The Rudd removal
I can’t claim to have seen the Rudd rolling coming up at all. When indications of a possible move against Kevin Rudd were written up by Dennis Shanahan and Patricia Karvelas in the Australian on Saturday June 19, I wrote this on another forum: “Given that the source is News Ltd, it’s probably a beat-up. If it’s real, it’s madness.”
When the narrative of a spill if the June 21 Newspoll was bad was busted by that poll being in Labor’s favour, despite poor leader ratings, I wrote, among other things, “Today’s Newspoll is 52-48 for Labor so I reckon most of the media hype of the last two weeks is now fit for use only as toilet paper.” Oops. In all this I made one false assumption and one dubious one. The dubious assumption was that a spill would only sanely occur if it was likely to improve Labor’s election chances.
The false one was that any leadership spill would be like others Australians are used to, involving prolonged disunity and faction-fighting followed by a close leadership vote leaving tensions that lasted not only for months but years and that would be certain to scar the next election. Keating’s disposal of Hawke, which took several months and two challenges, and the messy and strategically complex undermining of Turnbull by Abbott, Minchin and Andrews leading to Abbott’s election as leader in an indecisive three-way shambles were among the models I had in mind.
Yet, in this case, within hours of the impending spill becoming clear public knowledge it was obvious that Rudd would be removed, and the spillers were able to cause momentum for his replacement with Julia Gillard to snowball in such a way that Rudd had no serious support.
Furthermore, while it was soon obvious a plot to unseat Rudd had been brewing for weeks, there was no apparent trace of Gillard’s fingerprints, until she said yes to the spill thus more or less immediately sealing Rudd’s fate. Furthermore, for all that has gone on in weeks afterwards, this has remained the case: people may be uneasy about the NSW Right’s role in the removal of Rudd, but very low disapproval ratings show that they do not blame Gillard for accepting the job.
The historic nature of the Rudd dumping has been widely noted. While leadership challenges in Opposition are common and often successful, only two Australian Prime Ministers, Hawke (as mentioned above) and Gorton (rolled by McMahon in the famous tied vote, the third challenge against him) have been removed from office as a result of a formal leadership challenge within their own party. Both were on borrowed time. Hawke was clearly past his use-by date and had broken a promise to hand over the position. Gorton owed his less than stellar prime ministership to an at least partly personal veto on the even less stellar McMahon by Country Party leader John McEwen, and was rolled during his second term, almost the instant McEwen was gone. So really, nothing quite like this has happened before. Menzies avoided a similar fate by resigning his first tenure as PM, but Menzies had only barely won his first election anyway. Lyons would almost certainly have soon been rolled by Menzies had he not died in office. But to find a case where a PM’s own party specifically removed him as leader before he could defend the position one needs to go back to Billy Hughes, who split the party over conscription and was expelled from it but managed to take enough of his mates with him to remain PM. Kevin Rudd had no such luxury.
Would Rudd have won had he remained PM? I for one believe that he probably would have done so. It is true that his leadership ratings were bad (as bad, incidentally, as David Bartlett’s was before the 2010 state election) but Tony Abbott’s weren’t significantly better. All of Hawke, Keating and Howard won elections from dreadful leadership positions (in Keating’s case a net approval rating of –44, and Hawke and Howard both from worse than –30) and furthermore, they often had to recover from well behind on the two-party preferred vote in order to do so. Rudd experienced only three polls worse than 50-50 in his entire prime ministerial career, and was probably tracking at about 50.5-49.5 as an average of all credible polls (adjusted for likely “house effects” in the case of Essential Report and Morgan face-to-face) when he was rolled. Perhaps things would have got worse for him, but I do not believe so. The message that Rudd and his government had saved Australian conditions from the GFC was an easy one to sell and would have blown away all noise about the ETS, the mining tax and border protection with even a semi-competent campaign. The history of Australian elections since the advent of opinion polling is that the only reliable predictor of defeat for an incumbent government is a significant 2PP deficit appearing more or less consistently from well before the election to the final weeks of the campaign. Even governments that are several points behind on 2PP at some stage of their rule tend to survive. All governments go through some rough patches and undecided voters tend to wash in the government’s favour if they perceive a real chance that their votes might actually throw that government out. Bernard Keane in Crikey had some excellent comparative Howard-Rudd first prime ministerial term graphs showing that the trajectories of the two PMs in the polls over the same lengths of time in office were very close to identical. Yes, Howard had a lot more of a buffer to play with; not every sitting PM can lose eighteen seats and stay in office, but at the same time Howard in 1998 was going to the people with a far more risky policy (the GST) than Rudd was going to do. For all the hoopla about big mining industry campaigns against the RSPT, I doubt that at the election something so esoteric and removed from everyday quality of life would really have bothered all that many voters.
