Qld Coal country

At the end of my first look at the concordance between the Australian Swing map and the location of Australian mines, I promised some more ‘solid’ analysis. Something more scientific than — it looks pretty good on a map!

I’m going to start with Qld — because the Adani issue was the heart of the campaign on both sides. And as you can see from the chart below, the electoral outcome in Queensland was simply VERY different.

For the purpose of the above analysis, I selected all the seats have at least one booth that is less than 500kms from Adani (approx E 146.125, S -22). The seats are: Maranoa, Kennedy, Capricornia, Flynn, Dawson and Herbert. The two-party-preferred swing map is below (recall that blue = swing to LNP and red = swing to ALP). Again, it looks pretty good on the map: these booths all swung to the LNP.

Let’s get a bit more scientific about this: I’ve chosen seats that have a booth within 500kms of the Adani project. But this isn’t science. If this theory works, we should see a relationship between swings and the distance of booths to Adani … the closer you are to the proposed mine, the more it will influence the way you vote. And that’s exactly what we find.

Note that the chart below shows all the booths in the state of Qld (not just the close ones). The only thing I have done is to exclude booths with less than 100 total votes (as these can give erratic swings).

The closer a booth is to the proposed Adani site, the more likely it is to have swung toward the LNP at the May 2019 election (in two-party-preferred terms).

An expert in these matters has pointed out to me that the AEC two-party-preferred calculation has problems, and that it would be ‘cleaner’ to look at the primary votes by party. The party that Adani seems to have cost is the ALP — traditional ALP booths swung away from the ALP.

If we look at their primary vote (by booth) against the distance to the proposed Adani site, we find a very clear relationship. The closer a booth to the proposed Adani site, the greater the swing away from the ALP. Every 100kms further from the proposed Adani site, the ALP primary vote rises by 1.6ppts … until ~700kms from the mine the Qld swing looks a lot like the national swing.

What about my hypothesis that it’s not just Adani, that it’s mines in general? That’s not so clear. It works (a bit), but most of the effect (in Qld) comes from coal mines. The nearer to coal mine (not just Adani, but any coal mine) the larger the swing away from the ALP.

The fact that the rural seats are vast but linked to mining — and that city seats are sometimes close to mines, but focused on other things, confounds the data. So this time i focused on the booths that are close to a mine (less than 40kms away, say an hour’s drive). Note that there are no booths this close to the proposed site of the Adani mine. It’s just too remote.

If you look at the below chart, you can see that the closer a booth is to a coal mine, the larger the swing against the ALP.

So if a booth is very near a coal mine, it probably swung ~10ppts against the ALP; and every 10kms more distant from an operating coal mine, the ALP vote improves by ~2 percentage points. This passes the common-sense test — the LNP ran heavily on coal in Queensland (advertisement below).

It’s worth noting that eighteen of Queensland’s 30 seats have booths that are less than 40kms from an operating coal mine. The seats are: Blair, Bonner, Bowman, Brisbane, Capricornia, Dickson, Flynn, Forde, Griffith, Groom, Lilley, Maranoa, Moreton, Oxley, Petrie, Rankin, Ryan, and Wright. There are three Adani seats that are not in this list — so the total rises to 21 of the 30 seats in Queensland.

So coal is an issue for a significant part of Queensland. Perhaps it’s no wonder that the outcome of the vote in Queensland was so different.

The LNP coalition holds 77 seats in a 150 seat house, and the Queensland seat of Herbert is prime Adani country (won 58.4% LNP v. 41.6% ALP, on an 8.4% swing to LNP). Without Herbert they would have only 76 seats — the bare majority needed to Govern in their own right. And the high primary vote in Queensland for the LNP meant they got three senators, which has greatly increased their power to make law in the Senate. The ALP’s slump meant they got only one senator from the six available in Queensland.

Finally, here’s a chart of the ALP primary vote swing by booth, with the location of coal mines in black.