History is written by the victors

My day job is a mix of quantitative and fundamental investment analysis. What that essentially means is that I spend my time looking for situations where the numbers don’t match the narrative – and I can’t help but notice that in the Australian Federal election result the narrative and the numbers are miles apart.

Scott Morrison and the Coalition are chalking this victory up to a splendid campaign and a reconnection with middle Australia. The numbers suggest:

  • voters abandoned both the Liberal party (-1.0%) and the Labor party (-0.9%) in droves, but due to preferencing (particularly by Clive Palmer) those votes ended up back with the Liberals.
  • the Liberals have done much better than Newspolls suggest, which is being treated as a strategic masterstroke by Morrison rather than an indictment of polling results.

Remarkable question is whether Newspolls have been wrong for years, my guess is that they probably were. And so that probably meant two things (a) much of the Coalition instability was due to measurement error rather than a fundamental disconnect with voters (b) Labor falsely assumed that Bill Shorten was doing a good job despite his poor personal polling.

History is written by the victors

The history that the Coalition seems to be trying to write is that Morrison is respected and admired by the population which led to a +0.7% two-party preferred swing to the Coalition and their re-election. The first preference numbers tell a different story:

The data suggests:

  • Morrison won far fewer votes first preference votes than Turnbull did last election
  • The National party did much better than last election
  • Shorten helped by winning far fewer votes than he did last election
  • Morrison was effectively bailed out by Clive Palmer and One Nation preferences. The state that the Coalition did the best in (Queensland) was also where Palmer and One Nation votes were strongest
  • The distribution also mattered – a lot of the votes lost by the Liberals turned safe seats into less safe seats. Ironically, the Liberals seem to have lost a lot of “the base” that Abbott/Dutton et al are so keen to reconnect with.

Morrison deserves credit for bringing stability to the imploding Liberals. However, the real heroes for the Coalition look to be:

  • Pollsters who convinced the Labor party they were on the right track
  • Clive Palmer and the United Australia Party preferences
  • Michael McCormack (for those of you who don’t recognise the name, he is the current leader of the Nationals and deputy prime minister)

Issues for Pollsters

In the aftermath of the Trump election, there was a lot of talk of undercounting of conservative views and pollster errors. There are already similar conclusions being drawn for the Australian election. This is not the same as the Australian election. Our mistakes are different.

To start with, the overall US voting numbers weren’t that far off the opinion polls (Hillary won the popular vote convincingly), but the key pollster error was not enough attention paid to the distribution of voters in key states which led to a Trump win.  Relatively slight underestimating of the Republican vote in those states was enough to confuse the pollsters into thinking that Trump would certainly lose.

There is a lot of soul searching among pollsters following the election over what went wrong, and it is clear that there are significant problems with both election intention polling and exit polls.

Australian Election Polls

Mark the Ballot had the interesting observation prior to the election about the above table:

And what a remarkable set of numbers they are: The 16 polls taken since April 11, when the election was called, have all been in the range 51-52 for Labor and 48-49 for the Coalition.

If we assume the sample size for every one of these polls was 2000 electors, and if we assume that the population voting intention was 48.5/51.5 for the entire period, then the chance of every poll being in this range is about 1 in 1661.

Now maybe we happen to live in the timeline where the 1 in 1661 chance happened to occur.

But there would also seem to be a relatively strong possibility that:

  • polls are having with problems getting representative samples
  • pollsters are resorting to statistical adjustments to adjust for non-representative samples
  • these statistical adjustments are leading to clustering – no one wants to be an outlier and so statisticians make adjustments until they aren’t outliers

I also suspect that Clive Palmer created issues for the polls as his party did not exist at the last election. If Clive opts not to spend $60 million next time, then there is likely to be the opposite problem for polls.

Regardless, it would seem that polls should be treated with much more scepticism going forward.

Investment issues

Politicians have far less effect than many think on the performance of stock markets and economies. This is partly because there is less difference between the policies of the different political parties than the politicians want you to believe, and partly because they have only limited effect on economic outcomes – central bank decisions are more important in most countries.

This is more true for Australia than most, as a small open economy Australia is highly dependent on global commodity prices which political leaders have even less effect on.

So, don’t overplay the effect of elections in general. Having said that, I do expect a positive short-term effect and a negative medium-term effect from the re-election of the Coalition.

Short-term effect

The Labor party had policies that would have been negative for house prices which would have increased the possibility of house price declines sharpening. That crisis has not been completely averted, but the chance of a negative tail risk event has decreased. Our expectation is that Morrison will also be more aggressive in trying to bail out the housing sector as house prices continue to weaken.

Medium-term effect

One of the structural problems with many global economies is a lack of demand and what is being termed “secular stagnation”. Our assessment is that much of the secular stagnation is due to inequality and stagnant wage growth. Labor had a number of policies to address those risks in Australia, the Coalition does not. We expect low wage growth and secular stagnation to continue to be a feature of the Australian economy.