The case that Rudd was going to lose, concerned Abbott catching Rudd in the preferred PM stakes according to Dennis Shanahan (in part), and “toxic” internal polling which was semi-leaked to Andrew Bolt, according to the rollers. To dispose of the first one first, although Abbott was closing the gap this established only that he had finally reached the point where his own side thought he was OK. In any case, if the preferred-prime-minister gap is big enough to predict the result of the election, there will be bigger fires burning elsewhere. Hewson led Keating as preferred PM in 1993 and lost. Keating led Howard as preferred PM in 1996 and was thrashed. Beazley led Howard as preferred PM in 1998 and lost. All these things happened in the final Newspoll of the respective campaigns! Labor’s low primary vote was also cited by Shanahan as a major part of the problem, but it was artificially lower because of a high registered intention to vote for Greens, the Green surge coming in large part from voters who were soft as butter in their voting intention (as shown by Essential Report) and did not embrace Green attitudes on major issues anyway.
The internal polling (see Bolt here if you can stomach his irrelevantly sexist reference to Kate Ellis) is supposed to have shown Labor poised for a net loss of eighteen marginals under Rudd’s leadership. But there have, as yet, been no figures released to back Bolt’s summary. Who conducted the survey? When was it conducted? How many electors were surveyed in each electorate, and hence what is the margin of error? What are the numerical equivalents of Bolt’s “big loss”, “narrow loss”, “huge loss” and so on? Without knowing these questions, we cannot know whether the conclusion that Rudd was in real danger of losing that many seats is justified, and the failure of the plotters to leak the full figures suggests the figures might not be as conclusive as they are claimed to be. (That or they were so bad that they feared that leaking them would place Labor in a bad position no matter who the leader!)
Furthermore, the sample focussed on Labor seats within four points of changing hands and Coalition seats within 1.5. But assuming that the 2PP swing against Labor would likely be a point or two at most, if Labor was massively losing so many marginals it is likely there would be swings going all over the place and they would be winning Coalition marginals to compensate. An unlucky distribution is always possible if all the government is doing is buffering its margin in ultra-safe seats (with close to even 2PP figures either party might in theory win by 20 seats or more) and William Bowe in a comment at the Poll Bludger noted that looking at the patterns suggested by the semi-leaked polling, there is a hint that Labor support was strongest in low-income electorates, many of which are rock-solid Labor seats. However, there are very few historical cases at state and federal level (excluding malapportionments) of an incumbent government being diddled out of office by an unlucky distribution of the popular vote. The general pattern is that even governments that marginally lose the popular vote will usually still retain office. Extraordinary claims deserve extraordinary evidence, and a list of electorates with no stated sample size and no stated figures isn’t it. If Labor goes on to defeat, the replacement of Rudd should be recorded as a monumental blunder; if they win comfortably it should be recorded as a success, and if they win by a handful of seats then the verdict of history should be “what on earth was all that about then?”
I don’t propose to add much to the mountains of analysis already available on why Rudd was rolled, and the personality and political defects that got him into a vulnerable position in the first place, but I do want to add my own spin on this.. My own view, in hindsight, was that his autocratic style within the party made him so many enemies and so diminished the payoffs of factional influence, that once the chance of winning without him was seen as not much worse than the chance of winning with him, his troubles started. A factional figure might rationally see an 85% chance of retaining office with a leader who respected and upheld factional processes, even from a different faction, as a better deal than an 90% chance of retaining office with a leader who sidelined the factions and diminished their influence. Thus, to assume that Rudd would not have been rolled had he not been rightly seen by the party as headed for a likely defeat is unwarranted. In the absence of firmer evidence that the SS Rudd was really going down with all hands on deck, I believe that Rudd was rolled not only out of fear that he would lose, but equally as much (by the primary instigators) out of fear that he would win, and further cement his presidential style as superior to the factional system that represents the traditional career path within the ALP. The plotters did not have to convince the whole party Rudd was on the way to defeat, either. They only had to reach the point where it was clear that a challenge was likely and would severely injure Rudd’s leadership before the momentum for quick change to avoid prolonged bloodshed became unstoppable – a clean switch to Gillard was much better than months of uncertainty.
