Dr Kevin Bonham

Labor will almost certainly win this weekend’s federal election, and win it pretty comfortably. The only thing that prevents me from making an absolute call in advance of the commencement of voting is that this has been a rather strange election, and hence the outside possibility of a major last-minute shift of a sort generally not seen in other elections cannot be ruled out categorically. Indeed, the last Galaxy poll points to a close contest, just as the opposing Nielsen one suggests a landslide, but both are probably relative outliers just within their respective margins of error, and the truth is very likely somewhere in the middle.

I will start by dismissing the myth that the task Labor faces – gaining sixteen seats – is especially formidable. Sixteen seats is just over one-tenth of the size of the current parliament. In the last forty federal elections, there have been twelve instances where the Opposition has gained that proportion of the seats or more. In eight cases, the Opposition that did so won government. In two of those cases it did so from further back than Labor is now, and in a third (Bob Hawke in 1983) from about the same position. Furthermore, Labor’s current weak position in the parliament does not represent its baseline popularity over the past several years, but is an artifice of a risky leadership decision, and general campaign strategy, that backfired very badly at the previous election.

Secondly there is the myth that, as Howard himself put it in his curious attempt to play psephologist, the historic conditions for a transfer of government do not exist. Transfers of government are relatively rare events in modern Australia, and to attempt to extrapolate from the four that have occurred to any universal statement that there are no other ways it can happen is a mug’s game. Indeed, downplaying the prospect that the voters would throw him out for largely non-economic reasons may well have driven another nail into Howard’s coffin. The 1993 and 1996 elections should have shredded the myth that economy is everything once and for all in any case. With the economy in tatters, the Keating government survived in 1993 by running a successful scare campaign against the Opposition’s alternative proposals, only to lose in 1996 primarily over issues of mistrust and ideology when the economy was already in robust recovery. There are many echoes of Keating’s “L-A-W law tax cuts” failure in the Coalition’s failure to deliver on its completely unnecessary gloat that it could keep interest rates at record lows. The difference is that Keating in 1993 was a desperate man who had little alternative but to buy back the Lodge with money he didn’t have, while in 2004 the Coalition continued rubbing salt into Labor’s wounds even two days out from the election when its man was already beaten. If doing so won them the poisoned chalice of total control of the Senate, it is now they are paying the price.

Whatever other historical arguments one might wish to make about winning being unusually difficult for Labor can all be answered by a universal “but”. That “but” is this: “but unusual things happen in unusual circumstances.” Never before has an Opposition Leader in an increasingly presidential polity been as popular for as long as Kevin Rudd. Never before has a Prime Minister gone into an election under a forced promise to reluctantly retire, offering the nation a handover to a deputy who many voters do not like, and those who like cannot be sure they will get. Lastly, never before has a government managed to place itself at risk simply by being so boring. Labor’s overwhelming advantage on the issue of leadership alone means that precedent is irrelevant and all a psephologist can do is consider the opinion polls, with a note that there is no absolute guarantee that they are right, but no compelling reason to expect otherwise.

The leadership issue alone has made things hard enough for the Coalition, but I also believe it has failed to campaign effectively because it has had nothing to campaign effectively on. Labor’s biggest tactical (as opposed to leadership or policy) failure in 2004 was a dismal inability to play hardball with the Coalition’s constant use of scare campaigns, even though that issue was highlighted in the Hawke/Wran review of the 2001 defeat. In 2007, Labor has been on the ball regarding this tactic, and has consistently pushed the theme of the Coalition’s use of scare campaigns whenever they have appeared. Because the Coalition has found little other than scare campaigns to campaign effectively on, once voters are alerted to the tactic, they can become quite cynical about its constant repetition.

Not only has the exposure of the Coalition’s use of scare campaigns blunted its effectiveness, but also the scare campaigns themselves have been on dubious ground, since all of them are too easily thrown back at their authors. The Coalition has been unable to campaign on interest rates without having its (steadily expanding?) nose rubbed in its failed promise to retain them at record lows, and it has been unable to campaign on union power without Labor implicitly pointing out that with radical workplace “reforms” like WorkChoices being introduced as non-mandated policies, union power might not be entirely a bad thing. In this latter sense, the Coalition’s inability to restrain itself ideologically during its three years of control of both houses has left it in a seriously weakened position. WorkChoices may or may not be as bad as the public debate makes out, but it is a difficult sell and a very easy target. The result is that Labor, which has still failed to appropriately modernise its power base away from the union movement (something I identified as a danger for it at this election three years ago), is getting away with it, and may not have to confront the issue now for several years.

