But before discussing what is going on, I should explain some changes in the way we are approaching the polling, and introduce the FPI to you. I should also say thank-you to Sirromet who sponsored the wine research we did at the same time. Worth clicking on their label to the right.

As we are planning to poll monthly, we are going to track the movements between surveys. While our sample is not representative of the general population, we also assume that it is not unrepresentative either. That means that while percentages from our surveys would not be expected to exactly mirror those in the general population, there should be some relationship between changes in support in our sample versus wider changes

So I am introducing the First Preference Index as a way of tracking changes in party support without being distracted by the actual voting percentages. This index takes the September result for each party as representing a figure of 100 and will then express each month’s results in relation to the base. So, if there is a 3 percent increase (that’s percent increase, not percentage point increase) in support for a party, the index should be 103. A 3 percent decrease would give an index of 97.

From what we know of party support in the general community, to lose an election the ALP index would need to be around 88, and for the Libs to win their index would need to be around 112. These figures may vary if there is a substantial change in support for the minor parties.

At the moment, the ALP index figure stands at 112, and the Liberals at 95, which means the ALP improved its position over the course of the last month. The graph below tracks only the first period, and as a result doesn’t look like much at all. Once we’ve got a few months going, it will take on wormlike characteristics and be much more interesting. I’m also only tracking the major parties and the Greens on the graph because minor party votes will be more volatile in percentage terms and create distortions in how you see the data.

Which brings us back to the leadership. Malcolm Turnbull started with good approval ratings for an Opposition Leader. While 31% approved of him, only 18% disapproved. In the last month or so, based on the responses from our sample of 1716 across Australia, that approval figure has actually grown, but his disapproval rating has grown much faster.

In comparison, Kevin Rudd has rising approval ratings and falling disapproval ratings.

Rudd has won over some Greens voters, energized Labor voters, and also stolen some Liberals. Turnbull has fallen short, not because his approval ratings have dropped, but because his disapproval is up more than 200%.

As noted in our last analysis, Malcolm Turnbull had better approval ratings than Brendan Nelson specifically because he appealed to Labor and Greens voters. The problem he faced was enticing these voters away from Labor. While they might have liked him, they like Kevin Rudd even more, so Turnbull’s approval rating wasn’t a good measure of his electability.

In fact, in some ways it wasn’t a measure of him at all, as it was a reflection of the fact that voters hadn’t had enough time to assess him.

Our qual suggests that Turnbull is taking the fight up to Rudd. The corollary of that is that voters have more information on which to assess him. This information is that he is scoring points against Rudd, their favoured one. Rather than elevating Turnbull in their eyes, it tends to degrade him. The result is that Greens and Labor voters are moving from “Undecided” and “Unsure” to “Disapprove” and “Strongly disapprove”. At the same time, Nationals voters are much more supportive of Turnbull because they like what they see, and were not inclined towards Rudd in the first place.

What this polling shows is that if Turnbull wants a quick march to the lodge, it won’t be Malcolm up the Middle – not unless Rudd stumbles in a big way – but Malcolm sweating-it on the right. For Rudd, it means that he is perfectly positioned, and the looming recession has provided the cement that he needed to reinforce his election win.

This is a problem for the Greens who have benefited from disaffection with Labor. Perhaps it will be only short-term. The last election was the first for decades to be won substantially on an environmental platform – greenhouse. At the moment, the electorate is more interested in the economy. But when the economic crisis has run its course, they may well turn their focus back to the environment, where it is difficult to see either of the major parties meeting the expectations of potential Greens voters.

No Liberal leader in my lifetime has held government securely by governing from and for the centre. All successful non-Labor governments have needed a significant percentage of conservative blue-collar votes, and they have won these votes by accentuating the philosophical differences between them and Labor. (The now-sainted Malcolm Fraser was the first Prime Minister that I remember consciously espousing this theory apparently under the influence of his then advisor David Kemp).

The excitement about Turnbull is as much from the centre and the left as it is from the right. This is reflected in the relative lack of enthusiasm for him of National Party voters.

When asked what they like about Turnbull the majority of voters instance him being a small “l” liberal, and his views on the Republic and Climate change. They frequently use the word “Liberal” with a capital “L” as well to indicate that he is not like the party of Howard, Costello and Nelson. A lot of his support therefore comes from people who will like him only so long as he represents a repudiation of the Liberal Party.