(I note in passing that, while I find the case for removing Rudd to have been rather threadbare, the hackneyed Letters to the Editor complaining that we the people elected Rudd and not Gillard are more than a little bit tedious. Despite all attempts by parties and their agencies to turn Australian elections into Presidential-style contests, it still remains the case that the voters elect local members, who are members of parties, and those parties go into elections under particular leaders, but there is no guarantee that they will keep them. Voters elected the Labor Party to govern, not the Rudd Party, and the Labor Party has decided now to be led by someone else and to go to the election with that leader. Deal with it, or vote for someone else.)
Gillard vs Abbott: Battle of the unelectables?
Enough of Rudd – what of the chances of the Labor Party under Gillard? Initial polls have mostly shown a substantial bounce to Labor, equivalent to about two points of 2PP, with a single Morgan phone poll with a sample size of only 600 voters the lone exception to the trend. Comfortingly for Labor, the spike in the polled Greens vote has returned to the Government, which may seem a mystery given that Gillard’s initial statements on gaining office were that she would try to build consensus on climate change before acting and that she sympathised with those who were afraid when they saw pictures of boats on their TVs. Politically naïve as some Greens voters may have been about the rise of Gillard, I believe the real basis for this change is that these “Greens voters” were never real Greens supporters in the first place, but were actually Labor voters who could not yet bring themselves to support the Coalition, but believed that Labor were losing the plot under Rudd. They were telling the pollster they’d vote Green as a kind of glorified “dunno”, but at the election would never have actually done so.
The “honeymoon effect” – that every new leader of a party appears to be a genius for a time – is well enough understood that I hope no-one needs to be told not to take the 2PP surge all that seriously yet. If an election is called very soon, it is really the polls once the campaign is on that will matter the most, and if it is postponed til later than August then we will have time to see whether the initial bounce wears off. All the polling so far shows is that the switch was neither an immediate disaster, nor an immediate overwhelming success, and that it has a fairly good chance of working. What is more concerning for the Coalition is that Gillard herself appears to be viewed well, especially within Victoria, and leads Abbott heavily on perceptions of leadership attributes – in a way that Rudd did not (Essential Report showed Rudd somewhat ahead on most positive attributes, but around as bad on most negatives, and being seen as more complacent, superficial and out of touch than the current Leader of the Opposition might be said to take some doing.)
The personality contest between Gillard and Abbott will be fascinating. Both represent radical choices for their parties. Both have attracted the label “unelectable” a fair few times (Abbott far more often than Gillard) and both are a long way from the political centre, though Gillard will probably do a better job than Abbott of disguising this. Abbott’s negatives are that his Catholicism, matched with a shaggy-dog moral life story, his several remarks endorsing outdated social attitudes and his known track record of imposing his beliefs on public policy as Health Minister, all make him easily seen as both sanctimonious and hypocritical. Indeed, only one-third of voters consider him trustworthy. There is also the difficulty of shaking off the perception that he is Howard’s unrepentant headkicker, even more extreme than the former Prime Minister’s discredited final-term government, and this is a particularly telling problem for him in the area of industrial relations. Although it has recently been revealed that Abbott voted against Workchoices, his gaffe-prone nature has made him an easy target for negative-campaigning soundbites suggesting that he would reintroduce it.
Gillard is a radical choice mainly because it is enough of a political culture shock for a sleepy old-fashioned polity like Australia to finally get its first female Prime Minister, without that Prime Minister also being a childless atheist who has never been married. So far Australians are showing a refreshing indifference to this personal story, especially given that it lacks an obvious influence on her policy views (something that cannot be so readily said for Abbott) but if things really get nasty during the election campaign it will be interesting to see what voters (and some Opposition politicians) really consider to be off-limits.
Attempts to tie Gillard to a socialist past don’t appear to concern anyone in the slightest (largely because the claims of radicalism at the time really don’t amount to all that much under scrutiny), and Gillard’s Left faction membership does not predict her ideology reliably. However, I have little doubt that Gillard is at heart a regulator and a unionist who will be pragmatic for the good of the party but has no ideological sympathy with free-market type positions. She is an excellent negotiator and debater, but not a consistently good communicator of policy, and her policy formulation record over the last few elections is actually very patchy. As I write, Gillard seems to have adeptly defused the mining tax debacle, leaving Abbott whinging in the wilderness by continuing to oppose a tax that the industry has largely accepted, but her asylum seeker solution seems sketchy at best and a potential target for criticism of her policy ability, to add to the BER blowout and, for those with long memories, the failure of Medicare Gold. Whether all that will matter, or whether the election will just turn into a presidential-style referendum on the new PM’s likeability vs the Opposition Leader’s dodginess, remains to be determined.
