Spot the difference: A tale of two safe seats

Most analysis in this election will focus on marginal seats. But safe seats are the backbone of our electoral system. What makes them safe? Let’s dive in!

Check out these two streets. One street is from one of the most Labor-voting seats in the country. The other sits in the heart of one of the safest Liberal seats in the country.

But can you even tell the difference? The two cul-de-sacs certainly look similar from the air. The houses are medium-sized, and there is a sprinkling of pools.

Shall I give it away? The one with more trees is the Liberal heartland, the seat of Mitchell in the north-west of Sydney. The one with more orange roofs and fewer trees is in the seat of Brand, south-west of Perth.

The website Australian Election Forecast has Brand as 99.9% chance of being won by Labor, which is as high as the ratings get. Mitchell is a 99.8% chance of being won by the Liberals this year.

The birds-eye view looks so similar. What makes them different? Why are these two parts of the country at loggerheads?

Spot the difference

Let’s start with the most quintessential of Australian concerns: property prices. Sydney is far more expensive. It has been ten years since a house in the above street in Sydney sold for under a million. The most recent sale was for $2.5 million. But in Perth, those houses have only recently cracked the half-million barrier — the most recent sale was for $625,000. When you compare these two seats, the Labor voters certainly live in far less expensive homes.

The Sydney suburb is 35km from the centre of town, which is a middle-ring suburb in Sydney terms. The Perth suburb is 50km from central Perth, which is pretty much the boundary of what people consider Perth. On the periphery of the most isolated city in the world? That’s Labor territory, it turns out. 

The census doesn’t collect property price data by electorate, but it does collect mortgage repayments. Here’s a neat chart of mortgage repayments vs two-party preferred votes. I’ve marked Brand and Mitchell for you, as well as a select few other seats. You can see the much higher mortgage repayments paid in Mitchell. 

But property prices are an outcome, not just an input. They are an expression of something deeper.  

Who are these people?

A simple summary is that Brand is white and less highly educated; a traditional Labor scenario. Very few households speak a language other than English, and quite a few households have people who dropped out of school in Year 10. People there are likely to have been born in Australia or to have parents from England.

In Mitchell, most households speak English too, and most people were born in Australia. But a significant minority has two parents born in China, with fewer people who dropped out of school.

I wondered if electorates with a lot of people who have Chinese ethnic backgrounds were more prone to voting Liberal, but it doesn’t look that way. As the next chart shows, there may be slightly more Labor electorates where residents say both parents were born in China.

In Brand, the median income is $1,722 a week, and the median mortgage is $1,733 a month. In Mitchell, the median income is $2,715 a week, and the median mortgage is $2,990 a month. 

Income matters for voting, though perhaps not in quite the way you might expect. The big picture is that Australians are generally more likely to vote Labor the more money they make, as the next chart shows. But the top few electorates for income favour the Liberal Party over Labor (and some are now leaning teal).

The trends in all such analyses are present but subtle. Neither wealth nor income wholly determines how people vote, nor do ethnicity or religion. And that’s a good thing. When parties cleave wholly along ethnic lines, society gets divided really fast. 

It’s actually good that Labor has to appeal to the white working class of south Perth and Albo’s woke neighbours in Grayndler, because it means the party must compromise. No party that wants to win can take anyone or anything for granted, and that helps Australia avoid extremism. Even geography doesn’t determine votes, as we saw when the Greens made big inroads in Queensland last election. 

Safe seats can be complex things, as the Liberal Party is finding in Kooyong. The demographics of the seat are changing somewhat, but the allegiance of the seat is changing dramatically.

What people want can shift, and a seat can go from safe to at-risk in a heartbeat. In ten years, Brand could be a seat that Labor has to work hard to retain, while it might have Mitchell in its back pocket. 

What do you think is the biggest indicator of voting patterns?

We want to hear from you. Write to us at letters@crikey.com.au to be published in Crikey. Please include your full name. We reserve the right to edit for length and clarity.