Wednesday, July 2, 2014

Victorian Liberals: Going, Going ...

This is an overview of the state of play in the leadup to the Victorian election, prompted by recent polling.  I expect to make more detailed modelling attempts much closer to that poll.  In summary, this article argues that the Baillieu-turned-Napthine government's first-term status provides no argument against the reasons why it could lose.  While the defeat of this government is not yet a done deal, psephologically, it has at least one foot in the grave.  There is still time to climb out, but it won't be easy.

One of the themes I've been following on this site is the way in which federal politics contaminates state voting intention.  As noted in Is Campbell Newman Actually In Trouble? life gets much harder for state governments when they have a party of their own persuasion in power in Canberra, especially if the federal party isn't polling well.  This is also well covered in Peter Brent's current piece, Dog Days For State Conservatives.  Of the four conservative state governments elected while Labor was in office federally, all are showing signs of wear and tear.  We still have astonishingly sparse state polling from Western Australia since Colin Barnett polled a very poor personal result, but it wouldn't be at all surprising to see Labor in the lead there soon.  The NSW polling picture is still re-sorting itself after the shock loss of a Premier, and in Queensland the LNP have only a narrow lead (Queensland polling updates added) and are still struggling greatly to build their vote share to the level needed to make their unpopular Premier's own seat safe.

That leaves Victoria, which goes to the polls on November 29, unless an election is forced sooner - which doesn't seem as likely as some think.  The Liberals achieved a  big upset in Victoria in 2010 when they pipped the incumbent Brumby government at the post with 45 seats to 43, despite trailing badly in the polls until quite close to the election.  However their initial wafer-thin majority hasn't served them well, and largely as a result of the behaviour of a single rogue MP, they've now lost their majority, a Premier, a Speaker and a good percentage of the plot.

Some might say that it's really not their fault, and invoke the Abbottesque defence "Shaw happens".  However, the Liberals are responsible for renegade Geoff Shaw in the sense that they preselected him.  While Shaw's past may not have given strong indications of how much trouble he could be, due diligence should have turned up that he was at least a little fervent in his social-issue views, and might also have found evidence of some of his inglorious past.  It is not as if Shaw was a no-hoper candidate who got lucky on an electorate with a huge margin either: Frankston was a plum marginal with a buffer of just 2.7% before the last state election.

First Term Status Is Not Relevant

The usual template in forecasting an election like the upcoming one is that first-term state governments rarely lose.  To quantify that, in the last sixty years, 25 (78%) of 32 Australian state governments seeking re-election for the first time have continued in office.  However, if the seven defeated administrations are considered, it turns out that four of them went to the election as a minority government (like this one) and another went to the election on its second Premier (ditto).  Of course, not all governments facing the people for the first time in minority are defeated, as the massive re-election victories of Steve Bracks, Colin Barnett and Mike Rann showed.  But those were governments that enjoyed reasonably orderly relations with their crossbenches, and that also had the benefit of being of the opposite party to the party of the federal government of the time.  Moreover, none had been so careless as to lose their own majorities in office.

In fact, six of the seven first-term state governments to have lost in the last sixty years were of the same party as the party of federal government at the time.  The sole exception to this was the Walsh/Dunstan Labor government in South Australia, which after changing Premiers mid-term lost with 53.2% of the 2PP vote in 1968 as a consequence of malapportionment.  Moreover, at that time voters were hardly inclined to lash out at Canberra with new PM John Gorton immensely personally popular and his government with 10-14 point leads in the polls.

Thus, even if one knew nothing about the polling history of Victoria's current government, one could say that if ever a government should buck the historical trend of first-term governments getting back in, it is this one.   I believe the above points have shown that there is not just sufficient but overwhelming reason to discard the government's first-term status as any predictor of anything at all.  It might even be that based on past election results alone, the government should be expected to lose, but I'm a bit wary of overfitting the past trends based on a small number of data points here.  Let's just say there is more than enough reason to set aside the good record of first-term governments, and look at the polls.

Polling History And Recent Polling

Thanks especially to a massive data-dump of monthly goodies by Essential, it has now become possible to aggregate state polling data for Victoria.  This has been done by William Bowe. The new Baillieu government enjoyed a honeymoon bounce but lost the lead about 18 months into its term.  Labor enjoyed 55:45 leads in late 2012 and some slight improvement did not prevent the removal of Ted Baillieu in early 2013.  There was a honeymoon bounce for Denis Napthine, which propelled the Liberals back into a narrow lead, but it didn't last long at all, and since the middle of last year the government's polling has been increasingly dicey.

