investor daily logo

Why the opinion polls got it so wrong

By James Mitchell
4 minute read

As one of the few politically conservative journalists in a newsroom chock full of left-leaning voters, it’s comforting to know that most Australians are on your side.

I must confess that I took a secret, almost childish delight in witnessing how wrong the opinion polls got this election. Politics can be like that – secretive, childish and wrong.

While Morrison’s victory meant no change of government, it also meant no change in the trend of incorrect opinion polls, something that has become an unignorable issue since the 2016 US election.

Personally, I could see Donald Trump becoming President of the United States. I remember my colleagues at the time calling me a fool and a few other four-letter words as a result. I don’t agree with everything Trump says, far from it in fact, but nor am I outraged by him in the way that many of my contemporaries are.

And it is this outrage, I believe, that is a contributing factor in some of the major polling fails we’ve seen in recent years. 

The outraged are far more likely to share their views in the opinion polls than the silent majority who don’t wear their political colours on their sleeve. Comment pages on news websites also tend to attract the outraged and the heavily critical. In many circles, it is almost an unwritten rule that one doesn’t reveal which party one votes for – ‘reveal’ being the key work here. Nailing your political colours to the mast invites a barrage of criticism and even some nasty name-calling. Such is the result of the democratic process. 

But the gap between how we were all led to believe the election would go and how it actually went down speaks as much about our reluctance to share conservative views as it does about the validity of opinion polls. 

Three weeks ago, political scientist Dr Andy Marks declared that “Labor will win this election. I think that’s virtually unquestionable.” This statement from a man who is assistant vice-chancellor of strategy and projects at Western Sydney University and holds a PhD in political science. Naturally, I was eager to find out why he was so certain. And ultimately incorrect. 

“The qualifier for that prediction was based on the analysis of primary vote polling,” Dr Marks said. “That’s what’s so alarming about the result we saw on Saturday and the difference from polling trends.

“We do experience some latitude between the two-party preferred polling estimates, they’re never that reliable, but generally primary polling figures are more reliable. That was certainly the case in the 2016 election. I think this is the first federal election where we have seen such a strong deviation against primary vote polling. 

“That’s an indication that something is very wrong with the methodology. The major polling houses are struggling to attract sufficient samples.”

There have been other explanations of what went wrong with the opinion polls: the lack of landlines, a 1.5 per cent margin of error, a failure by polling companies to ‘de-bias’ their results. Dr Marks said opinion polls haven’t been able to track single-issue voting patterns like Adani, suggesting pollsters need to track sentiment and voting intention across a broader scope of platforms. Social media would be the big one. 

I liked the explanation Liberal senator Arthur Sinodinos gave on an ABC panel on election night: that, effectively, the traditional Liberal and Labor voters as we once thought of them are quickly disappearing. The well-healed seats have the luxury of not worrying about energy bills and house prices and can instead afford to vote on environmental and social concerns. The workers, who may have voted Labor in the past, are actually aspirational and conservative. 

Traditional voting patterns are clearly changing. If you’ve read this far, you may be asking yourself, as a fund manager or financial planner, why this is important to an audience of financial service professionals. It’s important because despite what we were led to believe, on election day Australians revealed that they are conservative people who want lower taxes and little, if any, change to the way things are: we want our franking credits and we want our negative gearing. 

As Dr Marks points out, the coalition was able to characterise Labor’s economic reforms as something to be apprehensive of: “The coalition didn’t offer anything as an alternative, other than tax cuts. This was an election that played out against a backdrop of economic apprehension. Globally, we have a trade war occurring between two of the largest economy and a general sense of instability,” he said. 

“The conservative message, which is something the coalition put forward, is one that appealed.”

The markets also found the Morrison government victory appealing on Monday. The ASX closed $33 billion higher, hitting a 12-year high.