With both politicians having been seen as risky electoral goods in the past, a contest between them is like the proverbial irresistible cannonball hitting the immovable post: something’s got to give. There is a view about that Australia has suddenly grown up, that Gillard’s perceived personal risk factors don’t have any traction and that Gillard is some kind of “red-headed Hawke” against whom Abbott simply can’t compete and that the Coalition under Abbott is wrongfooted and will now be horribly mutilated. Certainly, Gillard has a popular accessibility that the PMs since Hawke (and most of the Opposition Leaders as well) have all lacked, and there are parallels with the sudden (and successful) rise of Hawke to the Labor leadership in the shadow of an election. But while it is possible that Labor will enjoy a very strong victory and even increase their majority, they will need to write some history to get there.
Since Federation, there have been ten cases prior to 2007 in which a new government has been formed by winning a majority of seats at an election from Opposition (as opposed to a new government being formed on the floor of the House between elections, a la Curtin’s Labor government in 1941). In the early days, new majority governments had a torrid time of it – three of the first four were defeated at the next election and the other one didn’t even survive its full first term before collapsing. But since the UAP/Country Party coalition under Joseph Lyons retained its majority in 1934, things have settled down a lot and the last six new majority governments have all retained office when they sought re-election. All of them have done so with a reduced majority, the largest losses (coalitions under Lyons in 1934 and Howard in 1998) each being about a tenth of the seats in the parliament at the time. However, both those governments had a lot of seats to lose.
Whatever Abbott’s faults as Opposition Leader, the leaders faced by the last six PMs seeking to renew their government’s majorities were no world-beaters either. Lyons, Menzies and Fraser even had the luxury of return bouts against ex-PMs they’d already discredited once. If Labor is to make electoral history by increasing their majority at this election then rolling their own PM just months out from the election would be a very very strange way to be doing it. In the absence of an overwhelming backlash against the Gillard elevation, at the moment I suspect that Labor will follow the historical pattern and be returned but with a slightly reduced majority. However, as the sudden change of leaders has created a situation unlike any we have ever seen before, the margin of error on such a prediction is a large one – for the moment. Anything could happen, but probably won’t.
Tasmanian seats
At this stage I think it is likely that Labor will retain all five Tasmanian lower house seats.
Bass
Although Bass is theoretically very marginal and has had a lot of attention for this reason, four different polls (EMRS, Roy Morgan and both Labor and Liberal internal) have been reported, all showing Labor retaining the seat, at least three with an increased majority. While none of these polls are individually statistically up to scratch (and indeed the sample size of the Labor internal poll and its figures remain unknown) what I believe this shows is that the Bass result in 2007 was artificially close. The sitting member, Michael Ferguson, did a remarkably good job to keep the margin of defeat as low as he did, and a less popular sitting member would have lost the seat by three or four points on preferences. The polling is showing that the Liberals have their work cut out to really give Labor a scare in this one, and Labor is fortunate that its somewhat disaster prone MHR Jodie Campbell is not recontesting, thus giving the party a clean slate.
Labor Senator Helen Polley got a big freebie at the expense of Liberal incompetence when she quite correctly
smacked them down for seeking to make capital out of Bass internal polling that actually showed the Libs only level on primaries, from which position the Libs would be badly beaten given the high Green vote in the electorate. The curious thing was that the Liberals had already made this mistake before, gloating about a three-point primary lead in the EMRS Bass poll, when that would also have resulted in a swing away from them, as I pointed out in a previous column. The Libs seem to have forgotten that they were 6.3 points up on primaries in 2007 and still lost the seat on preferences.
Braddon
There has been not much attention on the contest in this seat so far. During the RSPT debate it was considered vulnerable because of mining interests in the seat, especially on the West Coast, and the Liberals released internal polling claiming they were seven points ahead (presumably on primaries) in the seat. I did not see any mention of the sample size of that polling and suspect that it was small. If it was 200 votes (the standard sample size for Chickenfeed grade electorate sampling in Australia, it seems) then the margin of error is the same as the claimed lead and the poll is therefore inconclusive, even as a guide to sentiment at the time that it was taken. I think this seat is still shaky for Labor despite the resolution of the mining tax debacle, and should be watched with interest as more polling becomes available.