Damien Klassen is Head of Investments at the Macrobusiness Fund, which is powered by Nucleus Wealth.

The information on this blog contains general information and does not take into account your personal objectives, financial situation or needs. Past performance is not an indication of future performance. Damien Klassen is an authorised representative of Nucleus Wealth Management, a Corporate Authorised Representative of Integrity Private Wealth Pty Ltd, AFSL 436298.

Follow me
Latest posts by Damien Klassen (see all)


  1. Really like your stuff Damien, thanks. Didn’t follow the UAP ads too closely so not sure on their position on immigration. Definitely for One Nation with preferences flowing the way they do this is effectively a way for the LNP to capture lower immigration sentiment without antagonising their donors. As this issue was largely under the radar for this election it would be good to see the split on lower immigration stance parties and their respective preference flows.

  2. Ironically the one poll to get it bang on the money was GetUp!’s poll in Warringah a couple of weeks ago (which I participated in) which had it 56-44 to Steggall. In that poll, you were asked who you would preference out of the two. They also polled 800+ people in an electorate of about 110,000 so presumably a much more significant sample size than 2000 nationally.

    • kannigetMEMBER

      2000 is a very large sample, no matter the size of the sample relative to the population your surveying. IF all other factors are dealt with then 1600 will give a margin of error of 2.5%, a sample size of 2000 is only 2% and a sample size of 800 would be roughly 3.6%.

      There is exponentially more expense chasing more accuracy. The issues are more in getting a representative sample. People these days are more inclined to tell you what they think you want to hear instead of what they actually believe. I think this is more inclined to be the reason for the opinion poll discrepancies. Only the truly passionate ON supporter will tell you they voted that way, because the rest dont want to be seen as xenophobic and socially isolated.

      • Thanks.

        Maybe robocalling becomes more accurate then (the Warringah one I did was a robocall) – does anyone care if a machine knows you are voting for One Nation?

      • I don’t tell anybody…partner, friends, colleagues my actual views on lots of subjects, due to my concerns about their reactions. I certainly wouldn’t go spilling my guts to someone who rang me out of the blue.

      • kannigetMEMBER

        While I think Robocall may result in less “fear of ridicule” related answers I think people hate them enough to make people answer them “randomly”. Add to that the fact that people who hate them and are more likely to hang up are potentially from a similar mindset and you may find they are less representative.

        Personally I think the demographics of Warringah have changed enough over the years and they have finally had enough of Tony. It may be the only poll of the last 6 years that was accurate.

        What questions are the pollsters asking? Who would you support ALP or LNP? or more based around preferences and policies? The problem may lie in the questions and not the method itself.

      • I would never take part in an opinion poll. If people who behave like me in this regard are politically biased one way or another, and if we are a significant fraction the population, then what hope is there of getting a reliable result? The only way to correct the data gathered would be by taking a poll of people who won’t take a poll. Good luck with that.

  3. “Morrison was effectively bailed out by Clive Palmer and One Nation preferences”. A lot of my acquaintances voted ON and Palmer , then LNP, just to send a message that the LNP need to lift their game. If ON and Palmer had not been on the ticket, they would have voted LNP.

    • TighterandTighter

      It’s interesting those votes leaking to UAP and PHON. People I was with on Sat said they couldn’t understand why people would vote for racist Pauline or Clive . I tried to point out lots were disaffected and feeling left behind, bereft of real choice etc. “Nope, it’s racist.” I quickly gave up.

    • Torchwood1979

      I’m shocked that the talking heads have barely mentioned this. Reading all the analysis and coverage gives the impression that ScoMo won this election for the Liberal/LNP single-handed, when his primary vote went backwards! Even the LNP vote in QLD went backward.

      Clive Palmer may not have any Senate seats but ScoMo owes Palmer and Hanson his majority government and I doubt they’ll let him forget it.

  4. An excellent analysis. The confirmation that Labor has bled votes to UAP/PHON should be ample evidence on where they should direct their new policy agenda although I doubt this will occur.

    OTOH, I have no doubt that the government will adopt Labor’s election policies to deal with the medium term effects and secular stagnation and happily claim them as their own.

    • I agree excellent analysis but yes the ON / UAP votes were Coalition votes with a message and not unconditional support. I agree Morrison will steal the ALP ideas he likes as he looks to hog the middle ground. Standard old school middle of road government. And the Opposition will hate it.

  5. You just wait until the victors start rolling out the narrative that the ALP’s failed attempted murder of negative gearing and imputation was the “root cause” of the falling house prices and hence is to blame for much of the current predicament of Strayans.