The obverse of this is that the same issues that recommend him to one set of voters – republic, environment and social attitudes – are instanced by the other set of voters, as reasons to hesitate in voting for him. The problem for Turnbull is that the second set of voters is the one that most strongly supports his party’s brand of politics.

A traditional approach to this problem would be to turn his back on the progressives. The other approach would be to assume the support of conservative voters and try to move the Liberal Party more into the centre. This tactic would probably strengthen the National Party, who would be well-placed to pick-up the disillusioned on the right, but would it result in many in the centre or left shifting their allegiance to the Liberals when they already have a republic and climate change champion in Kevin Rudd?

Perhaps the answer for Turnbull is to concentrate on those characteristics which all constituencies can find attractive. Voters also associate concepts of strength, leadership, and intelligence with Turnbull. They see proof of this in his past as a successful businessman, and also in his presentation. He is articulate and presents confidently. (Click here for the Leximancer map which demonstrates this).

Rudd’s qualities are a lot softer than Turnbull’s. He is less charismatic. Labor supporters like the way he is addressing issues. They also like the fact that he is a hard worker. Liberals concede that he is intelligent and thoughtful. Greens give him marks for keeping his promises. Rather than a superstar politician, the qual on Rudd suggests he is seen as an honest trier. (The Leximancer map is here).

Hesitations about Turnbull centre on his personality, the policies he stands for, and the Liberal Party. Arrogance is certainly one factor, but it is minor compared to general systemic concern about the interests that he represents. Surprisingly, his history in business is not a major issue. As noted earlier his small “l” credentials also work against him with some voters. (Confused? Try the Leximancer map.)

Rudd’s problems are also systemic – the Labor Party is not held in any higher regard than the Liberal Party. Then there is Rudd himself. Is he strong enough, does he have his mind on the job, or is it all overseas trips? A theme that comes through in a number of permutations is that he is “all talk and no action”. The word “talk” is associated with “action” 56% of the time! “Substance” is also an issue, as is “time”.

“Time” is hard to tie down. For some, it is that he has not had a long enough time, but for others it is time that he started to do something. Either way, there are temporal expectations that he has to do more before the next election. There is a lot of good will towards Rudd, but it is not limitless, or should I say, timeless. Another hesitation for Rudd is that he is already being seen by some voters to back away from policies that he espoused during the election, so that climate change isn’t always a positive for him. (To see a map of Rudd’s hesitations, click here).

To sum up. Turnbull is seen as energetic, decisive, intelligent experienced and determined, with a social conscience. As a Prime Minister he would make a good captain of industry. Rudd is seen as hardworking, intelligent, experienced and determined, but not decisive, with a social conscience. As a Prime Minister, he makes a good bureaucrat.

Because we ask qualitative questions, we are also in a position to fill-in the motivation that might lie behind the answers the quantitative pollsters get.

Newspoll has a fascinating poll in which they ask voters about character traits of the leaders. They also ask who is best to handle particular issues. Turnbull is very competitive with Rudd on the questions of Experience, Decisiveness and Understanding the Major Issues. He's not very competitive on issues like Cares for People, Arrogant, Likeable, In Touch and Trustworthy.

When it comes to who is best to handle particular issues, Turnbull just shades Rudd on the economy, although he lags on everything else.

The SMH AC Nielsen poll has Rudd as the preferred Prime Minister by 56 to 33 percent. Not as extreme as our lead, but still in advance of his party's vote.

I'm planning to have our qualitative analysis up by the end of the day.

Being best to lead the Liberals is one thing, but when it comes to leading the country, our respondents prefer Kevin Rudd whose popularity is ahead of that of his party. Turnbull in comparison lags the popularity of his party.

(Note this is not a random sample so it is indicative only of what may be the case in the population at large. It under-represents women and younger people as well as being biased towards a middle-class and higher socio-demographic, however as it is drawn from a sample of 2,366 it is significantly more representative than any other sort of qualitative research).

1. Turnbull easily the “people’s choice” for leader with 71% saying they would have voted for him rather than Brendan Nelson. Highest support was amongst Liberals (80%) and second highest from Greens (71%). Nationals supported him by 64%, and ALP voters by 63%. These are figures for what people would have done in the party room, of course, not figures for whether or not they approve of Turnbull.