The last two polls show especially large leads for Labor.  Probably the last ever Nielsen of Victorian state intention has Labor ahead 56:44 (59:41 by respondent-allocated preferences) while Newspoll has 54:46. Note that the Newspoll is taken over two months, while the Nielsen was only over a few days. I agree with William's aggregate that the average of those two (55:45) is pretty much where it's at, with a little extra for the Nielsen being the more recent sample.

We are well used to "miracle" recoveries by incumbent federal governments from this sort of polling only five months out but state governments are not usually as resilient to bad polling as federal governments.  There is only one recent precedent I can find: the Gallup Government in WA trailed 56:44 in the last Newspoll quarterly in 2004, and was doing not much better in other polls, only to win re-election with barely a scratch in February 2005.  Again, it had the benefit of having an opposing party in power federally.  Also, whatever the cause of that government's loss in support, it seems to have been a brief backlash, rather than systemic over much of the government's term.

Leadership ratings show that Denis Napthine is now a mildly popular Premier, with a Newspoll net rating of +9 (albeit down eight on the last quarterly) and Nielsen scoring him at +4.  Opposition Leader Daniel Andrews weighs in at solid but far from meaningful scores of +5 and -2, in both cases with the high neutral rating that ALP Opposition Leaders are especially prone to attract.  For once the beauty contest scores actually might tell some sort of story, because Napthine leads Newspoll's Better Premier contest 42:29 (quite a big lead given that his party is trailing) while Nielsen's Preferred Premier question has Napthine up by a solitary point, 41:40.  Probably the more ALP-leaning nature of the Nielsen sample is the main culprit for this difference.  However, it could also be that voters consider Napthine to be a good leader of a bad government: some may think he is better at being Premier than Andrews would be, but that they would prefer Andrews to be Premier because they want Labor in office.  (Napthine has rated quite well, given that his elevation was widely seen as a desperate move given his previously failed political life as Opposition Leader a decade ago. I add a personal note that I have found Napthine's refusal to allow Geoff Shaw to influence abortion policy commendable.)

The Coalition has also been polling truly terrible federal results in Victoria lately.  While the current Bludger Track reading of nearly 60:40 is exaggerated by a rogue Nielsen subsample, it's likely the real level is still in the high 50s and that a federal election "held now" would see a large swing in the state.

The Lie Of The Land

Recently the South Australian election made most of us psephologists look a little bit clueless, and the betting markets look incredibly clueless, by returning a government that had been trailing by similar margins in the leadup to the poll and was largely written off.  But the Weatherill government was playing on a very favourable deck because of the electoral geography of South Australia, so that a 47% 2PP was enough to give it a chance; as it happens they got that and got lucky with the swing distribution.  In contrast, in Victoria, 55:45 to the Opposition actually means what it sounds like: that the government must recover serious ground or it will be very badly beaten.

Following a major redistribution that has made things decidedly messy, the Liberals will go to the election as notional occupants of 48 of the 88 seats.  However five of those will have Labor incumbents, meaning that Labor gets much of the benefit of incumbency (a matter complicated by it not always applying to the whole of the electorate).  A new electoral pendulum can be seen on Antony Green's site (scroll down a bit over halfway).

The classical "uniform swing" read of the pendulum gives Labor the seats it needs on any swing above 0.9% (from the 48.4% it polled at the last election).  However, the surrounding seats on that pendulum are quite messed up by the redistribution.  The Liberals have a double sophomore effect in both Carrum (0.3%) and Bentleigh (0.9%), but they also notionally own seats that are occupied by ALP incumbents.  One of these is a seat where Labor has a first-term member and hence a sitting member advantage.  So despite the theoretical three-seat gain to the government from the redistribution, at least one of those gains isn't real.  Frankston, which could be especially difficult for the government to recover, becomes ultra-marginal, and there is also a slight concentration of very close seats on the Government's side.  This post is intended as a general overview post rather than the thorough modelling attempt that will follow (by others if not by me) but there is nothing in the arrangement of seats and margins that will give either side a great advantage.  It looks possible even that Labor could win with slightly less than 50% 2PP.

The Greens were considered likely to win seats in the inner city at the last election but failed miserably to grab even one, in part because of Baillieu's decision to preference Labor over them (which in my view contributed to Baillieu's victory, as it was seen as a bold and principled move.)  This time around it will be interesting to see whether the growing vote of Adam Bandt in Melbourne has any spill-over effect into the state election in the seats his electorate covers.

Is Andrews A Problem? Who?

Probably the most widespread reservation among the political classes about a Labor victory involves nonplussed views of the ability of the Opposition Leader, Daniel Andrews.  There is a view about that he is deeply mediocre, and a thin CV (uni - electorate officer - party apparatchik - politician) that screams out "Labor hack" does nothing to deter that perception.  A detailed profile in The Age may have put some flesh on the mystery personality but he will probably still be "Daniel who?" for a year or so even if he wins.