  6. Liu MianzhiMEMBER

    The ALP won NSW (24/22), VIC (21/15), SA (5/4), ACT (3/0), NT (2/0). TAS (2/2) was a draw.
    The Coalition only won WA (11/5) and QLD (23/6) – it absolutely smashed the ALP in QLD.
    I know people are trying to draw all sorts of conclusions about particular policies, but the single biggest measurable factor was the sheer size of the margin in QLD and how it contrasted with almost every other state. So unless a hypothesis can validly explain why a particular policy or factor was of such overwhelming concern to Queensland, to a slightly lesser extent to WA, and not at all a concern for any other state (even running the opposite way), it is unlikely to be a valid explanation of the result.
    At a state level, the Morrison government is the government that QLD and WA really, really want, but no other state wants.

  7. The polling failure is interesting. Is it the case that current polls are conducted via landline phone calls? I haven’t had a landline phone for years, and neither do my kids or lots of other people I know, so that would be a source of bias right there. The incorrect results from the exit polls is even more interesting. Are people telling lies to avoid being marked in public as a deplorable racist, xenophobic extreme right fascist?

    I think it’s telling that Morrison himself proclaimed the “victory” to be a miracle. Honestly I think it’s going to be a pyrrhic victory, because Scomo and his crew will get to own the coming economic troubles in this country.

    Overall, as Nikola and a few others pointed out over the weekend, there are some silver linings to this, including Abbott out of parliament, Shorten out of the ALP leadership, Clive Palmer not in Parliament and the LNP getting to carry the can. And best of all, the ALP’s madly destructive parental reunion thing won’t happen. Unless the LNP decide it was a great policy that they need to copy, I guess.

    In the meantime, I suspect that the ALP are going to draw the wrong lessons from all this, and disappear even further up their virtue signalling fundaments, with a consequent continuing loss of electoral relevance. Until they can appeal to actual human beings with normal ideas they’ll be loosers, and I don’t see how the change is going to happen as they continue to be dominated by neocon fake left inner city tossers.

    • truthisfashionable

      I can’t recall which group it was but I did receive a call to my mobile for a poll in this federal election. Also the NSW state election.
      So at least one of the polls had broadened from just landline phones.

      • I treat any poll related call with the same disdain as any other cold call, especially the robo-calls. My confidential vote is my business, and that information is not on offer free of charge to an anonymous pollster using the info to make their profits. I suspect a lot of the polls are largely bs.

        Pretty sure I wouldn’t be the only one hanging up on these jokers, so how representative can their intel really be? I mean, they aren’t really going to come out and admit that their entire business model is broken because the only respondents were the passionate true believers, while everyone else either hung up or told their robot to go and fornicate elsewhere.

  8. GeordieMEMBER

    “We expect low wage growth and secular stagnation to continue to be a feature of the Australian economy.”

    A most miserable outcome.

  9. SupernovaMEMBER

    Agree, Morrison was bailed out with ON and Clive Palmer’s UAP preferences. Palmer could bank roll ON in 3 years and that would cause considerable problems with ABC and SBS, both of which are heavily into identity politics. Mark Latham of NSW ON has read the electorate correctly, that identity politics is distancing the ALP from it’s true believers who have very little in common with identity politickers. In fact the ALP is really more divided than the Liberal coalition and this election proved it….most of the people got it right!
    However, contrary to the extreme leftwing media, it was Peter Dutton who saved the Liberal coalition and deserves credit for eradicating Turnbull. Eradicating Turnbull actually stabilised the party when most of Australia’s media published/promoted this as destabalisation. People are turning off and getting their information from other sources more in keeping with their own beliefs.

  10. The pollsters got it wrong. Failed miserably due to lack of competence. It’s the story of Australia in general.

    These pollsters are no longer staffed by skilled and intelligent young men and women with technical expert managers. They are now staffed by imported exam cheats, incompetent gay men and incompetent females and incompetent non-whites hired on the basis of diversity rather than skill as the main criteria. The managers are no longer technical experts but more likely a non-technical woman promoted primarily because of her gender and “managerial expertise”.

    The talent leaves Australia early on in their careers.

  11. “To start with, the overall US voting numbers weren’t that far off the opinion polls (Hillary won the popular vote convincingly)”

    Thank God for the Electoral College very smart for the founders to come up with this so all votes mattered in all States not just the big cities.

    I am sure if the Democrats won by the EC they would be loving the system. You take out the illegals and dead people that the Dems had voted Trump would have won the popular vote but doesnt really matter as the EC is doing its job and what it was designed for.

  12. drsmithyMEMBER

    Scott Morrison and the Coalition are chalking this victory up to a splendid campaign and a reconnection with middle Australia.

    Not just them. Pretty much every media outlet (including MB) is presenting this as some sort of catastrophic ALP wipeout.

    Rather than trying to do any sort of actual analysis, they’re simply taking what they thought was going to happen, treating it as the baseline, and then talking about how badly the ALP lost from a position they never had in the first place.

    It really is difficult to overstate the depth and breadth of the fourth estate’s sheer abrogation of responsibility over the last decade or two.