2. Turnbull’s approval is only 37%. A relatively minor 18% disapprove, giving him a net approval of 19%. Approval tends to go with political support, however the opinion of Labor and Greens voters is remarkably benign. While Greens and Labor disapprove of Turnbull, close to half of them "neither approve nor disapprove". The qual, to be published later, suggests that this is because Turnbull is seen as being a small "l" liberal, and to be more in tune with the centre than his predecessors John Howard and Brendan Nelson. How long this can be maintained will be interesting to watch.


3. Rudd is approved by 53%, but with 37% disapproving, his net approval is only 16%. This is relatively good for Turnbull whose net approval is higher. The major difference is that there are 35% of voters whose opinion of Turnbull is parked under “Neither approve nor disapprove”. This figure will decrease over time, and in fact is probably an indication from voters that they have not seen enough of Turnbull yet to have an opinion. It should be noted that ALP voters are more strongly behind Rudd than Liberals are behind Turnbull. Again, the qual will show that some of Turnbull's strengths are actually weaknesses when it comes to his key supporters.


4. Peter Costello has been haunting the debate, and while he thinks he would have won the last election if he had been Prime Minister, voters don’t think he is as good as Turnbull to lead the Liberals at the moment. 46% of all voters prefer Turnbull, compared to 22% who prefer Costello. Support for both is reasonably consistent across party supporters with some exceptions. Costello receives his best support from Liberals (33%) and his worst from Greens (8%). Turnbull is most popular with Liberals, followed by Greens.

   Joe Hockey scores better than Brendan Nelson, who ties with his erstwhile deputy Julie Bishop on 6%.



5. Putting aside the navel gazing as to who would be best to lead the Liberals, Turnbull is trounced by Rudd as preferred Prime Minister with 62% preferring Rudd. This is 5% ahead of the generic Labor vote which is 57%, indicating that Rudd has it over Turnbull at the moment.
   

That is not to say that the Liberals couldn’t have won this election, but they haven’t adopted the tactics they needed to.

We had 275 responses to a questionnaire posted at http://polling.nationalforum.com.au/index.php?sid=66776&lang=en. While the response rate was similar to other polls I have conducted, the skew was even more heavily towards the Greens than usual. At 20% the Liberal vote was where it normally is, but the Labor vote has collapsed to be almost level with it. In crude terms Labor voters have moved to the Greens who are at 40% on our poll were well above where they have been previously. Western Australia has compulsory preferential, and almost all of these votes will come back to Labor. So, while there is a protest vote in WA it is not going to the Liberals, it is going to the Greens and will return to Labor via preference flows.

(I should emphasise with these percentages that this is a qualitative poll. They are indicative of what is happening in the general electorate, but they are unlikely to accurately measure the exact magnitude of change).

Winning protest votes is the only option for a party that is not thought fit to govern. The Liberal Party finds itself in this position. This can be done best when voters think that the government is misgoverning, but that the opposition is unlikely to win. They can therefore send a message to the government via a vote for the opposition, without the risk that the opposition will win.

When we asked respondents what party they wanted to win, they overwhelmingly nominated Labor. However, 20% of those voting Labor said they wanted the Liberal Party to win, while on the other side, 11% of those voting Liberal said they wanted Labor to win. There was also 18% across the board support for the idea of a hung parliament. This is dangerous territory for Labor with their own supporters lacking enthusiasm for them and a significant minority wanting them put under the pressure of being a minority government.

On top of this, while there is disquiet with the thought of a Labor victory, everyone expects them to win. Only 14% expect the Libs to win, compared to 52% who expect Labor. It is in circumstances like these where voters often vote for the presumed loser to put a check on the presumed winner.

Labor understands this dynamic, and their advertising is targeted at raising a protest vote against the Opposition for failing to even meet the standards of being a good one. Colin Barnett is a “risk to your future”, and they are particularly targeting him on his plans for nuclear power. This should play well to the doubts about Barnett that voters have. They still remember his promise from last election to build a canal to bring water from the Kimberleys to Perth. They also understand that the only reason he has returned from the brink of retirement is that the Liberal team has no other talent. In fact, some also speculate that he will not stay long after the next election, whether he wins or not, leaving them at the mercy of the rest of the Liberal team.

This voter sums the Liberals problems up neatly. Their hesitations in thinking of voting for him were “His [Barnett’s] performance at the last election. The lack of any real leadership in the coalition. The lack of attention to social responsibility.”

Labor’s takeout message is “How could you trust Colin and the Libs to run the State?”.

In their turn the Liberals are playing into Labor’s hands. Their campaign is based around the theme that after 8 years Labor has had long enough to fix the problems that WA has, and it’s time for someone else to have a go.The first half of this message undoubtedly resonates. The issues that people raised as being important to their vote were frequently to do with problems arising from a failure to plan for the mining boom, or what will happen after the boom is over. Labor is not seen as competent in this area.

But as the Liberals are seen as even less competent, this pitch is not a vote winner.

Interestingly, while the media, at least here on the east coast, has most frequently concentrated on corruption in WA governance, particularly related to Brian Burke, this does not show-up as a top of mind voting issue. Probably both sides are seen to have these sorts of problems, meaning that they are neutralized for Labor.

So, while polls show that the Liberals have closed the gap with Labor, there is nothing in this polling to suggest that they will close it enough to win. Voters don’t want the Liberals to win and the Liberals are not putting an argument to them that they can accept as a reason not to return the government. They all know that things should be better in WA, but they also know that voting for the Liberals is likely to make things worse, rather than better. Carpenter wins by default.

There were more males than females (64% to 36%) and most respondents were 55 years and over. The skew to the political left was not as strong as usual with 21% Greens, 27% Labor, 15% Liberal and 17% National.

Findings

By contrast we’re optimistic about our own circumstances and on balance are more optimistic than pessimistic. The increase in the cost of fuel, food and housing is being felt by some, but not many, and these are more likely to be lower income and living in regional areas.

In terms of development, we favour those that are “sustainable”, although what this word means is a little unclear. It certainly doesn’t seem to mean oil shale or coal mining, which are the worst regarded. However, coal seam gas and LNG are quite well-supported, although many see them as a fuels that buy us time while we find more sustainable forms of energy.

There is also a sense that the world needs energy and that we cannot deny it to them, so that there is an imperative to export our resources.

When it comes to who is best to exploit those resources, we prefer our own large companies, with companies like Queensland Gas, BHP and Origin being well-regarded.

The takeout message from this research to me is that, while we are experiencing growth pains, we want to see growth continue. Development is not a dirty word in Queensland, as long as it is done right.

One thing is certain, Bligh will not be rushing to an early election.

Recent Galaxy polling suggested that the Coalition and Labor were neck and neck on primaries. I thought it was a rogue sample, but when our polls, which traditionally lean to the left, say much the same thing, then something is happening. In the BCC elections I successfully picked the result by adjusting our results 10% further towards Campbell Newman. If I do that in this case, Springborg and the Coalition would be on more than 60% of the two-party preferred result, an improbable result.

Politics is dynamic, and with an election not due until the end of next year, Anna Bligh can change course and rescue her government. She's already started to do that, trying to distance herself from Peter Beattie.

In fact, she will have to behave like Peter Beattie to win the next election. Beattie was notorious for apologising for the mess he or his party had created in the first place, and then run against his own party declaring that he was the only one who could fix the problem. He got away with that partly because the opposition was regarded so poorly that voters were torn between punishing them and rewarding Beattie yet again.

Similar dynamics are in play here. When we look at the issues that are important, they are all about quality of life, as embodied in physical and social infrastructure like water supply, roads, transport, health and education. They stem from the strong population growth in south-east Queensland, and the failure of the government to plan how to deal with it. It's a mess of Labor's making.

As a result most respondents think that the state is heading in the wrong direction, including Greens voters. This leads them to mark Anna Bligh down, and blame her for being too close to Beattie. Even those who are favourable to Anna think she needs more time, a reflection of the way Beattie still defines Labor and its leadership.

However, when they look at the opposition they are not impressed. They tend to like Springborg, but even so, many feel they don't know him well-enough, despite his 20 years in public life. They are also concerned that he does not have a team behind him, and especially express concern about the behaviour of the Liberal Party. In this context some mention the proposed merger as a positive.

In terms of issues, the political climate is very much like it was in the local government elections. Environment and quality of life are central, and candidates who can credibly promise to safeguard and improve them were rewarded. In most cases, when there was a clash between independent and party candidates, such as on the Gold Coast or in Townsville, the independent won.

Not surprisingly, "party" turns up as a negative for both Bligh and Springborg."Greener" voters may well be the key to this election, and they are particularly antipathetic to the mainstream parties.

I've put some tables below. We had 924 responses, which was fantastic.

Key points:

1. There has been around a 10% swing against Labor since the last election On our poll this results in a 52% majority to the Coalition after preferences.
2. A majority thinks the state is heading in the wrong direction, including 62% of Greens voters
3. The most important issues are infrastructure, water, health, transport, roads, and population.
4. Most (61%) disapprove of the performance of Anna Bligh, and her biggest negative is Peter Beattie.
5. 41% approve of Lawrence Springborg, and 31% disapprove, and his biggest negative is his team and the Liberal Party.
6. 38% of voters each prefer Bligh or Springborg, with the remainig 24% uncertain.
7. 65% expect Labor to win, but 42% would like the Coalition to win.

However, suffice it to say that it is not good news for Anna Bligh. I think she knows that from the activity that I can see in the news.

You can download the map of attitudes to Rudd by clicking here, and to Nelson by clicking here.

Contrast this Labor voter:

"Even though I am a Labor voter, I had Kevin Rudd pictured as a somewhat uninspiring bureaucrat with a savvy media image. The budget dramatically improved my assessment of him as a genuine Labor Leader"

with this Coalition voter:

"Kevin Rudd is all talk and no action. Before the election he gave the impression that he could drive down inflation and the cost of living. It is now obvious that he can't."

Nelson suffers from two major handicaps - the former Howard government and Malcolm Turnbull. While the top ranking concept is "Nelson" it appears mostly as a negative. The discussion is about his future or his performance rather than the budget. He is seen as lacking substance so is defined by what he inherits, or what he is not.

There is also a large degree of entrenchment of views. "Vote" is the second strongest concept which appears either in the context of "I always vote Labor" or "I always vote Liberal". In other words, Nelson is relevant only in so much as he is part of the institutional arrangements. This is to be expected. We've just had an election (six months is "just" to the average voter) so the budget is going to have a minimal impact on perceptions because elections, particularly those when you change government, are psychologically decisive. We don't want to go and revisit the decision too soon afterwards.

I've tried doing this one with a pop-up image instead of an embedded one, so hopefully you will get better resolution. So, click here and you should get a usable map. I've used final preference as the variable, so Coalition voters are on the left (the software makes the decisions, not me) and Labor on the right.

There are some lessons for both sides here. For starters, "Alcopops" does not appear as a concept or a theme. There is a good reason for this. "Alcopop" only turns up 15 times out of the whole sample. All the time that the Coalition has spent on this issue has been wasted on the basis of this research. To follow-up, the strongest turn-off for Labor voters was "change". That word has been associated in all our polling in the last 12 months with "climate", and this polling is no exception. Labor was elected on a platform of fixing climate change, and some of their strongest supporters think they are not doing enough.

"Families" is another theme with risks in it for the ALP. Many of the Liberal voters are being turned off by the phrase "working families", because they feel that Labor is too narrowly focussed on a group which doesn't include them. If they're older, or divorced, or if the wife is at home, they feel as though they aren't included. "They keep saying working families as though they are the only people that are important to the economy, there are plenty of people that are working hard that don't have a dependent family and aged pensioners don't even get a mention."

"Tax" cuts all ways. Some feel they are paying too much, and others think that the tax cuts should have been directed to better services. In contrast, "Pensioners" is unambiguous. It may have surprised most commentators that pensioners demonstrated in Melbourne against the budget, but on the basis of this research it shouldn't have been unexpected, even if it couldn't have been predicted.

The "Baby Bonus" made an appearance, and again, for a number of reasons. Some thought that the rich should be encouraged to have more children, because those children were likely to be more successful than average. Others thought there shouldn't be a baby bonus, while still others thought the means test was too generous.

This is the first of the qual. I've analysed it using Leximancer, which produces the concept map below.

I've used people's final preference as the variable to which the concepts are related. In the bottom left-hand corner you have those who would preference Labor first, and in the top right-hand corner those who would preference the Coalition. The "Undecideds" are on the bottom right. Resolution is not as good as I would have liked, so I might need to do something else in future. Please bear with me.

The theme  with the largest area is "tax", and this sits closest to the Coalition voters. This is largely what you would expect, as the tax issue resonates most strongly with Coalition voters. Within the "tax" theme sit concepts like "families" and "tax" and "cuts". So, voters were largely happiest in general that the tax cuts were honoured and that there was support for families. These things are winners across the board.

What is a little surprising is that the theme most closely associated with Labor voters is "rich". When you examine the quotes, this theme is very closely linked to the one next to it - "testing". "testing" is actually a compound of two words - "means" and "testing". So taken together these concepts indicate that the means testing of benefits is what is most pleasing in the budget to those preferencing Labor. There are a couple of explanations for what is behind this. For some, means testing allows money to be saved for those who really need it. "Means Testing. If your rich then why would you need payments? Its a good way of reducing the countries costs. And the money going into education is good." Contained in this concept is also a sense of equity. Why should someone who doesn't need the benefit get it? I also detect a satisfaction in some responses that people who were seen to undeservedly prosper under the Howard regime are not doing so well under Rudd.

The other area of approval related to the budget is "education" which also contains within it the concepts of health and infrastructure. The Education Revolution still resonates, and symbolises the important social welfare/social equity issues to Labor voters. Hovering around the edges is the word "fund" or "funding", both of which predominantly refer to the three statutory funds for health, education and infrastructure. Both sides tend to approve of this concept. Labor voters associating it with a government that they see as at last planning for the future, and Liberal voters giving Rudd a mark for adopting what they see as Liberal policies.

Do you approve of the job Kevin Rudd is doing?

Our sample suggests that Rudd's approval rating is now fairly similar to a successful Labor Premier, and while extraordinarily good, is starting to come back to earth.

Would you be more likely to vote for Kevin Rudd because of the budget?

What this table demonstrates is that the budget will intensify voting intention, but make little difference one way or the other. Liberal and National voters are significantly less inclined to vote for the government because of it, but the Liberal and National votes are also reduced to a hard core because of the extremely large share of the total vote that the ALP has at the moment

Do you approve of the job Brendan Nelson is doing?

These are devastatingly bad figures for Dr Nelson. While his approval rating on this poll is higher than on other polls, his disapproval rating is more than 50%. There is limited room for him to improve as only 2% of people don't have an opinion, and 28% neither disapprove nor approve. His "strongly disapprove" figure is 27%, while his "strongly approve" figure is only 3%.

Would you be more likely to vote for Brendan Nelson because of the budget?

This poll was taken before the address-in-reply so it may have improved after that. However, what it says is that while the budget didn't win Rudd any votes, it lost Nelson some. You'll have to wait for the qual to find out why, but my strong suspicion would be that it is to do with his performance.

We also have figures on what percentage has changed its vote since the federal election. I won't reproduce the table, but 5% of last time Labor voters would vote Liberal if an election was held now, and 2% of Liberal voters would vote Labor. No Nationals have changed their mind. So there has been a small erosion of the Labor vote. Unfortunately there aren't enough of these people to be too sure what they represent, but I will have a look at the data later and see what we can make of it.

Is the country heading in the right direction?

This is an almost complete reversal from previous polls with ALP voters ecstatic (can't think of a better word for it) at the direction of the country. Liberals are fairly concerned and Nationals even more so. There is also a degree of apprehension from Greens voters.

Do you approve of the budget?

This is similarly related to voting intentions, and reflects a similar, although less accentuated division of opinion as the direction question.

Is your personal position better or worse?

To some extent approval or disapproval flows with perceptions of whether the budget has improved personal circumstances, although the close alignment with voting intention suggests that a large proportion of perception is reflected in voting behaviour.

Will it be good for the nation?

Some intriguing and minor divergences here from whether respondents approved or not of the budget. For example, while 89% of Labor voters approve of the budget, only 79% think it will make for a better country. Perhaps the deficit can be explained on the basis that they don't think it will make it better or worse, perhaps a tacit approval of the way things were.

I've been intriqued that while The Australian's respondents have pulled my sample to the right, it still leans heavily towards the left of the spectrum. There is something about the online world which begs to be analysed.