Concerns about Andrews' strategic judgement arise from his handling of the Shaw crisis.  He seemed quite out of his depth in suggesting that he and Napthine meet with the Governor to discuss the situation (which was simply a matter for the Parliament).  Labor's tactics of trying to have Geoff Shaw expelled have also raised some eyebrows, as serious as Shaw's misdeeds are, they are far from unique and the expulsion of a member not even found guilty in a court of law is something best relegated to the history books.  My suspicion though is that the way posters on political websites see Andrews has little to do with how he is perceived more broadly, and there is nothing in Andrews' polling figures to suggest voters think that he is hopeless.

What is conspicuous in Andrews' case is just how long the "meh" factor has lasted.  While the uncommitted score in his leadership ratings dropped to a "mere" 28 in the last Newspoll, in the one before that it was 35.  While "uncommitted" ratings for state opposition leaders are generally high at the moment (possibly signalling a declining voter attention level to state politics in general) Andrews has sustained these kinds of ratings for longer than any other Opposition Leader in the country in the last 30 years bar one.  The one is Peter Collins (Liberal, NSW) who still had an uncommitted score of 35 after three and a half years in the job (at which point he was rolled without facing an election, which ended up doing his party no good at all.)

How Might The Government Survive?

As my comments above have made clear, I think that based on the objective polling and historical electoral evidence, the Napthine Government's survival chances can be considered rather slim.  But there are always factors that are not captured in the polls at any given moment, so that objective evidence doesn't necessarily tell the whole story.  With several of the trade unions at Labor's base currently being investigated by a Royal Commission, there is always the possibility of a breaking scandal affecting any Labor opposition at any time.  Thus far though, Andrews hasn't been effectively tarnished, despite persistent attempts to tie him to the CFMEU.

Another possibility is that voters who see the state government as a chaotic mess might suddenly decide they don't care so much about that; after all, there will be no Geoff Shaw in the next Parliament, and some party or other will probably have a majority in it.  But for that to make a difference the government would need to manage some combination of running on its own record successfully, and running a scare campaign against the ALP to suggest it is not ready to resume government yet.  The current tactic is to claim that Andrews stands for nothing and has no ideas, but I have actually never yet seen that one work, since Oppositions always release some kind of policy platform closer to the election.  Also while I haven't closely examined the Victorian government's policy delivery performance (particularly on public transport, seen as one of the causes of its predecessor's doom) I doubt there are really light years in difference between public perceptions of the performance of this government and the one before it.

That is not to say the government is hated.  A recent Essential poll on trust in government showed that Victorians have a high level of trust in their government compared to Queenslanders.  But maybe that's just saying that the government is not competent enough to be feared.

Going into the 2010 Victorian election, it didn't seem to many observers (Peter Brent an exception) that the Brumby government - old and tired as it was - had done enough wrong to even make it a close contest.  But I think at that time the power of federal-government drag on state election performances was something that had not received nearly enough attention.  Now we know that while Geoff Shaw may or may not still be a real problem for Napthine come election day, Tony Abbott almost certainly will.

Further Reading: Antony Green has a much more thorough roundup of the Liberals' marginal seat issues at this election.

For more on federal-state government drag see my slightly later article What Kills State Governments: Age Or Canberra, which provides considerably stronger evidence that the Napthine government is in a position from which survival shouldn't be expected.


Sidebar poll: there is a poll on the sidebar to get reader feedback on the format of federal polling articles, and the results will also influence how I cover state polling for states other than Tasmania.  

Til now I have preferred to avoid writing lots of little short federal polling posts without much meaning that just relay current polling results and a bit of interpretation, and tried to go for more substantive theme-based coverage.  The problem is that I can't write an article that is actually about something significant based on every week of polling, and the dribble of updates bugs some people who don't like to scroll down.  Also, perhaps the adding of updates creates the impression that there is not much going on on the site when in fact articles are being frequently updated (as seen in the updated article list on the sidebar).  I don't want this site to become something that is basically a copy of the Poll Bludger format but with fewer commenters, and I would get bored silly writing endless mini-articles entitled "This Week's Polls", but I do acknowledge that frequently scrolling past stuff you've already read can get tedious.  Hence the poll, which will run for up to two weeks, but may be "called" early if the result is lopsided.  If this interests you, please vote.

No comments:

Post a Comment

The comment system is unreliable. If you cannot submit comments you can email me a comment (via email link in profile) - email must be entitled: Comment for publication, followed by the name of the article you wish to comment on. Comments are accepted in full or not at all. Comments will be published under the name the email is sent from unless an alias is clearly requested and stated. If you submit a comment which is not accepted within a few days you can also email me and I will check if it has been received.