Lacking any other effective basis for its push for re-election, the Coalition has run primarily on its economic record, as indicated by its slogan “Go for Growth”. But even here it has found itself damned if it does and damned if it doesn’t. When the economic debate is broken down to the concerns of everyday people, the picture looks less pretty than when it is expressed in statistical abstractions that omit part of the story. For instance, employment is at very high levels, but some of that is casual employment and underemployment, and even the full-time employed are often struggling with the challenges of home affordability. To relate the economic debate to its impacts on Rudd’s proverbial “working families” would raise the question of why more had not been done for them besides getting them off the dole queues during eleven years in power, so the Coalition has tried to simply stick to the general broad picture. But Barnaby Joyce is right that this has failed it, for in so doing it has wandered off into a feral abacus’s garden in which eleven years of Liberal rule is somehow supposed to have turned bogan battlers from the boondocks into economic policy wonks. Kevin Rudd connects with those people now (even though he is a policy wonk himself); John Howard still has their serious respect, but no longer really does so. The real economic wonks know that John Howard sold the pass on economic restraint many years ago and is primarily a fiscal popularist. It is hard to argue that there is any major difference between the two parties in terms of their likely economic impact, and focussing on the past history of interest rates has only highlighted the extent to which they are driven by conditions of the time not much within an Australian government’s control.

The government has continued harping on the economic management issue because it continues to hold a major lead in polls on the issue – it looks like a strong suit, so they stick to it. The problem is that while people believe the government to be strong on the economy, there is little evidence that it is really affecting their vote. To invert Bill Clinton’s favourite saying, this time, it aint the economy, stupid.

I did expect the government to narrow the gap to Labor in the months leading up to the campaign, and was even quite jittery during May about having declared Labor entitled to favouritism in an article published around that time (http://walleahpress.com.au/FR35Bonham.html) in “The Famous Reporter”. I still believe the Coalition should have reduced the gap some time ago, but they needed to tie the economy to the fate of millions with advertising showing how they would use strong economic circumstances to help individuals. Instead, they have rarely addressed the theme of “going for growth” in a positive and convincing fashion, usually giving it only a token airing before retreating into more criticism of the Opposition (and hiding it entirely when the Reserve Bank nudged up interest rates). This has made the slogan seem hopeful and tentative rather than confident, whereas with Labor, voters have known New Leadership is what they will get – even if it is just a younger, softer, cheekier clone of the incumbent.

Enough of that waffle and onto some specifics. Unlike local and council elections I claim no particular expertise when it comes to national forecasting, so these are just the guesses of another interested observer who is hopefully vaguely politically astute, but could easily be way out.

I do not expect the 2PP vote to close greatly from the nine-point gap currently showing in the polls, especially not with a fraudulent-document prank gone wrong enveloping the Coalition in the final days. I do expect that the Coalition may show resilience in some otherwise losable seats because of the strength of its high-profile performers within them. Many seats may be saved against the swings, but I also suspect that there will be some high-profile casualties anyway. In order, I consider the list of the most plausible outcomes to be: a comfortable Rudd victory, a massive Rudd victory, and a close Rudd victory. Barring a colossal last-minute scandal attacking Labor, Howard’s government can only survive (even in minority) if there is an unknown glitch in opinion polling as it is conducted today, that has only become apparent within the last few years. The “secret Howard voter” who denies they will vote Liberal until they reach the ballot box is one hypothesis along these lines, but there were no signs of this beast in 1998 or 2001, and only the most arguable evidence of it to some modest degree in 2004.

I am currently predicting that Labor will win around 87 seats, but this may yet prove quite conservative (or, less likely, too optimistic for them). Additional to all those it already holds, I currently consider Labor the favourite, whatever Centrebet says otherwise in some cases, in each of the following:

NT: Solomon. TAS: Bass, Braddon. SA: Makin, Kingston, Wakefield, Sturt. VIC: Corangamite, Deakin, McMillan, LaTrobe. WA: Hasluck, NSW: Parramatta, Lindsay, Eden-Monaro, Page, Dobell, Paterson, QLD: Bonner, Moreton, Blair, Petrie, Flynn, Leichhardt, Bowman.

To that I add a net balance of two “wild card” seats not on the list. Of course, Labor will probably not win all of those I have listed, but they may instead win more others I have not. Or they may lose up to three in WA, but win more somewhere else in the country. If the swings are really on in the right places, at least another fifteen seats are vulnerable, perhaps more.

As a starting point for considering the Tasmanian seats in detail, I have shamelessly pooled the last two EMRS polls to create c.400-vote samples for each seat, redistributed the undecided and then applied my own fine-tuning (or not) to attempt to second-guess a few results. It will be interesting to see if my tweaks are in the right direction or are wrong, but they are not very large ones in any case.

Denison

Duncan Kerr will very easily retain this seat.

The 2004 result was Labor 49.5, Liberal 32.6, Green 14.5, others 3.3. (2PP 63.3 to ALP) The pooled EMRS sample result is Labor 50, Liberal 29, Green 20, Other <1 (2PP 67). I predict Labor 50, Liberal 28, Green 19, Others 3 (2PP 67). Franklin

Julie Collins will almost certainly, and probably very easily, inherit this seat.

The 2004 result was Labor 46.4, Liberal 38.7, Green 11.2, Others 3.7 (2PP 57.6 to ALP). The pooled EMRS sample result is Labor 43.6 Liberal 34.6 Green 20.2 Others 1.6 (2PP 62.5). I doubt that the Green vote in Franklin is actually equal to that in Denison or that Vanessa Goodwin is doing quite that badly. I predict Labor 45 Liberal 36 Green 16 Others 3 (2PP 60).

Lyons

Dick Adams will almost certainly, and probably very easily, retain this seat.

The 2004 result was Labor 44.6 Liberal 42.3 Green 10.0 Others 3.1 (2PP 53.7 to ALP). The pooled EMRS sample result is Labor 46.7 Liberal 32 Green 16.3 Others 4.9. (2PP 59.5) However, a specific Lyons-only poll (c.400 voters) conducted to explicitly include Ben Quin has Labor 45, Liberal 29 Greens 13 Quin 8 Others 2 (3 undecided). I think the latter is likely to be more accurate. I predict Labor 46 Liberal 30.5 Green 13.5 Quin 8 Others 2 (2PP 59), although Quin could poll up into double figures. Additionally I make a specific prediction that should Quin be excluded before the Greens, over 65% of Green preferences will go to Dick Adams.

Bass

Jodie Campbell is very likely to, probably fairly comfortably, win this seat – but watch the Green vote, which will be up several points (but not, I think, into the twenties) on account of the Tamar Valley pulp mill.

The 2004 result was Liberal 49.1 Labor 39.2, Green 8.1, Others 3.5 (2PP 52.6 to Liberal). The pooled EMRS sample result is Liberal 39.8 Labor 39.4 Green 18.7 Others 2.1 (2PP 54 to ALP). I predict Liberal 40 Labor 39 Green 17 Others 4 (2PP 54 to ALP).

Braddon

Sid Sidebottom is very likely to recover this seat, probably fairly comfortably.

The 2004 result was Liberal 47.4 Labor 43 Greens 5.6 Others 4 (2PP 51.1 to Libs). The pooled EMRS sample result is Liberal 43.8 Labor 45.8 Green 8.8 Others 1.6 (2PP 53.5 to ALP). I predict Liberal 43 Labor 45 Greens 8 Others 4 (2PP 53 to ALP).

The Senate

Before I start on the Senate I would like to demolish two psephelogical myths I hear over and over again.

The first is that Christine Milne’s win over Family First in 2004 was especially close. It appeared close on election night and the Greens’ claiming of it then was most certainly premature, but subsequent shifts in the counting of postal voting pushed the effective margin for Milne over Petrusma out from what would have been around 1000 votes to what could have been around 6000 had all votes needed to be distributed. This is actually less close than Robert Bell getting in in 1990 (just over 3000 votes), Bell losing to Bob Brown in 1996 (about 2200) and Louise Crossley missing out in a three-cornered contest with Harradine and Labor in 1998 (around 4000).

