to look at the graphs that involve ‘ALP swing’, is to see them as an insight into the power of the contribution that Greens and FFP preferences had on those swings. The smaller the level of preferences that delivered the swing to the ALP, the higher more the ALP were responsible for the swing via increasing their primary vote.

  9. smokey said

    I agree with D.Richards. Labor and the Greens now have common interests which leads to mutual benefit in the polls, much more so than Family First. Whether that means specific seats or overall percentages is less relevant than Labor selling itself to the god squad.

  10. caf said

    I think you need to redo the graphs, plotting “ALP TPP Swing per unit of Coalition TPP vote” instead of raw “ALP TPP Swing”. Otherwise all you’re seeing is that seats with high ALP TPP vote (correlated with high Greens primary vote) swung less, which is entirely to be expected.

  11. HarryH said

    my impression is i agree with caf.

    it seems entirely predictable that the swings were bigger in non Labour/Green strongholds this election. Rudd ran a Conservative campaign. That is all fine and dandy when your base are desperate to get rid of Howardism, but if Labour align themselves in any way with Family First in the future they are courting big trouble.

    For the forseeable future(at least 2 elections) it is in Labours interest to keep an alliance through prefs with the Greens.A Labour/Green vote for an incumbent Rudd Government is unbeatable. This is political reality. Of course moral reality should dictate that Labour never even contemplate dealing with Family First but we won’t go there…..

    Bottom line is a Labour/Green vote for an incumbent Rudd Government will be minimum 53% TPP and probably a bit more.

    Don’t forget Fielding will be gone next year which will commit FFP to further irrelevency.

    Anyway this is pretty premature speculation because the Conservative Parties could be very different by next election and diff power plays will come into calculations.

    If the mooted new Lib/Nat party becomes more moderate to challenge Labour there could be a breakoff social conservative party rump.

    who knows

  12. caf said

    HarryH: Even under conditions of uniform swing, it’ll still be higher in seats with a lower TPP base. This is because a swing to the ALP represents some proportion of Coalition voters changing their vote to the ALP – so it will naturally be higher in seats where there are more Coalition voters to begin with.

    By way of illustration, imagine a seat with a TPP of 80% to the ALP. A swing of 5% to the ALP here would represent 1 in 4 Coalition voters changing their vote. Now imagine a seat with a TPP of 40% to the ALP – a swing of 5% here would represent just 1 in 12 Coalition voters changing sides.

    Oh, and Steve Fielding was elected in 2004, so his term runs until 2011.

  13. HarryH said

    yeah my mistake on Fielding.

    i meant next election not next year.

  14. Ron said

    Your q

  15. Ron said

    Possum ,stick to the charts & diagrams. Political hardball & preference negotiations are a separate science.

    The ALP pre election achieved their actual objectives.
    FF is like the prize at a ‘bucks night’ 1 day before a wedding

  16. Rod said

    Time for a new thread, perhaps, but one seat where Greens prefs may have made all the difference, McEwen, is back “in play”.

    THe ALP have announced that they are challenging the McEwen result in the Court of Disputed Returns – http://www.abc.net.au/news/stories/2008/01/29/2149232.htm

    Bailey, who alleged corruption, malfeasance and the like after the initial count gave it to Labor (and demanded a recount and threatened court action otherwise), is now saying that the ALP are bad sports and refusing to “abide by the umpire’s decision”. I reckon those of us who enjoy cryptic crosswords should take note of the way that you can spell “politics” using the letters “H_Y_P_O_C_R_I_S_Y” 😉



  17. Grumps said

    A very interesting set of blogs Possum which I have enjoyed.

    I am not at all up to arguing the merits of candidates^2 or not or the basis of your mathematics (It does hurt my head 😉 ) but you do touch on a point I feel I need to vent about.

    Took the time to read [a href=” http://www.aph.gov.au/house/committee/em/elect04/subs/sub073.pdf”%5DAntony’s%5B/a%5D submission. Suggest a few others take the time to read it and pass it on to our local federal reps. A good submission Antony!!! Bloody horrible way to include a pissy link hope it works.

    Between the two of you, you touch on point about the cynacism with which both major parties treat the electorate, through the electrol act.

    I am one who has used the Langer vote and advocated its use to friends. I watched with dismay as ‘El Rodent’ abused the system by removing the validity of the Langer vote. I fill in the senate below the line and ensure a mistake to cease my preferences after my consideration, but ensure I fill all squares after this point with a consecutive number.

    Up until the changes of the electrol act the Langer vote was considered valid as it indicated first preference and where you ceased allocating preferences. It was shown clearly within the scruitneers hand book on the AEC site. (after ferreting in the AEC site that only a dedicated nerd would bother to carry out)

    Obviously ‘El Rodent’ was concerned that this peice of democracy was to his electrol detriment, so it was made illegal. (tried looking at the AEC site this year to ensure my senate voting system was still valid and the ferreting by the dedicated nerd was greater and the bastards had changed the information to be less user friendly)

    Clearly the system is designed to distort the advantge of the major parties. Your work possum demonstrates this.

    Even more clearly the electorate must be given the chance to vote for whom they want. I least of all want to vote for some god bothering extreme right wing party based in a dance hall on a hill in well to do sydney suburb (or the secret variety). Nor do I wish to vote for an extreme left wing femo marxist drug addled grouping.

    In my view the electrol system is broke. Change is required along the lines of optional preferential If we are all required to vote, (I agree with the legislated requirement to vote). Further a common voting system is required for all federal, state and territory elections.

    Keep up the good work possum.

  18. Possum Comitatus said

    Thanks Grumps – and I’m with you.

    Optional preferential is IMO a far more representative way to conduct democracy. If people do not wish to preference a group they despise, but must on the basis of there being a group they despise even more – they ought to be allowed to stop their preferences where they choose.

    It might make things a little more complicated for some others – but democracy ain’t neat, and I don’t thinks it’s meant to be.

  19. Ron said

    Grumps agree , I suggested those points in an earlier blog
    (to reduce informals) but got zero agreement.

    95% believe the informals problem is NES’s but I contend the problem is more numeracy related than literacy and the solution is to allocate each party a SINGLE 2 digit voting number.
    eg ALP 42

    For both the Federal Reps & Senate and all State Houses , the above the line voting option be mandatory
    requiring a 2 digit number 2 be put……..eg ALP say number 42

    (voting for number 42 automatically allocates prefs per ALP HTV
    in both Reps & Senate and in every house of every State Election

    The result would be for all above the line voting Australiawide:
    nil donkey vote ,
    negligible 1’s taken as ticks being informal
    and negigible overall informals in total
    (and quicker voting for the Public)

    The option of below the line voting can STILL be offered as well

    eg. the 30% approx of Greens and FF etc who currently vote against their Partys HTV preference wise , can still do so

    ..but I suspect education on the above the line easier option would separate the majority of this 30% to the above the line
    option anyway as their prefs are unlikely to be policy driven

    I expect the cultural warriors to be horrified by the above

  20. caf said

    I’ve done what I suggested earlier and scatter-plotted Greens Primary Vote versus ALP TPP Swing normalised as a percentage of the Liberal/National vote before the swing (ie roughly “the percentage of the Liberal/National voters who changed sides this election”). You can see the plot here.

    The negative correlation is still evident – in seats with a higher Greens primary vote, less Coalition supporters swung to the ALP. What’s interesting is some of the outliers – in Throsby (NSW) and Calwell (VIC) more than a quarter of Coalition voters switched to the ALP!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: