What does the latest Roy Morgan poll tell us? Peter Tucker thinks that the Liberals can dare to believe they have a chance against Labor. The crack in the door, opened a little by the EMRS poll last December, has just grown a bit wider. Not quite wide enough for Will Hodgman to stride through just yet, but one foot has been firmly thrust forward.
The Morgan Poll released this week shows the highest primary vote for the Liberals since 2002, and the worst primary vote for Labor over the same period.
Regular readers of my poll analysis know that I am guarded about reading too much into any one poll; however, this Morgan result lends weight to the thesis, that I cautiously put after the December EMRS poll, that Labor has taken a serious hit since the March 2006 election. ( Peter Tucker archive )
Worse for Labor is that it appears voters have not just moved off the government, they have started to switch their allegiance to the Liberals. This is a very important point: undecided voters as such are not necessarily a drama for Labor, providing the opposition stay low on the primary count. The government can always woo undecideds back. As long as the gap between the two parties is about a quota — in the mid teens — the government is happy. We saw this leading up to the last election: Labor scored consistently in the low-mid forties on primary votes, but the Liberals stubbornly failed to score much above thirty percent. Come the election, those who had been undecided or had “parked” their vote with minor parties simply went back to the devil they knew.
But this Morgan poll, when read with the EMRS poll, starts to indicate a brighter picture for Will Hodgman’s Liberals, and should set alarm bells ringing in the Labor camp. The table below shows primary support for the 2006 election result alongside the latest Morgan and EMRS polls (undecideds excluded).
As mentioned, the 42.5 percent for Labor and the 34.5 percent for the Liberals is, respectively, the worse and best results in a Morgan poll for four years. To illustrate the importance of this to the Liberals, it is instructive to look at the differential — the gap — in primary percentage points between the two major parties over that period. The closer this margin, the closer we are to a hung parliament. The table below illustrates.
There are some interesting points to draw from the above data. The differentials in the mid-high twenties coincide with Jim Bacon’s premiership. The drop in popularity for Paul Lennon is marked by the mid-2004 slump to the mid teens — and he never took the party back into the twenties. Now, are we seeing the start of a new pattern where the differential holds in single figurers?
Too early to be that conclusive. Yes, the latest Morgan poll is just one and we could easily see the gap move back in Labor’s favour as the year progresses. But I am steadily if cautiously moving to the view that voters may have, at last, decided to look more closely at our alternative government.
Morgan and EMRS use different polling techniques (face-to-face versus phone), have differing survey sample sizes (1,000 versus an impressive 1,200), and differing survey collection timetables (polling over several months versus one week). I think this is a benefit as it allows us to compare and contrast the results. As my first table shows, the two polls are reasonably correlated, which given the difference between the two methodologies, gives weight to the basic story they tell.
I know that many observers will start pointing to this event or that (yes, the Pulp Mill or Bryan Green) as the “cause” for Labor’s poor showing. My personal view is that broad voter opinion is not that event-specific, rather it is built incrementally over time. Voters rarely “sweep” parties in or out. They are not perpetually “scoring” the government and opposition parties. The swings at Australian elections are nearly always under five percent; even movements of 2-3 percent are often written up as “landslides”.
So, the Liberals need to see, over the next two years, a steady and persistent gap in single figures between them and the government. If they can be within 4-6 percentage points leading into the 2010 election, then they have some chance of negating the “hung parliament” bogie that Labor will inevitably put up, as both parties would be in striking distance of a majority.
Notwithstanding, I can not see the Liberals getting the 12-15 percent swing they need from the 2006 election to even have a chance of gaining majority government in their own right. There has scarcely been a swing in double figures at any Australian election since WWII for an opposition to defeat a government. But the Tasmanian Liberals are now in their best polling position for ages, and that has to be good news for them. This far out, I think a hung parliament is clearly the favourite result for 2010. About $1.15 if I was a bookie (which I’m not!)
Just a note for those who go to the Morgan site to see the poll results. Gary Morgan says: “However, if a Tasmanian State election had been held in the September to December period, the ALP would have won.”
Rubbish. Well, they might win the most seats, but they would be very unlikely to win a majority, and who knows then how a government would be made up? Under Hare Clark, a wide range of outcomes is possible at any given primary vote allocation. The Morgan poll would probably see a 12/10/3 split, but a Labor majority is a slim possibility only.
For illustration, I discuss probabilities at given primary vote outcomes here: http://www.onlineopinion.com.au/view.asp?article=3812
Morgan Poll: http://www.roymorgan.com/news/polls/2007/4127/
Peter Tucker has worked in Tasmania as an advisor for the Liberals in opposition and in ministerial offices for both Labor and Liberal governments. He is currently a free-lance political commentator and PhD candidate at the University of Tasmania’s School of Government. http://www.utas.edu.au/government/postgrad/PhD/PT.html