Anyway, for those of you who are interested in the Benchmark Poll you can read the relevant material here:

Benchmark Poll

• Questionnaire
• Report
• Fear factor now in Labor corner - Graham Young, 25th October, 2007

In the last election Howard’s promise that interest rates would always be lower under the Liberals than Labor may have been crucial to him holding mortgage belt seats.

My colleague John Black found no statistical relationship between the swings that actually occurred and average mortgage sizes, but he found a relationship in other seats. His theory was that the interest rate issue didn’t so much affect those who were buying houses, as it affected people, like their parents, who worried about the home buyers. Our questions aimed to test these hypotheses.

We also wanted to know how real "mortgage stress" is. Are those who spend more than 30% of their income on mortgages rich enough to afford it, gearing into investments, or desperately hanging on to a toe-hold in the housing market?

Last, we wanted to know whether voters blame Howard for high interest rates and whether nervousness about them made him more or less attractive to voters.

Our sample swings to the left, and compared to the average is older, slightly richer and more likely to own a home, and 27% of those with mortgages were making payments putting them in "housing stress". Despite being more affluent than average they gave us valuable leads on these questions.

Interest rates were a significant issue for all groups, except for those renting. Just more than a third of renters thought it was a significant issue, and about the same percentage thought it wasn’t. For those with a mortgage the figures were 54% and 17%, while for those who owned outright they were 41% and 29%.

If you were paying more than 30% of your family income in rent, then the figures were 65% significant, 11% insignificant. Safe to say that even if these respondents have bought boats, holiday homes and investments they’re still worried about the holding costs.

There wasn’t much altruism evident. When we coded qualitative responses from swingers, 23% of them came up with a variant of "I’m alright Jack" while 21% said "I’m suffering", or words to that effect. 11% were primarily worried about the effects on people other than themselves. These were mostly aged between 55 and 64, the age when you are most likely to have offspring buying a house. Another 6% actually saw a personal benefit in higher interest rates, and were also generally older, so were probably net lenders.

A problem for Howard is that 11% of swingers volunteered that Governments had no effect on interest rates, and a further 6% said it was a significant issue because he had lied about them last election. "…I believe the Liberal Party will again lie and claim they'll keep interest rates low, or lower, or at record lows (pick this week's lie)."

We also looked at those in mortgage stress and the personal impact was overwhelming – 47% were personally suffering. But only 8% nominated coalition lies. It seems that while mortgages are hurting, the government isn’t necessarily being blamed.

So, who is more likely to keep interest rates lower? The answer to this was fascinating. 80 percent of Coalition voters nominated the government, but Labor voters had no confidence in their own side with only 13 percent nominating them as the best. The apparent discrepancy is made up by those who believe neither side has an edge with 69% of Labor voters calling it a draw, compared to only 15% of Coalition voters.

A closer look at swinging voters makes this better for the government. Those voters who have crossed to Labor believe that Labor is better able to handle interest rates than the Coalition, by 21 percent to 7 percent, but those who are unsure, or moving from Labor to Coalition more strongly nominate the Coalition over Labor. So, while "Neither", a neutral position, is the answer for 50% of swingers, a quarter tend in the government’s direction as a result of interest rates.

The reason for this is clear looking at the qual. The Liberal Party enjoys a good record in voters’ eyes of keeping interest rates low, while the Labor Party doesn’t. Those who favour Labor to keep rates lower rely on abstracts like the parties’ policies. They also point to recent rate rises under Howard.

We’ve also analysed these responses with the Leximancer software. As the attached map shows, "Howard" is a negative for the Liberals on the issue, but "Rudd" isn’t even present, strange given his presidential approach. "Labor" is negative for the opposition and is associated with the Liberals being best to lower interest rates. The debate is dominated by perceptions of risk and looking for the party least likely to make rates higher, rather than likely to make them lower. IR is also absent as a concept, suggesting both that job insecurity is not an issue with interest rates, and also that the Liberal Party has not been successful in linking IR reform with a better economy.

All in all, this is not a strong hand for Howard, but it is an even weaker hand for Rudd. In the no-holdem game of political poker, if I were Howard I might be prepared to play this game to the end, particularly given the number of dud alternatives hands there are lying around.

Tables