Lyons
Another federal election, another round of nonsense about Lyons being at risk of falling to the Liberals. This was another seat where the Liberals claimed to be ahead on the basis of internal polling supposedly showing Dick Adams three points behind (on primaries, presumably), but again small sample size must be suspected, and in any case from such a position he would just win on preferences anyway. Labor has almost a 9% margin in this seat and although that was inflated by Liberal candidate disarray in 2007, this one isn’t going anywhere while Dick Adams remains the MHR. After he goes, it might get interesting someday.
Franklin
The Liberals didn’t even let their internal polling for Franklin get “leaked” to my knowledge, but you can bet it shows them losing it. Although Franklin became a marginal at the last election, this was mainly because Labor went from a long-term MHR to a farcically messy replacement process, and were up against a strong Liberal candidate in an electorate slightly cheesed off by all the attention on the northern contests. While Julie Collins did recently make an incredible fool of herself in response to a simple question about the point of the Rudd version of the mining tax, her opponent, Jane Howlett, is not only off her usual political patch but no world-beater anyway, having now twice failed to be elected to State Parliament. That said, her second attempt was not bad, but I see nothing to suggest the Libs will threaten in this seat.
There has been idle speculation the Greens might threaten, coming off a strong result in the 2010 state election in which they polled 27% in the electorate. It won’t happen and there’s no real chance of them even coming second in this one. Yes, they did poll 27 in the state poll, but federal Greens votes in Tasmania tend to lag behind the state Greens votes by a few to several points, and of that 27, at least four points was a personal vote for Nick McKim that failed to stay in the ticket. Given that the Greens polled a feeble 14.4% in Franklin federally in 2007, they’ll have their work cut out getting all that close to 20 this time around.
Denison
Denison is the most complicated seat at this stage, but it is still overwhelmingly likely to be retained by Labor barring a major candidate malfunction. I strongly recommend Peter Tucker’s article on why the Greens just won’t win Denison and why the “mainland hype” is wrong; with the somewhat late endorsement of a not particularly high profile Greens candidate it doesn’t look like they really think that they can win it either. That said, if you really want low-profile try the Liberal candidate Cameron Simpkins (barely present on the political radar in the past beyond some involvement in Matthew Groom’s outstanding state campaign) or Labor’s anointed heir apparent Jonathan Jackson (who? Oh that’s right, it’s Tasmania, he’s related to a former politician, well, that’s OK then, who needs profile?) The strange thing about Denison now is that by far its highest profile candidate is the Independent Andrew Wilkie, and in all likelihood he’ll only finish fourth.
I actually think that in the very outside chance that there is still an upset in this seat it is more likely to be pulled off by Wilkie than the Greens. If he can only somehow get himself into third position, he is likely to be preferenced by everyone and to make big gains on candidates ahead. However, as I pointed out in comment 8 of one of the Wilkie threads this requires heroic assumptions on primary votes, and preferencing, and still a lot of luck. It would be remarkable if Wilkie could more than double his state vote in the federal campaign, which is what he needs to do to really have even a ghost of a chance.
On his blog here Peter posts an email from a reader suggesting a way in which Wilkie might yet make it: Liberals realising that they cannot possibly win the seat might vote tactically and vote 1 Wilkie on the grounds that that at least creates some chance of someone taking the seat off Labor. But as Peter points out Wilkie will have trouble convincing voters who voted for him in the state election that he is any kind of threat worth voting for in this one (the narrative of a “wasted vote” for a third or fourth party candidate dies hard although in a preferential system it is absolute nonsense.) I also do not think Tasmanian Liberal voters are used to this kind of tactical ploy or likely to vote in this manner (and would they really want a former Howard nemesis as a local MHR just for the sake of annoying Labor anyway? It would turn into a case of “watch what you wish for”.)
One thing that does surprise me is how little effort we are seeing from Labor to build up and promote their replacement for the long-serving Duncan Kerr. After 23 years with one MHR, one would think a transition strategy would involve frequent releases from the new candidate, perhaps an early advertising blitz (OK his photo appeared in Kerr’s electorate newsletter, but not much more), something to ensure that longterm Labor voters feel they know a bit more about the new guy before they start thinking about voting for anybody else. Labor should not be too complacent about the seat just because it is safe, since that is how political accidents can happen. Nonetheless, I too think that Labor’s chance of winning this seat is about 95%.
Senate
A bit too early to say much here until we see which way the national dynamic is going, but suffice to say that at this stage anything other than 3-2-1 (3 Labor) or 2-3-1 (3 Liberal) is very unlikely.