The second is that very positive polling figures for Labor at House of Representatives level can readily be translated into an easy 3-2-1 for Labor in the Senate. In both 2001 and 2004 the figure (Liberal primary minus Labor primary) was much higher in the Tasmanian Senate than in the Tasmanian House of Reps seats. It was also higher in the two elections before that, although not massively so. In 2004 this gap actually exceeded a Senate quota. So an easy 3-2-1 should not be taken for granted.

The 2004 result was Liberal 46.1 Labor 33.5 Greens 13.3 FF 2.4 Murphy 2.2 Democrats 0.8 others 1.7.

Above I have predicted the primary vote gap between the parties in the House of Representatives to narrow by about 6.5%. For the Senate I predict it will narrow by much more than that. This is firstly because the difference between the Senate and House figures was, in my view, blown out by the Liberals lacking any House of Reps incumbents in 2001 and 2004. Secondly, the Liberal Senate team at this election is rather low-profile and faceless, consisting of a solid and well-regarded but relatively anonymous lead Senator (Colbeck), a new Senator in the second spot and a relatively unknown #3. This compares with the Labor team headed by the very high profile Nick Sherry and actually including a mix of genders.

Let’s say it narrows by 13%, and that the Greens increase their vote by almost as much as I expect them to in the House, and see how we go. Say: Liberal 38.5 Labor 38.5 Greens 17 FF 3 the rest united 3. In this scenario each major party has .7 of a quota left over, the Greens have a spare .19 of a quota, Family First have .21 of a quota and the rest is just debris … or is it? Green preferences would go to Labor, Family First’s would go to the Liberals, and the outcome would then be determined by the preferences of whichever miscellaneous groups had done the best (which is extremely difficult to say, as none of them are psephelogically promising), plus the below-the-line leakage. That is not to say that this is what will happen, but just to illustrate how much ground Labor must recover to be sure to capture a third seat. Two things are apparent: firstly that the Greens poll better in the Senate than the House and do so more or less entirely at the expense of Labor, and secondly that there have been many Tasmanian voters who like their Labor Reps incumbents (perhaps an issue in Lyons and Franklin in particular, though I have not checked) but prefer the Libs in the Senate.

Some polling (based on rather small sample sizes) has suggested that Labor will do better than even on the primary vote, which if so would make a third seat for them extremely likely. A 291-vote Morgan sample in October (a rather small sample size) returned Labor 45.5 Liberal 30.5 Green 20.5 FF 1 Other 2.5, which is a straightforward 3-2-1. It is also a swing of over a quota from last time, and I doubt the vote can swing that much, especially not when the two combined EMRS polls only show a House of Reps swing of 3% to Labor, and the swing is actually stronger in the seats with Liberal incumbents.

With much trepidation and an expectation of being way out, I take a stab in the dark and predict Libs 37.5 Labor 39.5 Greens 17 FF3 the rest united 3 with Labor winning a third seat with a modest margin, but the Liberals still have a serious chance of winning three senators, a far better chance than they have in any Tasmanian House of Reps seat. Some scenarios other than 3-2-1 one way or the other in the Senate are still vaguely conceivable as freak events, but I explicitly exclude a Family First win from that unless they can poll several percent. The factors that helped them get close-ish last time (primarily, Murphy and Labor and micro-party preferences) are no longer apparent.

As for the national Senate picture, I expect the Coalition to narrowly lose its majority, but beyond that I try to avoid playing with sharp objects too much without having tried to model every state in depth. It’s highly likely that Nick Xenephon will be elected despite his rather oddly-behaved SA #2 attacking him in the press this week. I won’t be surprised should the Greens again fail to grab quite the bag of seats that everybody expected them to get last time (I think they will still do rather well, and probably win four or five). Lastly, the possibility of a Christian party gaining a seat may be almost non-existent in Tasmania, but it cannot be dismissed entirely on the mainland where preference deals place Family First and even the CDP in with some chance in some states.

Any questions? Make em quick, I will be offline most of Saturday!

Kevin Bonham apologises for giving his readers an election preview that will have all the lifespan of a delinquent mayfly, but has been very busy lately.